Large format HF drivers [Archive] - Lansing Heritage Forums

doodlebug

10-12-2008, 07:12 AM

But I am still in doubt, because some people here on LH say that you need to compensate the falling HF response of the compression drivers in the x-overs. They do that in S3100 I know, but on the JBL 4673 the x-over doesn't have any kind of compensation. It is just a straight 12db/oktav filter, as I see it.

Is that because the 2445 2" drivers do not roll of in HF like the smaller 2426 1" drivers ??? Are they more linear in their response?

Vernb

If you take a look at the 3160 crossover - attached - used in the 4670 series theater systems, you'll find a compensation switch, S1, that does this to compensate for the rolloff. It works quite well in my setup although the physical size of the crossover is a bit overkill for a residential system.

To play around with the whole thing, though, I'm seriously considering just going with an electronic crossover as others have suggested. This gives me the freedom to dink around with crossover points and levels that hardwiring a dedicated crossover will not.

If you think about the cost of replicating something like a 3160 you'll find you will have as much or more in a dedicated crossover than an electronic one, assuming you have the extra amps around already.

Cheers,

David

vernb

10-12-2008, 10:27 AM

If you take a look at the 3160 crossover - attached - used in the 4670 series theater systems, you'll find a compensation switch, S1, that does this to compensate for the rolloff. It works quite well in my setup although the physical size of the crossover is a bit overkill for a residential system.

To play around with the whole thing, though, I'm seriously considering just going with an electronic crossover as others have suggested.

Cheers,

David

Hi David
Thanks for pointing out the idea with the S1. I thought it as basically just a volume pot for the 2445 driver (the complete HF range).
I hope that is the same thing with the 3115 x-over? The reason I'm asking is that I have exactly the 2225 and 2445 drivers (JBL 4673 components), so I would have an impedance problem with the 3160 x-over, I think....

If you went active, you wouldn't have the S1 as a HF compensator. Wouldn't that be a problem?
Vernb

doodlebug

10-12-2008, 11:24 AM

Hi David
Thanks for pointing out the idea with the S1. I thought it as basically just a volume pot for the 2445 driver (the complete HF range).
I hope that is the same thing with the 3115 x-over? The reason I'm asking is that I have exactly the 2225 and 2445 drivers (JBL 4673 components), so I would have an impedance problem with the 3160 x-over, I think....

If you went active, you wouldn't have the S1 as a HF compensator. Wouldn't that be a problem?
Vernb

I don't think the 3115 xover has a similar boost circuit built in. I believe it has a level control which, in the 316- is done as a series of jumpers.

If - actually when - I go to an electronic xover there will be no need for the passive xover at all. I'll put it on the shelf for later.

Note that the 3160 xover has a couple of other issues such as the 4 ohm woofer requirement and the 16 ohm horn requirement. If your drivers work then great but not all have that setup. They come up sometimes on auction sites.

BTW, S1 is a reference to the switch in the 3160 schematic where you can switch in and out the boost circuit into the horn tweeter's path.

Cheers,

David

Zilch

10-12-2008, 01:42 PM

What crossover is used in 4673 I linked above? :blink:

3115A has a three-position HF Boost switch and compensation circuitry:

http://www.jblproservice.com/pdf/Network%20Schematics/3115A%20Network.pdf

Yah, the schematic in the system tech sheet is the older 3115 without that good stuff:

http://www.jblproservice.com/pdf/Network%20Schematics/3115%20Network.pdf

Many active crossovers have a CD Comp button to do similar.

If you want to build 3115As yourself (with no tapped inductor), there may be an equivalent circuit posted in the forum somewhere.

Link it here for the benefit of others if you find it, please.

Otherwise, I presume y'all know who to ask to kindly post one.... :yes:

Zilch

10-13-2008, 12:54 AM

If you want to build 3115As yourself (with no tapped inductor), there may be an equivalent circuit posted in the forum somewhere.O.K., I put some effort into searching, and the only thing I found re: 31xx crossovers was this equivalent for 3110:

http://audioheritage.org/vbulletin/showpost.php?p=52084&postcount=3

There are some clues there as to where to start, but he's subsequently also observed that a simple inductor swap (even if we know the DCR) plus L-pad does not produce a true autotransformer equivalent.

Once a true equivalent IS generated, however, the same boost circuit might be usable with it, or an improved version developed.

This "generic" approach is not generally favored, but a true LX5 (and LX5 plus HF boost) equivalent that anyone can build might be seen as having merit and deserving the requisite effort.

If not, the options are clear: go active or find 3115As and recondition them.... :yes:

Footnote: I've never done much with large-format 2" drivers, but on the one occasion I ran 2450s on 2380A with 3110A (800 Hz) using HF boost, it sounded surprisingly good. Sorry, I did not measure the response, alas.

SMKSoundPro

10-13-2008, 08:56 AM

Boy howdy!

I am sitting here watching the snow fall, and the small juvenile moose in our back yard paw at the ground for some chikweed...

thinking about exactly this!

I still think about building some 4507 boxes for the small dance floor using a 2225 or 2226 and a 2425/2370 combo. I know that this little two-way kit gets a lot of press here, and am really ready to start cutting plywood.

I am lacking the 15's at the moment, though. I was thinking of 6 or 8 boxes to start. Fill them with components as I can. Use a two way active crossover with subs underneath.

I just need to learn more.

Scotty.

4313B

10-13-2008, 09:14 AM

I am sitting here watching the snow fall, and the small juvenile moose in our back yard paw at the ground for some chikweed...It's a damn good thing Palin is out of the area at present or you'd be grinding up moose burgers about now... nice photos!

SMKSoundPro

10-13-2008, 09:18 AM

Moose hotdogs!!! YUM!

Zilch

10-13-2008, 09:49 AM

It's a damn good thing Palin is out of the area at present or you'd be grinding up moose burgers about now... nice photos!No thoughts on an LX5 (+HF boost) equivalent circuit? :dont-know

SMKSoundPro

10-13-2008, 10:01 AM

I am thinking of 4671

2225 with 2425/2370 and I see 3110 xover I donot have.

I always have 5232 with 800 cards.

Please advise me otherwise.

link: http://www.jblproservice.com/pdf/Network%20Schematics/3110%20Network.pdf

doodlebug

10-13-2008, 11:11 AM

O.K., I put some effort into searching, and the only thing I found re: 31xx crossovers was this equivalent for 3110:

http://audioheritage.org/vbulletin/showpost.php?p=52084&postcount=3

There are some clues there as to where to start, but he's subsequently also observed that a simple inductor swap (even if we know the DCR) plus L-pad does not produce a true autotransformer equivalent.

Once a true equivalent IS generated, however, the same boost circuit might be usable with it, or an improved version developed.

This "generic" approach is not generally favored, but a true LX5 (and LX5 plus HF boost) equivalent that anyone can build might be seen as having merit and deserving the requisite effort.

If not, the options are clear: go active or find 3115As and recondition them.... :yes:

Footnote: I've never done much with large-format 2" drivers, but on the one occasion I ran 2450s on 2380A with 3110A (800 Hz) using HF boost, it sounded surprisingly good. Sorry, I did not measure the response, alas.

Thanks to our all-seaching Zilch, I see this is an interesting bit of detail re: CD horn compensation. Actually, all this got me motivated to pick up a Behringer CX3400 active xover yesterday to play around with. I had a choice between the dbx or the Behringer and chose the not-well-regarded-due-to-infantile-failures Behringer mainly for the added features such as CD compensation and adjustable delay on the low freq. outputs, which the dbx did not have at the similar price point. I'll just run it continuously for the next few weeks to make sure it doesn't die on me.

Am playing with it now along with the 4507/2226H sub boxes underneath the Quad ESL63s but will swap in the 2380/2445J 16 ohm horns when I get the Mac MC-225 tubed up again in their stead. Zilch, I agree, these do sound surprisingly good with the 3160s. My original searches at LH seemed to indicate that I shouldn't expect much.

BTW, first thing I did was take the cover off the Behringer to inspect for possibilities of improvement. Yep, lots of extra space for a replacement power supply, which consists of your basic 7815/7915 +/- voltage regulators and what appears to be +/-21vDC voltage rails. Most everythng is surface mount on the board that houses the adjustment pots and control switches. I wonder what's involved to lay hands on the service manual - if there is one.

Should be fun......now, let's see...what's involved in setting up a reasonable calibration setup with a laptop/Presonus Firepod audio interface and some calibration software? I'll likely use an AKG 1000 or Rode mic.

Cheers,

David

4313B

10-13-2008, 11:23 AM

No thoughts on an LX5 (+HF boost) equivalent circuit? :dont-knowI don't have time to get into that kind of thing anymore. Honestly? You'd do better to just come up with a new design specific to your driver set.

Zilch

10-13-2008, 11:41 AM

Please advise me otherwise.It's the same drill. 3110A with HF boost.

http://www.jblproservice.com/pdf/Theatre%20Series/4671.pdf

If you look at the N4671 schematic, it's 3110A with max boost hardwired.

2370A is alledged not to require HF boost, but it's an underperformer (for Hi-fi, at least) without some compensation. Whether that's true for dance, I don't know, but JBL clearly added it for theater. You can easily build the 3110 equivalent I linked above and give them a try. Another option is to add a UHF "cluster" above the dance floor to provide "sizzle" if it's missing. I'd build that with the $9.90 PT waveguides (4) and use 2407, 2408, or the BMS equivalent nd tweeters. See JBL's Dance series design:

http://www.jblpro.com/pub/cspels/CBT4%205-5-03.pdf

Footnote: JBL used a similar hardwired 3110A network in 4660, initially with an 8-Ohm HF driver, but ultimately went back to 16-Ohm "J". I've posted pics in an earlier thread showing what happened. The note says "Please use D16R2425...." :p

http://www.jblproservice.com/pdf/Theatre%20Series/4660.pdf

http://www.jblproservice.com/pdf/Theatre%20Series/4660A.pdf

Footnote 2: A member recently worked out a compensated network for 2370A, detailed in another thread here. I've never built it, and don't know if it's appropriate for your application, but, again, as a matter of principle, you'll want to rebuild any vintage crossovers you might use, particularly in a professional installation, so building new may be the preferred option. :yes:

I always have 5232 with 800 cards.For HF comp, you'll want to use the FFBREQ cards, maybe. :dont-know

I don't have time to get into that kind of thing anymore. Honestly? You'd do better to just come up with a new design specific to your driver set.Thank you, 4313B. I believe that's where this is headed....

doodlebug

10-13-2008, 01:46 PM

Thanks, folks. This is all I needed to get started on a new series of projects.

Cheers,

David

SMKSoundPro

10-13-2008, 02:10 PM

Thanks Mr. Zilch.

I have been searching the web and forum for more info.

A 3110 on the bay for $124 BIN, but not the A model.

Also looked for JBL M553 active xover. Not too many leads there.

I have considered a 4507 cab, filled with 2225/2226 low, 2425/2370 combo and 2405 uhf. I thought 3-way active with a m553 or dbx 234.

Call it good enough with a couple of double 18 subs underneath.

Does this get it done for the little dance floor?

I am visuallizing my L200b's with a slot concept mounted to the walls around the place. Just a good three-way full range job box that would take care of just about everything the dj's could throw at it.

Thank you!
Scotty.

Zilch

10-13-2008, 03:45 PM

I have considered a 4507 cab, filled with 2225/2226 low, 2425/2370 combo and 2405 uhf. I thought 3-way active with a m553 or dbx 234.Gotta mod the M553 to run 2405 on it three-way. The XO frequency is too low, as I recall. Oddly, it'll work 4-way, but you need a pair of them, then. :banghead:

[Check the specs to verify.... :yes: ]

robertbartsch

10-14-2008, 02:31 AM

Yikes, I think my thread has been high-jacked!

Anyway, I'm in Malaysia on business for a couple of weeks and it is good to see some discussion here on big HF drivers for home use.

So is a passive crossover at 500hz too low for a set up consisting of
2225 and 2445 drivers?

While dining on local curried fresh fish guts and pickled minced sprimp heads, I got to thinking about bi-amping this rig. I purchased a Denon 2200 and may use it with an older Denon 3300 reciever I have lying around that has separate pre-amp "outs," power amp "ins."

...any opinions on this idea?

JBLRaiser

10-14-2008, 05:26 AM

It's a damn good thing Palin is out of the area at present or you'd be grinding up moose burgers about now... nice photos!

the Elephants and Asses are mucking up the place.

Robh3606

10-14-2008, 07:18 AM

No thoughts on an LX5 (+HF boost) equivalent circuit? :dont-know

Just drop in a low value cap with a low value inductor like the 4430/4435 or the L200B crossover. Use a resistor to set the level. You drop it in before the attenuation used to match the compression driver too the woofer level. Wayne adds a low value capacitor with the 2370 in systems he builds using that horn. Going to need measurements of the horn and compression driver.

http://manuals.harman.com/JBL/HOM/Technical%20Sheet/L200B%20ts.pdf

Rob:)

Zilch

10-14-2008, 10:47 AM

Yikes, I think my thread has been high-jacked!Consider it a "Sidebar." The principles are the same.

So is a passive crossover at 500hz too low for a set up consisting of 2225 and 2445 drivers?No, your chosen horn must support that low frequency, though. An advantage of large-format drivers is that they will play that low competently. Some here would advise that even 500 Hz is too high for any 15" woofer, and that you still need a midbass driver.

I got to thinking about bi-amping this rig ... any opinions on this idea?I believe we said that. You still have the HF comp issue to resolve for 2-way.

vernb

10-14-2008, 12:45 PM

Well, spoke to the shipping company and my JBL 2350 horns are on their way to me from Belgium as we speak. I have the 2445 drivers (seems they measure around 12 ohm????).
I have two 2225 to use (both 8 ohm), and a pair of 2205 (both 16 ohm) if that would work better.

I am not sure if Zilch meant that there is already CD comp. in the good old LX5 filters? Do I actually need it with the 2445 on my 2350 horns?
Why isn't there any CD comp. in the 4673 (3115 x-over)?

I really would just like to build a simple 2-way x-over at 500hz with CD comp. myself (no L-pads please), but need a schematic to help me through.

This is an interesting thread now. Thanks for all the inputs.

----------------

If for some reason I shouldn't build a two-way, then I would like to use 2205 LF, 2482 MF on 2350 horns, 2426 HF on 2345 horns. and ,make a three-way .(I have all the parts, but thought that would be even harder to make x-over for??? and it really doesn't seem like anyone else has done that combination of drivers. x-over would be 300hz and 5000hz

Vernb

grumpy

10-14-2008, 12:54 PM

Robertbarsch-
Was the intended use ever expressed? (could make a nice SR or party speaker).

... or how large a horn can be accommodated? Not defining the horn or use, kinda
makes speculation on crossover design, moot (moose-meat?).

Vernb-
2350 doesn't need CD eq... it isn't CD, although you might want to boost the treble
or live with it rolling off w/2445. sort of wasting that 2426 on the 2345 using it above 5KHz.. ???

Zilch

10-14-2008, 01:01 PM

I am not sure if Zilch meant that there is already CD comp. in the good old LX5 filters? Do I actually need it with the 2445 on my 2350 horns?

Why isn't there any CD comp. in the 4673 (3115 x-over)?If there is HF comp in LX5, somebody'll have to show me where.

I don't know 2350 from a hole in the ground, but I would guess it needs HF comp, yes.

4673 uses 3115A WITH comp. I thought I clarified that above.... :dont-know

SMKSoundPro

10-14-2008, 01:19 PM

I agree with Vernb.

Kind of a waste of a 2425 above 5k. Why not just go to a slot? (or 2404 baby butt cheek tweeter?)

I can see where there are many differing opinions here that all need to be discussed.

In my operation, I am looking for a multi-purpose speaker box that will take care of a small 30' x 30' dance floor that is jam packed every weekend. This would not be considered true hi-fi, just good old bump and sizzle with a wide variety of musical input devices...cd's, mp3's and such.

I have 2441 and 2445 drivers and 2380 horns available also; but thought that was just over-kill in a single 15" + horn + tweet box for this little space. Subs are already in the room for oomph.

Question: With this is mind, does the 4507 box have enough room to encase these three items, 2225+2445/2380+2402, and do the job well? Or is a larger enclosure needed?

Scotty.

ps. I never intended to hijack someone's thread, and if I did...I am very sorry!!!

grumpy

10-14-2008, 01:27 PM

2350 is a radial horn, somewhat self-equalizing on-axis... Applying generic constant-
directivity compensation would probably not be optimum. Compensating for the HF
limitations of the 2445 might be helpful if done mildly. When used as an element of a
2-way system, passive or active crossover and essentially no EQ, I preferred the 2446
driver... this was outside, mid-throw, party-time :bouncy:.

Zilch

10-14-2008, 01:30 PM

2380 will fit in 4507, but you've either got to move the woofer/port location down a bit or notch the bottom 2380 flange.

Should be able to find a layout that will fit 2402 in there, as well.... :yes:

******

The "A" series of 31xx crossovers has two switchable HF comp contours available:

http://www.jblpro.com/pages/pub/obsolete/3105-3115A.pdf

CRAP! They screwed up the pdf and it doesn't show the curves. :banghead:

They're posted elsewhere in these forums, though.... :yes:

http://audioheritage.org/vbulletin/showthread.php?t=3838&p=41315

http://audioheritage.org/vbulletin/showthread.php?t=3838&p=42419

grumpy

10-14-2008, 01:31 PM

scotty,

I'd think you'd want more 15" drivers... the 4507 w/ one driver doesn't really
keep up with the 244x/2380 in my experience... atop a 4560 is a different case, but then
low bass can be an issue.

(unless planning on getting the 4507's off the floor, putting the horn inside might
protect it, but won't get the sound up to where it's needed).

SMKSoundPro

10-14-2008, 01:43 PM

Thanks for the comeback.

The boxes are hung from the 9' ceiling where I have unistrut everywhere for lighting cables...etc.

I am still thinking of the 4507 with a 2425/2370 and a bullet. I have all of the parts, and would probably biamp this box at 800-1200hz, then use a 3106 between the 2425 and 2402. Or triamp with a dbx 234.

I believe this would give me the best bang for my buck that will fit the space available. I keep thinking of the 4333 monitor and how well that works. I just don't have 2311 or 2312 horns and lenses, but have 8 2425/2370 combos on my shelf.

grumpy

10-14-2008, 02:00 PM

think that'd be a more natural cab match, w/ or w/o the bullet. Lots 'o dj cabs were built with
essentially those parts (15" bottom, 2370+driver on top) and sounded pretty good in
that environment.

SMKSoundPro

10-14-2008, 02:41 PM

Thanks Grumpy!!!

That's all I needed!!

scotty.

vernb

10-14-2008, 02:47 PM

4673 uses 3115A WITH comp. I thought I clarified that above.... :dont-know

Yes, thats right, I was mistaken. So basically I could build 3115a. I might do that.

vernb

10-14-2008, 02:52 PM

2350 is a radial horn, somewhat self-equalizing on-axis... Applying generic constant-
directivity compensation would probably not be optimum. Compensating for the HF
limitations of the 2445 might be helpful if done mildly. :bouncy:.

Dumb ass me. I thought CD comp. meant Compression Driver (HF roll off) compensation, not constant directivity.

So with this in mind, and what I read from the 4430 literature, basically the compression drivers do not roll off, but the horns they are on do, or at least some of them.

Zilch

10-14-2008, 02:58 PM

Dumb ass me. I thought CD comp. meant Compression Driver (HF roll off) compensation, not constant directivity.

So with this in mind, and what I read from the 4430 literature, basically the compression drivers do not roll off, but the horns they are on do, or at least some of them.No, it's the compression drivers that roll off. Some horns like exponentials compensate for that by progressively beaming the HF energy. Constant directivity horns do not. Other types may be intermediate.... :yes:

Yes, thats right, I was mistaken. So basically I could build 3115a. I might do that.We seem to be going around and around here. The PROBLEM is that 3115A, like LX5, uses a tapped autotransformer, not available. That's why I did the "Equivalent" number, above....

grumpy

10-14-2008, 03:00 PM

Oh, they roll off... :) The CD horn response just looks more like the plane-wave
tube response. The radial horn family sort of gives the response a 'lift'.

(Z beat me to it :))

vernb

10-14-2008, 03:00 PM

To clear up some things here I will explain that I want to use my upcoming 2-ways in my home stereo for music in a room 13x15 foot.

The reason of my idea of using 2426 (on 2345 horns) as HF over the 2482 mids (in a possible 3-way), was that I don't have any 2405 or 2404, and they are a bit pricy, and I already have the 2426 that I haven't been able to sell.
And in my first DIY (see my avatar) they are being used all the way up to 20.000hz without problems (it is really 276ND but that is the same unit with a different magnet as I see them). So I thought they would do fine as HF over the 2482 that doesn't go any higher than 6000hz. The 2426 also have higher sensitivity than 2404, to match that of 2482

Vernb

grumpy

10-14-2008, 03:06 PM

Does sound like fun, and having parts makes it possible (vs. chatting about it).

Might have some weirdness in that the pattern on the 2350 will start to narrow up
as frequency rises, then you crossover to the 2345 and it pops wide(er) again.
Worth looking at patterns vs. frequency when deciding on crossover frequencies
and slopes.

Time alignment might be difficult to achieve without introducing other problems...

vernb

10-14-2008, 03:24 PM

We seem to be going around and around here. The PROBLEM is that 3115A, like LX5, uses a tapped autotransformer, not available. That's why I did the "Equivalent" number, above....

Sorry. I am not sure I understand what a tapped autotransformer is. Anyway I looked in your linked threads and think I understand a little of the problem, but I still can't find a 500hz crossover layout, with HF boost, but no tapped autotransformer and no L-pads.

Can I actually just use my JBL S3100 x-over and replace the values to lower the x-over point?
That filter sounds very good to me on my 1400PRO/276ND DIY and in the S3100 speakers too.
I don't know if 1" drivers roll of in the same way as 2" drivers do???

robertbartsch

10-14-2008, 03:32 PM

grumpy

10-14-2008, 03:55 PM

Vernb... I'd expect the 2350+2445+2225 to need a different crossover from the S3100.
I believe the 3100's horn -is- more in the Constant Directivity family (albeit asymmetrically).
I'd also expect your crossover could be made with much less complexity.

Autotransformer? an inductor wound with taps... the taps act act as regular transformer,
but with a common (not isolated) leg. They are custom designed and wound, and
therefore not readily available in particular inductances and tap values.

Robertbarch will need something at least similar to a 3115A or 3110A... or go active,
or completely custom.

Zilch

10-14-2008, 04:15 PM

http://audioheritage.org/vbulletin/attachment.php?attachmentid=11580&stc=1&d=1134693309

DID measure with 2446J, tho:

http://audioheritage.org/vbulletin/showthread.php?p=81644&#post81644

Robh3606

10-14-2008, 04:39 PM

The 2350 is an old school Radial. Very little compensation is needed as it uses the directivity to compensate on axis. With the extended response of a 2445 compared to a 2440 you may not have to do all that much beside the main crossover point.

http://www.lansingheritage.org/images/jbl/specs/pro-comp/radial-horns/page3.jpg

Rob:)

johnaec

10-14-2008, 05:28 PM

think that'd be a more natural cab match, w/ or w/o the bullet. Lots 'o dj cabs were built with
essentially those parts (15" bottom, 2370+driver on top) and sounded pretty good in
that environment.Here's a pair essentially ready to go: (no affiliation): http://cgi.ebay.com/JBL-SR4725-PAIR-15-1_W0QQitemZ260300515980QQihZ016QQcategoryZ47094QQs sPageNameZWDVWQQrdZ1QQcmdZViewItem

John

Ducatista47

10-14-2008, 07:16 PM

Autotransformer? an inductor wound with taps... the taps act act as regular transformer,
but with a common (not isolated) leg. They are custom designed and wound, and
therefore not readily available in particular inductances and tap values.

If a picture is worth a thousand words, here are two (I think) on a board in the 3145 (for the 4345) Thanks to Ian for the image, found here:http://audioheritage.org/vbulletin/showpost.php?p=142746&postcount=4 The 3145 is a very large, complicated piece. It feels more like an amp than a crossover in your hands.

These are c.1983 vintage parts, many from Mexico. Perhaps that explains the "TOLTEC" on them. ;)

4313B

10-14-2008, 07:25 PM

55030 Tapped Autotransformer
(http://www.audioheritage.org/vbulletin/showpost.php?p=42824)

Tapped Autotransformer (http://audioheritage.org/vbulletin/attachment.php?attachmentid=1396&stc=1&d=1070555776)

JBL used them as voltage drop devices in old crossover networks to match very high efficiency compression drivers to much lower efficiency dynamic drivers. Since they are reactive devices they present impedance problems unlike fixed resistors. The fixed resistor packs as used in the L250 and 250Ti are a superior solution. Note that the L250 also used a tapped autotransformer on the 044. I think that was the last time Greg used such a device in his crossover network designs.

robertbartsch

10-15-2008, 12:21 AM

I bought a pair of JBL SR4725 a few months go. I took the components out and mounted them in a furniture-grade cabinet.

This is a kick-butt system that I enjoy very much.

vernb

10-16-2008, 12:09 AM

I bought a pair of JBL SR4725 a few months go. I took the components out and mounted them in a furniture-grade cabinet.

This is a kick-butt system that I enjoy very much.

that's a great two way for the home. Did you do any mods to the x-over?
Vernb

vernb

10-16-2008, 09:48 AM

So did anyone ever find a schematic for a two-way x-over at 500hz with CD comp (HF boost), without tapped inductor and without L-pads?

I sure hope we do one fine day.....:barf:

Vernb

4313B

10-16-2008, 10:52 AM

So did anyone ever find a schematic for a two-way x-over at 500hz with CD comp (HF boost), without tapped inductor and without L-pads?Nope, I don't like 500 Hz crossover frequencies so I wouldn't waste time looking for one. Maybe someone else will.

robertbartsch

10-16-2008, 11:36 PM

vernb

10-16-2008, 11:55 PM

So, with the horn I'm planning to use with the 2425, the HF roll-off would be significant and would require a tweeter like a 075, 077 or 2402, etc?

The cabinet I'm using is on the narrow and short side, so adding this would not be easy, I suppose.

well even though the HF roll-off is significant, you can easily use the 2425 without a tweeter. Both the 4430 and S3100 speakers do this and sound great. I copied the x-over from JBL S3100 to use in my 2-ways and it works absolutely great without the need for a tweeter.
Both 4430 and S3100 have incorporated a HF boost in the x-over.
Vernb

jerv

10-17-2008, 02:29 AM

FWIW, this is a filter (no L-pad, no transformer) I made for the 2226H and the 2425J on a 2370 horn.

http://www.audioheritage.org/vbulletin/showpost.php?p=197944&postcount=18

toddalin

10-17-2008, 10:33 AM

I worked out a high frequency compensation circuit for the 2425J/2370 combination when I was doing my surrounds.

I started with the basic L200B circuit to match my LCR, but instead of using a separate circuit for a 2402 as my LCR do, I kept it a 2-way.

I felt that the HF compensation was definitely necessary and greatly improved the sound.

I worked it out through trial and error using parts I had at hand and the Behringer. I kept my resistor values consistent with the Mills offerings and caps consistent with the Solen offerings at PE.

I initially made up resistor values to simulate their offerings, then once I had decided on the values, ordered the Mills equivelents. I used caps I had around and once these values were decided, I ordered Solens with Theta by-pass caps.

I actually worked out the compensation for both the JBL 2370 and P-Audio "9-inch butt-cheek" replica. Wow, I never would have thought they would be so different to get a similar hign end response! Actually, the 2370 did a little better than the P-Audio horn for extension and smoothness. I would have used them except for the WAF of the horns hanging under the four surround speakers in the living room. The external horns also helped the bass response by virtue of the increased cabinet volume. Alas, aesthetics trumped fidelity.

These are being used with low efficiency, 12-ohm woofers (JBL W10GTI) and are really smooth. The crossover is ~800 Hz.

When I use the same crossover on my 2235s, I don't get the woofers to go all the way up to the horn. I attribute this to the fact that the 2235 was reconed at 8 ohms and the original L200B was designed for a 16 ohm woofer. If I put an 8-ohm resistor in series with the woofer, it extends the woofers extension so that it better meets the horn. (But doing so will get you flamed around here.)

I would think that using an 8-ohm woofer, you could use this as the 500 Hz woofer crossover. To extend the horn a little lower, I would up the size of the first horn cap from 16.5 mH to 20 mH. When I install the 2312 horns to replace the 2307s, I'll increase these caps to 18 mH as was done in some of the JBL offerings with the 2312.

If this is of interest, I'll try to draw up and post the circuit.

4313B

10-17-2008, 10:46 AM

If I put an 8-ohm resistor in series with the woofer, it extends the woofers extension so that it better meats the horn. (But doing so will get you flamed around here.)You're free to post whatever you feel the need to post so long as it remains within Don's guidelines. You can put all kinds of things in series with your woofers and post about the experiences as you see fit.

vernb

10-17-2008, 01:27 PM

FWIW, this is a filter (no L-pad, no transformer) I made for the 2226H and the 2425J on a 2370 horn.

http://www.audioheritage.org/vbulletin/showpost.php?p=197944&postcount=18

Thanks Jerv. Just what I needed
Vernb

toddalin

10-17-2008, 03:33 PM

After many hours of testing, looking at the Behringer, and listening, this is what I came up with. It works very well with a low efficiency woofer (W10GTI ~81 dB@1 watt), but I only run the L-pad at less than 1/4 turn of the way so would probably be fine with something on the order of 90+ dB. To bring the horn in a little lower (for your intended ~500 Hz duty), I would replace the two 8.2 mfd caps with a single 20 mfd.

http://www.largescaleonline.com/eimages/lsolpics/Team_Member_Pics/toddalin/2-way.jpg

robertbartsch

10-17-2008, 05:32 PM

speakerdave

10-17-2008, 06:01 PM

Is there a good reference book on reading electrical schematics and building simple circuts?

I've search the web and the stuff I'm finding is too advanced for my knowledge level (zero)?

Where do you buy electronic components, resistors, capicitors, boards, ect.?

If you really want to understand, take the course. A classic textbook is: Robert L. Boylestad, Introductory Circuit Analysis. Been around for thirty years at least, any edition will do. Old cheap used paperback copies can be found at Abebooks for $5 shipped. Current edition $108 at Amazon. You need to have a sense of an ohm, an amp and a volt going in. You might begin, though, with a textbook on technical math, and you should have a good calculator. I took half the course long ago (passive elements) and dug out my copy a while back and have been spending what time with it I could. It's difficult, but in the long run, I think this is the only way to go.

Parts Express, Madisound, Michael Percy Audio.

Ian Mackenzie

10-17-2008, 07:10 PM

You're free to post whatever you feel the need to post so long as it remains within Don's guidelines. You can put all kinds of things in series with your woofers and post about the experiences as you see fit.

I did that with an Altec 515 a long time ago to trial a diy system. I put an 8 ohm resister in series and it raised the Qt considerably and extended the apparant extension in a Bass reflex setup. I then changed out the 515's , sold them for a grand and bought a pair of 2245H's. The improvement was a no brainer!...LOL

Zilch

10-17-2008, 07:24 PM

When I use the same crossover on my 2235s, I don't get the woofers to go all the way up to the horn. I attribute this to the fact that the 2235 was reconed at 8 ohms and the original L200B was designed for a 16 ohm woofer.Nope, wasn't that:

http://manuals.harman.com/JBL/HOM/Technical%20Sheet/L200B%20ts.pdf

Zilch

10-17-2008, 07:34 PM

Is there a good reference book on reading electrical schematics and building simple circuts?

http://sound.westhost.com/beginners.htm

toddalin

10-17-2008, 07:41 PM

Nope, wasn't that:

http://manuals.harman.com/JBL/HOM/Technical%20Sheet/L200B%20ts.pdf

OK, but I still say that, at least in my case, a 16 ohm woofer would have worked better with it.

4313B

10-17-2008, 08:13 PM

So is a passive crossover at 500hz too low for a set up consisting of 2225 and 2445 drivers?You need to measure both drivers in the system you intend to build and then determine the correct crossover frequency.

Well. It is hard if you don't know what to search for. I have been looking for a long time for a two way system for the 2445 drivers - with out luck.You really need to build your networks based on your specific application. I know it's a real PITA but that's the way to do it right.

jerv

10-19-2008, 08:02 AM

Thanks Jerv. Just what I needed
Vernb

Hope you can use it.
If you decide to build it - please let us know what you think;)

vernb

10-19-2008, 10:45 AM

You need to measure both drivers in the system you intend to build and then determine the correct crossover frequency.

You really need to build your networks based on your specific application. I know it's a real PITA but that's the way to do it right.

But my specific application and drivers is/are the same as Jervs and the same as JBL 4673, and since I have no measuring equipment, I will build Jervs x-over and work my way from there. The reason for not building JBL 3115 x-overs from the 4673 is that they have tapped inductor, L-pads, and no CD comp (HF boost). Jervs doesn't.
That's what I did with my first project: Drivers and horns were close to JBL S3100, so I made a S3100 x-over and then fine tuned it to the 1400pro driver (the main difference to the S3100).
Vernb

vernb

10-19-2008, 10:50 AM

Only thing that now puzzles me is that most of the "so called" JBL 16 ohm compression drivers I have, actually do not measure 16 ohm. My 2445 measure around 12 ohm and my newly arrived 2482 measure around 10 ohm. Even my 2426J measure around 12 ohm.
I don't know much about building passive x-overs, but I thought that the impedance of the driver meant quite a lot to the x-over point. So that a 500hz x-over for 16 ohm drivers will actually not x-over at 500hz if the driver measures only 10 ohm. Am I right ? :blink:

Vernb

4313B

10-19-2008, 11:41 AM

But my specific application and drivers is/are the same as Jervs and the same as JBL 4673, and since I have no measuring equipment, I will build Jervs x-over and work my way from there. The reason for not building JBL 3115 x-overs from the 4673 is that they have tapped inductor, L-pads, and no CD comp (HF boost). Jervs doesn't.
That's what I did with my first project: Drivers and horns were close to JBL S3100, so I made a S3100 x-over and then fine tuned it to the 1400pro driver (the main difference to the S3100).
VernbOk.
Only thing that now puzzles me is that most of the "so called" JBL 16 ohm compression drivers I have, actually do not measure 16 ohm. My 2445 measure around 12 ohm and my newly arrived 2482 measure around 10 ohm. Even my 2426J measure around 12 ohm.
I don't know much about building passive x-overs, but I thought that the impedance of the driver meant quite a lot to the x-over point. So that a 500hz x-over for 16 ohm drivers will actually not x-over at 500hz if the driver measures only 10 ohm. Am I right ? :blink:

VernbRight. I addressed this long ago. JBL used 10 ohms for the 077/2405 and 12 ohms for many of the compression drivers for rudimentary modelling purposes.

You shouldn't just pick 500 Hz out of the air. You may want to exploit the natural roll-off of the horn/compression driver combination. It should be 12 dB/octave which means you'll only need a 12 dB/octave electrical filter to get the 24 dB/octave acoustic roll-off.

Honestly, if you're just tossing darts blindfolded then you might be better off picking up a cheap active filter like a dbx 223 and start dialing frequencies until it sounds good enough to you. Or pick up one of those dbx DriveRack PA's and play the whole day away. I think they're like $450. I don't think you can even pick up SoundEasy and a good eight channel sound card for much less than that.

Zilch

10-19-2008, 12:43 PM

But my specific application and drivers is/are the same as Jervs and the same as JBL 4673, and since I have no measuring equipment, I will build Jervs x-over and work my way from there.I'm probably getting lost in concurrent discussions of three different systems here, but Jerv's system uses a different XO frequency, HF driver, and horn than 4673. Ultimately, you've got to start someplace, and I've documented at least a half-dozen trips down this path in these forums.

I'm kinda with the member formerly known as Giskard on this stuff. It was fun the first couple of times, but, in the end, it's obvious nobody gives a flying whit.... :blink:

http://audioheritage.org/vbulletin/attachment.php?attachmentid=20655&stc=1&d=1163897565

http://audioheritage.org/vbulletin/showthread.php?t=12420

vernb

10-19-2008, 01:12 PM

I'm probably getting lost in concurrent discussions of three different systems here, but Jerv's system uses a different XO frequency, HF driver, and horn than 4673. Ultimately, you've got to start someplace, and I've documented at least a half-dozen trips down this path in these forums.

I'm kinda with the member formerly known as Giskard on this stuff. It was fun the first couple of times, but, in the end, it's obvious nobody gives a flying whit.... :blink:

http://audioheritage.org/vbulletin/showthread.php?t=12420

I do read your threads with great interest Zilch. As I read it the 4673 uses 2225 and 2445 in a 2-way setup. I also want to use 2225 and 2445 in a 2-way setup. I just want to use 2350 horns instead, which if I listen to all above, might not need quite as much HF boost as the Constant Directivity horns. I misread Jervs post 2425 for 2445 (too many numbers :banghead:).
The reason for choosing 500hz x-over frequency, is because it sounded very good on my JBL S8 system and because JBL chose to use it in the 4673 speaker.
I will keep searching, and not bother you any more, sorry
Vernb

4313B

10-19-2008, 01:37 PM

The reason for choosing 500hz x-over frequency, is because it sounded very good on my JBL S8 system and because JBL chose to use it in the 4673 speaker.And I'm saying that pretty much anything would sound better than those old solutions. JBL was very well known throughout the industry back in the days of the S8 for putting really nice transducers with marginally acceptable filters.
I will keep searching, and not bother you any more, sorryI don't think you're bothering anyone. We're just tossing out suggestions. I won't bother doing an equivalent circuit for the LX5 like Zilch asked because the network isn't worth the trouble. It just isn't.

vernb

10-19-2008, 01:51 PM

I don't think you're bothering anyone. We're just tossing out suggestions. I won't bother doing an equivalent circuit for the LX5 like Zilch asked because the network isn't worth the trouble. It just isn't.

So basically, now that you know my exact components: 2225 (8ohm), 2445 (16 ohm, but really 12) and the 2350 horns. What crossover frequency would you recommend? I tried 800hz in the S8 system as well, also good. But as said earlier, I would like a x-over layout that I can actually build myself, no L-pads and no tapped inductors. I like using passive filters.
This is only my second project, and that is why I am still very slow/inexperienced.

Thanks

Vernb

4313B

10-19-2008, 02:26 PM

What crossover frequency would you recommend?I don't know where your drivers want to cross over.

Here's an example:

2122H and 2307/2308/2421B wired in parallel with an 8 ohm L-Pad on the 2421B to adjust its output to match the output of the 2122H. Notice where the 1st notch is? That's where the crossover frequency should probably be. This was done in a huge hurry so I would check everything several more times before going with it. Furthermore I don't know if a "classic" filter would yield the results I want. I might want one pole on the 2122H set at something like 800 Hz and the second pole set at something like 950 Hz, whatever would yield the best summed response. That's why traditional active filters don't always give the best result either (As an example look at the voltage drive specification for the active filter used to bi-amp the Everest II). With a DSP you can do alot more.

I believe JBL set the acoustic crossover between the 2122H and 2307/2308/2421B at 1.3 kHz in the 4345. Different baffle arrangement than this example and they measured a little bit further away as well.

I like using passive filters.I understand that. They can be very nice.

vernb

10-19-2008, 02:45 PM

Since I already have build S3100 x-over (crossing over at 750hz), I might just try that for the fun of it and see how it sounds. Biggest problem is probably that the 2225 is less efficient than 1400PRO and 2445 higher efficient than 276nd that I use, but since the x-over is biampable I can compensate for this with the amps.
vernb

4313B

10-19-2008, 02:50 PM

Since I already have build S3100 x-over (crossing over at 750hz), I might just try that for the fun of it and see how it sounds.Ok, let us know what you think.

vernb

10-19-2008, 03:03 PM

Sorry 4313, I didn't see your post until after I had posted mine. I sort of see your point.
Vernb

Chas

10-19-2008, 03:19 PM

[quote=Zilch;224602
it's obvious nobody gives a flying whit.... :blink:
[/quote]

Yes they do!:yes:

Zilch

10-19-2008, 08:26 PM

http://www.partsexpress.com/pe/showdetl.cfm?&Partnumber=270-204

A link to the horn I'm planning to use is above. My cabinet width is a tad narrow so the 15 inch horn width fits OK.Uhmmm, says it's 17.7" wide. 2380A is a tad narrower at 17.5", actually, at least the one I just measured.

So, with the horn I'm planning to use with the 2425, the HF roll-off would be significant and would require a tweeter like a 075, 077 or 2402, etc?Uhmmm, 2425 or 2445? :banghead:

If you use the "correct" components, it'll likely be better than your repackaged SR4725s, even without supplemental UHF.

As I read it the 4673 uses 2225 and 2445 in a 2-way setup. I also want to use 2225 and 2445 in a 2-way setup. I just want to use 2350 horns instead, which if I listen to all above, might not need quite as much HF boost as the Constant Directivity horns.Correct, apparently.

I will keep searching, and not bother you any more, sorry.Nah, it's me (and others) doin' the bothering.

I would like a x-over layout that I can actually build myself, no L-pads and no tapped inductors.Not gonna happen, thus far, nope.