Zilch's AK Design Collaborative - Econowave Speaker

[Russellc]

Just for giggles, and because of all of Waynes's contributions and so forth, I hooked up the waveguide he uses (eminence h290 I believe) in the 4Pi and screwed on a JBL2425H:

Results? Surprisingly good, but the top end seems a little polite, odd since the 2425H is said to be 110db efficient, while I believe the Selenium is around 107? Am I missing something here?

I thought it wouldnt be enough padding, but kind of sounds like too much. When I screwed the 2425H onto the JBL wave guide, it seemed slightly worse!
Go figure. Obviously values need tweedled with. so back to the Selenium unit.

Russellc

 

[dshoaf]

Holy crap! it looks just like my livingroom! With the 2225H, I had much better sound with all ports open. With the recones to 2235H, (WOWIE!)
I am deciding between one port and two closed, albeit with make shift rolled up microfiber towel standing in for second Zilch plug. Try it playing some nice music that really pressurizes the room, and pull the plug on the 2226H, I think youll see what I mean. Its all a matter of personal taste of course, but I though it was a little dead compared to all four open. With 2235H, the mids are a little muddied with only one on SOME recordings.

Have fun, I'm watching as I have the exact same setup, and its kickass for sure!

Dude, its not _your_ living room....mine has different color walls...but that's about it. You need to get that MC225 fixed up and on the Econowaves, too.

Its gonna be a while for the 2235s and the 2226s are doing just fine. I do see what you're talking about, though, with the punch with the Zilch-plugs out. Is that what is meant by 'having all the stops pulled out", perhaps?

Zilch, I can now understand that some folks might not like the CX3400's CD boost implementation, (+3dB at 3.5kHz then rises 6dB/octave out to 22.5KHz, for those interested). I'll be building out the crossovers soon but I'm still playing around with the crossover controls to see what I can get out of it. I can see that some foam stuffing could. I may well set up the compensation network and just run the crossover as Jack G noted some pages back.

The next major step is actually to get the measurement jig up and running. I'm flying blind without it.

Cheers,

David

 

[Wayne Parham]

Just for giggles, and because of all of Wayne's contributions and so forth, I hooked up the waveguide he uses (eminence h290 I believe) in the 4Pi and screwed on a JBL2425H:

Results? Surprisingly good, but the top end seems a little polite, odd since the 2425H is said to be 110db efficient, while I believe the Selenium is around 107? Am I missing something here?

I thought it wouldnt be enough padding, but kind of sounds like too much. When I screwed the 2425H onto the JBL wave guide, it seemed slightly worse!
Go figure. Obviously values need tweedled with. so back to the Selenium unit.

Good choice of words. The JBL 2425 and 2426 are very polite. They're smooth enough, but don't reach much above about 15kHz or so. The drivers show the textbook 6dB/octave mass rolloff curve, with a well damped diaphragm that delivers smooth response. So it is important that you use a full 12dB of top-octave compensation.

Don't pay attention to the charts below about 4kHz. I did these measurements a long time ago, and never intended to publish them anywhere. My workroom at the time was very small, and I windowed for a first reflection that was pretty close. It wasn't terribly important to see any lower in frequency because my real focus at this time was on the top-octave compensation, and I wasn't interested in the frequencies below mass rolloff, around 4kHz.

Here is the JBL 2426 on an H290 with 12dB top-octave compensation. Again, pay attention only to the curve from 4kHz upwards, where top-octave augmentation is in play.

Just for reference, here's a chart of the JBL 2426 with only 8dB augmentation. You can see from the raw response chart, that's not enough. But you get some HF boost:

 

[Russellc]

Good choice of words. The JBL 2425 and 2426 are very polite. They're smooth enough, but don't reach much above about 15kHz or so. The drivers show the textbook 6dB/octave mass rolloff curve, with a well damped diaphragm that delivers smooth response. So it is important that you use a full 12dB of top-octave compensation.

Don't pay attention to the charts below about 4kHz. I did these measurements a long time ago, and never intended to publish them anywhere. My workroom at the time was very small, and I windowed for a first reflection that was pretty close. It wasn't terribly important to see any lower in frequency because my real focus at this time was on the top-octave compensation, and I wasn't interested in the frequencies below mass rolloff, around 4kHz.

Here is the JBL 2426 on an H290 with 12dB top-octave compensation. Again, pay attention only to the curve from 4kHz upwards, where top-octave augmentation is in play.

Just for reference, here's a chart of the JBL 2426 with only 8dB augmentation. You can see from the raw response chart, that's not enough. But you get some HF boost:

Wow! no wonder, I see what your saying from the compensation on the old Pro 4 Pi. I will try it with the "full 12 db". Wayne I can not say how much it is appreciatde that you give your input. My trying the 2425H on the Eminence wave guide in and of itself was not just for giggles, I knew it would work as it does to great effect in the 4 Pi pro version, when JBL was used. What I did for giggles is trying it with the crossover/compensation inteneded for the selenium and JBL wave guide, I apologize if it sounded condesending in any way.

By the way Wayne's site uses several things similar, yet different from my econowave. For instance: My 4647 box is a large 5 cu ft. Waynes Pro 4 Pi that can use JBL 2226H or 2225H as well, uses a different alignment with a smaller box. He has a downloadable software that will compute the Pi alignment in these smaller boxes. Some will find this of use with econowave if they cant find the "right size" box but are desireing similar performance out of these or other drivers. All of this is over my head and many may already be aware of all this, if so sorry to go over it again, but I felt it was important.

russellc

 

[TnT2]

Wow! I am glad I started watching this thread last week. A lot of new development though I haven't gone through pages 7-40ish yet. Too much info.

Russell, I am glad your bought up your JBL setup. I recently got a set of 4671OK. This one has the 2225H/2426J drivers in the 4507 box and stock crossover and horn. Depending on which amp I use, the combo is polite on both top and bottom so I have been thinking of upgrading it by replacing parts. Been thinking of getting a pair of 2235H to replace the 2225H. If you do the 12dB top-octave compensation as Wayne suggested, could you give us your impression of the top end? I don't mind getting the PT waveguides but would rather not have to spend more to get the D220ti unless I have to.

Wayne, I am still reading through your material here and haven't even touched the stuff on your website yet. Thanks for imparting your knowledge on this thread.

 

[Zilch]

http://www.jblproservice.com/pdf/Theatre Series/4671OKA.pdf

L2 & C4 provide HF boost in the stock network.

 

[Russellc]

Wow! I am glad I started watching this thread last week. A lot of new development though I haven't gone through pages 7-40ish yet. Too much info.

Russell, I am glad your bought up your JBL setup. I recently got a set of 4671OK. This one has the 2225H/2426J drivers in the 4507 box and stock crossover and horn. Depending on which amp I use, the combo is polite on both top and bottom so I have been thinking of upgrading it by replacing parts. Been thinking of getting a pair of 2235H to replace the 2225H. If you do the 12dB top-octave compensation as Wayne suggested, could you give us your impression of the top end? I don't mind getting the PT waveguides but would rather not have to spend more to get the D220ti unless I have to.

Wayne, I am still reading through your material here and haven't even touched the stuff on your website yet. Thanks for imparting your knowledge on this thread.

First, and sorry if you are already aware, but your 2225H has the EXACT same motor as 2235H, and reconing converts them nicely, and its a brand new driver, as surround, cone,spider , voice coil and all are replaced. I did it with my spare pair. It was a tad over 400 for both recones. Dead 2225H baskets can be had cheaply.

As to the 2425H it really is outside the econowave thread, but hooked up "as is" it is a little polite on top. It is very smooth without the breakup of the Selenium driver. The Selenium breakup is pleasant and is said to provide much of the sparkle in econowave. With the stock < 95 db crossover, the JBL isnt very
sparkley, but sounds fine. As Wayne has been so kind to point out, that driver needs a full 12 db of compensation, you can see his chart of components a few pages back. this would give more top end extension, and I intend on trying this as an experiment, plus I have drivers to build some of the PI designs, something I also intend to do. If you havent seen the Pi site, it is a good resource for the use of that driver, as they used it for years
(2426H), ( basically identical to 2425H) in the Pro PI offerings. He has since moved on to the B&C DE 250, an excellent driver I have intention of trying as well. The Seleniums are only 42 buck apiece, give them a try and play along. I personally guaratee you will be impressed as hell, and it can always be used for something else later. You have an excellent base for many killer projects with the 4647/4507 boxes and their four ports. As Zilch has pointed out, 4 open tunes the 2225H to 40 hertz with big time slam, add 2235H and close one port, and you're tuned like JBL 4430, or close another to tune a little lower and remove a little boom ( nice boom however), close three and I believe you are tuned like Jbl 380 subwoofer. Its a killer box.
Russellc

 

[Zilch]

As to the 2425H it really is outside the econowave thread....

I would agree, except as to the question of whether other compression drivers are interchangeable in E'Wave, which has arisen here before.

The answer is, at least with respect to 2426H/J:

 

[Bauhausler]

Totally off topic

but I haven't fallen off the earth completely. I'm still around and I read a bit of this from time to time.
DNAGLE came over today and correctly identified a funky little pair of speakers that I just bought for their T35 tweets and X36 crossovers as EV Baronets. I have no idea what I'll do with those, but we did confirm that a JBL LE10A and an EWave will fit in there. Frightening.
I haven't meant to shun everyone here. It wasn't just you. Yesterday I got two emails from the people who run the blog bOING bOING asking if I was mad at them or something because I hadn't been around for 9 weeks (I checked, and they were right about the 9 weeks). I was flabbergasted that they noticed I wasn't posting any more.
Anyway now that the election is over and I can stop obsessing about that I hope to get back to speaker projects and AK.
Thanks for the phone calls. I'm OK, really.

 

[Zilch]

Guess who....

He might have some other E'Waveguide ideas for you to try, too.... :yes:

 

[TnT2]

Russell and Zilch,

Thanks for the feedbacks. So the 2426J is out as Zilch's plot clearly shows. Looks like I really should try out the D220ti and the waveguide, though the BMS driver is pretty nice as well.

I was aware that you could recone the 2225H with the 2235H recone kit but I wasn't sure if the magnets are the same on them. Now I know. Did you do your own recone or sent it out? I am intrigue by this possibility since it will be pretty much new everything. I didn't realize the 4507 cabinet is so versatile. At one time, I was thinking of just keeping the drivers and giving the boxes to my friend. Glad I never got that far.

Ok, I am going to do some more readings. Thank you both again.

T.

 

[mrTT]

The last two days, my router was down, again! I had that dumb router that's been giving me heck for the last few months. Now that my new router is running, I should be fine.
I just received a few resistors and a few more to come. Since I have 16ohms set up, I am going to follow Waynes suggestion and set R1 at 16ohms, R2 at 50 ohms (or is it the other way around? but u know what I mean.) then the c1 will be .33uf (oils baby, one russian and one american. so the cold war is over. it is said that these russian caps were used in missiles. yike! I am glad they are going into my crossover instead. ) As soon as this is done I will report.

 

[Russellc]

Russell and Zilch,

Thanks for the feedbacks. So the 2426J is out as Zilch's plot clearly shows. Looks like I really should try out the D220ti and the waveguide, though the BMS driver is pretty nice as well.

I was aware that you could recone the 2225H with the 2235H recone kit but I wasn't sure if the magnets are the same on them. Now I know. Did you do your own recone or sent it out? I am intrigue by this possibility since it will be pretty much new everything. I didn't realize the 4507 cabinet is so versatile. At one time, I was thinking of just keeping the drivers and giving the boxes to my friend. Glad I never got that far.

Ok, I am going to do some more readings. Thank you both again.

T.

I sent it out to the local JBL service center. I dont think JBL will sell just the kit? Last I checked kits were around 180 ish, so for the extra change I let an expert do it. Magnets are Identical b/w 2225H and 2235H, just the cone kit. for that matter, I believe its the same as 2234H if you leave off the mass ring?

Russellc

 

[Russellc]

I would agree, except as to the question of whether other compression drivers are interchangeable in E'Wave, which has arisen here before.

The answer is, at least with respect to 2426H/J:

Which one is the purple line, Selenium?

Russellc

 

[Zilch]

Which one is the purple line, Selenium?

Yes. It's actually black on top of purple, 'cause I ran it both at the beginning and end of the tests to be sure that the filter had remained the same throughout.

[Not that anything so untoward as a loose wirenut ever occurs in ZilchLab, of course... ]

 

[Russellc]

Russell and Zilch,

Thanks for the feedbacks. So the 2426J is out as Zilch's plot clearly shows. Looks like I really should try out the D220ti and the waveguide, though the BMS driver is pretty nice as well.

I was aware that you could recone the 2225H with the 2235H recone kit but I wasn't sure if the magnets are the same on them. Now I know. Did you do your own recone or sent it out? I am intrigue by this possibility since it will be pretty much new everything. I didn't realize the 4507 cabinet is so versatile. At one time, I was thinking of just keeping the drivers and giving the boxes to my friend. Glad I never got that far.

Ok, I am going to do some more readings. Thank you both again.

T.

I would encourage you to get the Seleniums and play along. However, take a look at the JBL 4430 monitors curves. These old compression drivers can be coaxed to give more extension. I believe they are about at the max that can be had from them in 4430. Zilch will know more about this, I believe he has a pair.

For the price of admission, get a set of the Seleniums, they sound quite good in this application, and there's lots to learn here that transfers to all sorts of other projects.

Russellc

 

[Russellc]

Yes. It's actually black on top of purple, 'cause I ran it both at the beginning and end of the tests to be sure that the filter had remained the same throughout.

[Not that anything so untoward as a loose wirenut ever occurs in ZilchLab, of course... ]

Hey! I've been playing with wire nut crossover building and its a great solution! Easy to undo, and it came as quite a surprise that they dont screw up the leads at all! My CC are made this way....speaking of which, you said before that the value of the resistors wasnt that critical....is there a range that can be used? I didnt find 1.5 meg at radioshack, I think I saw 1 meg in a multi pack type of thing.

As to the 2425/6 curves, seems like I remember it doing a little better in the JBL4430 monitors. Or am I remembering wrong? I thought I had more extension than I am getting now when I tried that driver with the JBL wave guide used here when I put it and 2425H on top of the A7s back in the begining. different crossover and compensation, which was suggested to me for the Altec 802 8G. Oddly, I think the compensation cap was 6.2 in that network! I'll look when I get home. Seems way to large a value compared to what is being used here.:scratch2:

Russellc

PS Looking on the JBLPro site, I see the specs were out to 16 khz, + or - 3

 

[Zilch]

As to the 2425/6 curves, seems like I remember it doing a little better in the JBL4430 monitors.

PS Looking on the JBLPro site, I see the specs were out to 16 khz, + or - 3

Yes, it's what I said. In detail, each driver/horn/waveguide performs differently. JBL worked out their own custom compensation approach for 4430. If you look at the distortion curves, you'll see why they stopped at 16 kHz. The same driver is run higher in other products. Altec used a different approach to compensation in their designs.

It'd all sure be easier if everything behaved generically and according to theory and smiplifying assumptions, but it doesn't. Once you get your RTA up and running, measure this combination and compare to my results. Are all six of my 2426s defective?

 

[Russellc]

Yes, it's what I said. In detail, each driver/horn/waveguide performs differently. JBL worked out their own custom compensation approach for 4430. If you look at the distortion curves, you'll see why they stopped at 16 kHz. The same driver is run higher in other products. Altec used a different approach to compensation in their designs.

It'd all sure be easier if everything behaved generically and according to theory and smiplifying assumptions, but it doesn't. Once you get your RTA up and running, measure this combination and compare to my results. Are all six of my 2426s defective?

No I doubt your 2426Hs are defective, but I understand these curves were with the econowave crossover and compensation, no? And like you said they dont behave generically. Yes, the distortion goes up seemingly exponentially above the 16 khz point. But this driver has a very smooth sound, or polite, as I like to say.

The compensation on my A7s isnt Altec, it a modified Eminence with compensation another member gave me, very similar to whats used in the PI filters document, and here, but the compensation cap, if I remember correctly was 6.2, way too high under the charts we are using here.

Yes it would be nice if they all behaved generically, as long as thats based on a good sounding example!

I need measurement equipment badley. I search craigs list daily for both the DEQ 2496 as well as the CX 3400 crossover, no luck yet.

Russellc

 

[TomSc]

Hi all,

Just wanted to chime in to say that I'm still on the bandwagon, but very short on time to tweak on my EW's. Too many house related projects that involve plumbing and drywall. Luckily, one of the projects is finishing some in-wall shelves for my collection of gear!:thmbsp: But I have been doing some reading after the kids are in bed and I'm not allowed to make any more noise in the shop. I have keeping up on the progress of others here, wishing I could play too. This design shows so much promise from what I have heard so far and I am chompin' at the bit to get back at it! The weekend is already full, but hopefully next week will bring more time to E-Wave....:yes: