Noob ESL builder looking for advice

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi. I'm launching headlong into an ESL project for the first time. I really appreciate the wealth of knowledge here on diyAudio. Hopefully I can make higher quality loudspeakers through our forum discussions.

The project is two ESL panels, built on as reasonable a budget as I can manage. I am not looking for full range, and will add the lower end frequencies with a separate subwoofer.

I have a steel perf sheet on the way, which can yield panel sizes of either:
91 cm x 25 cm
or
101 cm x 22 cm

Is that so narrow a difference that I should just go with what looks best to me, or would the taller, skinnier version have a little better dispersion in the higher frequencies?

I want to go with the dual transformer design for stator step-up mentioned by Calvin, Jazzman, and others. Is there a good supplier for those right now? I tried Antek and Newark, they are sold out of the 6v version.

Many thanks in advance for the help!
 
Hi. I'm launching headlong into an ESL project for the first time. I really appreciate the wealth of knowledge here on diyAudio. Hopefully I can make higher quality loudspeakers through our forum discussions.

The project is two ESL panels, built on as reasonable a budget as I can manage. I am not looking for full range, and will add the lower end frequencies with a separate subwoofer.

I have a steel perf sheet on the way, which can yield panel sizes of either:
91 cm x 25 cm
or
101 cm x 22 cm

Is that so narrow a difference that I should just go with what looks best to me, or would the taller, skinnier version have a little better dispersion in the higher frequencies?

I want to go with the dual transformer design for stator step-up mentioned by Calvin, Jazzman, and others. Is there a good supplier for those right now? I tried Antek and Newark, they are sold out of the 6v version.

Many thanks in advance for the help!

Just PM'd you
 
It is Great to see a new ESL project getting started!!!

The difference of 3cm in width won't make a whole lot of difference in the dispersion pattern.

But what will/can make a difference is the width of the sections you choose to use.
2 or 3 sections is good for your width of material.
I would go with 2 sections.
Keep the middle spacer narrow if you can ( 1/4" to 3/8") as this will help to increase your radiating surface area to a maximum.
3 sections may be a bit narrow as each section would be only about 2 5/16" (6cm) wide.
This may limit your lower midrange response due to excursion limiting from being too narrow.

My little panel has a diagphram width of 3.75" and smallest I think I would go maybe even 3" unless it was made strictly for a tweeter section.


jer :)
 
Last edited:
Vertical sections vs horizontal. I see ML using horizontal, which seems a necessary step with the arched stators. But a good number of the DIY planars have vertical sections.
What is the maximum safe section area? I will stretch the Mylar with a bicycle tube table. (Although, since I have never done that before, it will be interesting to figure out how to know when it it stretched "enough".) D/S spacing about 1mm.
 
Vertical sections vs horizontal. I see ML using horizontal, which seems a necessary step with the arched stators. But a good number of the DIY planars have vertical sections.
What is the maximum safe section area? I will stretch the Mylar with a bicycle tube table. (Although, since I have never done that before, it will be interesting to figure out how to know when it it stretched "enough".) D/S spacing about 1mm.

Sanders' Cookbook defines the maximum span between support spacers as 100X the d/s (100 times the diaphragm-to-stator spacing). I prefer to go more conservative and limit the span to 80X d/s.

For 6-micron film on a flat panel, I would stretch the film with a bike tube jig, to 1.5% elongation. Before stretching, I put reference marks on the film exactly 12" apart in the width dimension, then apply stretch until the marks are 12 3/16" apart for 1.5% elongation. I also recommend 1/16" d/s for a hybrid panel (that's about 1.5mm).
 
Last edited:
Yes, that is a very good general rule of thumb.
Especially for diagphram widths larger than say 5 to 6 inchs.
However this rule quickly runs into practical considerations as the panel gets wider than say 9" or more.

The lowest frequency of operation must be taken into account as well.
As not too run out of room for large excursions at the lower frequency's of operation.

There was a good chart that I knew of that suggested the minimum D/S as per lowest frequency of operation.
However I am having a difficult time trying to find it on the web or the original document has been modified since and no longer has the data that I am looking for.
I do have it in my archives somewhere though.

Even with my little panel with a width of 3.25" and aprox. D/S ratio of 1:45 I have been able to clip the diagphram into the stator at Frequency's lower than about 200Hz to 250Hz even with a spacing of .072" (+/- .003").
Even more so when I hit the resonate frequency of the diagphram of about 70hz to 90hz.

May I add that this much excursion for my little 3.25" X 9.75" diagphram is quite very loud!!

A 1mm spacing may be good if you are planning on crossing over at a rather high frequency of say 1Khz or even 1.5Khz or higher and definitely if it was intended to be strictly a tweeter panel.
IMHO .0625" or as Charlie had mentioned a spacing of 1.5mm to 2mm is a very good place to start and should be good for those peaks at around 300Hz.

Remember that the excursion doubles for every octave lower in order to produce the same SPL as the starting fundmental.

Here is a very good read on the subject although it pertains mostly to headphone drivers,

HeadWize - Project: Notes on DIY Electrostatic Headphones by Chu Moy

And here is a much more detailed version as well just scroll back to page 138 for the beginning,

Loudspeaker and Headphone Handbook - Google Books

And here as well,

Audio and Hi-Fi Handbook - Ian Robertson Sinclair - Google Books

This is a very good thread that had touched on the subject as well,

http://www.diyaudio.com/forums/plan...sl-suitable-x-over-frequency.html#post2266701

The last time I was able to run my little panel I found that I got about +/- 30 degrees of usable dispersion off of the center.
And at about 45 degrees off the center there was noticeable drop off of the high frequency's as expected.

Enjoy !!

jer :)
 
Last edited:
Excellent information--exactly what I was hoping for. This helps my design. :)
I have the 6 micron film on the way from Asia. Metal should arrive today.
Since I am shooting for about 300 Hz as the crossover frequency, maybe 1.5 to 2mm is a better way to go, as you both suggest.
In the U.S., what stores carry the 3M double-stick in that thickness? Anyone know?
 
Excellent information--exactly what I was hoping for. This helps my design. :)
I have the 6 micron film on the way from Asia. Metal should arrive today.
Since I am shooting for about 300 Hz as the crossover frequency, maybe 1.5 to 2mm is a better way to go, as you both suggest.
In the U.S., what stores carry the 3M double-stick in that thickness? Anyone know?

If you go to my website and click the link "Building the ESL Panels", you will find links there to all the materials needed, including the "multipurpose polyurethane foam tape" from McMaster-Carr (which is 3M tape). A .063 x .75 x 36yd roll will cost you $41.
 
If you go to my website and click the link "Building the ESL Panels", you will find links there to all the materials needed, including the "multipurpose polyurethane foam tape" from McMaster-Carr (which is 3M tape). A .063 x .75 x 36yd roll will cost you $41.

Another option is local office supply stores(STAPLES, Office Depot, etc). They generally carry the 36yard rolls of 1/16" thick, 3/4" wide 3M foam tape. Not sure how pricing compares now, but 5 years back it was similar to the McMaster price.
 
Here are some guides as to what the have available when you do your search of suppliers,

3M? VHB? Tape - High-strength bonding tape is a proven alternative to screws, rivets, welds and other forms of mechanical fasteners - 3M US

http://multimedia.3m.com/mws/mediaw...evTSeSSSSSS--&fn=VHB Design Guide Low Res.pdf

Just about every where you go you can find the stuff,Home depot,Menards,Lowes,Craft and Hardware stores.

The above PDF shows the various model numbers and thickness's to be looking for that are available.

Uline has various types and thickness's as well,

http://www.uline.com/Grp_240/3M-VHB-Tapes

The stuff is not exactly cheap!!
Although 36 yards is a lot of material.

jer :)
 
Last edited:
The price of that foam tape went up about $6 a roll since I last bought some two years ago... prolly about the same everywhere else. I figure if you have to buy copper foil tape from McMaster-Carr (for the charge ring), you have to pay a shipping charge anyway so you might as well order the foam tape there too-- and you'll have it in about two days.

BTW, I think it's great to see a new builder going for it. I just wish I could be there to see the look on your face the first time you hear your new speakers!
 
Last edited:
Metal arrived today. Got a few bends in it, but should straighten out OK. Ordered the Mylar. Will be here in a month or so... Gives me enough time to get a tension table put together. Now to find someone with a band saw. Or, since the metal is thin enough, might try a Dremel.
 
Maybe some metal cutting blades in a hand held jig saw would suffice as you will eat up a lot of fiber wheels in a dremel.

Just make sure that you support the metal well so that you don't bend it should the blade get caught and stick on you.
That won't happen though providing that you have it properly secured on the work table and the piece being cut off supported as well.

If you have a larger Die Grinder type tool this may be a better choice than a Dremel and would take less time as well.

And do use proper eye protection as well as some gloves when using Fiber wheels.
The last time I used one I got a piece fiber or metal splinter some how in my finger tip and this kept me from being able to play my Guitar for nearly a year.
This happened back in January while I was building my Variable HV bias supply.
I have just recently been able to start playing again not to mention your eyes of course.

Make sure that you round off any sharp edges before you even think about starting to coat your stators.
This can never be stressed enough !!!
The past has shown that this is the root of most all failures that occur for noobies on their first DIY ESL build, Sharp edges and stator coating thickness.

You have a lot of work ahead of you but it is all worth it in the end.

Just as Charlie put it, It would be priceless to see your face the first time you fire those bad boys up and are working with no issues!!!

It has been done before and I just get so excited when I see a new build through.

Good Luck !!!

:cheers:

jer :)
 
Metal arrived today. Got a few bends in it, but should straighten out OK. Ordered the Mylar. Will be here in a month or so... Gives me enough time to get a tension table put together. Now to find someone with a band saw. Or, since the metal is thin enough, might try a Dremel.

I use big metal scissors 10.5" long and over 400 grams in weight to cut my perforated sheet, had no problems even at my age (74). No need for band saws or dremels. You can cut any shape you want with scissors. It,s safe as well no flying objects no splinters etc. I have built over 30 full range planars using scissors and have had no accidents, touch wood yet.
 
Last edited:
As Jer pointed out, grinding off sharp points along the edges is critical. I don't mention this on my webpage but when I fist fired up my speakers they played great for about 30 seconds before all that voltage found some sharp points and weak insulation around a stator edge and started arcing, which smoked the panel and shutdown my amplifier. I ended up having to completely strip off the paint coating from the stators and start over again. It was horrendous but it taught me a lesson about taking the time to properly prep the stators.

One more thing: You don't use a hammer on a hard surface to flatten dents and creases out of a metal stator, or you will likely stretch the metal and give it the flops (oil canning), in which case the stator would be ruined. To straigten dents/creases, I use a wooden mallet with a rubber router mat between the stator and work table.

Have fun!
 
Cool !
I expected that tin snips might induce a little bending.
As long as the didn't induce any major warpage in the panel you should be okay as the edges won't effect anything, it is the overall panel flatness that is important.

I have been doing some more research on section widths and I found that on page 3-5 of this document,

http://www.pispeakers.com/ssdm_99.pdf

And using this calculator,

SOUND FREQUENCY & WAVELENGTH CALCULATOR

Seems to support my experience of useable dispersion of -/+ 30 degrees with my panel width of about 3".

To give you an idea 2.7"width= 1 wavelength at 5khz and by using the polar graphs in the document will give you a good idea of your dispersion factor.

On my next build little panel build I am contemplating a 1" tweeter strip with a 4" to 6" midrange section just to see how it will work out as I can't hear much of anything above 16.5Khz anyhow.

This is meant to be for some very good info and don't let it confuse you with your current build.
You will still be quite amazed once you get it finished and running!

Cheers !!

jer :)
 
Last edited:
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.