PDA

View Full Version : Altec 9844-8B



Pages : [1] 2

Skywave-Rider
11-10-2007, 02:53 PM
Hi,
I just got a pair of Altec 9844-8Bs. They use two 414 16E woofers and a 902 HF driver
with 811B horn.

I'd like to refresh the crossover and add a tweeter, of course that would require a custom network.

Well, I have a pair of K-77s I took out of my Klipsch Heresies which were replaced by
the Bob Crites tweeter. I could use those. I might, at some point in the future upgrade
to a better tweeter (Beyma/Crites/?), so hopefully it would be compatible.

Since I have no design skills, but can solder like a champ, :barf:
I wonder if anyone can point me in the right direction for a crossover design that would work well.

Thanks!

CONVERGENCE
11-10-2007, 03:43 PM
Yes you can build a 3 way second order reverse polarity crossover.

Parts List

Capacitors
C1 = 1.54 uF
C2 = 16.83 uF
C3 = 1.37 uF
C4 = 12.31 uF
Inductors
L1 = 0.4 mH
L2 = 3.3 mH
L3 = 0.32 mH
L4 = 3.22 mH

2nd Order Reverse Polarity

6400 Hertz / 800 Hertz

8 Ohm Tweeter / 8 Ohm Mid / 8 Ohm Woofer
2.45 db Bandpass Gain, Spread = 8 : 3 octaves

Skywave-Rider
11-11-2007, 12:35 PM
Thanks Convergence!
I am currently specing out the components at Parts Express.
I am in your debt.
Best wishes!:applaud:

Earl K
11-11-2007, 12:56 PM
Thanks Convergence!
I am currently specing out the components at Parts Express.,,, snip ,,,,,,

- FWIW, you should add to your order a qty of ;
- ( 4 ) 50 watt , 8 ohm variable Lpads .
- You'll need 2 per crossover .
- Ask Convergence where to locate them in the circuit & how to wire them in .

:)

Skywave-Rider
11-11-2007, 01:27 PM
Thanks, I was just about to press "submit order." Convergence, what do you think? Do those L-Pads go in the tweeter and squawker circuits? Or function as his/los? Can I do without them? I like the idea of going without them, but that might be unrealistic?
Thanks in advance.

CONVERGENCE
11-11-2007, 04:39 PM
LPad (Driver Attenuation Circuit)

Z = 8 Ohms

A = 2 db

Parts List

Resistors
R1 = 1.65 Ohms
R2 = 30.9 Ohms

AFTER L3 AND C3 FOR HORN AFTER C1 L1 FOR TWEETER

Skywave-Rider
11-12-2007, 07:25 AM
Thank you Convergence and Earl. I will add the L-pads. Did a little reading about them last night, so I think I understand how they work. You know what they say, a little knowledge....

I understand the K-77 (EV T 35) may not be the best tweeter, but I have a pair on hand, and I'd consider changing to a better tweeter if I think the 3 way setup has potential for me.

There's always a chance I'll go back to the original 2 way network setup with new caps. But I've got to try this and I'll post when I get it done.

Greatly appreciated, great board.

:bouncy:

Skywave-Rider
11-26-2007, 10:43 AM
OK, the crossovers are done. I'm going external because I want little as possible to change on the 9844-8Bs. I also made a rig to mount the tweeters which go on top of the cabinet. When I test them I'll report back and post pics of the entire rig. If these work out, I will want to test one speaker with a 2 way crossover for comparison. If anyone has suggestions on how to do this without making up an entirely new crossover, that would be great. I am assuming the original crossovers are no good. Wish me luck.

Skywave-Rider
11-28-2007, 09:36 PM
Yes they work. I have not tested the speakers in a room, so I'll try to do that next week and get pics. My initial impression is that the K77 tweeter is not blending well. It seems to muck up the imaging, which is what I remember from my Heresies; I changed the Klipsch out to a Crites tweeter and that was solidifying. I have the K-77s (EVT35) mounted on the Altecs in their proper vertical orientation. I hear phase problems; but I will have to listen closely next week.

I have a feeling after I put them in a listening room I'll either want better tweeters (APT-200?) or will revert back to 2 way. If I revert, does anyone have an opinion as to whether or not I should copy the original network, modify my 3 way, or do something completely different?

Thanks!

djgaloot
11-29-2007, 10:02 AM
Skywave,
Nice looking crossovers. Though I do not have any input, I am watching with great interest as I have similar speakers (Malibus) and also plan on making a 9844 clone. Some have suggested active crossovers as an alternative. I was thinking of a ribbon tweeter for the HF. Right now mine are all stock with the old stock 800hz crossovers.

hmolwitz
11-29-2007, 01:12 PM
http://www.audioheritage.org/vbulletin/showthread.php?t=10252

"The T35 is not a very good tweeter. Most start taking a dive at about 10K Klipsch would select those that would make it to 15K and send the rest back. I have a pair tested in a anechoic chamber just makes it to 10k I replaced the diaphrams & now get 14k. Diaphrams are inconsistent due to phenolic weight and thickness. The newer T35/ k77 today will get 15k to 17k. This tweeter should have been retired along time ago."
__________________
Maron Horonzak


The specs on that would not be as good as the 902, you would be better off leaving it alone.
The 902 should be fairly linear out to beyond 15K, but this thread:
http://www.audioheritage.org/vbulletin/showthread.php?t=14690

Will give you the full powers of Mr Zilch brought to focus on the lacking of the Altec HF horns, along with the assorted wisdom of many other members.
Part of the beauty of that speaker is the simplicity of a 2 way.

Many find the 902 a wonderfully acceptable driver, it is probably marginally better than the earlier versions.

Enjoy the road, it is not the destination that matters.
Nice job on those crossovers
Harry

Art J.
11-29-2007, 05:15 PM
Yes they work. I have not tested the speakers in a room, so I'll try to do that next week and get pics. My initial impression is that the K77 tweeter is not blending well. It seems to muck up the imaging, which is what I remember from my Heresies; I changed the Klipsch out to a Crites tweeter and that was solidifying. I have the K-77s (EVT35) mounted on the Altecs in their proper vertical orientation. I hear phase problems; but I will have to listen closely next week.

I have a feeling after I put them in a listening room I'll either want better tweeters (APT-200?) or will revert back to 2 way. If I revert, does anyone have an opinion as to whether or not I should copy the original network, modify my 3 way, or do something completely different?

Thanks!


Hello gang, that is a great looking crossover for sure.
It seems that phasing problems bother you as much as
they bother me.
Adding a tweeter causes chronic phasing problems anyway
but you could help the situation by getting the electrical phasing
in line by using the "cascade method" on that crossover. The
modified diagram shown should be self explanatory when thinking
in terms of phase shift.
In your case, the no cost experiment means moving the C1 input side
to the C2 output side keeping the shown polarity.
That will help for the most part but to fine tune it,
you will need an oscilloscope and audio oscillator.
It may improve overall balance too.
Have fun. Great soldering.......

Skywave-Rider
11-30-2007, 10:25 AM
Hello gang, that is a great looking crossover for sure.
It seems that phasing problems bother you as much as
they bother me.
Adding a tweeter causes chronic phasing problems anyway
but you could help the situation by getting the electrical phasing
in line by using the "cascade method" on that crossover. The
modified diagram shown should be self explanatory when thinking
in terms of phase shift.
In your case, the no cost experiment means moving the C1 input side
to the C2 output side keeping the shown polarity.
That will help for the most part but to fine tune it,
you will need an oscilloscope and audio oscillator.
It may improve overall balance too.
Have fun. Great soldering.......

Thanks Art, and to all.
I intend to do the cascade mod next week. I understand how feeding the input of C1 from the output of C2 adds delay, but I'm not clear on why that is required -- I think it's because C2's value is much higher though the electrical delay must also be frequency dependent. I don't know how that can be adjusted other than changing the capacitance value if the crossover points remain constant. Clearly I need to read a little about this which I will try doing.

At this point it's a learning experience, and since I can go back to 2 way, I feel pretty good about the mods because I'm not molesting the speakers.

One reason I can't do changes too rapidly is because the Altecs are currently not in my home; they're at my office. I live in a tiny studio apt..:banghead:
Tiny studio apts are not Altec friendly.

Thanks again!
._._.

CONVERGENCE
11-30-2007, 05:06 PM
Nice job on those crossovers. Now before doing any other changes try them with your system first. The 3 way Altec monitors is not a new invention. I can think of The Mastering Lab with their 411, 511E and EV tweeter. And a similar model designed in the 70's.Rod Stewart had a pair.

Art J.
11-30-2007, 05:42 PM
> I intend to do the cascade mod next week. I understand how feeding the input of C1 from the output of C2 adds delay, but I'm not clear on why that is required -- I think it's because C2's value is much higher though the electrical delay must also be frequency dependent. I don't know how that can be adjusted other than changing the capacitance value if the crossover points remain constant. Clearly I need to read a little about this which I will try doing.

Well, I'm not a good explainer but the idea is to treat the mid and tweeter as a separate
second order reverse polarity circuit. This gives hope for phase alignment at the highest
x-over frequency where phasing effects would be most noticeable. C1 L1, C3 L3.
Then working from the top down, the woofer circuit is added where phasing differences
will not be as noticeable. Their will always be a phase conflict with passives, so put it where
it will do the least damage.
With the strait 3 way band pass arrangement, I see no hope for
getting any form of alignment.. It seems to be a non-issue with most. Don't know why.

> At this point it's a learning experience, and since I can go back to 2 way, I feel pretty good about the mods because I'm not molesting the speakers.

You did such a great job building the x-over, I'm hoping you can make the best
of what you got.
I tried finding a link on cascade filters without much luck. I picked up on these before the
internet and built with good results. All I found was this link that may be helpful and
confusing at the same time. The more you study it, the worse it gets.

http://www.trueaudio.com/st_xov_1.htm (http://www.trueaudio.com/st_xov_1.htm)




:barf:

Skywave-Rider
12-01-2007, 11:50 PM
Thanks again to all.

I am reading and puzzling and hopefully not blundering. Too much.
News: I took the plunge and ordered a set of APT-200s.
http://www.partsexpress.com/pe/showdetl.cfm?&Partnumber=290-536
If they don't do the job, they can go back. (I realized the k-77s are not adequate.) I hope to try it all out next week. Kind of exciting.

Wishing for a RTA and oscilloscope.

Meanwhile, Rod's speakers are (were) awesome. Are those double 12" like my 9844s, or 15" woofers? They seem to be placed right on the floor, or nearly -- at least for the ad. That should help the bass. I will try that at some point.:)

Enjoy the rest of the weekend.

Skywave-Rider
12-04-2007, 01:00 AM
Tweeters are in transit.
In the meantime sorting out the loading cap/diaphragm question. Trying to find out if I have symbiotic diaphragms or all aluminum. The covers from my supposed 902 drivers are not there, so it's hard to tell what's what. I have determined for sure though that I do have loading caps.
If you can help, I have pics at this thread:
http://www.audioheritage.org/vbulletin/showthread.php?t=19156
Thanks!

Earl K
12-04-2007, 03:13 AM
Trying to find out if I have symbiotic diaphragms or all aluminum.

http://www.audioheritage.org/vbulletin/attachment.php?attachmentid=29789&stc=1&d=1196725297

- Your diaphragms are definately all aluminum ( not symbiotic ) .
- As was pointed out in the other thread, talk to Bill at GPA about buying the proper, bigger back-caps that have a felt liner ( & then retire these "reduced volume" loading caps ) .

:)

Skywave-Rider
12-04-2007, 10:15 AM
Thanks Earl K!
Precisely the affirmation I was looking for. U made my day.:D

Skywave-Rider
12-14-2007, 08:45 PM
Got APT 200 tweeters rigged up onto the mounts I made for the k77s; very good results. I was able to listen only about 20 mins, but the insane imaging of the k77s is gone and it's become tight and precise. I was thrilled to discover that a lot of the severe phasing and comb filtering seems to have gone away. I did notice issues with smeared transients and experimented with positioning the tweeter fore and aft. I found forward such that both horn bells are in line to be the best (so far.) I would have imagined that lining up the diaphragms would be the way to go, but I have a lot to learn. Preliminary results: pretty happy.

I ordered a Behringer DEQ2496 to look at some RTA; and hope to try it next week. Received 902 driver covers and felts today from Great Plaines Audio, so also looking forward to disposing of those loading caps.

Let you know when I can get all this done and give a longer listen.

Skywave-Rider
12-20-2007, 07:53 AM
I've gone through 2 Behringer DEQ 2496 RTA/EQs, both have been defective and returned. Going for a third today from a different vendor, if in stock. I have a feeling there is a bad bunch of units out there.

On another front, removed the loading caps and installed GPA covers successfully. Happy to report I did not run a screwdriver through any diaphragms. :)

When/if I get the RTA I'll post images.:(

Skywave-Rider
01-02-2008, 12:12 PM
Got my 3rd DEQ2496 and reporting it is in continuous operation for 24 hrs with no known issues. So I think this one works. I see the production code on it is about 8mos newer (from Musician's Friend) than the previous one I got from B&H. Since the Altecs are not in my apt., I have not RTA'd them yet. I did check my Heresies and to my surprise, they are flatter than I thought they would be. (Note that I have Bob Crites tweeters installed, which are Eminence APT 50s on custom horns to fit the Klipsch.) Tried "AutoEQ" function on them and hated the sound. Sounds "plasticky." Frankly, I think the Heresies have no major issues in my current placement, and since they are horn speakers as well, have less problems with room reflections in the mid and upper octaves compared to:

My EV Sentry 100s,

Also in my apt..
I just refurbished these and will be using them for mixing on my ProTools rig. I RTA'd and "AutoEQ'd" these and I will say in this application, this DEQ2496 worked well. Evened the frequency response. Showed me where alternate placements would be better, and makes the bass work from these monitors -- flatter. The sound of the Behringer unit is still there, but I can work with that. The Sentrys are for working, and a representative freq. response is more important to me when using them. For listening while not working, which is when I use the Heresies more, I won't use the EQ, or have it in line at all.

I use digital EQ all the time in ProTools, and can't live without the Waves stuff, so it's not like I'm against DEQ. But when it's not necessary to do D to A/A to D, or vice versa, I'd rather not.

At home I use an old Technics SU 8600 ss amp on the Sentrys and currently a Magnavox 88 02 00 6v6 pp amp on the Heresies. For the Altecs at my job, I have a recently acquired Crown PS 200.

I intend to have some RTA of the Atec 9844s in the next few days. I know I will need expert advice from you guys, but I will try not to draw this out unnecessarily.

In the long view, if I find myself living in a larger place, the Altecs will be my room mates, and will probably (maybe) be powered by one of my little tube amps.

Side note: A few weeks ago I used the 9844s at a video/film screening with the APT 200 tweeter and the Crown amp. It sounded very good -- but they are certainly bass shy -- as they were with the original crossover.

Skywave-Rider
01-07-2008, 04:19 PM
This is one of my 9844-8Bs pictured with a Genelec 1031 (boo,) and a Westlake BBSM-8 (yay!) behind. Please no comments on the ill-placed Sonex foam you will see scattered about, it is not my doing as this is not my control room or studio. I don’t own the other speakers either. In fact I did own Genelecs and happily sold them last Summer. If I can get the Altecs running satisfactorily, I will try some mixing with them in this place to hear how the result translates.
http://i183.photobucket.com/albums/x316/skywaverider/Aunt%20Molly%20Tribute%20Gathering/Altec%209844-8B/IMG_1755.jpg

Inside one cab showing new GPA covers over the 902 drivers:
http://i183.photobucket.com/albums/x316/skywaverider/Aunt%20Molly%20Tribute%20Gathering/Altec%209844-8B/IMG_1731.jpg

Inside the other cab shows one 414 appearing different, though it is stamped 414-16. Does this seem correct?
http://i183.photobucket.com/albums/x316/skywaverider/Aunt%20Molly%20Tribute%20Gathering/Altec%209844-8B/IMG_1732.jpg

Close up of renegade woofer:
http://i183.photobucket.com/albums/x316/skywaverider/Aunt%20Molly%20Tribute%20Gathering/Altec%209844-8B/IMG_1733.jpg

Compared to its room mate:
http://i183.photobucket.com/albums/x316/skywaverider/Aunt%20Molly%20Tribute%20Gathering/Altec%209844-8B/IMG_1734.jpg

As you can see the previous owner had port tubes which I pulled off. Once I get things sorted out, I might want to try that to help the low end.

I measured and aligned the rear of the voice coils for the 902s and the Eminence APT200s. The tweeter’s setback position will cause bad reflections off of the cab:
http://i183.photobucket.com/albums/x316/skywaverider/Aunt%20Molly%20Tribute%20Gathering/Altec%209844-8B/IMG_1739.jpg

So I went ugly and put on a sound blanket:
http://i183.photobucket.com/albums/x316/skywaverider/Aunt%20Molly%20Tribute%20Gathering/Altec%209844-8B/IMG_1742.jpg

I made the next several unscientific measurements using my new DEQ2496/companion mic, from a 3 foot distance with the mic height exactly between both horns. Kind of aimed at the blanket. The room is a studio which has non parallel walls, though the ceiling and floor are. I used a Sonex gobo to help kill reflections from the close wall:
http://i183.photobucket.com/albums/x316/skywaverider/Aunt%20Molly%20Tribute%20Gathering/Altec%209844-8B/IMG_1741.jpg

Continues in next post....

Skywave-Rider
01-07-2008, 04:21 PM
Continued from previous post.....

This is the result with both L-pads turned to minimum output:
http://i183.photobucket.com/albums/x316/skywaverider/Aunt%20Molly%20Tribute%20Gathering/Altec%209844-8B/IMG_1744.jpg

The 902 brought in:
http://i183.photobucket.com/albums/x316/skywaverider/Aunt%20Molly%20Tribute%20Gathering/Altec%209844-8B/IMG_1745.jpg

And the APT 200 brought in:
http://i183.photobucket.com/albums/x316/skywaverider/Aunt%20Molly%20Tribute%20Gathering/Altec%209844-8B/IMG_1746.jpg

Since I don’t really know what the hell I am doing, I tried to get levels across the spectrum as even as possible, but I was looking from the other side of the room and should have done better. Here are the L-Pad settings:
http://i183.photobucket.com/albums/x316/skywaverider/Aunt%20Molly%20Tribute%20Gathering/Altec%209844-8B/IMG_1747.jpg
Left pot is 902.

I then rolled the whole mess into a typical rectangular room with the idea of giving a listen. I also took some readings there as well. The room (a classroom) has hf absorptive panels on the left wall -- which you can’t see -- so I brought in the Sonex gobo for the right side against the blackboard:
http://i183.photobucket.com/albums/x316/skywaverider/Aunt%20Molly%20Tribute%20Gathering/Altec%209844-8B/IMG_1764.jpg

For the next two readings, I can’t remember what the mic placement was...
I think it was about the distance between the speakers; first with 902s in “normal” phase, which is reverse polarity:
http://i183.photobucket.com/albums/x316/skywaverider/Aunt%20Molly%20Tribute%20Gathering/Altec%209844-8B/IMG_1756.jpg

And now with 902 leads reversed so that they are actually in phase now:
http://i183.photobucket.com/albums/x316/skywaverider/Aunt%20Molly%20Tribute%20Gathering/Altec%209844-8B/IMG_1757.jpg

Now for some potentially bad news. I decided to do a close check in this room for each speaker, which I did not do in the studio room (don’t know why.) So I went back to the 3 ft placement as shown previously, but stayed in the classroom. Left speaker:
http://i183.photobucket.com/albums/x316/skywaverider/Aunt%20Molly%20Tribute%20Gathering/Altec%209844-8B/IMG_1761.jpg

Right Speaker:
http://i183.photobucket.com/albums/x316/skywaverider/Aunt%20Molly%20Tribute%20Gathering/Altec%209844-8B/IMG_1758.jpg

Looks like the left 902 is 5 dB down at about 2.5k. And generally the curves are not as similar in that range as I would expect. But I have not done this before, so I don’t know how significant that actually is, given room placement differences. I did not roll each speaker into a single location. I simply measured them positioned as they were in the classroom image. Here’s the approximate L-Pad settings:
http://i183.photobucket.com/albums/x316/skywaverider/Aunt%20Molly%20Tribute%20Gathering/Altec%209844-8B/IMG_1762.jpg

Right now I’d like to hear your opinions. After digesting what you guys say, I will see what I can do to make the 3way as good as I am able. I may also try the cascade filter modification suggested by Art J. Then I will definitely compare one three way to one 2way. And for this I don’t know if I should modify one of my existing crossovers or do you think I should build a copy of the original network?



I now see an attractive benefit of going active --- being able to change things on the fly.

Thanks for all the help!

Zilch
01-07-2008, 04:41 PM
You want to measure on tweeter axis. There's a rationale for this, but I'll have to look it up. Those tweets should be putting out much higher VHF than you're showing.

Forget the tweeter time alignment for now. Move them to the front and compensate with phase inversion, if required. I believe you'll find the top of the cabinet and the pad are messing you up.

It looks like something's up with phase in the left speaker between the mid and tweet making the notch at 6K3 Hz. Measure each driver in each system individually (running on their respective filters) to see how well they match.

That'll also show whether the filters are operating as expected. Then, it should be easy to see the effect of inverting the phase of each in the display, woofers plus mid, mid plus tweeter when you run those together.

The bass response certainly changed when you switched rooms. No matter, you're going to have to measure that nearfield to get any reasonable idea of what's going on down there. Your first curves are more what I'd expect to see, typical, in this respect.

Make notes in a notebook what you do with pic numbers. It quickly becomes difficult to remember what's what without that when you're trying to figure out what's going on and changing stuff....

Skywave-Rider
01-07-2008, 07:59 PM
Thanks, Zilch.
When I made the crossovers I checked and re-checked and checked the wiring some more because I know I am prone to these errors if I get distracted. I will check it again :)

I will also check the connections to the 902 diaphragms for verification. The APT200s have a factory red dot, which is hard to screw up.

When I do this again I'll go back to the studio, stick with the 3ft distance and roll each speaker into the same location for the checks. I'll align the bells of the horns and measure at tweeter height. I'll also use the same output of the PS 200.

When checking individual drivers is it OK to disconnect as opposed to cranking down the L-pad(s)? And for the woofers, which are somewhat less directional, should I move the mic down? Sorry for the basic questions.

Thanks for the notepad tip. :o:
I can't tell you how many hours it took me to do what I did -- but it's fun.

Zilch
01-07-2008, 08:26 PM
Key to understanding what's going on is parsing the system into its individual elements, and then looking for the expected behavior when you "assemble" them in measurements. If that doesn't happen, ask yourself the appropriate question: "HUH, what's up with THAT?"

Example: Right and Left look similar except at 6K3. Why? Well, first look at the tweeters alone and the mids alone. Are their responses the same? Why aren't they summing in the Left one?

Turning down the L-Pads is a good way to isolate the drivers, actually. The load stays in place, being provided by the L-Pad alone instead of the driver or a combination of the two.

If you're not biamped, put an 8-Ohm resistor in place of the woofers as dummy load when you disconnect them to measure the mids and/or highs. OR, if the filter sections are separate, just disconnect the LF filter at its input.

The bass is being made by both the woofers and the ports, and the best way is to measure each at a very close range, like 1/4". You then do a calculation to sum the relative contributions to the total response at each frequency within the range of interest, weighting each source according to its diameter. The volume of the pink noise must be turned down considerably for this, of course.

If you want just one reading, find a location that's central to the woofers and ports, and back maybe 3 - 6". That won't be as accurate, but it will provide an indication of the LF response better than from 1M or more away, which is likely to incude additional room effects.

CONVERGENCE
01-07-2008, 09:00 PM
Nice looking set up you got there. THe 414 are 16 ohms connected in parallel = 8 ohms.

Quote from your post"Inside the other cab shows one 414 appearing different, though it is stamped 414-16. Does this seem correct? " YES It could be a Z series they din't bother with the lid on some series that were
sold with cabinets.

Keep us informed this is interesting.

...............

Skywave-Rider
01-10-2008, 04:17 PM
I’ve redone all my measurements in a more systematic way as Zilch suggested. Thank you for all your guidance on this Zilch, your expertise has been invaluable and is greatly appreciated.

I returned to the Studio room, and added more sound blankets. I also kept the pink noise levels consistent excepting for the woofer/port close micing. I was extremely careful about the speaker/mic positions. All tests were done at tweeter on axis from a distance of 3ft from the front of the cabinet (except for close woofer test.) I used only one side of the amp. The last session gave me a lovely few days of ear ringing, so I grabbed my range muffs. Best Idea I had all week :)

I used dummy loads when removing drivers from the tests because I found there was always some output even at minimum settings. Additionally, as Zilch suggested, this was needed for taking the woofer out of the equation, since there is no L-pad for that.

The tweeters are moved forward such that both horn bells are aligned. Although this should mean phase problems due to time alignment, the whole rig sounds much better this way. No more blankets on the top.

Here’s the setup:
http://i183.photobucket.com/albums/x316/skywaverider/altec%202/the_setup.jpg
Dummy loads are 8 Ohm, non inductive, L pads remained at this setting.

I found after I did all the image cutting and pasting, that there are no big discrepancies between the drivers in both speakers and I think the response is very similar.

I don’t know what I can realistically expect from the crossover. But I would like to improve it as much as is possible. I don’t know how good or bad the curves appear to you guys. I want to learn and am open to your suggestions.

Here are the curves:

Zilch
01-10-2008, 04:39 PM
Well, there it is. The mid phase is correct, but, with the tweeter in this position, you need to reverse its phase. Do you see that?

And also that reversing the phase of the woofer produces the same notch between it and the mid as reversing the mid?

That's occuring at 630 Hz, it looks like, and I'm surprised the mids are playing so strongly down at 400 Hz. Your crossover designer may recommend a change to that particular filter.

The woofers are indeed rolled off below 50 Hz. As you have reported, there's no extended bass to be had here; need subs for that. You might achieve better balance with what's there by dialing down the mids and tweets more. It's looking somewhat "forward" overall.

The notch at 3.55 kHz is in your mid/horn combo, apparently. Do your new rear caps have felt or foam damping in them?

I don't normally run RTA pink noise so loud as to require muffs, though that certainly gives good signal to noise performance.

You should be able to observe the notch between the mid and tweeter increase and decrease as you move it back and forth from the front. It'll repeat in multiples of the wavelength at crossover.

Your results are certainly much better, and I would say also more reliable, now that you have the variables under control.... :yes:

Skywave-Rider
01-10-2008, 05:04 PM
Yes, it looks much smoother, the ~3k notch has become a mere depression with the phase reversal between mid & tweeter. But there's that 4K spike, it looks bad, doesn't it? Must be approx 5 dB. And it's a narrow Q. If I moved the tweeter front/back slightly, do you think that spike would move or be cut? If I had the speakers home, I'd try that right now. So I'll play with the tweeter positioning, I should be able to move it a few inches without getting into reflections off the top -- at that wavelength were talking short -- and try to get the best notch and see what happens to the spike.

Oh wow, I realized ur editing ur post as I type, so I'll hold off a while until the final version...:)

Earl K
01-10-2008, 05:23 PM
Hi Skywave-Rider,

Zilch has covered a lot of points / here are some more ;


I don’t know what I can realistically expect from the crossover. But I would like to improve it as much as is possible.

- From my perspective, you do need to do a bunch of crossover redesign, to make that 3-way setup listenable for long term listening .


I don’t know how good or bad the curves appear to you guys.

- Honestly, IMHO ; those FR curves are fairly mediocre looking.

(i) Overall, the lowpass section is not performing as designed with those "text-book" values. The woofer section has a 3 db down point that is in the 1250 hz area ( rather than the 800 hz area as originally intended ) .
- You may just want to keep this lowpass section "as is" ( live with it ) & focus on fixing the hipass section on the 902/811 horn circuit .
- Here's one reason "why" you didn't achieve a 800 hz lowpass on those woofer pairs .
http://usr.AudioAsylum.com/images/4/46616/414imp.jpg

- The pic above shows the raw impedance curve for a single 414-16z ( & I emphasize, "curve" ). You'll see that the woofers' impedance is only 16 ohms over a very narrow frequency area. The area along this curve where the cap is doing it's filtering , is actually more like 25- 30 ohms ( or 12.5 to 15 ohms when two 414s are paralleled ) . This pic is from This Thread at A.A. (http://www.audioasylum.com/forums/hug/messages/13/130132.html). ( I know I've linked into it, half-way through ).
- I'd suggest that you read the whole thread .

(ii) The 315 hz midrange spike ( from the twin 414 woofers ), definately needs to be tamed. Perhaps Zilch can suggest a "parallel type" LCR notch filter ( inserted "inline " , to notch it out some.

(iii) The hipass on the horn circuit ; your pic shows a 3 db down point ( Xover area ) in the 500 hz range.
- I'd suggest you take it up to 1250 hz, to match the existing lowpass on the woofer .

(iv) The UHF ; Well, at it's best, the 902a can get to up around 10K in a pretty linear fashion ( when a RC "bypass circuit" is implemented ). I would then roll in the tweeter at around 10K ( at either 18db or 24 db per octave ) .
- Such a high point will undoubtably loose most of that "hump" that you see below 10K ( in the raw tweeter RTA ) .



I want to learn and am open to your suggestions.

Okay, I've just made some suggestions ( of a few remedial actions that I would want to implement if I owned that setup ) .

:)

ps ; I'll leave the rest to Zilch

Zilch
01-10-2008, 05:51 PM
I'll leave the rest to ZilchI've suggested that Convergence might have some recommendations. ;)

It's 902s with Symbiotic, Earl, I believe, yet to be confirmed. If they were 806s, with which I am somewhat familiar, I might be able to be more helpful. Clearly the 902s have the same excessive midrange response, which must be tamed. How they behave at the high end, I don't know, but Skywave can easily determine that by bypassing the lowpass portion of the midrange filter.

Gonna go read the AA link now. I don't know 414 from a hole in the ground.... :p

CONVERGENCE
01-10-2008, 05:54 PM
Altec drivers don't like reverse polarity. So that design might not be suitable . Make a polarity check of your woofers with a 2.5 volt battery.
The speakers should be pushing not pulling just in case.

Bi amping again with todays prices is considered a real improvement over those passive networks especially with 3 way designs.

Most Passive 2 way networks from Altec VOTT will use the same values as the computerized models. On the other site we have not encountered such problems cause most owners have replaced the caps in their speakers with outstanding results.
.

The reason for reverse polarity in a 3 way butterworth design has to do with phasing of HF driver.

There are pro calculators around 100$ these will give you the right values according to RTA of the speakers.

..........................

Earl K
01-10-2008, 05:57 PM
It's 902s with Symbiotic, Earl, I believe, yet to be confirmed. If it were 806s, with which I am somewhat familiar, I might be able to be more helpful. Clearly the 902s have the same excessive midrange response, which must be tamed. How they behave at the high end, I don't know, but Skywave can easily determine that by bypassing the lowpass portion of the midrange filter.

Yes thanks, I've noticed my mistake and I'll edit it appropriately .

:)

Skywave-Rider
01-10-2008, 06:03 PM
I have 902 drivers with aluminum diaphragms. My covers (which have felts) say 902-Bs (which I believe are Symbiotiks) but I don't have Symbiotiks. Those were the stickers GPA had available.

I am reading all your posts.

Skywave-Rider
01-10-2008, 06:09 PM
Altec drivers don't like reverse polarity. So that design might not be suitable . Make a polarity check of your woofers with a 2.5 volt battery.
The speakers should be pushing not pulling just in case.

Hi Convergence. I did check the woofers in that way when I was sorting out the wiring. Positive voltage gives forward motion of all woofers.;)

I wish I could do math. Gonna be reading a while....

Zilch
01-10-2008, 07:03 PM
Bring this forum over to http://www.hostboard.com/cgi-bin/ultimatebb.cgi?ubb=forum&f=3729 and see what the experts will say.Yeah, well, I just posted over there. :p

[Rereading THIS thread, now, to figure out how it came to this....]

Edit: O.K., Skywave, do this: Hook up your factory crossovers and make the same measurements without the tweeter. Let's see the woofers and mids alone, and running together. Do nearfield on the woofers again, as well.

Also, what is the model of the stock crossover, and do we have the schematic somewhere? Provide what component values you can from your originals, please....

Skywave-Rider
01-10-2008, 10:09 PM
I just found this shot of the rear panel (the speakers are an hour commute from home -- thank God for digital cameras provoking indiscriminate photos.)

BTW, I am truly grateful to everybody who has helped. Can I say that again? Just did. And let's get along..........:p

I believe the original N-800Fs are "potted." I'll look for a schematic. When I cut out the wiring I left just enough to reconnect in just such a circumstance. But those orig. caps have got to be crap. I will take NEW measurements using the OLD crossovers in 2 way. Actually it sounds like fun to see what the RTA will look like.:)

Zilch, do I need to really remove the N800s and melt out the wax? I guess it depends on finding the schematic. The reason I'm asking is because I understand a lot of the Altec documentation is not what was actually used in manufactured product.

Regards,
Vin
That's my name which u may use if u like.

Zilch
01-10-2008, 10:40 PM
Zilch, do I need to really remove the N800s and melt out the wax? I guess it depends on finding the schematic. The reason I'm asking is because I understand a lot of the Altec documentation is not what was actually used in manufactured product.Nope, we know N800-F (N-800F, N-800-F) pretty well. I just posted the link on the Altec Forum. Ironically, it was one of their members who had the original document:

http://audioheritage.org/vbulletin/showthread.php?p=160631&#post160631

Member Bfish over there indicates that crossover was used in many Altec dual-414 designs despite the lowpass filter being used for 16-Ohm drivers in other products such as Valencia 846A. See also Malibu and Carmel.

In theory, it's "wrong" for use with 8-Ohm HF drivers, as well, but Altec engineers had more of a way of getting this stuff right than us textbook theorists sometimes give them credit for. Let's see how it behaves in comparison to what you presently have.

If you feel ambitious, you may want to read that thread in its entirety. Adding a tweeter is not necessarily the best answer to extending the HF. If you want to stick with your Altec HF drivers, the Model 19 (and others) "T-filter" compensation may get you where you want to be more seamlessly. Read the next post there.

Shine a flashlight into the throat of your 902 drivers. Do they have the orange plastic "Tangerine" phase plugs? If so, the prospects for success are good. We need to resolve the woofer and crossover frequency issues first, tho....

Skywave-Rider
01-10-2008, 11:19 PM
Awesome. Yes, I know the 902 has good HF extension with the aluminum diaphragm, which I have. And I do have the tangerine phase plug as well. Several members, the first of which was Tom Brennan, tried to persuade me not to go 3-way. To my ear, the original setup was not open sounding, however. (But admittedly, that's listening to an aged crossover.) So I thought it would be fun and instructive to try 3 way. And it is.

I think I have photographed every inch of that speaker but inside the wax laden crossover. I will read that post. And I will reconnect one of the N-800Fs as soon as I am able. Cool.

Skywave-Rider
01-10-2008, 11:31 PM
Nope, we know N800-F (N-800F, N-800-F) pretty well. I just posted the link on the Altec Forum. Ironically, it was one of their members who had the original document:

http://audioheritage.org/vbulletin/showthread.php?p=160631&#post160631

Haha, ur a funny guy, Zilch...:D

Zilch
01-10-2008, 11:52 PM
Awesome. Yes, I know the 902 has good HF extension with the aluminum diaphragm, which I have. And I do have the tangerine phase plug as well. Several members, the first of which was Tom Brennan, tried to persuade me not to go 3-way. To my ear, the original setup was not open sounding, however. (But admittedly, that's listening to an aged crossover.) So I thought it would be fun and instructive to try 3 way. And it is.That's still a possibility, but let's see what's going on first. Key to taming the Altec compression drivers is to knock down their forward midrange. That brings both the lows and highs up and better balances the system overall. N800-F does a bit of that with 16-Ohm drivers; we won't know what it does with 8-Ohm ones 'til you measure the combination.

If you read that next post, you know to take along a 6 uF capacitor (yeah, 6.2 uF is fine,) and an 8-Ohm resistor. Put those in parallel, and then the pair in series with the HF driver. That'll give us some indication of just how much HF extension your 902s actually have. :yes:

[More measurements.... :p ]

Skywave-Rider
01-11-2008, 12:08 AM
There are pro calculators around 100$ these will give you the right values according to RTA of the speakers.

Do you mean crossover design software? I was thinking about that. There's one Parts Express sells which seems to be popular. Unfortunately (fortunately) I have a Mac, but I can find a place to run that I'm sure. There's no way I'm going to build speakers from scratch given my living arrangements, but for the price it might be good for a guy with no mathematical abilities -- me:banghead:

I don't know if you guys know what the Regents exam is, but it is (or was) required in NY high schools. Big end of the year exam in every subject, state sponsored. You don't get a "Regents Diploma" if u don't pass. The year I had to take the Regents in Algebra (or whatever the one above that is) I knew I was doomed. I studied my ass off. My friend tried to tutor me, but he was a numbskull as well. My brain is not wired right for numbers. But I wasn't going down without trying. And tried I did.

Well what happened was one of the few really lucky things that has ever happened to me. (Besides building this crossover, LOL.) That year, some angelic thieves stole the Regents exams, I don't know why, perhaps they were academic kidnappers holding out for ransom? Who cares. EXAM CANCELED. NO MAKEUP REQUIRED. Haha. I did not have to take that frigging test. And here I am, for better or worse.:hate-pc:

Skywave-Rider
01-11-2008, 02:28 PM
If you read that next post, you know to take along a 6 uF capacitor (yeah, 6.2 uF is fine,) and an 8-Ohm resistor. Put those in parallel, and then the pair in series with the HF driver. That'll give us some indication of just how much HF extension your 902s actually have. :yes:

[More measurements.... :p ]

OK, so conduct all the measurements with the N-800-F plus those components? No measurements with N-800-F alone?

That's 1st order compensation filter, yes?

Zilch
01-11-2008, 02:54 PM
OK, so conduct all the measurements with the N-800-F plus those components? No measurements with N-800-F alone?Gotta do both to know what's happening. Again, keep good notes so you can follow it later.


That's 1st order compensation filter, yes?Dickason calls it a "Contour network" (Loudspeaker Design Cookbook Fig. 7.155). Operatively, it's a response-shaping step filter. There's a way to make it adjustable, and it becomes the "T-filter" used by Altec in Model 19 and others:

http://home.earthlink.net/~jmarkwart/sitebuildercontent/sitebuilderfiles/n1201-8aRev.pdf

We're setting it to "Max" here, is what.... :yes:

Skywave-Rider
01-12-2008, 12:21 PM
Dickason calls it a "Contour network" (Loudspeaker Design Cookbook Fig. 7.155).

OK, I ordered that book and the caps.

Zilch
01-12-2008, 12:57 PM
Here's the "T-filter" standing alone, top. Altec used it to provide supplemental HF compensation after some stock crossovers.

As you see in the M19 crossover, R2 may be varied to adjust the amount of compensation.

At the bottom here is shown a Zobel, which you'll also find discussed in Dickason, used to flatten the reflected impedance of the woofer as illustrated by Earl, above, so that the main filter behaves more predictably.

Yeah, 6.2 uF for C1 and 4 Ohms for R1s are fine. R2 and the Zobel components vary with the application. You can see that if R2 goes to a high value, providing the minimum of 6 dB compensation, it's basically 8 Ohms in parallel with C1, what I have suggested you try for starters here, as there's already some compensation in N800-F (maybe).... :yes:

Earl K
01-12-2008, 01:15 PM
Zilch,

FWIW: The Zobel that's included in Altec(s)' 1209-8a network has these values ;

R = 7.5 ohms
C = 18 uF

<> :)

MrT
01-12-2008, 07:06 PM
Hi Zilch

I have been following this post and am learning a lot. If you don't mind I will be following this post and have learned a lot. Hey Earl if you responded to the my post at AA, I thank you.

Question. I found parameter for 414z (16ohm) at DIY which was measured in free air with added mass. Will it be helpful?

Tim

Zilch
01-12-2008, 08:12 PM
Question. I found parameter for 414z (16ohm) at DIY which was measured in free air with added mass. Will it be helpful? Yes, of course.... :yes:

MrT
01-12-2008, 08:46 PM
Hey Zilch

I've posted n800f question at the altec sight. Since then, I've been reading 846a/valencia thread. Wow is all I can say.

Anyways. This is the first time posting up here so let me try here.

Tim

MrT
01-12-2008, 08:48 PM
Hey I did it! I got some more! Again, i got these from DIY. The guys there deserve the credit. So here they are.

MrT
01-12-2008, 08:52 PM
Here comes another one!:) Zilch. I don't know what these graphs say. I would appreciate your help here. Basically I am trying to clone n800f for 414s and 806as. The one you pointed out to me. Hope I am not hijacking the thread.



Thanks.

Tim

Zilch
01-13-2008, 12:34 AM
Zilch. I don't know what these graphs say. I would appreciate your help here.First one is impedance and phase, basically confirming the one Earl posted earlier. The impedance is substantially on the rise between 200 Hz and 2 kHz, the area of interest. Note the glitches ~2 Hz, common to both curves, independently measured. Something's up with those drivers at that frequency. The T/S parameters derived from that curve appear at the bottom there.

The next one is a software analysis of a two-way crossover, but it seems not to be the stock N800-F, as, while there is a spread between the low and high pass filters (782 Hz - 1028 Hz,) that is not as wide as I believe it to actually be. Also, the Altec filters are not Butterworth. I've never measured the voltage drives, but I suppose I could do that for some N800-8K clones I recently built. For now, I'll take a look at my sims to see if that's how N800-F behaves with the "wrong" impedance drivers.

Last one is an RTA plot of the woofer, but not taken nearfield, apparently, and in an unknown alignment. Thus, it's only the higher frequencies which are reliable, and we can see that the driver's natural roll-off occurs well above the crossover region under consideration here. The response appears relatively flat from 100 Hz to 3 kHz....


Edit: N800-F looks like it could work, actually. Here are the simulations of the voltage drives, assuming resistive loads. Green and Yellow are 16-Ohms Cyan and Violet are with 8-Ohm loads. Red marker "a" is at 1.2 kHz, and there's 13 dB of HF compensation using 8 Ohms at that point. I suspect the acoustic crossover point may be somewhat lower than that, in which case 902-8As will likely be running out of headroom against dual 414s. Skywave's going to show us how it actually performs.... :thmbsup:

MrT
01-13-2008, 07:02 AM
Hey Zilch

Thanks for the analysis. I understand it better now. :D I know the RTA plot was done with the driver in a 120lt cab. The guy who provided the parameter xoed his at 1200hz. However, I did not understand how the value of these component correlates 3.9mh+8.2uf. Second order butterworth? I thought the crossover was some what close to n800f. It will be great to see skywave's findings. :applaud:


Tim

MrT
01-13-2008, 10:25 PM
Hi Zilch and Earl

I've been reading that Valencia thread over and over again. Now, I have a little better understanding of what filters are for. It took a long time and reading to understand the filter concept. While we are waiting for the Skywave's results I would like to ask some questions. The compensation filters you design: http://audioheritage.org/vbulletin/showthread.php?p=160329#post160329 Was it for bms drivers or can it be used for 806as without any modification in addition to n800f? If I were to copy m19's crossover, or do i need to if I follow your filter, what changes do I need to make for my drivers are 16 ohms. I bought my drivers in December. And I want to enjoy them as long as I can.

Your suggestion for the active crossover sounds very attractive. i have to pull out my old Marantz and start cleaning it up. But this has to be long term project.

Thank you.

Tim

Zilch
01-13-2008, 10:47 PM
It's for BMS drivers. Wait for the outcome of these evaluations, or move up your priorities to go active.

M19 is for 8-Ohm drivers. What crossovers are you using now, N800-F? If so, build the "T-filter" and add it between your crossovers and drivers. Start with no R2, as we are doing here. That'll give you 3 dB more compensation than is built into N800-F.

Unless you go active, you're going to run into the headroom problem with less compensation, even, as 806As are 3 dB less sensitive than 802/902.

In either case, you're not going to get as much HF extension as the Altecs with the Tangerine phase plug. Those curves are in the Valencia thread. If you decide to change diaphragms to eek out the max you can, go with the Radians, perhaps. I tested those there, also, but not in 806As.

We don't yet know how the N800-F is going to work with 414s. I like the LF rolloff I see in the sims, but only the measurements can tell us what the result is, acoustically, once that voltage drive is combined with the frequency response of the paired drivers....

MrT
01-14-2008, 07:39 AM
Hi Zilch

Thanks for the reply. I have not finalized my crossover yet. I got my 806s a month ago and 414 3 weeks ago as Christmas present. And, I am using a temporary mean to get by for a while. After reading your valencia thread at least 3 times, I decided to go with n800f clone for sure. I think holding onto 806 may be purely emotional for BMS looked real good. So.... I feel terrible for saying this. I am already thinking about buying new drivers. :D

But for now, after looking at Skywave's results I will add t filters and zobel accordingly.

Thanks

Tim

Earl K
01-14-2008, 07:56 AM
- Donaldpatton has done as much as can be accomplished to lift the HF response of the ( anemic ) 806a .
- He's presented the best ideas that I've seen ( to date ) for bringing the 806 into the 21st century.
( Of course Zilch might just junk this driver and encourage the purchase of a BMS / based strictly on FR response plots - that's hard to argue against )

- Donalds' excellent work can be seen Here ! ( at his website ) (http://web.mac.com/donaldpatten/iWeb/X-Over/X-Over.html)

http://web.mac.com/donaldpatten/iWeb/X-Over/X-Over_files/Xover405MF.jpg

Highlights ( in the Hipass circuit ):

- The resonant LCR circuit ( in the HF bypass part of the citcuit ) that's centered around 14K can provide an actual "boost" for frequencies beyond 10K ( where the 806 takes a HF dive ) . This approach is the only way to go when trying to get the last few db(s) of HF from the 806 ( & old 802 variants ).

- The LCR notch filter ( trap ) centered on the drivers fundamental resonance is another highlight that I like .

- I appreciate the fact that this is a single pole ( 6 db ) hipass / summing with an 8 pole Bessel lowpass .

Changes ( I'd make / & these are really personal preference ) :

- I'd explore losing the padded "drive" pots arrangement ( R2 ,R3, R4 & R6 ) . I'd redesign the padding arrangement to instead use a "T" type pad ( variable or fixed ) . This should offer a more "constant impedance" for the main filtering cap . Of course this would mean the HF contour circuit would need to be attached at either end of the "T" pad and the LC values altered accordingly for the new impedancce .
- I might add a coil ( a second pole ) for an 811 horn-driver combo / if the FR warrants the extra filtering . I'd likely place this coil "after" the "T-Pad" ( to take advantage of the lower working impadance in this area of the hipass / hence a smaller / cheaper part ) .

:)

ps ; Gone till the weekend ( so, no more posting by me / I'm hoping Zilch will take the bait and run with it ;) )

89-300ce
01-14-2008, 08:45 AM
Here's the "T-filter" standing alone, top. Altec used it to provide supplemental HF compensation after some stock crossovers.

As you see in the M19 crossover, R2 may be varied to adjust the amount of compensation.



Probably a dumb question. Since the R2 resistor in the M19 crossover is variable why wasn't the the 8 ohm side of the Lpad used to replace the preceeding R1 resistor like in conventional Lpad use?


Jorg

grumpy
01-14-2008, 09:17 AM
why wasn't the the 8 ohm side of the Lpad used to replace the preceeding R1 resistor like in conventional Lpad use?

http://www.colomar.com/Shavano/lpad.html

it isn't often that the leads are brought out to where this would be possible, if I
understand the question. -grumpy

89-300ce
01-14-2008, 10:19 AM
I believe we are talking about the same thing. I guess when Altec originaly built the crossovers the 2 gang potentiometers where not available yet or too expensive.

Jorg

Earl K
01-14-2008, 10:45 AM
Jorg, ( I leave in minutes but ,,, )

- Per your overlay of the ( 8 ohm ) LPad on Altecs fixed Tpad .

- If one uses a 16 ohm variable Lpad at that postion & then makes R1 , 4 to 8 ohms ( & then followed by any 8 ohm driver ) / one creates a variable 16 ohm ( more or less ) T-Pad that also has very usable isolation from the HF contour circuit ( some isolation is needed ).
- I use this configuration on the bench exclusively when mocking up circuits for 8 ohm drivers .
- If you used this configuration ( instead of your present 1201-8a padding arrangement ) / you would need to recalculate ( for a 16 ohm load ) the CLC elements preceeding the new padding topography . Alternately, just strap a 20W, 16 ohm conjugate resistor in front of the new padding ( & keep the present CLC elements ). I recommend you try it both ways .

:)

89-300ce
01-14-2008, 02:07 PM
Thanks Earl,

I'll try it once I get an RTA set-up of my own so I can see what I'm doing. I'm curious though. Why do you recommend a 16 ohm lpad? The R1 fixed resistor in the Tpad is/was a 4 ohm and the variable R2 resistor in the M19 is a 0-32 ohm. An 8 ohm lpad seems to be a closer match to the Altec design, just the R1 resistor being variable as well.

Jorg

Zilch
01-14-2008, 05:54 PM
I ran sims, and that kinda works. The shapes of the curves change somewhat as the values are adjusted, but not fatally.

With an 8-Ohm L-pad, you begin at 3 dB of compensation, and the midpoint is 6 dB, so the useful range is in the lowest settings.

With a 16-Ohm L-pad, the starting point is 6 dB.

Changing the fixed R to 8 Ohms with that gets 10 dB compensation at midpoint, and 15 dB max (all at 1.2 kHz,) which I consider a useful range.

I'd have to build it up and measure the reflected impedances; too tough for me and the calculator to figure.

Aside from its inherent symmetry, (you appreciate that the left side of the "T", your connection to terminal 3, varies from 8 (or 16) Ohms to zero Ohms (CW) depending upon the control setting,) I consider the original circuit to have the advantage of but one wiper contact when made adjustable, with the parallel leg of an L-pad easily serving as the adjustable R.

I am clueless as to how the reflected impedance of that varies with the setting, as well, and would have to measure. Note that Altec typically used this filter AFTER the main crossover attenuation L-pad, i.e., between that and the driver, providing an element of isolation as Earl recommends above.

And finally, this compensation filter should not be conceived or used as attenuation, which must be otherwise provided for balancing. Presumably, the amount and character of compensation should not vary with level, and once correctly established for a particular driver/horn combination, attenuation should be idependently adjusted for the desired balance with the woofer.

The Model 19 crossover does not operate this way, and worse, attenuation and compensation adjustments are interdependent. Hardly anybody gets how to use it, and it was widely ridiculed by pros who didn't understand it. Markwart apparently abandons M19's attenuation adjustment and uses compensation for level balancing.

For the truly gluttonous, here again is my M19 XO analysis:

http://audioheritage.org/vbulletin/showthread.php?p=128826&#post128826

With respect to Skywave's project, I expect this is all academic; he's likely out of headroom with the compensation already provided in N800-F....

MrT
01-14-2008, 06:36 PM
Hi Zilch

Just finished cloning n800f with the parts I already had in my hands. Turned on the music. Sounds good. Turned on my laptop and opened up the Lasing Heritage forum. What do I find? More options. :blink: I noticed that N800f has attenuation as you said they were 25ohm pot and 20ohm parallel to the LF. My question is does T filter goes before or after the attenuation or does it matter? I am confused.

Thanks

Tim

Zilch
01-14-2008, 07:37 PM
I noticed that N800f has attenuation as you said they were 25ohm pot and 20ohm parallel to the LF. My question is does T filter goes before or after the attenuation or does it matter?After's fine. I'd suggest you try what Skywave's doing, just the 6 uF with 8 Ohms in parallel, the combination in series with the HF driver. In your case, that provides 3 dB more compensation. If the HF no longer plays loud enough to balance with the LF, you're out of headroom.

The 8-Ohm version of that crossver, N800-8K, uses an L-pad for attenuation in lieu of the pot and fixed resistor. Use a 16-Ohm L-pad with N800-F as alternate for a 16-Ohm HF driver, or an 8-Ohm one with an 8-Ohm HF driver....

MrT
01-14-2008, 08:10 PM
Will do!

For Zobel for 16ohm LF, am I doing the same thing as Skywave?

Thanks.

Tim

Zilch
01-14-2008, 08:20 PM
Skywave's not doing Zobel yet, I don't think.

But for dual 16-Ohm 414s in parallel, what Earl said.

18 uF and 7.5 Ohms for starters.

8 Ohms's fine, too.... :yes:

MrT
01-14-2008, 08:54 PM
Ok. I will wait then.

Tim

Gary L
01-14-2008, 09:26 PM
I know you guys are having fun here and I began reading this thread because a buddy has 9844s and we just fixed them.

He was not happy with the HF and felt they needed an additional tweeter!

I found a bunch of info in the "Look at my new babies" thread and let him sit here and read.

Neither of us are up to the level of building XOs and testing the results.

He took a much easier route and bought a set of BMS 4550s.

Even I can hear a remarkable difference in them now and he is completely satisfied with his new 9844s now. So much so that he has me veneering the cabinets for living room duty as his wife wasn't going for the battleship grey.

I would love to see, while you are at it, what the equipment will show if you take them back to the two way XO and just add the BMS's.

Gary

Zilch
01-14-2008, 10:49 PM
Hi, Gary, and thank you for coming forward with your experience.

I believe we've pretty well established that there are better options than slapping a tweeter on vintage Altecs. Many enthusiasts are skeptical regarding those more radical approaches and instead remain committed to getting the best they can out of what they have. For others, the mere consideration of alternatives is insulting: they are perfectly content to enjoy Altecs for what they are; there's nothing to improve. I just got reamed again on another site for suggesting otherwise.

There are certainly performance enhancements to be achieved within this context, and Altec themselves showed how to do it with their own subsequent product improvements. Beyond this lies even more potential for those desiring to continue the process and explore outside the Altec boundaries. It's great good fun, learning all the way, and there is always further satisfaction in knowing that others are benefiting from the experience as well.... :thmbsup:

Zilch
01-14-2008, 10:53 PM
Ok. I will wait then. Naw, just do it.

We need input here.... :yes:

I'm not clear on one point: Are you using dual or single 16-Ohm 414s with your 16-Ohm 806As?

Skywave-Rider
01-14-2008, 11:54 PM
Hi Guys.
I'm transfixed on this thread.
Still waiting for parts. Hope to have them in a few more days.

Zilch
01-15-2008, 12:39 AM
Hi, Skywave! :wave:

No problem.

We just keep rattling here while we wait.... :p

hmolwitz
01-15-2008, 06:41 AM
I have a pair of these as well, and am watching this eagerly./
Thanks Zilch
Harry

MrT
01-15-2008, 07:16 AM
Ok. I will give it a try. I have one per side. The enclosure is altec 614 style. Mine is about 5 Sqft. But take a way the speakers, bracing, etc, The internal volume will give me about 4ft which is very close to 614. I will try zobel. But I can't measure anything. Me and my brain.

But I used to run the LF with 3.9mh+8.2uf and the HF with first order 10uf motor run. As of yesterday, I changed the xover to n800f. The LF has not changed much. But HF sounds more balanced. I tried 20ohm+25ohm pot combo verse 16ohm Lpad. The 20ohm+25ohm pot sounds more controlled.

The first order sound much warmer, meaning a lot of mid?

Hey Zilch. I have more speakers that I am hanging onto just for the emotional value such as JBL d216s which I believe is 96db and very clean mid. Should I turn these into 3 way? Oh. I forgot. The valencia thread seems to say that you favor 2 ways.

Ok Zilch. I will try zobel. 2 weeks ago, zobel-door bell sounded the same to me. Did not ring a bell to me. I am not a tech person. You know? My Single ended friends will call me crazy. But I have to try everything . :D

By the way. I am running my speakers with a couple of eico 14s. Top to bottom-tubes. Hummmmm.... You say mush. I say warm.:D

Tim

MrT
01-15-2008, 08:13 AM
By the way.

Am I using the same values for the zobel as Skywave's?

Thanks

Tim

Russellc
01-15-2008, 12:12 PM
Here's the "T-filter" standing alone, top. Altec used it to provide supplemental HF compensation after some stock crossovers.

As you see in the M19 crossover, R2 may be varied to adjust the amount of compensation.

At the bottom here is shown a Zobel, which you'll also find discussed in Dickason, used to flatten the reflected impedance of the woofer as illustrated by Earl, above, so that the main filter behaves more predictably.

Yeah, 6.2 uF for C1 and 4 Ohms for R1s are fine. R2 and the Zobel components vary with the application. You can see that if R2 goes to a high value, providing the minimum of 6 dB compensation, it's basically 8 Ohms in parallel with C1, what I have suggested you try for starters here, as there's already some compensation in N800-F (maybe).... :yes:

What wattage should the resistors in the "T filter" be?

Russellc

Zilch
01-15-2008, 12:55 PM
What wattage should the resistors in the "T filter" be?Not much power happening with compression drivers, so you could probably get away with something smaller, but I'd use non-inductive 10W Dayton or 12W Mills....

MrT
01-15-2008, 01:28 PM
Hi Zilch

I tried the 414 with zobel 8.2uf+16 ohm. Here is the graph. I have to say that I did not use any fancy program or microphone. So this is my disclaimer. If anyone of you object, I will erase this graph. I do not want to mislead someone with my ignorance. The first is 414 without zobel. And the second is with the Zobel.

Skywave-Rider
01-15-2008, 03:23 PM
How did you generate these graphs, T?

Zilch
01-15-2008, 05:01 PM
Seems to be rolling off better above the crossover frequency, as Earl suggested above.

Sounds better now, I betcha, without that nasty hump at 1250 Hz.... :thmbsup:

MrT
01-15-2008, 05:50 PM
Hi Skywave

Here is how I did it. An el cheapo mic, a laptop, white noise, and 12 inch from the woofer. :D

I think your gears will be much better for measuring the compression drivers.

Tim

Zilch
01-15-2008, 07:57 PM
We have some measured parameters for that woofer to work with, right?

Plug them into what Elliot says at 3.1.1 here:

http://sound.westhost.com/lr-passive.htm#6.0

I get 11.4 Ohms and 40.87 uF.

If you confirm, try something close to that and measure again.

Then, switch to pink noise, if you have it, and let's see that result, also.... :yes:

MrT
01-15-2008, 10:34 PM
the parts and pink noise i do not have. I want to shift my attention to 806s. They are good to 10000k, I think. A good and sweek midrange. With What I have 16 ohm vintage sansui horns, I may want to go three way. Or Zilch,
do you want to follow up with Danald's time delay xover as Earl suggested and see how far we can push the 806s? please! :bouncy:

And I will do another test with the values you suggested. But I am sure people are dying to see the show going. As far as I as concerned you are the show!

Tim

Zilch
01-15-2008, 10:50 PM
the parts and pink noise i do not have. I want to shift my attention to 806s. They are good to 10000k, I think. A good and sweek midrange. With What I have 16 ohm vintage sansui horns, I may want to go three way. Or Zilch,
do you want to follow up with Danald's time delay xover as Earl suggested and see how far we can push the 806s? please! :bouncy:Pink noise files are available online, and also on CD. If you use that, we can correlate your findings with the voltage drives in the sims.

Donald's LF filter with delay is for the 605A Duplex, with a specific offset. I'll likely do sims on the HF, tho.

Before I went three-way, I'd go active with CD compensation. That'll get you out to 14 kHz with the 806As, and dialable delay or sub augmentation facility.

Skywave-Rider
01-15-2008, 11:12 PM
Hi Skywave

Here is how I did it. An el cheapo mic, a laptop, white noise, and 12 inch from the woofer.

Tim

Looks easy to read and crisp to me. What software generated the readout?

Zilch
01-16-2008, 12:20 AM
Design frequency presumed to be 1.6 kHz, marker "a".

L6 DCR estimated 7.5 Ohms.

MrT
01-16-2008, 06:04 AM
Hi guys.

Skywave. The software is free! Google jdft 2.5. and download the program in 20 seconds. Useful but not too accurate. :D Then take screen shots with your computer. By the way. If possible use your gears. I bet they are much more accurate than mine. And we need accurate measurement.

Zilch. Active? I really am thinking about it. Do I have to biamp it? I found out that one of the Marantz side is blown. And lots of cleaning to do. But i am all with you. Let's work on the 806s. :applaud:

I have a doctor's appointment today. So, it may be late this evening before I can do anything.

Tim

MrT
01-16-2008, 06:08 AM
Zilch

By the way. Doesn't 605 use 414 (12inch) bases? My 806a reads 11.8 DCR.

Tim

hmolwitz
01-16-2008, 06:42 AM
The 605 is a 15" woofer with a 1" compression driver coaxially mounted.
The 601 is the 12" woofer, but it has a smaller HF driver the T3000.
Harry


http://www.lansingheritage.org/images/altec/catalogs/1963-pro/page09.jpghttp://www.lansingheritage.org/images/altec/catalogs/1963-pro/page09.jpghttp://www.lansingheritage.org/images/altec/catalogs/1963-pro/page09.jpg

Zilch
01-16-2008, 01:31 PM
Zilch. Active? I really am thinking about it. Do I have to biamp it?Yes, of course, but the power requirements of the HF are very modest. They'll even run on wimpy tubes, if you prefer "Organic." :p

MrT
01-16-2008, 04:45 PM
Hi guys

I have a correction to make. I did use pink noise not white. Sorry! And using 11.2ohm and 40.3uf this is the graph I came up with. Again, disclaimer. This is not a fancy or scientific measurement. Don't want to do this to you. :banghead:

I can't do active. Need an amp. My Marantz is out! :( So Zilch I need you to do some magic with 806as. :p

Tim

Zilch
01-16-2008, 07:01 PM
What you had for Zobel was better, looks like.

This seems to be rolling it off too early.

I'll try another approach for comparison. Shouldn't have used L1, probably.

It's instructive, though. Even with rudimentary measurements, it's apparent what's going on.

Get your HF running. Pull back to 1 M. Let's see how much headroom you have.

You can swap Zobels in and out while measuring and listening and not harm anything, BTW. The difference should be apparent in both.... :yes:

MrT
01-16-2008, 07:28 PM
Hi Zilch

I realized that in the past few days i've made 3 changes to my Xover. From 3.9mh+8.2uf to 3.5mh+10.5uf, to 3.5mh+10.5uf with 16ohm +8.2uf zobel, and finally 3.5mh+10.5uf with 11.2 ohm+40.3uf zobel. With some alligator clips i listened to three different settings.

By the way, are you around Berkley area? I used to live in So. Cal for about 10 years or so. And I used to drive up to Berkley area to meet up with my students. Having said that here are my impressions. I can compare the sound of a n800f clone with some what like High Sierras. There are highs and peaks that I can hear . And I can see why some can favor that type of sound. And the last one (the one with 11.2ohm, etc) I can compare with San Luis Obispo. The gentle and rolling hills. Yet, the bass was very clear and well defined. So, I would say 16 ohm+8.2 uf combo is some where in between but it leans towards high sierra side. Hope this make sense. I decided to use these expressions for listening to music and systems can be very subjective. Excellent experience!

I've been listening to John Lewis "the Bridge Game" vol. 2, Gilels playing Beethoven's Waldstein Sonata, Sychronicity by Police, and Jennifer Warnes' Famous Blue Raincoat. :D

Now I am reading David Weems' book.

So, there you have it.

Tim

dlp4341
01-17-2008, 12:36 AM
Zilch
L6 is 5.6 Ohms DCR ( # 28 AWG )
Your curves look great
BTW do you have 605As with 8 Ohm woofers
Don

Zilch
01-17-2008, 02:17 AM
Hi Don, and welcome!

It's YOUR filter generating the pretty curves; all I did was click some buttons. :yes:

5.6 Ohms makes for a deeper HP notch. :thmbsup:

I was only off by 34%. :p

No 605s here at all, alas.... :(

Why'd you roll it off at 14.5 kHz, Don?

Looking at the uncompensated response, 806As start taking a dive above there so it becomes a harder push.

I'm guessing they start sounding nasty if we allow any more boost above there?

MrT
01-17-2008, 06:53 AM
Hey Don. You found us! :applaud:

Skywave-Rider
01-17-2008, 08:18 AM
At the bottom here is shown a Zobel, which you'll also find discussed in Dickason, used to flatten the reflected impedance of the woofer as illustrated by Earl, above, so that the main filter behaves more predictably.

Yeah, 6.2 uF for C1 and 4 Ohms for R1s are fine. R2 and the Zobel components vary with the application. You can see that if R2 goes to a high value, providing the minimum of 6 dB compensation, it's basically 8 Ohms in parallel with C1, what I have suggested you try for starters here, as there's already some compensation in N800-F (maybe).... :yes:

I got my contour network parts today. I intend to do the measurements tomorrow. Zilch, since it takes quite a while to mail order components, do you think it's a good idea to order the Zobel parts referred to above? "18 uF and 7.5 Ohms for starters" are what you suggested for dual 16 Ohm 414s.
Also got Dickason's book, which I have begun reading.:blink:

Zilch
01-17-2008, 12:41 PM
Also got Dickason's book, which I have begun reading.:blink:Don't be put off by the math in there, now. It's all rather simple algebra. Go right to the chapter on crossovers for a better understanding of what's going on here.

Zobels are important primarily for the purpose of making a driver appear more as a resistive load so that the filter(s) behave in a predictable manner. As MrT has just demonstrated, they may also be used to optimize a specific filter/driver combination for a particular performance. However, as you'll see in Dickason, there's often (I'm tempted to say "Always," but I'd probably be wrong,) an equivalent filter design without employing a Zobel, per se.

Regarding parts, Timbers at JBL is rumored to have a "Crossover Cart" loaded with a selection of components of different values. I kind of have the same, acquired and accumulated over time. I've been wanting to build some switched substitution boxes to make the task easier, and also to prevent me from using my design components in builds.

So, yes, if you're going to do much of this, it's good to have some parts on hand to plug in and out to observe (and hear) their effect on your design. By your efforts, we'll soon know how well the stock design actually performs, and something more about the limitations of how it may be improved.... :yes:

MrT
01-17-2008, 05:48 PM
Hi Guys

Here are more graphs. Again, disclaimer. No fancy gears. Just some free program and el cheapo Mic. The Mic was literally right in front of the horn.

I took out the L-Pads. I realized that I got the wrongs ones. They are 8ohms not 16ohms, all soldered up. So, can't return them. Can I add additional 8ohms to tab 1 and 3and use them as 16ohm pads? :blink:Anyways. Here we go!

Tim

Zilch
01-17-2008, 06:58 PM
Add 8 Ohms in series with tab 3 only.

For now, just leave the L-pads out.

You can see the 806A "Shout" going away there.

The higher the resistance you put in parallel with the 6 uF now, the more compensation you will get.

At some point, the woofer will dominate, and you're out of headroom, the difference in sensitivity of the LF vs. HF drivers at the crossover frequency....

MrT
01-17-2008, 07:27 PM
Hi Zilch

Thanks for the reply. I hear you about the headroom. So, am I at the end of the road here? I may be able to pad down the mid more. Then, what happens to the over all balance? Stop now and be happy?

I would love to see what Skywave will come up with tomorrow just for the learning purpose.

Over and out for now.

Tim

Zilch
01-17-2008, 08:54 PM
The Markwart crossover(s), as example, set the VHF at max and adjust the balance by increasing or reducing the amount of compensation.

I don't believe you have enough headroom to do the Patten HF boost.

Build the "T" filter using tabs 1 and 2 of your 8-Ohm L-pads for the adjustable resistor, and use that to balance your lows and highs. Tweak your Zobel for the smoothest response through the crossover region.

With that, I think you're done until you go active.... :yes:

MrT
01-17-2008, 09:26 PM
Hi Zilch

That sounds awesome to me. I heard T sounds better than L-pad. So I will do. That's the same scheme we see in model 19 hf section. Correct?

Would you draw T filter schematic for 806As using 8ohm L pads, the values of the resistors, etc? I would really appreciate that.

414 sounds much better! That's for sure.

Thanks Zilch. This has been my first project. A lot of people helped me. I thank them all. :applaud::applaud::applaud:

Over and out for now.

Tim

Zilch
01-17-2008, 11:25 PM
The T-filter is shown in this post at the top:

http://audioheritage.org/vbulletin/showpost.php?p=197189&postcount=48

Substitute L-pad terminals 1 and 2 plus a 2-Ohm resistor in series for R2 to replicate the adjustable Model 19 version. 3.9 Ohms is not available from Parts Express in either Mills or Dayton, so use 4 Ohms. 6.2 uF is fine for C1, or a pair of 3.0 in parallel if you want to be spot on.

The contours generated by this filter are slightly different from those of a simple resistor and capacitor in parallel. I think the T-filter is better for your application....

MrT
01-18-2008, 06:34 AM
Thanks Zilch. I am going to finish it this weekend.

Tim

DonM
01-18-2008, 10:43 AM
The T-filter is shown in this post at the top:

http://audioheritage.org/vbulletin/showpost.php?p=197189&postcount=48

Substitute L-pad terminals 1 and 2 plus a 2-Ohm resistor in series for R2 to replicate the adjustable Model 19 version. 3.9 Ohms is not available from Parts Express in either Mills or Dayton, so use 4 Ohms. 6.2 uF is fine for C1, or a pair of 3.0 in parallel if you want to be spot on.

The contours generated by this filter are slightly different from those of a simple resistor and capacitor in parallel. I think the T-filter is better for your application....


Zilch,

As you know I'm also working on a new filter for my Altec 806's.

Could you post the final complete network drawing for the HF?

Once again thanks for all your hard work and research on these drivers.

Don

Zilch
01-18-2008, 06:19 PM
Could you post the final complete network drawing for the HF?Well, it's not "Final" anything, of course, merely the adjustable "T" filter added on to the N800-F HF filter.

With LP1 set to full 35 Ohms, it provides 3.3 dB more compensation, and set at 10 Ohms, ~6.3 dB

The combined contour varies somewhat from both the stock N800-F (yellow) and the T-filter alone, above.

R4 represents the stock crossover attenuation control set to "0"; it doesn't have to be there.

The intent here is to squeeze the most available compensation out of this crossover and 806A. Depending upon what woofer is used, there may not be enough headroom available to do even this.

Again, my current recommended approach to 806A is to active biamp and engage the CD compensation. With the inexpensive Behringer CX3400, that produces flattened response and HF out to ~14 kHz on the 811B horn. If you don't like the sound of that compensation, use the adjustable "T" filter with it, instead, or an alternative contour filter such as a derivative of the Patten 605A (806A HF, different horn) filter posted earlier.

DonM
01-18-2008, 08:02 PM
Zilch,

My current passive crossovers are very close to the N800-F. I'll take your diagram and add on the CD compensation.

Yes it is better to bi-amp. I also have the DCX2496 however my 4 channel amp left with my Son to University this year.

Thanks again,

Don

MrT
01-18-2008, 09:58 PM
Hi Don

Would you share your schematic of N800f? I am curious to know what type of mods you made. I would really really appreciate that.

I am finishing up mine. And Just wanted to compare mine with yours.

Hey Zilch. I looked at the active crossover at the Partsexpress. Really tempting. In fact, I am looking at my rack to see if there is a space for it and an additional amp. Mid-high: tube and bass: SS. Makes perfect sense.:D

Tim

Zilch
01-18-2008, 10:47 PM
Hey Zilch. I looked at the active crossover at the Partsexpress. Really tempting.

Guitar Center may still have it on sale for $99.

If not, just tell them you want it for that.... :p

DonM
01-19-2008, 10:06 AM
Hi Don

Would you share your schematic of N800f? I am curious to know what type of mods you made. I would really really appreciate that.

I am finishing up mine. And Just wanted to compare mine with yours.

Hey Zilch. I looked at the active crossover at the Partsexpress. Really tempting. In fact, I am looking at my rack to see if there is a space for it and an additional amp. Mid-high: tube and bass: SS. Makes perfect sense.:D

Tim

Tim,

Since I still do not have all the equipment required to test the response of my system (Altec 806a/416a/811b) I have been following all the tests that have been posted by Zilch.

My current HF includes a 2.4 uF capacitor (instead of the 4.0 uF), a fixed 8 ohm resistor in parallel (instead of the variable 35 ohms) and a 3.5 mH coil.

I'll be purchasing the parts recommended by Zilch in the above post.

Regards,

Don

MrT
01-19-2008, 07:58 PM
Thanks Don.

Tim

Skywave-Rider
01-20-2008, 02:24 PM
For these tests I reverted to 2-way using the stock N-800F crossover. I was only able to test one speaker, but I believe previous tests indicate the drivers are close. I set the HF pot to “2” because that is the factory ref. and also happens to be a spot which is not intermittent. I used a dummy load when removing drivers as previously. One set of readings with the stock crossover, another with contour filter added (6.2uF+8Ohm paralleled in series with 902 as shown.) I also reversed the phase of the 902 for comparison.

I used a distance of 3ft, mic positioned to center of 811B horn (this works out to approx. 3ft high,) excepting the woofer nearfield reading which is 3in., same as last time.

Listening impressions:
First thing I noticed was that the pink noise output sounded softer, I didn’t need muffs this time, though I wore them anyway. (I kept amp level and pink noise output exactly as before.)

Unfortunately there was no extra time to listen much, but I rigged mono signals to both the 2 and 3 ways. No time to listen with contour filter. The 2 way sounded great compared to the 3 way. Much more balanced mids, highs were actually better, less smearing there. Sounded well extended too. The 3 way was horn forward and “pointy” in that range. Sounded like it had less bass, and rather irritating in comparison.

This was a listening test which I did during cleanup, about 15 mins, but the change is large and easy to hear. Sorry I did not have time to listen with contour in. But to my eye, the measurements don’t show an enormous change with it in, though there is clearly an effect.

I was amazed to see how high the 902 goes. And I thought to myself, especially after listening, “why did I want to change this?” Well, please remember that previously I had loading caps on the 902s, and never heard this system in 2way with the bigger back covers on. I believe that is one factor. Another is the relative auditioning. The comparison to the 3 way which is not optimized. (I went back to look at the 3 way RTA and found it looks smoother overall, hmmmm.) One thing is for sure -- hearing this monitor 2 way for 15 mins has excited me again about the potential it has.

I’m still reading. Eager to hear all your comments and absorb the expertise. I’m definitely willing to experiment much further in pursuing the maximum potential from the 9844s. Zilch I know you’ve helped a lot and I don’t want to be presumptuous.

Pics below, and I think I can take the plastic off the DEQ2496 now....

Zilch
01-20-2008, 05:53 PM
Well, there it is again, I must say!

I worked quite a bit in the Valencia thread with trying to add a tweeter to 811B, and it just didn't happen worth a poop until I mounted a small one inside the horn mouth itself, and even then, there were issues, but none so major that, using this approach, Jackgiff was able to integrate one successfully. You're hearing substantially what I heard, most likely: standing separate, an add-on tweeter sacrifices coherency, and it's quite audible.

You've also found, as I did, that the 811B horn supports VHF, and with even the modest compensation provided by the N800F crossover, 902 will play fully out through 20 kHz. If you move up and down, you'll find that response is somewhat beamy, but the beamwidth is at least 30° vertical, and JBL's UHF drivers like 2402 and 2405 are about the same. The horn needing to be at listening height in order to hear the VHF adequately is the compromise. Bottom line on which it would appear we agree: Two way is better. Make it work. :yes:

What first struck me in these most recent measurements is the nearfield LF response. Compare what you measured to the simulated voltage drive of the N800-F, the blue line here:

http://audioheritage.org/vbulletin/showpost.php?p=197232&postcount=55

The filter has assumed control of the woofers, and the response is flat as Nebraska. We're quick to criticize Altec engineers for their apparent cavalier approach in using this 16-Ohm crossover with an 8-Ohm load. Here, again, is evidence there's more to Altec designs than might casually be observed; it's US who are the fools.

Tuning now to the HF, generally, with the attenuation at "2" you're still playing that too "hot," by as much as 5 dB, perhaps. Crank it down for better balance. Watch what happens on the RTA display as you do that. The HF should be just slightly lower than the LF for optimum balance. That's encouraging, because it says there's sufficient headroom for a better compensation contour, which I will suggest below.

We again have a notch at 3K15. Let's find out if it's real before doing anything about it. Two suggestions: get that on the display again, and then rotate the horn slightly off-axis, so that the mic is aiming in the first open area off center, i.e., not at the central vane, which may be shadowing that frequency, or generating reflective interference.

Second, lay several layers of fiberglass insulation (or a thick fluffy comforter, folded) on the floor between the speaker and the mic, 6" minimum, paper only on the bottom layer, against the floor. We may be looking at a cancellation due to floor bounce at that frequency. That's the first reflection with a different path length (think inverted triangle,) and it WILL interfere with the measurements.

If neither of those measures mitigates the notch, and you don't discover something else which does, that notch is then coming from the driver/horn, not the filter, and it should not be there. Order up a pair of the Radian diaphragms for these drivers from Parts Express, install one, and compare the response measurements.

There's one more thing I'd like you to try the next time you set up for measurements, and that is to disconnect one of the woofers and see how that alters the total system response. You'll have to crank down the HF more to restore optimum. Do nearfield measurement on the single woofer also. I'm suggesting you try this because there's a simple way to use the second woofer as a "helper" to augment the LF response at the low end.

Finally, with respect to optimizing compensation, Altec used the "T" filter in later iterations of these "Pro" monitor products as well as in Model 19. The schematic differs from M19 in that it does not rely upon a connection in the LF filter path, and thus may be used independently. I'll draw that up and post it here. You may want to build both the LF and HF sections, so you have yet another LF option to try. We can almost certainly improve upon the HF response N800-F is generating in your system.

Thank you again for performing these tests and posting your measurement results. Know that you are contributing to expanding the knowledge base regarding these vintage Altec products and designs, and it is appreciated.... :thmbsup:

Edit: Model 19 crossover with N604-8H VHF attenuation:

MrT
01-20-2008, 06:38 PM
Skywave

Thanks for the excellent post. I learned a lot. :applaud:I heard 902s are one of the better Altecs. And that they are clean and crisp. I've never auditioned them myself. But the graphs are saying a lot. On the other hand, I heard that 806As are warmer and smoother. Further, I did hear that 806As don't go up high.

Zilch. I have a question. My understanding is that N800Fs are used both for 8 ohms and 16 ohms. If so, how is that possible?

Thanks guys. To me this post helped me most in regard to 414s.

Tim

Zilch
01-20-2008, 07:27 PM
I heard that 806As are warmer and smoother.

Further, I did hear that 806As don't go up high.They're largely the same thing, I would say.


Zilch. I have a question. My understanding is that N800Fs are used both for 8 ohms and 16 ohms. If so, how is that possible?It's an 8-Ohm crossover when used with 8-Ohm drivers, and a 16-Ohm one when used with 16s, the primary difference being that the voltage drives, and thus the frequency responses, are different in the two cases. Where they cross acoustically, in combination with the drivers in the system, defines the crossover frequency, NOT the electrical drive.

See the link I posted immediately above.... :yes:

MrT
01-20-2008, 08:57 PM
I don't know how many times I read this thread. Finally, I got it. I keep going back to the online calculators which does not explain anything. But this thread does.

Further, no VHF high but all mid. Thus the warmer sound or is it the age or even both? I don't hear mosquito anymore. ;)

Many many thanks.

Over and out.

Tim

Skywave-Rider
01-20-2008, 09:41 PM
Thanks Zilch!
I’m trying to arrange getting in there late in the week to do the next round.

I just realized that I have been referring to polarity of driver connections as “phase.” I believe I should have said “reversing polarity” when swapping the 902 connections.

I calculate the floor bounce to be about 18.6 wavelengths at 3150 Hz, so I think that puts the reflected wave at almost 180 degrees out at that frequency. But I don’t trust my math. Nor should anyone else.

Fiberglass first,

hopefully new diaphragms are not called for. :)

MrT
01-20-2008, 09:46 PM
Hey Skywave

Once again, I thank you for letting me into your thread. Your graphs revealed so many things. Awesome! :rockon1:To me this is the most helpful thread for 414s. My 2 cents.

Enjoy your speakers. Play some music and let us know how they sound.

Thanks.

Tim

Skywave-Rider
01-20-2008, 10:03 PM
Thanks, Tim.
Maybe one day I'll live in something bigger than a studio apt. and will be able to take them home, LOL.

Skywave-Rider
01-20-2008, 10:10 PM
I calculate the floor bounce to be about 18.6 wavelengths at 3150 Hz, so I think that puts the reflected wave at almost 180 degrees out at that frequency. But I don’t trust my math. Nor should anyone else.

Just double checked my math....WRONG.
I think it's only about 70 degrees difference in phase between direct and floor bounce at that freq. Forget I said anything....:o:

Zilch
01-21-2008, 12:29 AM
Circuit's posted at #118. I checked it in simulation; should work.

DCRs for inductors are provided. Use 0.8 mH for L2.

LP1 is balance with LF, then set LP2 for VHF level.

I like some of the JBL compensation curves better, but we're sticking with Altec here.... :p


Unfortunately there was no extra time to listen much, but I rigged mono signals to both the 2 and 3 ways. No time to listen with contour filter. The 2 way sounded great compared to the 3 way. Much more balanced mids, highs were actually better, less smearing there. Sounded well extended too. The 3 way was horn forward and “pointy” in that range. Sounded like it had less bass, and rather irritating in comparison.I suspect you're looking at the HF response you measured and saying, "SHEESH, that's HORRIBLE!"

Well, look again. If that notch weren't there, you'd be +/- 2.5 dB, and it's sounding nice to boot.... :thmbsup:

Here's what I got with Gary's 802-8Gs using the T-filter. Look familiar?

You're damn close, is what:

http://audioheritage.org/vbulletin/attachment.php?attachmentid=24117&stc=1&d=1175586531

Skywave-Rider
01-21-2008, 02:05 AM
I like some of the JBL compensation curves better, but we're sticking with Altec here.... :p

So you're saying I can't put a DC bias on the network?!?! ;)

Is this inductor OK for the low end leg of the circuit:
http://www.partsexpress.com/pe/pshowdetl.cfm?&Partnumber=266-910
A "magnetic" core coil. It specs perfectly for both inductance and DCR.

Do you think I should buy all the parts to build a pair or make one as a test mule?

Cool!

Zilch
01-21-2008, 02:22 AM
Is this inductor OK for the low end leg of the circuit:
http://www.partsexpress.com/pe/pshowdetl.cfm?&Partnumber=266-910
A "magnetic" core coil. It specs perfectly for both inductance and DCR.

Do you think I should buy all the parts to build a pair or make one as a test mule?That's the one I spec'd, yes. The others you'll find are 18 ga Jantzens, I believe.

Build a pair. If they suck, with some minor rewiring, you'll have a nice pair of brand new Model 19 crossovers to sell to a Valencia upgrader.

[And a couple of spare parts.... :p ]

Skywave-Rider
01-21-2008, 02:29 AM
If that happens, the spares will go into my burgeoning "crossover cart." :D
Thanks again Zilch.

Zilch
01-21-2008, 12:17 PM
One more thing, Skywave. When you disconnect one of the woofers, short its terminals. Otherwise, it will behave as a passive radiator.

It'll still do that, to a limited extent, but will be heavily damped by the dynamic braking effect of shorting it on the cone movement. :yes:

Skywave-Rider
01-21-2008, 01:31 PM
Thanks Zilch. I'm re-reading "Valencia" and this thread to try to conform all that has happened. Thanks for the shorted woofer tip, I understand backwave emf.

Need to ask: I'm assuming I should present a 16 Ohm dummy load in parallel with the in-circuit woofer to fake a nominal 8 Ohm impedance for this test. Am I right? Or do you really want to see what a 16 Ohm woofer will do with the N800F? [It's safe to say I won't even have parts for the M19"Z" clones by the time I do this test.]


More questions/confusion to follow.
:blink:

Zilch
01-21-2008, 03:25 PM
We want 16 Ohms. MrT posted his measurements of 16-Ohm 414s on N800-F. They don't look bad; let's verify them. We can try it again later with M19Z, as well.

Where I'm going with this is a 20 mH Super Q in series with the "helper" woofer. We don't know the in-box woofer impedance curve, so it's hard to be precise with respect to what frequency that's going to bring it in. They're not cheap at $32+ apiece, but, if you're feeling flush, spring for that and give it a try, as well.

It's a technique used in several JBL products, including the latest flagship Everest II, to extend the LF response some, and these 9844-8Bs may benefit from that, so long as the single woofer response blends well with the HF at crossover. Yes, you'll lose the double-woof SPL, perhaps as much as 6 dB; that's a tradeoff only you can assess based upon how they sound and your intended application for the speakers.

A major upside is that you don't have the two woofers interfering with each other off-axis (horizontal) at higher frequencies. Wavelength at 1 kHz is 13.5", no? So with a 6.75" differential distance to the two woofers, you've got full cancellation at that frequency. It's simple triangulation to figure out how far off-axis that would be, depending on listening distance.

For caps, I don't see much point in paying a premium for 1%. When I play with the values in the sims, nothing changes much with such small differences in values. I just use the stock Solens, without bypass caps. If you like the sound of the cheaper Daytons, use those....

Skywave-Rider
01-22-2008, 12:03 AM
After hearing your explanation, I really want to try the "helper" woofer concept. The potential for extending the low end is cool, but I really like the possibility of having better horizontal phase coherency in the mids. And now that the super tweeter is off the table, imaging is tightening up. But I was thinking, in the crossover region, with the current design, the transition from horn to woofers is basically a vertical shift, and vice versa. Using one woofer may make the transition somewhat horizontal, which might be a compromise, but a better one. Am I looking at that correctly? However, the 811B is so wide at the mouth, this issue may not mean much.

I had assumed the Everest 2 used double vertical woofers. Just checked it. Wrong. It's beautiful. Wide horn mouth. It's a modern 9844 with mini supertweeter, LOL.

Decision made. I'm buying those 20 mh megacoils.

By adding the coil we are shifting the power distribution by relieving the helper woofer from some amount of mids, tbd. But the 414 will only go so low, so is this technique a trick to add a little emphasis where it starts rolling off?

And I just thought that the helper woofers should be the ones to the outside, far left, far right.

My intention for these speakers is for eventual home use. I will also try them for mixing music, but nothing abusive because I have a warped perspective and they are beautiful -- to me they really are......
I am also interested in the history of the technology and I know a lot of music was recorded and mixed on monitors like this; I want that experience. My career goes back to the age of Uries, I guess.

Probably the only other speaker I lust after at this point (besides those Everst IIs!! :D) is a 604E, because of the Gold Star studio connection. But not sure if I've ever heard them.

Enuf rambling.

Zilch
01-22-2008, 12:37 AM
By adding the coil we are shifting the power distribution by relieving the helper woofer from some amount of mids, tbd. But the 414 will only go so low, so is this technique a trick to add a little emphasis where it starts rolling off?Yes, and because it's only 6 dB/octave, depending on the woofer impedance curve where the action is, it may bump up the bottom end, generally, as well. Whether you'll like that or not depends in part on how you use them. As nearfield monitors, you'll probably want that; consider it baffle step compensation. At home, perhaps, too; it may obviate the necessity of using subs with them, depending upon your room placement and listening preferences. A simple toggle switch in your final design will allow you to switch among the three operating modes in situ.

Again, all of this is speculative. We're just not going to know until you perform the requisite measurements. For this, set up nearfield with the RTA in average mode. Run just the one woofer until the curve is stable, then kick on the second one with the inductor and watch the curve gradually lift down there, in full giggle mode. It should go up 6 dB; 3 dB from summing, plus another 3 dB from mutual coupling. Finally, measure the "helper" woofer alone to see what and where it's actually contributing to the system response.

[More RTA fun.... :barf: ]

Zilch
01-22-2008, 02:12 PM
I heard 902s are one of the better Altecs. And that they are clean and crisp. I've never auditioned them myself. But the graphs are saying a lot. On the other hand, I heard that 806As are warmer and smoother. Further, I did hear that 806As don't go up high.For the 806A owners following this thread, we have earlier discussed their limitations in comparison Skywave's 902s with larger magnets and tangerine phase plugs. Here's a direct comparision of Gary's similar 802s versus Scott's 806As on Altec horns, unfiltered:

http://audioheritage.org/vbulletin/attachment.php?attachmentid=24110&stc=1&d=1175567265

And here's Steve's different pair of 806As measured 7 months later, along with how they perform when actively crossed and 10 dB compensated, i.e., pushing the CX3400's "CD Horn" button:

Zilch
01-22-2008, 06:59 PM
And here's Steve's 806As on the N800-F HF filter:

MrT
01-23-2008, 04:54 PM
Hi Zilch

Were the Red and Green measured with T filter? Active crossover do make a huge difference!

Thanks.

MrT

Zilch
01-23-2008, 06:37 PM
Nope, no T-filters shown, only on Gary's 802s @126.

Biamping provides the headroom necessary to do the requisite compensation, active or passive.

Footnote: I am stunned by how similar Scott's and Steve's 806As measured @135. I had never previously juxtaposed them. It's as if I were remeasuing the same pair.

That's what 806As are on 811B horns, apparently. If someone measures them with a different result, I'd certainly like to know about it.... :yes:

MrT
01-25-2008, 10:16 PM
Nope. I did not go active. Not yet anyways. But I did finish putting up T filter to n800f. It sounds dramatically different. Better balanced, sound stage is better and I am really happy with the result. Time to enjoy the fruits.

Thanks guys. Thanks Zilch.

Tim

Skywave-Rider
01-26-2008, 11:49 AM
Mr. T,
I'm happy you got the T-filter well integrated and apparently tweaked. If you get a chance, post some pics of the monitors.

vuki
01-28-2008, 12:32 PM
Hey Zilch

Thanks for the analysis. I understand it better now. :D I know the RTA plot was done with the driver in a 120lt cab. The guy who provided the parameter xoed his at 1200hz. However, I did not understand how the value of these component correlates 3.9mh+8.2uf. Second order butterworth? I thought the crossover was some what close to n800f. It will be great to see skywave's findings. :applaud:


Tim

Hi,

those are the measurements of my 2-way Altec (414z in 120 Lit. BR + 902 + 450Hz round tractrix). First graph is free-air parameter measurement. Second graph is far field in-room RTA measurement of 414z in "Petite Onken" enclosure (ca. 165 lit.) without crossover.
One cannot "file" crossover based on its components values because crossover "works" with real loudspeaker's freq. response which is all but textbook flat.

Regards,
Vuki

MrT
01-28-2008, 02:12 PM
Hi Skywave

Here is what my speakers would look. I tried to stay with this plan as closely as I can. But this is for single 414. Yours may be different. Look through this sight if you are trying build your own enclosure. I hope this helps.

http://www.lansingheritage.org/images/altec/plans/1970's-lf-plans/enclosures06.jpg

Vuki. Thanks for dropping in and clarify your measurements.

MrT

Zilch
01-28-2008, 05:50 PM
One cannot "file" crossover based on its components values because crossover "works" with real loudspeaker's freq. response which is all but textbook flat.We'd like to buy or build textbook crossovers and have them work. They DO work, of course, but rarely all that well.

One would be hard pressed to find very many textbook crossovers among JBL products, and, as we've seen here, Altec deviated substantially from that path, as well.... :yes:

Skywave-Rider
01-30-2008, 01:06 PM
Parts arrived. Just tried to scavenge 16uf caps from old homebrew x-overs and learned a valuable lesson. That is: HOT GLUE IS #@%&ING STRONG. Just ordered a pair of those :banghead:
Will do the N-800F measurements part 2 tomorrow.

Zilch
01-30-2008, 02:03 PM
Yeah, hot glue's a bad idea until you're really, REALLY done. I never am, so it's cable ties on everything. The only way I succeed in breaking the glue loose is with a heat gun, utility knife, and a prayer that whoever used it did a crappy job. :p

I'm anxious to see how the "Helper" woofer thing operates in your setup. :bouncy:

Depending on where the 20 mH inductor kicks in, the main lowpass rolloff will revert somewhat to the 16-Ohm curve, which may work out because there'll only be one woofer playing in that region. :hmm:

There are two options: 1) Run the helper with its inductor off the LF output of N800-F as normal, in parallel with the woofer not having the additional 20 mH, or, 2) Run the helper with its inductor directly from the crossover input, in which case the main woofer gets the full 16-Ohm voltage drive curve.

Try both, maybe.... :yes:

Skywave-Rider
02-04-2008, 01:03 AM
Hi,
I opted for comforters to reduce floor bounce. I’m having some fiberglass issues (in my throat right now -- probably from opening/closing the cabinet repeatedly and digging around in there); as for the bedding’s effectiveness, I don’t know. I additionally fastened some foam onto the ceiling in the bounce zone.

For nearfield measurements in 2.5 way I placed the mic at 3” from the baffle equidistant between both woofers and ports. For full range 3’ readings I offset the mic so as not to be centered on the middle vane, as Zilch pointed out that might create the notch we saw in the 3k region. I “think” this helped and I posted the RTA for benchmark “centered” on the vane and the offset pointed into the first sector to one side.

This time I did not have the opportunity to listen at all. I left the 20mH coils installed in N-800-F output mode, so this week I should roll them out and give a listen.

I have all the parts for the next crossover, but have not put anything together yet.

Skywave-Rider
02-04-2008, 01:30 PM
OKhttp://www.youtube.com/watch?v=97CEcUEjJAE
Try removing the "OK" at the head of the above URL and paste it into a browser window.

Fun to watch the LF come up. Not very clear, though.:)
Nearfield with supercoil running from N-800-F output. HF is on a dummy load.

Zilch
02-04-2008, 01:36 PM
Well, you've certainly got BASS in the 2.5 mode with the helper running! :p

Though the nearfield with one woofer @ 16 Ohms looks the best, remember that nearfield exaggerates the LF response. I think of it as reflecting how the response would sound with full boundary reinforcement.

Missing in your measurements is nearfield of the helper woofer running alone. Get that first when you go do more.

As suspected, the notch at 3K15 would appear to be coming from shadowing by the central vane of the 811B horn on axis. Mitigating the floor bounce seems to have helped, also. I've seen the HF roller-coaster response you're showing in my own measurements here; I believe it's the horn vanes doing that. We'll ignore all of that henceforth, unless some surprise appears.

Adding the helper woofer doesn't do anything to the main woofer response in the lowpass region (the top end of woofer response,) the reason being that the inductor impedance is apparently high enough at those frequencies so as not to alter the filter frequency or slope. We're therefore presently running on the 16-Ohm curve rather than the 8-Ohm one.

If we were to stay with N800-F, we might choose to alter the filter for optimum response blending of the two drivers in the crossover region, but I think we'll be preferring to work with the modified M19 crossover, once you have those built.

The HF is still running too "Hot," but I appreciate that you have some difficulty cranking that down due to intermittency in the stock crossover attenuation potentiometers.... :yes:

Skywave-Rider
02-04-2008, 01:47 PM
I can't believe I forgot to get the helper alone, Zilch....
Yes, bass! Thanks.

Zilch
02-04-2008, 01:47 PM
Fun to watch the LF come up. Not very clear, though.:)

Nearfield with supercoil running from N-800-F output. HF is on a dummy load.

WAY cool, and quite amazing, actually, I must say. I am in full giggle mode here. :p

[Fun to hear the pink noise playing, too.... :thmbsup: ]


I can't believe I forgot to get the helper alone, Zilch....No problem; we'll get it next time. Both direct wired and on the N800-F filter, please. :yes:

We need to know that if we're going to tweak the bass extension....

MrT
02-04-2008, 09:44 PM
Hey Skywave

Thanks for the graphs. And thanks for measuring one 414 a side. You did not forget me. :bouncy: Awesome!

I am using 3.7mh+10uf+zobel (16ohm+10uf). I am using a set of selector to use the crossover and the first order, the second order, and the first order with zobel, and the second order with zobel. Each has it's merit, I think.

Zilch. I have not forgotten about the active crossover yet. At this point I am wrestling with my home brewed preamp hum. It seems like my system is full of problems.

By the way. With the helper zobel is not really needed. Right?

Thanks guys.

Timp

Zilch
02-04-2008, 10:11 PM
By the way. With the helper zobel is not really needed. Right?Unknown.

Zobel's action's not down where the helper is playing, rather, up in the main lowpass region.

We're not working there yet....

Zilch
02-06-2008, 01:21 PM
The video has enjoyed 105 views, and nobody has said "Boo."

Multiple choice:

1) Too esoteric; nobody gets it.

2) It's really, really dumb.

3) "You speaker geeks are WAY too easily amused...."


:p

Chas
02-06-2008, 01:59 PM
The video has enjoyed 105 views, and nobody has said "Boo."


More like: Very familiar to me, as I have spent a few Sunday afternoons seeing and hearing that......;)

Skywave-Rider
02-06-2008, 02:19 PM
To me, a combination of all 3. LOL.
Chas, good to know we're all in the same asylum.

Zilch
02-06-2008, 02:59 PM
More like: Very familiar to me, as I have spent a few Sunday afternoons seeing and hearing that......;)We're not letting you slip away that easily, Chas. :scold:

Is this going to work for Skywave's 9844s? Tell us more about your experience doing "Helpers," please.... :yes:

Skywave-Rider
02-07-2008, 10:38 AM
We're not letting you slip away that easily, Chas. :scold:

I'd love to hear about your experience, Chas.

Zilch
02-07-2008, 11:47 AM
Member Bfish over on the Altec Forum observes that, while entertaining, the video is unfair to the single woofer playing alone at the beginning in that the second, unconnected one, is functioning as a passive radiator and altering the box tuning characteristics. Even if the voice coil were shorted out, the tuning is still for two woofers, not one, and inherently wrong. What we're seeing there is multiple variables converging favorably to produce the effect.

Though the video was produced primarily for giggles, that's a valid criticism which suggests an alternative experimental design with the potential to yield additional useful information: Start with the inductor shorted, i.e., both woofers playing concurrently, then unshort it to observe the attenuation outside the region where the "helper" makes its contribution.

Do the dual woofers normally mutually couple at those higher frequencies? How much SPL are we actually "giving up" to achieve the extended bass produced by this scheme? Do we care even one whit about losing that in exchange for the apparent bass extension benefit? :dont-know

Skywave-Rider
02-07-2008, 12:08 PM
OK cool!
:applaud:
Question: Does it follow then that if one was to redesign a speaker like this that it would be a matter of course to look at partitioning the box unequally into 2 volumes tuned for each woofer and their respective ranges? (That is if one did not care about efficiency from 2 414s working in tandem in the same volume.) And don't speakers couple anyway, even in separate enclosures?

I gotta read all the box stuff in Dickason.

I'll plan to run those tests.
Edit: I just read your post again and now more fully understand what you said about the higher frequencies the woofers are producing together and how we must determine the attenuation when switching to 2.5 mode. Thanks!

Chas
02-07-2008, 12:25 PM
I'd love to hear about your experience, Chas.

Sorry, my time with Altec stuff ended in misery, so I gave it up years ago for JBL.....But, I do still kinda like the stuff.

I didn't try what you're doing, but I think it makes good sense. This thread is a keeper, for sure.

In fact, I am planning to make a switchable (parallel or helper) four 2235H (two per side) same concept similar to the 4435 except with four modular, independent boxes.

Now that I think about it, the 4435 shared woofer volume in the early versions according to a few folks here, but not for long. They apparently added a partition later.

Good luck guys, I'll be following closely.:)

Skywave-Rider
02-07-2008, 12:52 PM
Sorry, my time with Altec stuff ended in misery, so I gave it up years ago for JBL.....But, I do still kinda like the stuff.

Hahaha. Yeah. I'm goin' down, I know it now for sure. But flailing all the way.

I appreciate hearing about your experience. I'm learning a lot and could not have done any of it without Zilch and all the others down through to the bottom of this thread and elsewhere. I'm pretty sure I have the stamina to roll with it.

I know little about JBL. Though I'm learning and respecting.

My experience with JBL is as control monitors and I never liked them in that application. However, any working speaker takes time to get used to. I won't argue with anybody's preference.

Yeah, I could easily lust after 4435s.

Zilch
02-07-2008, 12:52 PM
Question: Does it follow then that if one was to redesign a speaker like this that it would be a matter of course to look at partitioning the box unequally into 2 volumes tuned for each woofer and their respective ranges? (That is if one did not care about efficiency from 2 414s working in tandem in the same volume.) And don't speakers couple anyway, even in separate enclosures?As Chas observes, 4435 may have been made both ways, but everything else that comes to mind is common chamber I believe: 4612, SK2-1000, Everest II, dual-woof 43xx (?), even the early horns (?), Westlake et. al (?). Mr. Widget or others may have done some experimentation relating to Project May.

In the helper schemes, it seems to me not to matter, at least with respect to the fundamental tuning: you tune for the dual woofs, because they are both playing in the region where box tuning has its primary effect. Outside there, I'm not sure but that isolation has some benefits; they're just unknown to me. Manufacturing costs may also play a role here, as well, when we look to actual product for design guidance, though the flagship statement Everest II could clearly have been made either way.... :dont-know

Chas
02-07-2008, 12:56 PM
In the helper schemes, it seems to me not to matter, at least with respect to the fundamental tuning: you tune for the dual woofs, because they are both playing in the region where box tuning has its primary effect. Outside there, I'm not sure but that isolation has some benefits; they're just unknown to me. Manufacturing costs may also play a role here, as well, when we look to actual product for design guidance.... :dont-know

Makes sense to me too.:yes:

MrT
02-07-2008, 10:38 PM
414 happens to be my favorite woofer. It will be a wonderful to see them being maximized. I am so excited to see you guys doing this. On a side note. There were a guy in town selling 4 9844 driven by 414-16Y for 500. I missed them by a phone call. I wish I could have bought them. So I could be part of the project.

As I get a chance I will post some pictures.

Timp

Skywave-Rider
02-08-2008, 11:15 AM
On a side note. There were a guy in town selling 4 9844 driven by 414-16Y for 500.

A much better price than I paid. Yeah, it would be great to have 5 of these in a surround setup, but I can't really even fit my Heresies in my apt. as it stands. Frankly, I feel lucky to have a decent pair of 9844s and a place to keep them, as inconvenient as that place may be.

Hope to do the next round sometime this weekend.

Skywave-Rider
02-08-2008, 08:50 PM
Zilch,
Here’s what I plan to do tomorrow, let me know if I’m missing something:

All tests nearfield and with 902 on dummy load. Tests 1& 2 with inductor at N-800-F output only, tests 3 & 4 from both x-over input and output.


1. 20mH shorted, both woofers playing.
2. 20mH in line, both woofers playing. (Video at change from short to in line.)

3. Helper alone, coil at N-800-F input.
4. Helper alone, coil at N-800-F output.

I think I should dummy load 16 Ohms at main woofer for #s 3 & 4, yes?


Unless I missed something, next one will be with M-19Z.

Zilch
02-08-2008, 09:04 PM
Yup.

You're s'posed to listen some, too, you know. :yes:

[I claim us measurers listen more and with greater precision than listeners do.... :thmbsup: ]

Skywave-Rider
02-08-2008, 09:13 PM
Yeah, :) thanks.
I will this time.

Skywave-Rider
02-10-2008, 11:42 AM
I was bumped yesterday and am on standby today. Worse comes to worse, will go Weds.... :(

Might be for the best because what I thought was fiberglass in my throat turned out to be a bad cold*, so I’ve been moping around recovering for almost a week -- I feel OK now, but my ears are still stuffed a little. And since I missed the listening test last time, have to do a good one this time.

So I turn my attention to a couple of points. First, when I look at the Valencia thread, jackgiff posts these FR readings:
http://audioheritage.org/vbulletin/showpost.php?p=167133&postcount=802

Comparing his BMS 4550s to my Altec 902s:
http://audioheritage.org/vbulletin/showpost.php?p=199436&postcount=146

I do not see the roller coaster in the HF coming from his 4550s on 811B horns. Mine look scary. On some of Zilch’s Clio graphs I think I see camel humps, but it is primarily on the 802 measurements.
http://audioheritage.org/vbulletin/showpost.php?p=197999&postcount=126

So, is this possibly the N-800-F ‘s effect on the 902 HF? Or is it inherent in this driver/horn combination.

Zilch, you remarked that you see the undulations at the Lab and attribute it to the 811B horn. When you see that, is a 902 bolted on?

Maybe this is moot until the Z-19s?

The other thing, about the dummy loads I’ve been using: This is a newbe question I know, but since I’m substituting resistances for nominal impedance, what relevance do those 8 or 16 ohms have? Am I right in thinking that if I want the crossover to function as it would when the actual driver is in place, then there is no substitution for the actual driver with its frequency dependent complex impedance?

So, If I have that assumption right, then when doing tests like we are doing, is there a way to insert a fixed resistance which is a number more effective than nominal? For example, if I want to look at the 414s upper bass FR near the crossover point as a focus, would it be better to look at the 902s impedance at about that frequency and substitute that value for the dummy load? Then the result would be good for that narrow range only. So I’m thinking a non inductive pot would be useful.

Or is the nominal figure really some complex mathematical average of the entire impedance curve the driver exhibits? So perhaps the 8 or 16 Ohm dummy is the best substitute.

And all of this assumes that the filters do interact.

I can feel you rolling your eyes....hehehe.

Getting antsy, sorry for the delays.



*Or a combination of both; pretending to be a lab-coat guy here, gotta watch the assumptions.

Zilch
02-10-2008, 03:24 PM
My most expert analysis of the roller-coaster thing at this juncture is, "I don't know," and I can only point to some of the same indications you have observed. I see it with Gary's 802-8G tangerines on 811 and 511, and I see very similar in the CBS Labs measurements of Model 19 from 30 years ago.

In my own situation, I'd be tempted to point to the suboptimal conditions under which I measure stuff, but I can put BMS drivers on PT waveguides and get stunningly flat response curves at any time, no sweat. Jackgiff has repeated that result with his Santana project.

As you've seen, I can also get respectably flat results with BMS on 811 and 511 without doing anything in particular to address individual response anomalies. Pi-speakers gets similar with an Eminence driver, often posted by others for comparison. I've never measured 902s; best I can do until then is assume they are similar to Gary's refurbed 802s, but I don't think there's enough evidence pointing to the Altec drivers as the source to condemn them.

I'm ignoring that in your measurements for now because, worst case, the response undulations are on the order of +/- 2.5 dB. The pattern is regular, and it's coming from something, somewhere; it's not a random error. There are several members here who can look at that and tell us what it's about, and I may be PMing them for their expert opinion if it persists under other conditions. I may also be proposing that we swap drivers for reciprocal testing at some point.

Regarding dummies, yes, using a resistor closer to the actual impedance of the woofer is probably desireable, and there are ways to construct a more accurate diriver simulator than a simple resistive load, but with parallel crossovers such as we're dealing with here, the filters do not interact but in detail, so it doesn't much matter.

The primary purpose is merely to approximate the loaded condition, and at the level of resolution we're able to measure, a fixed resistance of the nominal value is most likely good enough. As you suggest, one could use the driver in a second system isolated apart, closed in a neighboring room, for example, for an ultimately accurate "dummy," if desired.

RTA inherently provides a smoothed metric of what's going on, and is very useful in assessing that in large scale. Yes, we can get more accurate with greater resolution using alternative measurement technologies, but would likely be applying mathematical smoothing to make equivalently useful determinations at this level.

[I presume M19Zs are nearly ready for prime time by now.... ;) ]

Skywave-Rider
02-12-2008, 10:17 AM
Near field of helper woofer. 902 on dummy load. Main woofer on dummy load.

Listening planned for weds, current setup allows for normal woofer operation and 20mH in line via patch point insertion.

M-19Z should be this weekend. Hope to have seawolf’s Hafler pre amp by weds so I can easily do source switching and, actually, adjust volume like a person.

Edit: I just realized I forgot to short the main woofer for the helper woofer "alone" tests.

Zilch
02-12-2008, 01:15 PM
Arbitrarily picking -10 dB as criterion, the "Helper" is working primarily below 200 Hz, as we saw in the video. We could narrow its band of influence by making the filter 2-pole. Last time I tried that, my network impedance went to about 1 Ohm, as I recall, tho. That would take some further research. :p

I'm having difficulty seeing much of a difference above that, not what I'd expect to see from summing, at least. It's clear the lowpass filter is behaving differently, 8 versus 16 Ohms. Do we have a new video? I think we could see the differences more clearly there.

Yeah, shorting the main would make results on the helper more meaningful in comparison to earlier measurements, but I think we've got a good handle on what it's doing, and it's what we want. The time's better invested in listening at this point, and then moving on to the new crossover measurements. Gonna be some surprises there, I betcha.... :thmbsup:

Edit: Here's some results I'm currently getting with a different driver on 811B horn. The roller-coaster is considerably in evidence. M19 crossover highpass filter is Cyan:

Skywave-Rider
02-18-2008, 01:30 PM
Life has intervened, but I'm back on track. One M-19Z complete, the other in progress, please forgive the "tribute" Zilch. Crossover features construction site "sidewalk" lumber and a ton of lead. Not featured: hot glue on components :)

Zilch
02-18-2008, 02:14 PM
Looking good!

I decided to leave my vintage M19 crossovers alone, and just build this version new from scratch myself, so parts are ordered.

I've had a couple of inquiries; it'd be good if I was working with all new parts, as well.

[This had better work. My "I am a LAMER" bumper sticker looms in the background.... ;) ]

Skywave-Rider
02-18-2008, 04:43 PM
Thanks.
OK, but no matter what the result, this is the bumper sticker:

http://home.att.net/%7Eisokinetics/Project0.jpg

Skywave-Rider
02-25-2008, 02:17 AM
OK Zilch, here goes....

Measurements made at 3’ and slightly off axis, like last time.
Nearfield at 3’’ between both woofers and ports.

Unfortunately, I forgot to bring my dummy loads, so I couldn’t do any single driver measurements.:banghead:

In the L pad photo, the left pot is MF, right, HF. Yes, I suck at woodwork.

Big surprise to me is the narrow passband in the MF control. I thought it would be broader or even shelving-like in its effect, with the HF control working on the very upper end. Is this result expected? I went through my wiring to check for screw ups. There is one in the pic I posted last time, but I caught that early on.

I left one 9844-8B with original N-800-F and rolled both into a rectangular room for listening impressions. I fed both sides mono signals and played one at a time. I determined speaker placement options and several listening positions through the Room Mode Calculator. Program material included music and voice (speech.)

First, the good news:
The VHF gain, clarity and control via the L pad is exceptional. Now, this is like adding a tweeter. The top octave is even “airy.” HF transients sound clear, clean, lifelike. The N-800-F, in comparison, is not as extended sounding and might accentuate c. 8K where the 19 does it higher.

And the other news:
The mids have a few problems. The 19 honks out at about 6-800 Hz sounding like a narrow kind of forwardness making voices hard and fortissimo operatic passages difficult to play near “loud.” In comparison the N-800, much maligned by my initial impression, is smooth. Though It may be forward, it is not peaky sounding.

There is also a lower mid entanglement, 200-350 Hz approx., which makes bass instruments for example, hard to follow. Though the original network is far from perfect, it offers more clarity in this range.

Overall the new crossover sounds louder, but I’m assuming this has more to do with the pronounced mids as stated.

2.5 Mode: (These impressions apply to both crossovers, though I feel it sounds better in the N-800 cabinet.)
This gives a thump at about 80Hz. It’s exactly what Zilch predicted would result, a mode allowing the monitors to be used in a midfield position, without much reinforcement from wall boundaries. The los don’t seem to extend much, if at all, just congregate into a bump at 80Hz which works well on some material. On other material it can muddy LF clarity. During my listening session I used it more than I did not.

This mode engaged also removes some 200 or 250 Hz. Noticeable when playing electric guitar music. That’s where the body of the electric is, and I suppose the more paper the better when you’re emulating a 4x12 cabinet. However, shoveling some of the lower mids into the bass makes the mids leaner and clearer on other material.

Although I haven’t gotten into port tubes yet, I imagine these cabinets will never provide forceful low end. But I’m not expecting that either.

I have to head off to bed, I’m sure I’ll have more to say.

Zilch
02-25-2008, 12:52 PM
Okay, DROP the scissors and step BACK from the project cart! :scold:

The new N19Z crossovers have clearly assumed control of the HF, and your measurements indicate that the adjustability is very much according to the design sims. You apparently have sufficient driver response and headroom to achieve full VHF extension to 20 kHz, and adjustment range suitable for balance with either one or two woofers running. To use it most easily, adjust the MF control for optimum balance with the woofer(s), and then adjust the HF control for flattest VHF response.

Look again at this sim. There's shelving, but it is short-lived, up to 3 kHz only, above which frequency the HF control adjusts the slope of the compensation. If you crank the mid control up, it's only the shelf that gets boosted, and I certainly see that in your "MF max" measurements. I'll confirm once mine are built:

http://audioheritage.org/vbulletin/attachment.php?attachmentid=30602&stc=1&d=1200945041

Optimally adjusting the response appears somewhat difficult to do, as the roller-coaster ripple remains, well, "overwhelming." We now pretty much know it's not the crossover doing that. It'll be interesting to see what happens once you install BMS drivers; I'm still not entirely convinced there isn't something else going on we haven't figured out yet. :dont-know

We really need to see the response without the woofers running to know how the basic highpass filter is operating, and where. Dummy loads aren't essential for that, if your HF and LF filters are separately accessible; simply disconnect the input to the LF. I always build with separate input terminals to the two sections to facilitate this sort of testing or use, and simply bridge them for normal combined operation. Dummys on the output work too, as you know from experience, and probably offer a somewhat better approximation, at least from the standpoint of amplifier loading.

That's half of sorting out the midrange issues. The other half is what the woofers are doing. The lowpass filter in N19Z is simply that of Model 19 (2.7 mH and 21 uF). I used that as a starting point, since we're presumably operating at 1.2 kHz now, rather than the original 800 Hz, or wherever it actually was. See the sim, bottom, below; your midrange "honk" is clearly shown in 16-Ohm mode, red curve. As previously observed with N800F, this filter behaves very differently with 8- vs. 16-Ohm loading.

From your listening evaluations, it would seem that the lowpass in N800F (3.5 mH, 10.5 uF) may actually be more appropriate at one impedance or the other (2.0 vs. 2.5 operation). Compare the sims for both filters to see what's going on. I've reposted the N800-F immediately below. From looking at them, it seems your LF nearfields are done with the new crossover? If so, it looks fine with both woofers running, but you may want to try the N800-F LF in combination with N19Z HF. We can easily tweak the lowpass according to your findings.

In any case, it's now clear what "Helper" mode is doing. With just one woofer running, the response falls ~3 dB above 100 Hz, and everything below that, relatively speaking, is boosted accordingly. I maintain the thesis that the LF response is also more extended in 2.5 mode than it would be with one woofer running alone, other factors being equal, according to the rules of low frequency summing and mutual coupling, as outlned by Eargle and Foreman in JBL Professional Sound System Design Manual here:

http://www.jblpro.com/pages/tech_lib.htm

Still have a big hole at 3K15. :banghead:

Are you SURE there's no EQ or anything operating in that DEQ2496?

http://audioheritage.org/vbulletin/attachment.php?attachmentid=30466&stc=1&d=1200212143

Chas
02-25-2008, 01:53 PM
Guys, very impressive to see an 806 extend to 20KHz. Maybe I should have kept mine....

Zilch
02-25-2008, 01:55 PM
Guys, very impressive to see an 806 extend to 20KHz. Maybe I should have kept mine....Can't get there with 806, nope. These are 902s.... :yes:

Chas
02-26-2008, 06:23 AM
Can't get there with 806, nope. These are 902s.... :yes:

Ha! Okay, now it makes more sense!:D

Skywave-Rider
02-27-2008, 12:14 PM
Thanks a lot for your interpretations and for posting the data, Zilch.
I plan to go back and conduct the missing tests and re-evaluate the low freq performance of N-19Z. With and without helper.

I wonder when the BMS units will appear....

The helper woofer scheme makes the stock crossovers work much better.

Skywave-Rider
03-27-2008, 10:41 AM
I have completed missing RTA from last test, and made additional with BMS 4552ND on N-800-F and M-19-Z in 2.0 and 2.5 woofer modes.

Next I will do a listening shootout in mono between 902 HF on N-800-F and BMS on whatever x-over sounds best with that. Both 2.0 and 2.5 will be compared.

All of it will be posted together.

In the interim I have gotten fuzzy on the 2.0/2.5 modes which I was clear about last month.

Do I understand this correctly: In 2.5 with 20mH in line with helper woofer, the crossover sees 16 Ohms down to the crossover region, when lower frequencies engage the helper and the inductor essentially allows both woofers to play in parallel, 8 Ohms.

Looking at Zilch’s sims:
http://www.audiokarma.org/forums/attachment.php?attachmentid=82397&d=1205267758

Green and yellow are 8 Ohms. ERROR: PLEASE READ POSTS ABOVE!

In helper mode, the crossover region in the low pass appears to be significantly boosted by maybe 8dB in the upper bass, really lower mids, whereas when 1 woofer is operating (16 Ohm), you'd have an 8 dB contour cut there. A boost lower down would occur when both woofers kick in.

Do I have this right? ERROR: PLEASE READ POSTS ABOVE!

Zilch
03-27-2008, 12:46 PM
Green and yellow are 8 Ohms.

In helper mode, the crossover region in the low pass appears to be significantly boosted by maybe 8dB in the upper bass, really lower mids, whereas when 1 woofer is operating (16 Ohm), you'd have an 8 dB contour cut there. A boost lower down would occur when both woofers kick in.

Do I have this right?Nope, green and yellow are 16 Ohms:

http://audioheritage.org/vbulletin/showthread.php?p=197232#post197232

In "Helper" mode, the impedance only changes below 200 Hz. You choose a lowpass filter which provides the crossover and response for the single woofer and the compression driver largely independent of what happens when the helper kicks in.

Since your single woofer is 16 Ohms, in 2.5 mode, it's getting the green curve, i.e., boosted drive in the the crossover region. I think that's what you want, conceptually, since it's just one driver and not the acoustic summation of two drivers playing simultaneously in that region.

That would also appear to be the reason Altec used this "16-Ohm" lowpass with the 8-Ohm parallel pair, i.e., to provide LESS drive to them in this region, the cyan curve, and that's why I suggest that the N800-F may be a better lowpass for use with the N19Z highpass in either case, though it may be necessary to adjust component values to achieve the optimum crossover frequency and slope.

Recall that the N19Z lowpass is merely transcribed rote from Model 19 as a starting point, and provided primarily as an alternative for performance comparison.... :yes:

Skywave-Rider
03-27-2008, 03:06 PM
That makes sense now and correlates with the last listening test. No wonder I couldn't understand it. Thanks yet again for setting me straight Zilch. Please write a book....

I assume voltage drives vis a vis horn/cabinet and specific real world characteristics translate to a frequency response x amplitude graph. Is that why voltage drives are used in sims? Because the data is more straightforward?:o:

When the helper woofer is not engaged, that is, in a simplistic hypothetical wherein the main is playing in 16 Ohm mode alone, is the diaphragm excursion of the helper "controlled" or damped by the amplifier? It seems as though it would be. Otherwise the helper is operating passively and doing "something." I don't exactly know what. Without an acoustic phase reversal chamber, if there is such a thing for speaker cabs, it would have to detract from the output of the main.

Sorry, I'll look that up in Dickason.

I realize this may be a moot point since in the real world, especially when listening to music, there is usually program content below 200Hz. And therefore the helper voice coil would be energized and operating to some extent most of the time and within its unique range.

I've also realized that the impedance characteristics in this dual woofer system along with your 20mH helper scheme are fascinating!

Zilch
03-27-2008, 03:39 PM
I assume voltage drives vis a vis horn/cabinet and specific real world characteristics translate to a frequency response x amplitude graph. Is that why voltage drives are used in sims? Because the data is more straightforward?Yes, it's objectification of how the filter behaves electrically. How that transtlates through the other variables you mention to become the acoustic performance is indicated by what you are doing, the actual measurements. Bottom line, it's the acoustic crossover that matters, but understanding and controlling how the system gets there is essential to the design process.

These sims just show how the filter behaves with resistive loads, and thus merely provide a rough indication of what's going on with them. They can be refined more by using electrical models of the actual drivers, instead, and even better, that combined with the frequency response of each in the desired alignment, as is done with crossover optimization software such as LEAP, in which case the acoustic performance can be predicted quite accurately. Some of that may be done with your RTA, actually.

Regarding the "Helper" behaving as a passive radiator outside its passband, I agree that something is going on, but frankly, in detail, I don't know what. Considering the variety of designs incorporating this approach successfully, it's hard to conclude anything, since it's done both in common and separate volumes. Clearly, action as passive radiator in the conventional sense is not an issue, as the frequency is too high for resonance with the cabinet volume to occur.

Does an active driver give a whit about an open duct or port above resonance frequency? Does it care about a conventional passive radiator up there, either? That's also why, when the issue was raised about the cabinet tuning being different for one versus two drivers, I deemed it moot; at the frequencies of interest, both drivers are operative, and above that, the tuning does not affect the performance.

That's my analysis, and those are my assumptions. Others may know different and better, and hopefully, they'll step up to the plate and smack this around a bit here....


I've also realized that the impedance characteristics in this dual woofer system along with your 20mH helper scheme are fascinating!Another element not yet considered here is that when the impedance drops from the second driver kicking in, the amp will deliver more power, as well, providing more boost. That's less a factor with tubes, but I'll have to think through the implications.... :yes:

Skywave-Rider
04-23-2008, 04:02 PM
Listening report to follow shortly.
Please hold your fire.:)

Skywave-Rider
04-23-2008, 04:26 PM
I borrowed another ECM 8000 measurement mic and made a comparo as shown below. These measurements were not made in the same room I have been using for RTA; this is where I have done listening tests. (Note LF HVAC rumble, haha.) [RTA is 1 meter in this case, 4552ND/M19Z crossover for reference.]

Zilch
04-24-2008, 10:31 AM
You'll have to explain your legends.

I'm going to verify that those are actually Altec diaphragms in your 902s.

Holding fire.... ;)

Skywave-Rider
04-24-2008, 04:46 PM
Listening was done in the same room used previously and with 9844-8Bs placed as before, determined by the room mode calculator.

Program material was mono or stereo summed and panned to one monitor or the other. I made no stereo listening test.

In the previous comparison between N-800-F and M-19-Z, both with Altec 902s on the 811B horn, the N-800-F was preferable, though the M-19-Z had excellent UHF (see post #176.)

In that session I switched between conventional 2 way (2.0) and helper woofer (2.5) mode, and generally found 2.5 mode to be better balanced for most program material on either crossover. (This time I switched, but not as much, because 2.5 mostly sounds better, though at times it can provide some boom.)

Since I was able to get BMS 4552NDs in a group buy, I decided to pit those against the 902s, each on whichever crossover sounded best overall.

Prior to changeover, I balanced each crossover/driver combination with pink noise and RTA. I used no additional compensation for the BMS drivers (the FR curves indicate none required!) Since the BMS is hot, I placed an L-pad ahead of the driver to get into the proper adjustment range for the L-Pads on M19Z and the stock attenuator on N800. No 19kHz notch filter was employed for the BMS.

4552ND on N800F and M-19Z:

The 4552/N800 pair sounds smooth on vox/ mids.

The M-19 makes them forward and honk as was heard in the previous listening session with 902s.

The LF is clearer on the N800, same impression as I had previously. In 2.5 mode, this crossover will sometimes boom.

I did not notice boom on the M19 in 2.5 mode. The upper bass/lower mids however are not defined and that contributes to the lack of clarity in the low end.

The UHF range is preferable on the M19, which is what I found when using the 902 driver on this crossover as well. (By ear, this is 12-14kHz and up.)

HF is preferable on the N800, however.

Conclusion: 4552ND sounds best on N-800-F.

4552ND on N800F vs. 902 on N-800-F:

When switching back and forth between the two drivers, it’s pretty easy to hear the HF roller coaster or undulating, non linear response of the 902. Some notes are peaky - I presume from the hills you see in the FR graph.

The 902 will deliver more sibilance (4.5kHz-6.5kHz for males/5kHz-8kHz for females.) Sibilance is not necessarily bad. It’s some times realistic. But an overabundance is distracting. My judgment is that the 902 is not excessively sibilant; if the recording has sibilance problems, well; then watch out, it will be overbearing.

The 4552 is very smooth throughout its range. Esses are subdued in comparison. The BMS compliments the 414 mids, they blend, and produce a mellow characteristic.

The 902s are more dynamic, dramatic. I think this driver (and presumably its predecessors) does the greater share of providing a live performance reproduction, or as I hear others describe, the “Altec sound,” in the loudspeaker system.

In comparison, the BMS sounds as though it flattens transients. Perhaps “flatten” is a poor choice of words. Because it does not sound like clipping of course, maybe “compression of transients” is a more apt phrase. I think it would be interesting to try impulse tests on both of these drivers. Perhaps the 902 exaggerates transients; if that is possible.

Cymbal pings are often less prominent or in a few instances lost on the BMS. I don’t know if this is part of the “transient compression.” It may also be attributable to the 902 exaggerating the ping on an uphill of its undulating response.

Generally the 902 has more air. The 4552 is drier.

I had a female listener in attendance. Unprompted and without coaching, she tried to articulate some sort of issue in the BMS HF that was not present in the Altec driver. She stated she heard this on rides, tambourine, mostly metallic percussive sounds. As she is not a sound engineer, nor an audiophile, it was not easy to zero in on this without leading the witness. The best I can determine is she hears a sound in the HF associated with metallic percussion which stands out and is not normal or typical.

I did not hear this during the listening session. However, since I’m aware of the 19kHz spike in the FR of the BMS, I might suggest it has something to do with it. Not empirical, I know.

[As a sidebar, I have 4550s at home in an active setup. Prior to inserting Zich’s 19 kHz notch filter, I felt the presence of subtle “noise,” or irritation, gone after the filter was inserted. I am aware this could be a psychological expectation, but I don’t think so. BTW, I think that biamped system sounds great.]

Conclusions: At very low listening levels, I would prefer the 902s. Better transient response (jumping to a conclusion, there) makes them more sparkly above 6 kHz which in terms of equal loudness, plays better at low volume.

At mid to high levels, I would prefer the 4552s.

4552s on N-800-F win, though I won’t sell the 902s.

Crossover:

This is the hard part.

Perhaps I should do a recap of or rebuild the N-800-F. (See my initial query way down at #1, LOL.) If so, should I dig out the original Altec iron and wire wound?

It might be cool to add the UHF circuitry and adjustability of the M-19-Z to the N-800.

2.5 mode is a keeper. I don’t know if the bump at 250 can or should be dealt with. I would leave it switchable 2.0/2.5. It’s amazing how much more LF you hear, but it’s not as obvious on the FR graph as it sounds. It does cut lower mids/upper bass too, which is helpful.

The rise at 1k and the dip (with BMS) at approx. 2k: Big issue? How does it look? I tried to pad the “floor bounce” area a lot. But I don’t know. I’d love to hear from those with experience.

Somewhere I have the calculation for port tubes done. But I don’t know if that plays into the crossover LF mode issue.

Zilch, I’m very curious to see what you think of and measure from the 902s.

Skywave-Rider
04-24-2008, 05:20 PM
You'll have to explain your legends.

I'm going to verify that those are actually Altec diaphragms in your 902s.

Holding fire.... ;)

My apology for not being clear. I tried to edit this into yesterday's RTA post, but I can't. Is there a time limit on editing?:blink:

Anyway:

-"2.5" the 20mH is in line.
-"2.0" the 20mH is bypassed.
-"helper shorted/16 Ohm" means the helper is disconnected and shorted, the crossover sees one 16 ohm main.
-"main/helper shorted/8 Ohm/2.0" means the helper is disconnected and shorted, and a 16 Ohm dummy is inserted in parallel with main, no 20 mH.
-"main/helper shorted/8 Ohm/2.5" means the helper is disconnected and shorted, and a 16 Ohm dummy is inserted in parallel with main, plus 20mH in series with dummy helper.
-"helper/main shorted/8Ohm/2.0" means the main is disconnected and shorted and a 16 Ohm dummy is inserted in parallel with helper, no 20 mH.
-"helper/main shorted/8Ohm/2.5" means the main is disconnected and shorted, and a 16 Ohm dummy is inserted in parallel with main, plus 20mH in series with helper.Let me know if there are other inscrutables.

felixx
04-27-2008, 04:50 AM
Could you draw a schematic diagram with the final cross. network?Thx.

Skywave-Rider
04-27-2008, 07:58 AM
The M-19-Z is in post 118.
The N-800-F is here:
http://audioheritage.org/vbulletin/showpost.php?p=160631&postcount=484
I do not yet know what my final configuration will be. Do you have 9844s?
What is your setup?

Zilch
04-27-2008, 11:59 AM
I'm still working up an analysis of the latest measurements, but I'm popping in with an observation that we don't yet know for certain what the highpass filter is in Skywave's N800Fs, though I'm not certain we care, particularly, if it is not to be part of the final design.

In another forum, we have found two versions of N800F using significantly different values for the highpass capacitor. One provides some compensation for the excessive midrange of the HF driver, and it's indeterminate which Altec actually used in this 8-Ohm application....

felixx
04-27-2008, 01:40 PM
The M-19-Z is in post 118.
The N-800-F is here:
http://audioheritage.org/vbulletin/showpost.php?p=160631&postcount=484
I do not yet know what my final configuration will be. Do you have 9844s?
What is your setup?


I have only woofers 4x414-16B and 2x806A+811horns,also the N800-F crossovers.
I read your experiments on 9844 speakers.First I was thinking to make the same enclosures.
Also I'm thinking to use them into another setup configuration/enclosures.... ....more like K2 9500/5500 series design to give more impact on the bottom.I know ...probabily I loose some spl.Do you think your M-19-Z could fit to 806A drivers?
What do you think?
The 806A's could'nt touch more than 14-15khz max.;so I'm thinking to attach a supertweter like Fostex T90A.


Zilch
I can look inside on my N800-F,if you are interested.

Skywave-Rider
04-27-2008, 03:50 PM
Do you think your M-19-Z could fit to 806A drivers?
What do you think?
The 806A's could'nt touch more than 14-15khz max.;so I'm thinking to attach a supertweter like Fostex T90A.


Zilch
I can look inside on my N800-F,if you are interested.

The Model 19 used 802s, which might have a similar top end limit compared to 806s.
http://audioheritage.org/vbulletin/showthread.php?p=208190&highlight=802#post208190
I'm sorry, a lot my documentation went into the ether by way of a recent computer disaster. Double check that. The crossover Zilch designed (M-19-Z) is derived from the Model 19.

I suppose you've read my personal folly in trying to integrate a tweeter into this system. I'm not saying you won't make it work, but many swap the HF driver for one having an extended frequency response, the BMS4552ND I'm using is one example of such a driver. The Altec 902, of course, is the stock driver which is the other one under evaluation here.

There's also this thread:
http://audioheritage.org/vbulletin/showthread.php?t=14690&page=58
which everyone should read for the shear drama of tweeter integration.


I'm still working up an analysis of the latest measurements, but I'm popping in with an observation that we don't yet know for certain what the highpass filter is in Skywave's N800Fs, though I'm not certain we care, particularly, if it is not to be part of the final design.

In another forum, we have found two versions of N800F using significantly different values for the highpass capacitor. One provides some compensation for the excessive midrange of the HF driver, and it's indeterminate which Altec actually used in this 8-Ohm application....

I somehow expected I'd have to dig into the glop. Steve O, over at AudioKarma, kindly detailed his procedure for me.:(
It would be interesting to find out exactly what's in there, though.

Zilch
04-27-2008, 06:18 PM
I have only woofers 4x414-16B and 2x806A+811horns,also the N800-F crossovers.
I read your experiments on 9844 speakers.First I was thinking to make the same enclosures.
Also I'm thinking to use them into another setup configuration/enclosures.... ....more like K2 9500/5500 series design to give more impact on the bottom.I know ...probabily I loose some spl.Do you think your M-19-Z could fit to 806A drivers?
What do you think?
The 806A's could'nt touch more than 14-15khz max.;so I'm thinking to attach a supertweter like Fostex T90A.Altec made several products stacking the 414s vertically. This thread tells you how to get augmented bass out of them by using one of the pair as a "helper." Skywave's possibly the first to try the approach with these woofers, and I believe he's telling us it works, and that he likes the result.

The M19 crossover, in whatever form, is configured for 8-Ohm HF drivers. You may find some clues as to how to adapt it for 16 Ohms on the Markwart site. Another option is to install 8-Ohm diaphragms in your 806As.

As you know, however, 806As aren't going to get much past 14 kHz no matter what, and integrating a tweeter with them on 811B horns is problematic. Frankly, I'd abandon the Altec HF drivers AND 811B horns and just do this with dual 414s in 2.5 mode and BMS 4555 on JBL PT-F95HF waveguides:

felixx
04-28-2008, 04:51 AM
An waveguide have less problems.
How about the material from what is maded?I think wood or concrete cement will be the best compromises.

Check this one:
http://www.partsexpress.com/pe/showdetl.cfm?&Partnumber=270-312

Or an oblate from 18Sound:
http://www.eighteensound.com/index.aspx?mainMenu=view_product&pid=177

Or:
http://www.flickr.com/photos/12312992@N07/page2/

Skywave-Rider
04-28-2008, 05:53 AM
I was thinking of trying the 1" version of the 18 Sound horn you linked to
http://www.eighteensound.com/index.aspx?mainMenu=view_product&pid=179
in a different project.
But I went with the JBL waveguide Zilch showed.
Economy and proven performance made my decision.
And it's great.
I wouldn't change the horns in the 9844s.
They're vintage, and I like the sound.

felixx
04-28-2008, 11:16 AM
I think will be nice to make a wood rouded or horizontal horn.
The mark of the horn counts for me.
Do you have good reports about using metal or cheap "plastic" horns?Some ringing problems or artificial-plastic "sound"?

Another question....
Do you think will be differences between an horizontal horn and a rounded one?

http://www.flickr.com/photos/12312992@N07/page1/
http://www.geocities.jp/arai401204/Horn/A480FL/A480FL.html

http://www.geocities.jp/arai401204/

Zilch
04-30-2008, 10:29 PM
1) N800F w/902

A) Clearly, 902s have full extension with the stock crossover, but +/- 2.5 dB undulating response throughout the high-frequency range is unacceptable. Your 902s will arrive here in a few days, and if that appears in my measuring them, I'll try to figure out where it's coming from.

B) The notch at 1K25 disappears in 2.5 mode, indicating it is the two woofers interfering with each other at that frequency for the measurement mic location.

C) The bass response is flatter in 2.5 mode, and there does not seem to be elevated response in the very low frequency range where the helper is operating, i.e., it's not making them "boomy."


2) N800F Nearfield

A) The influence of the load impedance upon the acoustic response produced by the filter is quite apparent. At 16 Ohms, it follows the Green curve. With two woofers playing and summing, the midrange would swamp the bass if they weren't playing on the Cyan contour.

B) With the 20 mH in series with the 16-Ohm dummy load, the single woofer plays the Green curve, as the inductor's high impedance at higher frequencies decouples the parallel load from the filter. Thus, in 2.5 mode, the main woofer plays the Green curve, what we want, presumably, since there is no summing of two woofers playing in the midrange, only the low frequencies below 200 Hz.

C) Individually, helper and main show similar performance; they are well matched.

D) The 20 mH rolls off the helper above 200 Hz nicely. It's effectively "Off" above 500 Hz.

E) Both woofers/2.5 shows more elevation of the VLF than theory would predict, even with mutual coupling. That does not appear in the full-range curves, though, so my guess is the port output is exaggerating nearfield response measurements. In any case, it's beautifully flat down there. I'm calling 2.5 a successful and worthwhile upgrade from the measurements.


3) 4552Nd on N800F

A) Well, that's the name of THAT tune, I would say -- so much for 902s, at least your pair in their present condition. 2.5 and BMS substantially improve the performance, even using the N800F crossover.

B) Some minor tweeks of both filters could likely smooth out the response in the crossover region between 800 Hz and 2 kHz. I'm betting that's the best 9844s ever sounded.


4) 4552 on M19Z

A) Same with M19Z. Looks like even though you don't have the mid attenuation quite low enough for maximally flat response, the contour is better than that generated by N800F.

B) The lowpass filter is not optimum, and needs tweeking. Run the M19Z highpass with the N800F lowpass as I suggested might be a better combination above.

C) M19Z offers the benefit of adjustable HF contour, but it's almost necessary to have measurement capability to adjust it optimally. Those without that should stick with N800F, probably, built new, once you verify the HP filter capacitor value.

D) Open one of your stock N800Fs, please, and determine the highpass capacitor value, so we'll know which design to recommend.


5) M19Z Nearfield

A) A repeat performance. Flatter than N800F, and without the peak at 1K25.

B) Compare the lowpass filter components of M19Z vs. N800F. Tweak M19Z for best response through the crossover region.


You're nearly done with this project, and should be very pleased with, and proud of, what you have achieved with this.

I know I certainly am.... :thmbsup:

djgaloot
05-02-2008, 07:55 AM
I am following this thread with great interest though I do not have the electronic and acoustic technical expertise of you guys. I hope that in the end someone (Zilch, Skywave) would write a short summary with conclusions. I have a pair of Malibus which have two 414s arranged vertically and a 806/811 horn with the N800 crossover. I believe they are very early 1960's vintage and nearly perfect condition. They sound pretty nice but I was planning on rebuilding the crossover or perhaps, down the road, going the electronic xover/biamp route to maximize performance. I also have the parts to build a 9844 clone as a center channel. From the work done so far it seems that a BMS 4552ND driver will offer more extension with the 811 even with a stock N800 and that the 2.5 mode will smooth out the bottom. I am still a bit unclear on exactly what the 2.5 mode is. Is it running the 2nd 414 as a passive radiator or crossing it over a bit different with an inductor?
Thanks for all the work!
Dave

Skywave-Rider
05-02-2008, 08:57 AM
Hi Dave,
It's nice to hear from you. I think two vertical Malibus flanking a 9844 center would make an amazing installation. In 2.5 mode you will have less phase cancellation problems in the midrange from having the 414s arrayed horizontally. That's a good application for a 9844 type box. (Don't take my word for it, though. LOL)

Zilch should have my 902s today for testing. You may want to hear what he has to say after he scopes them. I'm sure whatever he finds will be posted here.

I will do a "final" summary eventually. But you've probably read my most recent listening report.

2.5 or "helper" mode means one of the 2 paralleled woofers has a single pole filter inserted in series with it. In this case the filter is a big 20 mH inductor, which moves the low pass point down significantly compared to the "main" woofer. It's operating at approx. 200 Hz. At frequencies above 200 Hz, essentially one woofer, a 16 Ohm load is operating, accounting for the reduction in mid/lower mid frequencies.

At frequencies below 200 Hz, both woofers operate at a lower impedance (8 Ohms) and therefore deliver more output, work together, and couple acoustically. BTW, I learned all of this from Zilch, who suggested this concept. (JBL apparently uses 2.5 in the Everest monitor.) You can also read about it in Dickason.

The other factor which nudged me into buying the coil was the aforementioned horizontal layout problem. I have not yet done much stereo listening, so it's hard to comment on whether or not imaging tightens in helper mode.

Stay in touch. :)

djgaloot
05-02-2008, 05:08 PM
Skywave,
Thank you for the explanation of the 2.5 concept; I am still trying to wrap my mechanically-minded brain around it. Do you get some bass interaction (boost?) from the now somewhat passive 2nd woofer above 200hz or are you separating the two woofers with a baffle arrangement. Seems like things could get "confused" acoustically. I will continue to lurk and learn and go back and re-read all the work you have done. FWIW, I have a pair of Boleros that use a passive radiator in a horizontal arrangement that will be my rear surround speakers for now.
Dave

Skywave-Rider
05-04-2008, 08:51 AM
Dave,
Zilch has a pretty good explanation of the passive conundrum here in this post:
http://audioheritage.org/vbulletin/showpost.php?p=205661&postcount=185

My two cents: Most recorded program material includes frequencies below 200Hz. So the helper is mostly active and energized. Does it radiate passively at higher frequencies? I don't know, but in a multi way speaker where drivers share the same volume, would that also be the case? For example a mid driver exciting a woofer in the same box. I don't know but Zilch's explanation seems reasonable to me.

It's a good question and I bet a lot of people have thought about it like we are.
:)

Dave, do you know where I can find pics or Altec lit. referring to Malibus?

djgaloot
05-04-2008, 02:34 PM
I have pictures of mine and some info. I sent a PM
dave

Zilch
05-04-2008, 09:25 PM
Skywave's 902-8Bs:

Skywave-Rider
05-04-2008, 11:02 PM
Cool.
Zilch, is #1 the driver I have marked (with tape) as #1? That's the one in my RTAs.

Thanks, I need some time to look.

Skywave-Rider
05-05-2008, 08:22 AM
I just started reading the Clio Clinic thread. I may have questions later.
In referencing my RTAs, so far looks about the same. The top curves are raw driver/bottom M-19, correct?

Is that a BMS-like spike at 19k? (Breakup?)

In the last RTA 1/6th octave, the cyan, "-3M" means 3 meters back from the violet measurement? Is that 4meters? Or does it mean something else entirely?
:o:

Is it possible to do a "transient response" test of these?

Most importantly, are you regretting ever getting involved with this old junk?

PS: I now believe I am somewhat colorblind. LOL

Zilch
05-05-2008, 10:19 AM
Yes, #1 is the one you marked as such.

Yes, top curves are the driver raw response with only a 47 uF protection capacitor in series. Bottom is each running on M19 XO.

"3M" is three meters. First thought was, as we have often seen others assert, that some distance is required for the horn to "integrate," so I pulled back about 10 feet and did RTA. That's not it, apparently.

Yes, I can do step response, but I'll have to read up on it.

I've attached two pics of the diaphragm in #2 below. The numbers stamped on there are:

35480
31060
979 8328

Perhaps someone can identify this.

http://audioheritage.org/vbulletin/showthread.php?p=14606&highlight=35480#post14606

There are indications on other forums that these are factory "Light" diaphragms for 902s. PM Dgwojo for confirmation.

I compared the response of both drivers on 811 vs 511 horns, bottom. 511 plays lower, and is a little smoother; it also rings considerably more than 811.

Using driver #1, I can render it listenable with M19 crossover, bottom curve. As I observed in the Valencia studies, it has a pleasant, "throaty," sonic character. The driver is altering the timbre across the high-frequency spectrum according to the response curves. Yes, I can hear that. It's wrong, and unnatural, but again, not unpleasant sounding.

Similar undulations appear in the CBS Labs M19 measurements from 30 years ago, and in the measurements I did on Gary's 802-8Bs last year:

http://audioheritage.org/vbulletin/attachment.php?attachmentid=24142&stc=1&d=1175669096

I'm suspecting the "TANGERINE Radial Phasing System" is the culprit here. Has no one else ever measured these Altec drivers and horns? Where are the curves? :dont-know

Zilch
05-05-2008, 01:33 PM
Most importantly, are you regretting ever getting involved with this old junk?There's certainly no dearth of wank and blather about it. ;)

Earl suggested in an earlier thread that the pattern may be sufficiently consistent that a series of passive notch filters would be a reasonable approach. I applied AutoEQ to test that, below.

The difference is clearly audible. Flat is a bit strident, room corrected is better, but now I hear the horn ringing.

Can it be fixed? I'd say, "probably."

[Not my job, tho.... :p ]

Skywave-Rider
05-06-2008, 07:16 AM
Can it be fixed? I'd say, "probably."

[Not my job, tho.... :p ]

Damnation!

geaugafletcher
05-06-2008, 08:54 AM
A series of passive notch filters?

Ugh, that's where I draw the line.

Zilch
05-06-2008, 10:36 AM
A series of passive notch filters?

Ugh, that's where I draw the line.JBL routinely does it, including in TOTL products.

However, with respect to vintage Altec, a driver swap is more, well, "expeditious...." :yes:

felixx
05-06-2008, 12:42 PM
I opened today my 806A-16 ohms drivers and I find this diaphragm series:
391046.
Do you know something about this diaphragms?I can't find any info about that.
Thanks for tips to all.
I measure with an ohmeter and the value is something close to 10ohms.

Earl K
05-06-2008, 02:08 PM
I opened today my 806A-16 ohms drivers and I find this diaphragm series:
391046.
Do you know something about this diaphragms?I can't find any info about that.
Thanks for tips to all.
I measure with an ohmeter and the value is something close to 10ohms.

- 391046 is not the diaphragm part number . What you have is most likely a 20275 diaphragm ( which is 16 ohm ) .
- The 8 ohm counterpart ( to the 16 ohm 20275 ) was the 21597 ( found in the 806-8A & the 802-8D driver ) .
- What numbers do you see written on the inside of the voice-coil ?

- You can learn a lot about Altec part numbers by familiarizing yourself with the wealth of info found here ; http://www.lansingheritage.org/html/altec/reference/_derived/1975-driver.htm_cmp_lansing-heritage010_bnr.gif (http://www.lansingheritage.org/html/altec/reference/1975-driver.htm)

:)

ps; here's a pic of a 23744 diaphragm / note where the number is written .

Earl K
05-06-2008, 06:03 PM
http://audioheritage.org/vbulletin/attachment.php?attachmentid=24142&stc=1&d=1175669096
http://audioheritage.csdco.com/vbulletin/attachment.php?attachmentid=32409&stc=1&d=1210007988

I'm suspecting the "TANGERINE Radial Phasing System" is the culprit here. Has no one else ever measured these Altec drivers and horns? Where are the curves?

- I don't know if I would blame the "TANGERINE Radial Phasing System" as much as suspension irregularities & network incompatabilities .

- The two pics of yours which I have included, show notches in different spots / with the 3K notch being common to the two drivers / horn combos .

- Below is a Plane-Wave measurement of the 909 driver . It shows an inherent notch at 3K and one around 9K . ( I doubt we'll ever see the goods for the 902-8B driver ) .

- FWIW; Yesterday I took my 902s and put them on about 8 different horns. One thing became apparent. The horn ( & it's classification type ) can either exaccerbate or minimize the notches. The included 909 info also reflects this somewhat.
- I was able to make ( with real CD horn compensation ) a very respectable line/curve ( with only 1.5 db notches at 3 & 8K ) on an RCF 9041 CD-type horn ( which is much like a smaller 2370 ).

- I came to the conclusion that the network will play a significant role in minimizing or expanding the notches ( into holes ) as large as what's seen with the M19 network . IMO, that network needs to be redesigned for use with the 902/811 combo.

- Here's someone who has measured ( using Clio ) an 802-8G on a 1000hz Tractrix. One can see that the notches have been minimized.

:)

Zilch
05-06-2008, 11:12 PM
Might we not expect the resonance frequencies to differ depending upon the diaphragm type as well, Earl?

Gary's 802s have standard diaphragms, whereas Skywave's are apparently 902 "Lites." :dont-know

Skywave-Rider
05-07-2008, 08:46 AM
Relaying some diaphragm info from Dave over at AK:



"Hi,
Those are 35480 aluminums, one of my favorite phragms from Altec in 1980, they're stock in the 902-8 ohm drivers. The only driver I like better is the BMS 4552ND for better treble, nice, Dave."

Skywave-Rider
05-07-2008, 09:48 AM
- FWIW; Yesterday I took my 902s and put them on about 8 different horns. One thing became apparent. The horn ( & it's classification type ) can either exaccerbate or minimize the notches. The included 909 info also reflects this somewhat.

Thanks Earl, that's interesting. I wonder if Zilch has tried this on the PT waveguide. I should have, but did not think of it b4 sending the 902s out.



- I was able to make ( with real CD horn compensation ) a very respectable line/curve ( with only 1.5 db notches at 3 & 8K ) on an RCF 9041 CD-type horn ( which is much like a smaller 2370 ).

Using standard 6 dB/octave CD horn compensation?



- I came to the conclusion that the network will play a significant role in minimizing or expanding the notches ( into holes ) as large as what's seen with the M19 network . IMO, that network needs to be redesigned for use with the 902/811 combo.

:)

Could the incorporation of series notch filters into either the N800 or M19 constitute the redesign, or do you mean scrapping, entirely, the HF section of the x-over, making something totally new?

Zilch
05-07-2008, 11:24 AM
Thanks Earl, that's interesting. I wonder if Zilch has tried this on the PT waveguide. I should have, but did not think of it b4 sending the 902s out.I'm doing additional measurements here, and can add that to the list.

Also wondering whether there isn't more to the loading cap thing than everyone presumes. You might want to check that once they get back.

Altec engineering has suprised us more than a couple of times now.... :yes:

Earl K
05-07-2008, 04:27 PM
Guys, due to work commitments, I'm out of this discussion till late Thurs ( or Fri. ) .

:)

Zilch
05-09-2008, 09:53 PM
I wonder if Zilch has tried this on the PT waveguide.

1) 902-8B #1 on 1" PT-F95HF and MR931 Model 14 horn, and both dialed "flat" using M19 crossover.

2) On 811B full range with LE14A on axis and 15° off axis.

3) RTA version of same for comparison to Skywave's measurements.

4) On 511A full range with LE14A on axis and 15° off axis.

5) Can 1/3 octave (31-band) AutoEQ fix it?

geaugafletcher
05-10-2008, 09:10 AM
More gnarly-looking graphs...

Thanks for the measurements, Zilch!:)

Zilch
05-10-2008, 09:52 AM
More gnarly-looking graphs....Again, and I believe Skywave has already confirmed this, it doesn't sound bad, rather merely has an inherent coloration, no matter the program.

If the same is demonstrable with standard diaphragms as well, it's a component of "That Altec Sound" characteristic of these later drivers.... :yes:

Skywave-Rider
05-13-2008, 10:38 AM
Briefly had Selenium D220Ti drivers on the 811s, without proper compensation.

Now that the BMS 4552NDs are back in the box, I have to say that the 9844s are beautiful sounding. Gorgeous. I'm lucky to have found them.

Won't be able to open N-800-F this week. Have to figure out how to do the goo-softening procedure at my job -- where they live. Seems this is best done outside due to fumes produced.

Zilch
05-13-2008, 12:02 PM
When Jackgiff opened his, it just kinda crumbled away, mechanically. In the "Babies" thread.

There's likely a double-cap flatpack in there, and the values are written on it.

Earl K
05-13-2008, 01:53 PM
Hi Skywave-Rider




- FWIW; Yesterday I took my 902s and put them on about 8 different horns. One thing became apparent. The horn ( & it's classification type ) can either exaccerbate or minimize the notches. The included 909 info also reflects this somewhat.


Thanks Earl, that's interesting. I wonder if Zilch has tried this on the PT waveguide. I should have, but did not think of it b4 sending the 902s out.




- I was able to make ( with real CD horn compensation ) a very respectable line/curve ( with only 1.5 db notches at 3 & 8K ) on an RCF 9041 CD-type horn ( which is much like a smaller 2370 ).


Using standard 6 dB/octave CD horn compensation?

- Yes, except that the compensation was what I would call "Type 2" CD horn compensation .
- FWIW, All the networks for the older Altec stuff incoporating radial/sectoral horns use "Type 1 compensation " .
- Really, the only difference is where the 3 db down point ( F3 ) of the cap is placed .
- Type 1s' "F3" occurs lower, and as such, counts on adding its' 3 db "HF-boost" to the @ "3 db HF boost" one gets from an exponentail horn ( as its' HF pattern narrows ). Together, they achieve the final 6 db type slope above the F3, set point .



- I came to the conclusion that the network will play a significant role in minimizing or expanding the notches ( into holes ) as large as what's seen with the M19 network . IMO, that network needs to be redesigned for use with the 902/811 combo.




Could the incorporation of series notch filters into either the N800 or M19 constitute the redesign, or do you mean scrapping, entirely, the HF section of the x-over, making something totally new?

- Yes, the HF section of the M19 or N800 would be completely scrapped .
- Last weekend I kludged together a network that included 3 series notch filters for use with ( exponential type ) horns .
- Overall, it's midband attenuation is around 12 db / making it ( I think ) only useful for someone who is biamping or using it with a 16 ohm woofer/8 ohm horn combo .
- I'll post some details this coming weekend .

:)

Skywave-Rider
05-13-2008, 03:39 PM
- Yes, the HF section of the M19 or N800 would be completely scrapped .
- Last weekend I kludged together a network that included 3 series notch filters for use with ( exponential type ) horns .

:)

Hi Earl.
It'll take me a while to digest your info, thank you.

I believe the 811B is a bi-radial horn. Are you working with an exponential horn to see what the filter is doing without having to factor in bi-radial or CD horn rolloffs? That is, to more directly observe the filter's effect?

I look forward to your results. :)

Note: Zilch corrected me, the 811B is a radial horn.

Zilch
05-13-2008, 09:26 PM
I was corrected on that one earlier. The sectorals are not constant directivity, and not biradial, rather simple radials, an exponential profile rotated horizontally about a vertical axis.

Altec's implementation of constant directivity was MantaRay....

http://aa.peavey.com/downloads/pdf/qwp1.pdf

Eric Best
05-13-2008, 09:35 PM
Skywave, How did the seleniums sound on your 811B?

Skywave-Rider
05-14-2008, 07:25 AM
I was corrected on that one earlier. The sectorals are not constant directivity, and not biradial, rather simple radials, an exponential profile rotated horizontally about a vertical axis.

Altec's implementation of constant directivity was MantaRay....

http://aa.peavey.com/downloads/pdf/qwp1.pdf

Thank you Zilch. I seem to have forgotten the difference between radial and bi-radial; and I'll go back to the Peavey .pdf to straighten myself out.

Eric, I put the Seleniums on for the hell of it, as I bought them for a different project. Since I did not have the proper compensation, only what is in N800F, stock, the FR was not optimal. Imagine too much 8kHz and rolling off above that. Zippy, a little sibilant, but not extended. My impression is that it would be good if dialed in.

Somewhere, someone must have a Selenium DT220Ti compensation filter worked out for the 811B; seems like a popular application. Try this thread over at AK:
http://www.audiokarma.org/forums/showthread.php?t=147170&page=8

Let us know if you do it.

Zilch
05-26-2008, 03:48 PM
Returning the drivers to Skywave, now.

Zilch
05-26-2008, 03:58 PM
Last one:

geaugafletcher
05-26-2008, 10:25 PM
Zilch, forgive my ignorance, but what are we looking at in that last graph?

Skywave-Rider
05-27-2008, 07:56 PM
Zilch, forgive my ignorance, but what are we looking at in that last graph?

Re. Zilch's last graph:

I'll try my hand at this but I know I've got something wrong, part of what I see does not make sense.

Sine wave for a duration of 5.58 ms, through the 902/811 horn combination. The crossover is active cx3400 Behringer, with CD horn compensation engaged. However, what I do not understand, also from Zilch's legend, is a frequency of 179.10 Hz. Can't be so low, but I am needing reading glasses so maybe I cannot see it well enough. The duration and and frequency don't seem to jive.

My conclusion: I'm not much help, sorry. I'm sure Zilch will fix us up soon enough.

Aren't those waterfalls interesting. Those I've recently learned to read pretty well, and the horn "ring" is visible. Yes, the 511 has smoother frequency response, but the 811 comparatively less ring.

I hope Zilch will be able to post same tests on same horns but with BMS 4552ND drivers.

Awesome graphs!
:applaud:

Zilch
05-27-2008, 08:29 PM
I hope Zilch will be able to post same tests on same horns but with BMS 4552ND drivers.Sure, but I have to catch up on other stuff first. I don't have answers on Step Response yet; have to read up.

In the meantime, here are more Waterfalls, this time normalized. You can see it has flattened the response across the top. This gives a more fair comparison, since the 20 kHz peak of 811B no longer skews its picture. 511A still loses at the high end, but it's better down low.

On gated sinusoidal, 511A rings with sustain on virtually every note, 811B only in the region of 1.5 kHz, if I recall correctly. It'd be interesting to damp them selectively, comparing the results, to ascertain where the problem areas are, and develop a solution.

[Not my job.... ;) ]

Zilch
05-27-2008, 09:16 PM
Here, for comparison, is the $9.90 JBL PT waveguide driven by a cheap Selenium driver, "EconoWave":

grumpy
05-27-2008, 10:24 PM
Question:

511A and 811B plots are non-baffled? (not mounted to a board)
Just that difference would be mildly more interesting... that metal horns
ring if not mounted should surprise no one.

Zilch
05-27-2008, 10:40 PM
Question:

511A and 811B plots are non-baffled? (not mounted to a board)
Just that difference would be mildly more interesting... that metal horns
ring if not mounted should surprise no one.

Certainly true.

Then again, 511 is rarely "baffled" in common use, and 811 only marginally secure in Altec products such as Valencia et al.

The primary source may be the "lips."

People have sawn out the vanes, and reported no difference listening, but nobody's actually measured, to my knowledge. :blah:

I've also seen one with the lips removed. Looks like a bowtie horn. More :blah:

It's all part of "That Altec Sound." Yet again more :blah: .... :yes:

grumpy
05-27-2008, 10:45 PM
If/when I get around to mounting either (probably 811b first), I'll try to
remember to give this (before/after) a go. Thanks for the clarification.

Skywave-Rider
08-25-2008, 06:54 AM
Tim, I wrote a response to your PM, but your settings don't allow me to reply. So I'll copy it here and hope you check back:
--------------------------------------------------------------------
NIce to meet you, Tim.

You should be able to make a great 2.5 setup with your Altecs.

I'm running, if I remember correctly, a 20mH coil in series to the helper woofer.
http://www.partsexpress.com/pe/showdetl.cfm?Partnumber=266-960&vReviewShow=1&vReviewRand=5229110#reviews

As for horns, my 9844s are vintage, so I am leaving the 811s, which I like. You could go with 511s, which are bigger and slightly smoother. Or you could go modern -- you might want to try these (I'm using them in another project: {JBL 1" thread-on Progressive Transition, Part# 338800-001, 90° x 50° dispersion pattern, flange dimensions 12" W x 6.5" H, 6.25" D including driver, $9.90 each. Order direct from JBL Pro Parts with credit card:}
http://www.jblproservice.com/forms/parts_order1.html

If you are interested, the 414s would be suitable for use in the "Econowave" concept; which incorporates that waveguide with a cheap Selenium driver. Look here:
http://audiokarma.org/forums/showpost.php?p=1856214&postcount=228

If you want to use a traditional 1200Hz crossover point, like I am with my Altecs, you can't do Econowave, as it crosses higher.

In the 9844 thread you will find schematics for both the N800F stock crossover which I am currently using (with 2.5 added,) and Zilch's M-19Z, which works great but the low pass needs tuning in my 9844s.

Let me know what u intend to do.
:)




Hi Skywave-Rider,

I found your thread 9844: http://audioheritage.org/vbulletin/showthread.php?t=18913 . I am very interested in your method of using the 2nd 414 woofer for 200hz and below. Would you mind emailing me a schematics for this setup. I currently have 4 414 woofers but dont have horn or drivers yet, what would you recommend for me to use since you have lot of hands on experiment with this setup. Thanks much.

Tim

MrT
09-06-2008, 06:44 PM
Zilch & Skywave

You guys went all the way with this project. I read it all the way. Great job! :applaud: 806a indeed has its limitation. Simply put it, the high is missing.

I am very interested in buying a pair of BMS4552nd. Where can I find them and how much are they?

Thanks for all the works you've put into the thread.

mrT

Zilch
09-07-2008, 12:32 PM
US Speaker is the distributor for BMS.

http://www.usspeaker.com/homepage.htm

They don't stock 4552nd, but will have them drop-shipped to you from the rep, Assistance Audio in SLC.

Tell 'em Zilch sent you.... ;)

MrT
09-07-2008, 01:19 PM
Hi Zilch

Thanks for the infos. Have you tried B&C de 250 or de400? How would they work?

On my side, I tried everything to boost up the high on those 806as. I give up. I thought about going with active but in the end taking that rout will cost more. So, buying a pair of new drivers will be the best way to go. So, here I am. It really had fun going through this thread. It is so amazing to see how this thread is keep on moving.

Timp

Zilch
09-07-2008, 01:47 PM
I have not tried B&C.

Note that BMS 4550 sounds about the same as 4552nd, if you prefer to spend a little less. :yes:

[Not as cute, tho.... :p ]

MrT
09-07-2008, 03:20 PM
Hi Zilch

I am all for saving money. What is the price difference? 10, 30, 100? If the difference is small, I will get the cuter ones. :D

Thanks.

Tim

Zilch
09-07-2008, 08:26 PM
~$50 each.

MrT
09-13-2008, 08:19 PM
Zilch

I called them. And I said, Zilch directed me to come to you. The guy said, who? :blink: Anyways, to cut the long story short, the sales guy was highly recommended me to buy something else. So, I didn't. Here I am being a little confused. I was locked onto buying 4550s.

Oh well.

MrT

Zilch
09-13-2008, 08:23 PM
4550s are fine.... :yes:

MrT
09-13-2008, 08:31 PM
are you here already? well. the guy said that they may be too fatiguing. and told me to buy something that are fuller.

anyways.

Thanks