My take on a stretch jig

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
not really. i know i have to go and finish it :( but the weather here is so bad its either raining or windy(or both), and i have to buy my wood and transport it by bike :( it gets wett or i go flying in the wind :) haha

i did order all rubber bands, and coated all the clambs for one side. so i have to make almost 4 times as much of these clamps. i also bought a mica heater i might use to heat the film. its a verry thing heating element 40x40cm i might need 2 of those in the final design. but first tests i will probable use a paint stripper. it should work as well.
 
Hi WrineX, really interesting reading this thread on the train to work. Especially the tests regarding Mylar tension & heatsetting v's tension/resonance. I have an old panel with its film intact. I have to measure its resonance and compare with your results. It's quite old so perhaps it has deteriorated? I will see. Would be great to hear more updates if you have made more progress with foils and perhaps the next stage; resistive coating. Thanks again for sharing your work!
 
Hi WrineX, really interesting reading this thread on the train to work. Especially the tests regarding Mylar tension & heatsetting v's tension/resonance. I have an old panel with its film intact. I have to measure its resonance and compare with your results. It's quite old so perhaps it has deteriorated? I will see. Would be great to hear more updates if you have made more progress with foils and perhaps the next stage; resistive coating. Thanks again for sharing your work!

Thank you, i must admit i left the project in one of the many boxes on the shelf :( i got to much to do and so little time :) i got half of the wooden pullers, but measuring the resonance could be hard since it might deviated because of the way it was measured. the measurements and comparisons are made withing the same method. so for me they are valid compared to an original panel on my measuring jig. you could have complete different numbers as long as you compare an original to a home made with the same jig, your good to go.

i will pickup the project somewhere as i usually do. :) im glad you enioyed it. by the way take a look at my youtube channel for more of these kinds of projects.
 
This is indeed a very interesting thread.
I've always wondered about the correct
tension as this will definately affect
frequency reaponse of the esl panel.
Has anyone experimented with tension
to attain optimum sound ?

Thks

the optimum is where quad used to put it. for 3 micron mylar that is just before breaking point. when tension is to low you can play lower but spl will suffer down that low and it will drop earlier then when the correct tension is used.

so the decline will start sooner but will extend further down when tension is to low. to high tension is almost impossible with 3 micron since it will rip before you reach much higher.

Problem is many use 6 micron, now 6 micron mylar is easy to get the tension TO high. since it it much stronger.

another downside of to low tension is the ability to play lower frequencies, but they need higher excursions, so you will hit the stator when you hit the resonance frequency. at much lower volume then normal. (i have this problem with one of my quads)
 
Thanks for the enlightment. I presume that 3 micronwill make a more efficient panels & perhaps the highs
will sound better as well.

well it does not do much with efficiency, the highs drop of slightly early with 6 micron, although its only 1.5 db at 18khz or something, so nothing to write home about. i think Bolserst had a nice graph somewhere of this effect. but there is this endless argue if one sounds beter then the other. and to me i dont really hear it i think ,but i would like to be it as close to original spec if possible.

oh by the way, im not a know it all, all i say here is my observations. i looked over the last posts and it sounds like this is how it is or it is a fact. but this is what i noticed and my theory behind it. just to be sure :)
 
Last edited:
4 um would be better maybe for the strength. An as wrineX says the membrane is loaded by the air which contributs most to the mass, and remember the coating that gets a couple of um as well. So increasing the membrane thickness would give a lot more robustness but contribute quite little to the mass increase in proportion.
 
4 um would be better maybe for the strength. An as wrineX says the membrane is loaded by the air which contributs most to the mass, and remember the coating that gets a couple of um as well. So increasing the membrane thickness would give a lot more robustness but contribute quite little to the mass increase in proportion.

still i am torturing myself to get it to 3 micron. and since there are 2 sets standing in my living room for over a year now, i might help myself and use 6 micron. witch would result my other time being wasted.... but might get things going and moving out of my living room for once. and get some spending money for other projects , like a bigger CNC table or more research into some of my projects i did , where i never wanted to spend to much money.
 
Slightly OT question. I've a pair of ML Odyssey made some tweaks
on the crossover but when I first saw the crossover I keep wondering
why the high pass has to be made so complicated I just dont understand.
The panel is suppose to be wide band from 250 to 20k . So why the need
to made the XO so complicated ? Would someone pls shed some light on
this pls

Thks
 
…I keep wondering why the high pass has to be made so complicated I just dont understand. The panel is suppose to be wide band from 250 to 20k . So why the need
to made the XO so complicated ?
Most of the complication in the crossover comes from the need to passively EQ the roll off in the response for frequencies where the wavelength is longer than the panel is wide. (ie dipole cancellation).

More details on ML crossover EQ here: (posts #46, #50, #56)
http://www.diyaudio.com/forums/planars-exotics/225959-toroids-esls-5.html#post3323339
 
I have read your thread a couple of times now Wrinex, truly inspiring.
Working on my (first) pair of ESL 57's, I tried to approach restoration of the panels also in a reproducible manner. Build a kind of stretching jig, also put a woofer underneath to measure resonance as a paramater of the tension in the diaphragm.

And then it became immediately clear that mechanical tension alone, like explained on many DIY sites, probably leads to rather unpredictable results in the long term: the diaphragm indeed slowly loses tension.

So I am looking at heat treatment now. Did my first attempts with diaphragm under tension, with a heat gun. Indeed this leads to more stability, but I am trying to get some control over this process. A heat gun is troublesome, and I have shot some diaphragms in the mean time of course (under tension they split in a second when heated too much).

I have looked many times at the pictures of the original Quad stretching jig, especially the oven. It is open at the bottom, and there seems to be some kind of shelve in it, which will be just underneath the diaphragm. I assume it is a kind of a flat bed heating. I have the feeling, looking at the construction, that this oven is probably not very hot. After all, it is a metal rig they roll in and out, and since they work on it constantly it will probably be relatively safe to touch without risk of burning yourself. Assuming they did quite some of these panels on a working day, one at a time, the time in the oven is probably rather (tens of) minutes than hours.

I have a flat heating panel on order now to do some experimenting (although I wonder if it will be hot enough).
Do you have any directions on actual temperature / time for the treatment?

Cheers,
Niels
 
Stretching

It is a long time ago. I also have experimented wich streching jigs and heat treatment etc to repair panels for a ESL (Beveridge 3s). I ended with a method that worked quick, simple and delivers controllable results:

(1) marking to rectangles on a table tennis table with white edding. One with the original size a second scaled by +2%.

(2) Then placing the mylar, attached with loose tension with tape

(3) mark the smaller rect on the mylar with an edding

(4) stretching the mylar starting by the halfpoints off all sides until evenly stretched

(5) when finished place first grid from top (used 3m double sided tape, for mirrors, still fine since years). Remove taped and join with second half.
 
I have read your thread a couple of times now Wrinex, truly inspiring.
Working on my (first) pair of ESL 57's, I tried to approach restoration of the panels also in a reproducible manner. Build a kind of stretching jig, also put a woofer underneath to measure resonance as a paramater of the tension in the diaphragm.

And then it became immediately clear that mechanical tension alone, like explained on many DIY sites, probably leads to rather unpredictable results in the long term: the diaphragm indeed slowly loses tension.

So I am looking at heat treatment now. Did my first attempts with diaphragm under tension, with a heat gun. Indeed this leads to more stability, but I am trying to get some control over this process. A heat gun is troublesome, and I have shot some diaphragms in the mean time of course (under tension they split in a second when heated too much).

I have looked many times at the pictures of the original Quad stretching jig, especially the oven. It is open at the bottom, and there seems to be some kind of shelve in it, which will be just underneath the diaphragm. I assume it is a kind of a flat bed heating. I have the feeling, looking at the construction, that this oven is probably not very hot. After all, it is a metal rig they roll in and out, and since they work on it constantly it will probably be relatively safe to touch without risk of burning yourself. Assuming they did quite some of these panels on a working day, one at a time, the time in the oven is probably rather (tens of) minutes than hours.

I have a flat heating panel on order now to do some experimenting (although I wonder if it will be hot enough).
Do you have any directions on actual temperature / time for the treatment?

Cheers,
Niels

Yes , the oven does not exceed melting temps of mylar ofcourse :) but it could be localized heating to. i have a ceramic heater i bought for this project its flat and large and i might needed 2 of them. idea was to make a jig so you can roll it over the stretch jig heat it and tension it then remove it again. but i must say i sort of stopped working on it. i try to fix my ESL's and be done with it. i have no energy for the quads left. i want a cool fresh project from start to finish :) these speakers are standing here for 2 years now and i had enough :(
 
Part of satisfaction is mastering the technique for me, I guess :) The jig I built uses simple clamps and bottles filled with metal punching caps. For heating I used 3 heating panels of 60x30 cm to cover the diaphragm. I place it over the diagram while stretching. It sort of hovers 2 cm above the mylar, and will get to around 110 degrees. I heat for 15 minutes.
 

Attachments

  • Stretching Jig.jpg
    Stretching Jig.jpg
    80.8 KB · Views: 269
  • Heating Panel 1.jpg
    Heating Panel 1.jpg
    7.2 KB · Views: 257
  • Heating Panel 2.jpg
    Heating Panel 2.jpg
    59 KB · Views: 259
Hey WrineX,

Did you make any more progress on this project?

I was thinking about my own bikes (have owned bikes since little child) and I have never seen an inner tube that doesn't lose some air pressure over time. Ok we are talking days when I notice the pressure leakage..

What pressure in psi are you pumping the tube to get the mylar to tension? Which makes me think of the next thing - the pump. These often leak some air also and a fractional amount when disconnecting the pump - was this an issue for you.

Maybe I should just try this method myself.

I brought a kit from ER Australia and there is no mention heat setting the Mylar. Perhaps it is already heat set type?

Finally, is there an agreed Fs value for stretch resonance? Seems like somewhere around 70Hz which is then lowered due to the charged panel and the sum of 4 panels in the speaker frame.

All the best.
 
Can I ask everyone here what the agreed consensus is for the resonance frequency of a single panel should be. I'm in the process of measuring some of the panel I have that are still intact. I have to look for a new software that gives me a clearing reading of frequency but shouldn't be too hard to find.

I have a kit with a quoted stretch tension of 2.6Kg which from the Uni maths days means it should in theory give me the required resonant frequency (if the suppler has indeed proved this) given the material is a known and the aperture is a known (disregarding local temperature and air pressure etc). My maths days also inform me that 2.6Kg is a fairly crude value and realistically when I actually stretch and tune my mylar the actual stretch figure I expect to be a more precise value - but I am yet to prove this.

I have also brought a spring tension gauge with a reading up to 350 milligrams so i will experiment with that also.

Would be great to hear people thoughts on what resonant frequency people have aimed for and achieved and also if heat setting has indeed produced stable results.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.