Yet another DIY Electrostatic Loudspeaker Design

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
As many of you know, I've been repairing Quad ESL's and other electrostatic speakers for nearly a decade now. I've always been wanting to build a set of ESL's from scratch that embody everything I want in a speaker system.

I've built a number of prototype panels and such, and I've finally gotten enough critical mass together to finally put together a system that is really working well.

If anybody is interested in it, I've got all the plans and construction details listed here:

http://quadesl.com/diy_esl1.shtml


Sheldon
 
stokessd said:


If anybody is interested in it, I've got all the plans and construction details listed here:

http://quadesl.com/diy_esl1.shtml


Sheldon

Very nice!

Are you using some specially formulated tape? All the polyurethane foam tape I have ever seen lasts about 6 months to one year, then turns to dust. It is even worse than the foam surrounds on some bass drivers, and exposure to ozone created by the high voltage in the speaker will only accelerate the deterioration. Also, when you stick a tensioned diaphragm to it the adhesive tends to get pulled into the speaker by the diaphragm, firmly sticking every piece of dust and dead bug that gets in there to the diaphragm right at the point where the stators, insulators, and diaphragms are closest together. I would expect a problem with arcing after only a few months of operation. Unfortunately, a vacuum cleaner won't help you clean it out because the dirt gets stuck to the adhesive.

As the adhesive gets pulled into the speaker, the tension on the diaphragm decreases. This may require that you drop the bias voltage (and so the speaker sensitivity) after a few weeks to keep the driver from oscillating.

I have found that contact cement bonds to the diaphragm and almost anything else you want to use for insulators. When the time comes to rebuild those drivers, you might want to give it a try. It is completely stable once it sets, which is a matter of minutes.

MR
 
Peranders-

I'm very pleased with the sound. One of the problems I've had with the other hybrid ESL systems is integration and efficiency. As you are probably aware, you get a first order rolloff on the panel due to pressure feedback as the reproduced wavelength (1/4 of it actually) gets longer than the shortest panel dimension. With a passive crossover you really only have a couple options to deal with that. first, you could knock down the high frequencies so that they match the lowest used portion of the panel. But this leads to a fairly inefficient speakers. Second, you can roll off your woofer fairly shallowly and fill in that roll-off region with the woofer. but this leads to some odd blending and integration problems. I've gotten around that by using an active crossover, where I can actively compensate for that roll-off and make the panel flat from about 40 Hz on up. This means that there aren't any crossover elements turning my hard earned amplifier watts into heat. And I can drive my panels to very loud levels with my Heathkit W5M 20 watt amplifiers. It also means that I can cross the panels off low and sharply to minimize driver interaction. the active crossover also allows me to time align the woofer to the panel. This really isn't a huge deal with the wavelengths the woofer produces, but if I ever wanted to cross over higher, it may be an issue.

So you could say that I think they sound quite good.


MR-

The foam tape I'm using is a eurathane foam with an acrylic adhesive from 3M. So far it's been quite good at withstanding the shear forces and has held up nicely over time too. these panels are about 6 months old now, with essentially no change in resonant frequency. I haven't seen any adhesive creep, the edge of the foam tape is still square and the adhesive edge hasn't separated from the diaphragm.

When these panels go south, I'll be making some lexan spacers and using either epoxy or contact cement. I wanted to try the foam tape method. there are several ESL makers who are doing that (martin logan and innersound to name two). It's so easy, you can knock out a pair of panels in like 25 minutes.

Thanks for the info, I'll be waiting for the foam to turn to dust.

Sheldon
 
Yes I've read the ESL cookbook. It's a very useful book for learning the ropes with ESL's. I do think that he is WAY too obsessed with extreme volume. I also think he's trying to kill the family pets with a sharing of the high voltage between speakers. :)

I am in violent agreement with Roger that TL enclosures are the way to go. But I think the 400-500 Hz is too high. Yes it's needed for absolute max SPL, but there is also a lot of musical information in that range, and frankly an ESL panel is more suited for any valuable musical information. So I am sacrificing SPL to get as little musical information on the woofer side of things. I also want the crossover point to be at a wavelength that is longer than the longest separation between the panel and the bass driver.

Sheldon
 
I agree on the crossover point. I am using a 4th order, 100 Hz crossover right now, which may be just a bit too low freq, but it keeps deep bass away from the ESLs and keeps vocals in the ESL.

The bass drivers are a double chamber affair, and they are pretty flat down to about 30 Hz. I have used TLs before and didn't find much difference, except that the TLs take up a lot more room. One of these days, I'll build something bigger that can go lower, but there isn't a whole lot to miss below 30 Hz, so it isn't a priority.

I built an experimental system once that used 10" PVC pipe to make a transmission like. It was C shaped, and stood up so the woofer was at about ear level. I then built an ESL that was an annular ring that fit over the pipe, so the speaker ended up looking like a flower, with the bass driver at the center and the ESL surrounding it. Looked weird, sounded pretty good. Took up too much space so I ended up tearing it apart. I wish I had taken some pictures...

MR
 
frugal-phile™
Joined 2001
Paid Member
MRehorst said:
I agree on the crossover point. I am using a 4th order, 100 Hz crossover right now, which may be just a bit too low freq, but it keeps deep bass away from the ESLs and keeps vocals in the ESL.

I found 125 Hz to be good when i added woofers to my Acoustat panels.

I wish I had taken some pictures...

Me too. I don't think what i'm imagining from your description is quite right. Can you do a sketch?

dave
 
Here's what it looked like. It was BIG. It used about 10 feet of 10" pipe. I had to create a mitre box to cut the pipe at the right angle to join the pieces together. What a pain THAT was!

The idea behind it was to achieve better integration between the bass and ESL drivers (a "normal" set up is like Mr Stokes- bass on the bottom, ESL on top. At high frequencies, the ESL is extremely directional, so the bass and high freq signals sound like they are coming from two different places (they are), one about 3 feet above the other. This isn't a problem in a large room where you can sit back away from the speakers, but in a smaller room, it can be distracting.

Anyway, the speaker was too big for a small room, and the annular shape of the ESL led to some interesting radiation patterns at high frequencies. Not a total success...

MR
 

Attachments

  • esl.jpg
    esl.jpg
    43.8 KB · Views: 1,010
The idea behind it was to achieve better integration between the bass and ESL drivers (a "normal" set up is like Mr Stokes- bass on the bottom, ESL on top. At high frequencies, the ESL is extremely directional, so the bass and high freq signals sound like they are coming from two different places (they are), one about 3 feet above the other. This isn't a problem in a large room where you can sit back away from the speakers, but in a smaller room, it can be distracting.

That can be an issue when the shortest wavelength that the low frequency driver produces is smaller than the separation distance between them. you get comb filtering and all sorts of fun stuff. I mentioned earlier that I wanted the highest bass frequency (IE shortest bass wavelength) to be larger than the longest distance between the panel and the bass driver. this will insure that there is never a perceived source separation between the two drivers. If I would have crossed at 500 Hz like Sanders recommends, there would be that problem.


Sheldon
 
open baffle vs. TL woofer?

Very nice project, Sheldon. I'm slowly working towards a design for my first DIY speakers. At the moment, I'm considering either ESL or something like the Linkwitz Orion... either way, with crossovers in DSP.

I'm curious about your opinion of open-baffle dipole woofers (such as the Orion) compared to TL, for pairing with a mostly-full-range ESL like yours...
 
diyAudio Senior Member
Joined 2002
Well,well...

Hello,


Welcome to you Mr.Stokes!

Having owned a myriad of incarnations (the incarnating not being just God's work but my humble doing) of the ESL57 I can only welcome you wholeheartedly.

Hard to beat a good ESL from the low mids up!

Sorry 'bout this overdue welcome,must have been discussings wine somewhere else.

Best regards,:)
 
Re: open baffle vs. TL woofer?

hifiZen said:
Very nice project, Sheldon. I'm slowly working towards a design for my first DIY speakers. At the moment, I'm considering either ESL or something like the Linkwitz Orion... either way, with crossovers in DSP.

I'd love to dabble in DSP crossovers a bit. I plan to do that in my future when I finally stop listening to my records. But as long as I've got analog sources, I'm sticking to analog domain crossovers. There are serious advantages to the DSP route though.


I'm curious about your opinion of open-baffle dipole woofers (such as the Orion) compared to TL, for pairing with a mostly-full-range ESL like yours...

At this point I'm still unsure. A have a friend here in town who is getting ready to build a set of hybrid ESL's himself and he is going through the same soul searching.

The dipole sub thing is a bit of a kludge. You need tons of excursion on your drivers to compensate for all the pressure feedback. And unless Dr. Linkwitz is doing something new, he is using production bass drivers for his dipole sub. I have yet to see a bass driver that is really designed for that type of duty. So you've got to pick a real overachiever of a driver designed for box mounting and beat it whithin an inch of it's life.

It's an interesting concept and for the upper range of the sub's operation I can see that it could have real advantages, but I really think that some special drivers (maybe servo moter driven drivers...) are needed.

Keep in mind that this is armchair quarterbacking, as I don't have direct experience with this methodology :bigeyes:

Sheldon
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.