Plan B (aka, fun with virtual graph paper)

[grumpy]

May this thread go where it will...

2234 x2 in ~10ft3, ala 4435
2345HPL on 2352, expect Zilch has a near drop-on recipe buried here
somewhere :D xover somewhere between 700Hz and 1.2KHz...
(kidding about the drop-in, some real-good starts, perhaps).

Room for ST front & center or placed back in canyon on top (not shown)
if desired.

tune to suit (ports), maybe increase depth a smidge...

if you imagine the horn, viewing from above, you'll see an approximation of
the "rams horn" ... I'm assuming all of the hard looking corners would
be eased to 1/2" radius or more.

Attached pic breaks a few "rules". Oh well.

Might setting the "MF" 2234 outboard help with time alignment if you
set toe-in less than might be normal otherwise? (a lingering DD66000
question)

-grumpy

[Zilch]

2435HPL OR 2452H-SL, each takes a different filter.

Both require redrilling the horn throat flange for driver mounting.

Still gotta figure out how to deal with the horizontal offset alignment at crossing.

4435 had them (HF and primary LF) vertical in line.

MTM is "cumbersome," I like Everest better, if we can make it work....

[grumpy]

cut/pasted images of parts for better visualization.
Either compression driver suggested would fit.
2435's happen to be what I have to work with.
Trying to -not- sink the bass drivers back 6" or so, if I can avoid it.
I'm going to go with my hypothesis that the side offset can somewhat
compensate for a depth alignment issue, although I don't yet have to tools
to make observations to confirm or deny...

-grumpy

[Zilch]

I'm envisioning the compound tilt of the plane of time alignment.

Using toe-in and woofer delay, with the primary woofer stationed outboard, it's possible to get the planes to intersect at any point in the center 3-D space.... :thmbsup:

[I forget, now, if they said outboard or inboard for Everest II. I don't see inboard working, as the planes tilt away from the center....

[grumpy]

I forget, now, if they said outboard or inboard for Everest II. I don't see inboard working, as the planes tilt away from the center....

I don't recall that being stated either... just that you could change it with jumpers.


I'm envisioning the compound tilt of the plane of time alignment.

Thar' be a lotta long words in there missy. We be naught but humble (design) pirates.
:rotfl: :rotfl: :rotfl:

but yeah... that's what I was getting at... seriously.

-grumpy

[Mr. Widget]

Why is the cabinet so shallow? I ask because in my experience deeper cabinets tend to have far less of a box sound.


Widget

[grumpy]

Fair question, reasonable concern (one I share)... also one of the broken "rules".

"Room intrusion" is at least a minor factor for me (depth of other furniture, or TV, or ...).

Width-min set by making the bass drivers adjacent and horizontal, ala 4435.
Height-min is set by adding a BF horn (that happens to meet my mental requirements
of not throwing away potential OA performance, attempting to minimize horn
depth, and shooting for a sub-KHz xover). I don't see a COTS horn that does this
and I don't have the resource to make one.
I don't really want to move the bass drivers closer to the floor, and the horn center
I'd like to have close to seated ear level.
I don't think I need more than 10ft3, hence the narrowish depth... BTW, add two more
inches & you've matched the DD66000 depth :) . Could push and pull here & there
a bit in the mock design without me popping a fuse :D (E.g., put a "butt" on it and
make the top modular)...

-grumpy

[Zilch]

Well, they got it wrong, is all.... :p

[Mr. Widget]

... BTW, add two more
inches & you've matched the DD66000 depth :) .JBL has, in my opinion, built their factory cabinets too shallow on the majority of their designs... their need to sell product can get in the way of performance. In the case of the DD66000... the curved rear panel is likely a big help in this regard... something you don't have.

You may be able to get away with the shallow cabinet if you use 3" Sonex... but I'd be inclined to avoid a design that is potentially compromised from the start.


Widget

[grumpy]

OK. Throw out that hypothesis... :) next! -grumpy

Widget, thanks for the feedback.

re cab depth... my first cut at this was deeper and taller... horn placement is more fixed
than is my determination to have a shallow design, but those woofers have to go somewhere,
need a certain amount of space, and perhaps have some benefit in proximity. Perhaps the
component choice is flawed (pushing compromised design choices), but I'll think on it.
I don't have a requirement for them to fit into an existing soffit location. I also don't particularly
have a need to show visual heritage, although it's a nice idea.

I do like the idea of continuing the general shape of the horn beyond it's boundaries... and that
takes some visual real-estate (perhaps a stick in the eye for some). One could consider a more
oval or ellipsoid shaped profile (looking down), in exchange for construction complexity.

Would be quite happy to see alternate "Plan B" considerations, whatever they may be.

-grumpy

[Zilch]

OK. Throw out that hypothesis... :) next! -grumpyNaw. Zilch will watch for the errata adendum.... ;)

[grumpy]

OK... made me look. posted pics are 2" too thin from intent... not that it gets you into golden
rectangle proportions :) s/b ~3' x 4' x 1.33' ...still agree that panel resonances could be a
tough nut and intentional resonances (ported box) could be less than ideal.

FWIW, MTM -looking- version ends up being ~5ft tall with depth
depending on where you decide to put it: ~20" if a plain big box, assuming you go
for a "thin" front profile (22"). -grumpy

[Ian Mackenzie]

There are some fairly well defined rules for box building in the Vance Dickason publication "The Loudspeaker Design Cookbook".

[grumpy]

Hi Ian,

Indeed. Thanks and good point. I've made nice inert boxes... they are no longer a
challenge and they look like, well... boxes. ;) Ran another design concept last night
for fun... looked like a mini-me dual El-Pipe-O with the horn between the verticals...
I and thought my posted pic at the start of this thread was aesthetically challenged...
:rotfl: -grumpy

[Ian Mackenzie]

Yes it was and it caught the attention of those so inclined.

Do you think it would be safe to let Me loose on your project.

Personally I think he would run rings around until we all got dizzy and it would be an embarrassment.

[grumpy]

Me's welcome :-) No judo chops, please.

My project? Hmm.. well, my components will certainly end up in a project and will
hopefully benefit from discussion here. It appeared that several others had a similar
component set and some interest in putting them together. Horns should arrive in
the next few days :bouncy:

-grumpy

[grumpy]

2352 horns came in. Just had to listen to those 2435's ... doing so now :D
Setting on 4430 bottom end with the HF peaking circuit dialed way back.
After a few recordings I'm pretty pleased with the potential of this pairing,
and having a hard time imagining (personally) needing more high-end
frequency extension. Will try to run a WT2 impedance plot tomorrow.

-grumpy

BTW, ain't gaffer's tape great ? (haven't drilled horn/driver flange yet)

[audition]

No dimention shall be less than 1/3 any other dimention of the enclosure.

[grumpy]

Thanks for the "rule of thumb" suggestion, er mandate
to aid in confining my dimentia. ;)

[Zilch]

What I'm using right now (1 kHz) for 2435HPL on 2352 with single 2235H in 4507, two ports closed. Tag an L-Pad on the HF out to adjust. Drop the bypass caps if you build with Solens (cheaper, actually). Inductors are Jantzens: L1 = 14 Ga., L2 & L3 = 20 Ga.

[grumpy]

Thanks Zilch. I'll have to see what's missing in my current parts bin :)
-grumpy

[UreiCollector]

Grumpy,

I like the looks of what you have planned. My personal guideline has always been to avoid any dimension that approaches 1/2 wavelength of my high frequency cutoff (plus 1 octave) for the given woofer. Given that your using such a large horn, I would imagine your cutoff will be pretty low in freq, thus allowing your largest dimension to be fairly big.

I have not experimented with varying box size/shape with constant volume, but have made many enclosures with internal depths in the 14-24 inch range without significant issues.

I would also recommend a generous amount of absorption on the rear wall of the cabinet, to try to “kill that reflection” off the rear wall, from bouncing back into the rear of the cone….perhaps even adding an angled surface behind the woofer to “redirect” that reflection away from the back of the cone.

I think personal preference will come in to play as well....some, for example, don't like horns at all, and prefer the sound of an open dome.......I personally love horns. But of course, we should not ignore the opinions of others. :)

I’m sure you have put a lot of thought into these, and just wanted to add my personal ideas. Good luck!



I think your desire to build a non-room-intrusive speaker is realistic, and should not be ignored, as you have to live with and love these when they are finished.:applaud:

[grumpy]

Thanks Frederick,

"a lot of thought" might be stretching things a bit ;) Concern about the
cab depth (lack thereof) and possible issues/solutions seems to be a primary
constructive criticism, and I appreciate the input/ideas. No concrete has been
poured yet, so I'm flexible :)

-grumpy

[grumpy]

OK, 2435HPL on 2352 and unterminated (have yet to build a test pipe)...

both on floor pointed to 8' ceiling, horn w/o baffle, just floor nearby.

Roughness on horn plot... room noise coupling back? or ??? (not exactly a
quiet environment at my house) Will try a higher drive level later.

-grumpy

[Zilch]

Earl can give precise guidance here, but it looks good for 1 kHz, maybe not much lower, tho....

[grumpy]

That's about what I expected, and what I gleaned from these plots as well.
Fortunately, that's how I expect to use them as well. :)
Will be interesting to experiment and see what effect boundaries/baffle
will have & compare w/ literature. -grumpy

[Zilch]

I did:

1) 2435HPL on 2352 - same result, basically.

2) 2452H-SL on same - will play lower, if desired.

3) My plots are 100 points, which has inherent smoothing. I repeated 2435 at 1000 points to see if it is actually raggedy.

Well, somewhat. ;)

I'm set at 2000 uA test current.

[grumpy]

Cool... thanks for the verification!
Need measurement tools before more compression drivers :)

-grumpy

Edit: Other than small bumps above 2KHz, unterminated looks a lot like the published
PWT impedance plot. Interesting (a little) that your example has elevated impedance
values on the top end when unterminated (or odd that mine does not). Will have to
build one (PWT) once I get a mic that's useful for that sort of thing ... just don't hold
your breath :)

[Zilch]

I didn't do unterminated (no horn, I presume). My second plot is a different (4" Aquaplas'd Optimized Aperture) driver, for comparison of cutoff frequencies of different horn/driver combinations.

[Fodder especially for Earl.... ;) ]

[grumpy]

Ah, well that makes much more sense regarding several features of the center plot :p 2451-SL? an SL-ized 2447? 2452-SL (the current Zilch darling :) )?

Yes, by unterminated i meant with no horn. -grumpy

[Zilch]

2451-SL? an SL-ized 2447? 2452-SL (the current Zilch darling :) )?It was on the horn, is what, so I measured it. :p

I don't know much about 2452H-SL, other than that it is "Next generation," according to the promotional lit for the products that use it. Johnaec first found it in new two-ways spec'd to 20 kHz. I merely said "Ah-HA!"

I have seen the JBL engineering docs, and the curves are very different from others. I think there's more involved than just the Aquaplas'd diaphragm, though I don't know specifically what distinguishes it from 2451SL-H, for example. It certainly measures and sounds better than anything else here.

Here's the link to my testing of it again:

http://audioheritage.org/vbulletin/showthread.php?p=123766#post123766

It "pushes" VHF gracefully, in defiance of conventional wisdom. (Who'd think a 4" diaphragm would play VHF better than the smaller ones?) Giskard suggested in these forums that large format diaphragms might be the ticket for such pursuits. Indeed, apparently.

It ain't real cheap, tho.... :(

[grumpy]

Doh:banghead: right in your post. I just need to learn to read.:o: -grumpy

[Zilch]

Doh:banghead: right in your post. I just need to learn to read.:o: -grumpyNaw, I just now linked to a different post in that thread, one with a PIC of the driver this time. :p

I also remember this was the driver used in the fictional Mauna Kea system. If I recall correctly, that was the incentive for me to order a pair. If it's all good, and becomes the "Plan B" driver, we have TimG to thank.

[OR, maybe we just kick his ass on principle, instead.... ;) ]

[Earl K]

[Fodder especially for Earl.... ]

That indeed it is / though I'm not thinking very straight right now ( kidney-stone / has been passed - yea ! )

-What does the FR response show of the smaller 2435H on those big horns ?
- Where's it's F3 ( for instance ) ?
- Somewhere recently I saw an indication that even the new Everest" has a mild bump-filter built into it's hipass ( it may have been in the "Requirements for BiAmping" document ).

<> EarlK

[Zilch]

Hi, Earl! Congratulations on the passing! ;)

The FRs are in this post, the blue vs. purple graph, no filter:

http://audioheritage.org/vbulletin/showpost.php?p=124591&postcount=87

Blue's the 2435HPL.

Surprisingly, they look kinda the same at the low end. I could redo it, if necessary, more carefully. There was no windowing used in that one; I was looking for the HF extension at the time.

Lemme go do that. It's kinda useless for this purpose, now that I look at it again....

[Earl K]

Surprisingly, they look kinda the same at the low end. I could redo it, if necessary, more carefully. I was looking for the HF extension at the time....

- Yes, you are right about the similarity in the lowend performance .

- It would be nice to distortion plots for two on that large horn ( or are they also in that thread / I better go look ):p

- Thanks for the link / I don't remember seeing those comparisons before . :D

:)

[Zilch]

I've never done comparative distortions on these, but I'll take a stab at it. That's sinusoidal, so we'll get to see the FRs both ways.... :thmbsup:

[Zilch]

I gotta go to the gun shop and get me some shooters' headphones. That was friggin' LOUD!!

Once again, they virtually superimpose rolling off at the low end.

Distortion raised 20 dB, not 30 like I said in the titles. Forgot I had changed that previously.

Protection cap = 33 uF. Mic's at 60" from horn slot:

[grumpy]

Hey Earl, yeah...congrats on the pass (ouch).

It would appear that above 5KHz, 2452 may have a ~5dB advantage
distortion-wise (if linear w/ level to accomodate EQ lift). Also wonder if
undamped/light mass of 2435 is related to the larger level variations in the
low KHz region (compared to the 2452 on the same horn) ? -grumpy

[Zilch]

2435HPL has ferro fluid. I don't know what frequencies that damps.

2452H-SL, ferro fluid unknown. Tech sheet doesn't mention it having any, tho....

[Robh3606]

Hello Zilch

I have a suggestion for you. Change your scales and focus into the area of interest. Makes it easier to read and do comparisons. This is a 2435 and aquaplased 2425. The 2435 is the clear winner with lower distortion levels. Both were measured of their system horns, the H91/2307 and PTH1010. 2435 is blue and violet the 2425 is green and red.

Rob:)

[Zilch]

Is that THD you're showing, Rob, as opposed to 2nd & 3rd harmonic?

Also, in doing sinusoidal, do you use stepped, gated, or both? Delay?

[Robh3606]

Hello Zilch

Thats THD and it is ungated. I just set the sweep range and go. I am in close though to try to keep the room out of it say 6" to a foot or so. Here is the 3rd as well.

Rob:)

[Zilch]

Thanks, Rob.

I'll try that next time. :thmbsup:

[Mr. Widget]

Hey Rob,

What are the different graphics settings between post #41 and #43? I like the sharpness and resolution of #41. Also what format are you exporting them as?

The image in post #41 looks really good... I still prefer the more typical 2:1 aspect ratio of a traditional plot... but that is easily remedied.


Widget

[Robh3606]

Hello Widget

Those were exported as .gif files

Rob:)

[Mr. Widget]

What was the difference in settings between #41 and #43? They don't look the same...




Widget

[Robh3606]

The only difference is line width that I am aware of. I have one at 2 pixel and the other at 3.

Rob:)