Measured my ESL attempt

[Jonhny]

So if I had 5kV bias and 7kVrms on the stators I Think that I have 10kVpeak between the stators giving 5kV bias and -5kV on the stator for 10kV over the air gap. With my old panel I Think I had 1.8mm D/S. It wasnt 1.5mm as I thought first.
Now the air gap is going to be less than that with 10kV pulling the film towards the stator. I also noticed this with the first panel I built (the one with yellow taped stators). The sparks would light up the room when the Music was intense in the lower frequencies. I dont have the spreadsheet that calculates SPL depending on xmax. I will assume 0.5mm film excursion for now. This gives 1.3mm between film and stator.

So 10kV over 1.3mm air is possible? This seems very much.
Having 12kV bias with ~1.8mm D/S is also very much.
I have tried 10kV bias with 1.8mm D/S too, so I know it works. We have both measured similar values. I just dont understand how. In my case some of the voltage must have been stored in the tape insulation, but when you had 12kV?

 

[geraldfryjr]

I had a whole concept of something all typed out and the power glitched and now it is all gone!!!
So, I will try to remember everything as it was a little later.

But I did have a few strange issues with my coating holding a charge as well, I will try and find that post describing it.

Upon first raising the voltage I got some strange effects and phasing like the sound was being emitted from the back of the panel, until I got up past a particular point and above.

Then after that it was all equalized and I was able to vary the voltage all of the way to 0v and back up.
The volume would change as expected depending on what voltage I had it set at with no strange effects after that.

jer

 

[bolserst]

Any thoughts on if there is a big difference in SPL/diaphragm size with perforated sheets versus wire stators? I want a small speaker that still fills a room with Music (reasonably loud).

If properly insulated, perforated sheets and wire stators with the same % open area will produce nearly identical SPL for the same size diaphragm and Vbias/Vstator voltages. If you want to use a small panel, the best way to maximum SPL output is build a hybrid. Remember that the natural response of a dipole ESL slopes down as you go lower in frequency, so raising the crossover frequency increases the SPL level at which you will cross to the dynamic driver.

Simultating it at quadesl.nl seems to give a maximum when the drive voltage is a bit below the bias voltage. Say 3k5V drive and 5kV bias. I haven’t thought of the actual voltage between the film and the stator, the drive is in RMS and shared on the two stators. It should be easy to calculate, but time for bed now!

If you haven't read it already, Section 3.2.9 of Baxandall's chapter on ESLs derives values for Vbias and Vstator to maximize ESL output.
What limits the efficiency of ESLs is the the breakdown voltage for air. He uses a value of Emax=4kV/mm for dry air, although it varies with D/S spacing(see attachment).

For a given diaphragm spacing d, the optimum Vbias and V(stator-stator) are:
Vbias = Emax x d / 2
V(stator-stator) = Vbias x 2

The reasons this is so is because the force is proportional to the product of the two voltages, but the break down is based on the sum of the two voltages. For example, suppose the airgap can sustain 10kV, and Force(for illustrative purposes) is just the product of the Vbias and Vstator.

Vbias...Vstator…Force
0...........10...........0
1............9............9
2............8..........16
3............7..........21
4............6..........24
5............5..........25
6............4..........24
7............3..........21
8............2..........16
9............1............9
10..........0............0

You can see that if we are limited by 10kV across the airgap, we get maximum force(peak SPL) before breakdown in the airgap when we set Vbias= 5kV and apply 5kV of audio signal to each stator.(out of phase of course, so Vsig=10kV from stator to stator)

Note that if we increase Vbias about 30% above this “optimal” value, we pick up +3dB in sensitivity while losing only -1dB in maximum SPL output capability. I have found this to be a good trade off and generally operate my ESLs 20%-30% above the “optimal”. Increasing much above this and dynamic range starts to suffer. FWIW, Quad ESL57 and ESL63 are both operated at the “optimal” Vbias levels.

Here is the reference.
P. J. Baxandall, “Electrostatic Loudspeakers,” in
Loudspeaker and Headphone Handbook, J. Borwick, Ed.
(Reed Educational and Professional Pub., Woburn, MA,
1998), pp. 106–195.

 

[bolserst]

I guess my Question is that if the ionization of the air is what limits our Peak SPL, can the efficiency still increase as long as we can contain the bias voltage without arcing when it is raised?

Ionization/conduction of the air results from exceeding Emax(the critical field strength) for air at some point in the air-gap between stators. Adding insulation does nothing to change this property of air. Typically we think of the electric field between the stators as being uniform, but there are definitely hot spots around sharp edges in stators and this is where the ionization starts. Insulation added to perforated sheet metal stators will reduce the field strength around these sharp hole edges. This allows you to apply higher bias and stator voltage before exceeding Emax. However, adding PVC heat-shrink insulation to round rod stators does not significantly modify the field strength around the stators and as a result the air begins to conduct at essentially the same voltage with or without insulation. The only difference is what happens when the conduction starts(ie arcing/sparking for un-insulated, corona discharge for insulated)

Note that depending on the bulk resistivity of the insulation material, you may lose some of your applied bias voltage across the insulation.(see Attachment#2 of post #23 for more details)http://www.diyaudio.com/forums/planars-exotics/264972-measured-my-esl-attempt-3.html#post4140681
This can confuse things because you may be fooled in to thinking you are applying more bias voltage across the air-gap than you actually are.

The best way I have found to confirm what is going on is to build an un-insulated stator panel with the same stator dimensions to compare SPL output against your insulated panels. Lower SPL is a direct indication of how much voltage is being lost across the insulation. This would tell you how much to increase your applied bias voltage to achieve a desired voltage across the air-gap.

I have made it to 12kv with some leakage starting but it ran and it was even more efficient.
But can we take it to say 15Kv or even 20kv or so of bias and get away with only using like 100v to 400v of drive signal or less and still have an ample amount of SPL?

First, running the bias voltage high enough to ionize the air in the gap is not a desirable situation for your safety and the longevity of your panels. The ozone generated is unhealthy for you to breathe, and attacks most materials and coatings...resulting in eventual failures.

If you still decide to run your panels with air in the gap conducting, you will be losing the low distortion properties of constant charge ESLs as the conduction behaves like a soft limiter, producing odd order harmonics. This is often the reason that things sound like they are continuing to get louder, when in fact harmonic content is just being added. This is easy to measure with single frequency tone testing.

The other thing that happens when the air in the gap ionizes, is that its resistance falls as conduction increases. This changes the ratio of the air/insulation resistances and more bias voltage will be lost across the insulation as the resistance of the ionized air falls.(again, see Attachment#2 of post #23) So, once you have reached a point where you are ionizing the air in the gap, you will have progressively diminished returns with further increase in bias voltage.

 

[geraldfryjr]

Thank you for clarifying this for me Bolserst.

As you recall I didn't have a SPL meter at the time so I couldn't verify the exact numbers of what I was hearing then (2010).

I was using test tones and all I had was my ears, my scope and my eyes to go by.

I was however able to measure accurately the bias voltage and the signal voltages on the HV side using my scope at the same time.

So I do know that my driving signals were quite clean.

I had an idea that what you described is what was happening.

But at those levels I had found that music was quite loud and more than enough to even also make it unbearable to enjoy, especially with just using test tones.

In my actual measured tests I was able to produce +105db with just my 100 watt amplifier at 1 meter!!

I started approaching 108-110db range with my larger amplifier but as you described the air was breaking down and I couldn't tell if was getting louder at a certain point.

Since I use them in a nearfield position this can be significantly higher at times!

At those even higher voltage levels (that were way more then enough to begin with) I ran in to all sorts of stator coating issues.

And, It was the final demise of my great sounding little panels.

It was a hard lesson to learned every time I had to take them apart and try and repair them over and over and over again.

Besides once you start getting up into the +10kv voltage range the charge goes everywhere and gets into everything no matter how hard you try to contain it!!!

I had found that about 8kv to 8.5Kv is the highest practical limit that anyone should be concerned about.

And, Yes, I am quite sure that about 2kv of that voltage is dropped across the stator's insulation resistance.

I wanted to set up a apparatus to prove and measure this but it is just to time consuming for me at this point in time.

Although, I did find out what I needed to know while sacrificing my Second born well built panels from 2003.

It was for the sake of DIY experimentation and research to answer some questions I had, and that it seemed nobody had the answers too at the time, or if they did, Would Not share and tell!! He,he,he,he

When I First started I was using a power supply that only produced 900v of bias and I could barely even get a sound out of the panel it was almost too low even hear it!!

So, it amazed me very much once I started using voltages about 4 to 10 times greater!!

Everyone else that was using voltages in the 2kv or 3kv range and complained that their smaller panels sounded thin or not loud enough. or not enough bass and lacked in overall SPL all together!

I had proved (to myself at least) that if done right that this is not the case for smaller panels.

Even I had my doubts at one point, but I pushed them just a bit more and got what I was looking for!!

Once I got that little panel to be able to compete with the SPL's of the 8" sub that I was using with it ( each with the very same drive voltage level) it was the most amazing sound experience I have ever heard since the First time I enjoyed the sounds I was hearing for the First time coming my Apogee Duette's!!

Better than that!!!

The ESL and the woofer worked in unison as it was just one small compact speaker system.

Just the same as it does (before I finally burned it) in my final setup using a smaller 5.25" Sony driver and it was't even in stereo yet!!

I didn't detect any localization of the drivers (very hardly at all) in either setup!!

It was a Grand learning experience along the way and ESL's are no longer a Big Mystery to me anymore!!

Thanks !!

jer

P.S. Here are pictures of the Two main setups I used throughout all of my tests and experimentation's and of course one of the latest build.