Tubes and ESL's - diyAudio
Go Back   Home > Forums > Loudspeakers > Planars & Exotics

Planars & Exotics ESL's, planars, and alternative technologies

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 7th June 2015, 09:01 PM   #1
diyAudio Member
 
Join Date: Mar 2011
Default Tubes and ESL's

Today was assigned to another exercise in futility. Organizing the lab. By good fortune I found a box of stuff that I haven't looked into for a number of years. I quickly forgot about the mission at hand and was transported back to those halcyon days of yesteryear. A couple of beers helped in the transportation effort. Anyhow, contained within the box were the mortal remains of an audio project from my early days of tube building. This particular boondoggle was aimed at the construction of an electrostatic loudspeaker. Well, more than a speaker (anyone can build an ESL), but rather an ESL system.

My plan was to drive each speaker with a DIY triode that could hold off 10 KV and drive the panels directly off the plates. This output tube would then be controlled by a DIY planar electrode triode with lower voltage specs. The goal was to make a small bookshelf size pair of panels that would easily fit on a desktop or workbench and spray the area with the glorious sounds that can only come from planar drivers.

To enhance this madcap scheme, I decided to make the diaphragms from stretched Mylar films that were then coated with a few micro-inches of evaporated copper. This coating was applied in a vapor deposition bell jar system. A coating applied by this technique is extremely controllable as to thickness and placement.

I decided to test the tubes to see if they were still under vacuum and was pleased to find that the filaments still work and the the vacuum left inside could still hold off high voltage. I was surprised that there was any life left in them because the tubes were sealed with epoxy when they were constructed. The panels are still useable for testing and I think I'll resurrect the project now that I'm into building vacuum tubes again. If anyone is interested in this reprise I'll report on any progress.
  Reply With Quote
Old 8th June 2015, 12:17 AM   #2
diyAudio Member
 
Join Date: Jan 2008
Location: Toronto and Delray Beach, FL
For many years I said nothing can compare to direct high voltage drive for ESLs using a Sanders-like amp.

(I've recently taken my 1975 ESLs out of the storage and am driving them with transformers and mic-based DSP EQ. Not sure what I have now is inferior to direct drive, but hard to compare listening today to memory maybe 10 yrs back when I last used my direct drive amp.)

Ben
__________________
Current projects: motional feedback woofing and large ESL panels. HiFi aspirations since 1956
  Reply With Quote
Old 8th June 2015, 01:54 PM   #3
diyAudio Member
 
bolserst's Avatar
 
Join Date: Jun 2005
Location: Fort Worth, Texas
Quote:
Originally Posted by George Schmermund View Post
...I decided to make the diaphragms from stretched Mylar films that were then coated with a few micro-inches of evaporated copper...
Was there a particular reason you chose copper for coating the diaphragm? Generally it is desirable to have a highly resistive coating so the ESL panel operates in the low distortion constant charge mode. Copper is one of the most conductive element there is. I guess I would also be concerned about the copper oxidizing.

Quote:
The panels are still useable for testing and I think I'll resurrect the project now that I'm into building vacuum tubes again. If anyone is interested in this reprise I'll report on any progress.
Am I understanding you correctly that you are building your own custom vacuum tubes designed specifically for direct drive of ESLs?
If so, I would be interested to see your progress reports in this thread.
  Reply With Quote
Old 8th June 2015, 05:43 PM   #4
diyAudio Member
 
Join Date: Mar 2011
Bentoronto - I'll have to agree with your assessment of the sound quality of directly driven ESL's. The small panels that were constructed sometime back were truly astonishing. I know that I'm on thin ice making my own claims, of course!
  Reply With Quote
Old 8th June 2015, 06:49 PM   #5
diyAudio Member
 
Join Date: Mar 2011
Bolserst - I chose to use copper because the vacuum vapor deposition system was already set up to deposit Cu onto some planar magnetic Mylar panels that I had been experimenting with. There are a lot of myths and legends that accompany the rules of constructing ESL panels. I chose my design parameters based on the B&k microphones that I use for lab measurements. These microphones use a solid metal foil as the diaphragm. To calibrate these microphone cartridges and preamps I use a B&K electrostatic actuator. This is about as good as it gets for linearity and low distortion. The actuator's frequency response deviation is +/- 0.6 dB from 2 Hz To 200 kHz. I just decided to experiment with a much larger version of it. I must say that I find many of the assumptions about how electrostatic and planar magnetic speakers work to be rather amusing in some of the descriptions I've seen. To each his own!

As far as the Cu oxidizing, It could be an issue over a long period of time. That's a whole other topic. My plan was to eventually replace the Cu with gold. At the time Au looked very cheap compare to the prices today. Even so, with today's prices the amount needed for depositing a few micro-inches of it are still compelling.

I've made many vacuum tubes over the years and still find if very rewarding. Here's a link to some of them and their characteristic curves. These are just toys from the past. The lasted variety will be much improved.

Testing DIY triodes

Last edited by George Schmermund; 8th June 2015 at 06:56 PM. Reason: clarity
  Reply With Quote
Old 9th June 2015, 11:13 PM   #6
diyAudio Member
 
bolserst's Avatar
 
Join Date: Jun 2005
Location: Fort Worth, Texas
Quote:
Originally Posted by George Schmermund View Post
…I chose my design parameters based on the B&k microphones that I use for lab measurements. These microphones use a solid metal foil as the diaphragm.
ESL microphones like the B&K units operate in a different manner than typical ESL loudspeakers, so using them as a guide may not be the best idea. For the B&K ESL pressure microphones(sealed back chamber), the upper bandwidth limit is set by the diaphragm resonance frequency. Hence the use of metal foil that can hold the required high tension for a 40kHz resonance and be stable for years. Below resonance, the output of the microphone is directly proportional to applied pressure. This is because the displacement amplitude of a stretched membrane for constant applied force is independent of frequency below resonance. So, the motion of the microphone diaphragm is exceedingly small over the whole operating range. This is why the distortion of the B&K microphones is low even with a fully conductive diaphragm operating in constant voltage mode.

ESLs used as dipole loudspeakers operate differently from ESL pressure microphones. For example the diaphragm resonance sets the lower frequency limit(not HF limit as with microphone). Also, for flat far field pressure response the diaphragm displacement amplitude must increase as you go down in frequency. Once diaphragm excursions start to exceed about 15% of the air-gap width, the inverse square law distortion starts to show up. If you keep the diaphragm excursion down by limiting the lower bandwidth of your design, constant voltage mode works just fine without producing too much distortion or diaphragm instability.

Quote:
I must say that I find many of the assumptions about how electrostatic and planar magnetic speakers work to be rather amusing in some of the descriptions I've seen. To each his own!
I hear ya, but hopefully constant charge operation of ESLs didn’t fall into that category for you.
The theory is well documented in texts/papers by Hunt, Walker, Baxandall, etc... and experiment confirms the fundamental inverse square law of electrostatic forces when operating in constant voltage mode. Attached image shows a comparison of the diaphragm force vs diaphragm position for the two modes of operation. As mentioned above, keep excursion down and constant voltage mode works just fine.

Quote:
I've made many vacuum tubes over the years and still find if very rewarding. Here's a link…
Very cool, Thanks!
Several of those you built look like ones from the 1920s radios I collect.
I recall building a triode in AP physics class in high school...used a test tube for the glass envelope.
What fun
Attached Images
File Type: png Force_CQ_vs_CV.png (16.2 KB, 200 views)

Last edited by bolserst; 9th June 2015 at 11:15 PM.
  Reply With Quote
Old 11th June 2015, 01:44 AM   #7
diyAudio Member
 
Join Date: Mar 2011
Bolserst - Thank you for the discourse. As mentioned, the ESL's were meant to be something that could produce a pleasing level of reproduction in a small space like a desk, office cubicle, or workbench. The diaphragms measure 7" X 9". I had no interest in typical listening room sound pressure levels. I have planar magnetic drivers for that job.

Following the model of the B&K actuator, but reversing the order, the bias supply was connected to the diaphragm and a single stator was used as the ground plane. I could then run the bias up until the diaphragm was at a desirable displacement position and then modulate the HV. This allows the diaphragm to use its own spring tension to act as the restoring force without the need of a second stator. Sort of like a single ended class A amp. Yes, the displacement was small, but the SPL was equivalent to a good table top radio that isn't being over driven.

As a side note I'd like to point out that typical B&K laboratory microphones are all vented behind the diaphragm. If they weren't they would be barometers and never stay in calibration. In fact, many of the vacuum gauges that I use are capacitance manometers. They are extraordinarily linear and indifferent to the gas or non-condensable vapor species they are measuring.

I would also like to thank you for confirming that a working triode can be easily fabricated even by kids in high school!
  Reply With Quote
Old 11th June 2015, 06:05 AM   #8
Calvin is offline Calvin  Germany
diyAudio Member
 
Calvin's Avatar
 
Join Date: Nov 2004
Location: close to Basel
Hi,

SE-ESLs have been used in thousands on old radios as tweeters.
Due to the non-linear drive excursions must be kept small for the THD to remain reasonably small.
This restricts the useable bandwidth of a SE-ESL to the upper-mids/highs.
As one probabely needs a means of protection for the delicate membrane anyway, a second stator comes handy, improving the behaviour of the panel at the same.

jauu
Calvin
  Reply With Quote
Old 11th June 2015, 11:10 PM   #9
diyAudio Member
 
bolserst's Avatar
 
Join Date: Jun 2005
Location: Fort Worth, Texas
Quote:
Originally Posted by George Schmermund View Post
Following the model of the B&K actuator, but reversing the order, the bias supply was connected to the diaphragm and a single stator was used as the ground plane. I could then run the bias up until the diaphragm was at a desirable displacement position and then modulate the HV. This allows the diaphragm to use its own spring tension to act as the restoring force without the need of a second stator.
Just curious, did you use a rear chamber like the B&K? or were you operating your ESL as a dipole.

Note that for single ended ESLs, the bias induced spring tension(often called negative compliance in texts) is also non-linear. In fact if I recall the Hunt text correctly the displacement position involves a cubic function of the bias voltage. You will notice this as a rapidly increasing diaphragm deflection beyond a certain point as bias voltage is linearly increased. But, as Calvin mentioned and you experienced, performance can still be quite good as long as you keep diaphragm motion small. If/when you are in interested in experimenting with higher SPL, I would recommended taking Calvin's advice. Add a second stator and a high value resistance between bias supply and diaphragm. The negative compliance term becomes linear, and distortion performance is improved when diaphragm motion is not negligible.

Quote:
As a side note I'd like to point out that typical B&K laboratory microphones are all vented behind the diaphragm. If they weren't they would be barometers and never stay in calibration.
Yeah…I originally had a comment in my post about the mic back chamber having a tiny vent tube but then removed it to avoid overcomplicating things with unnecessary details. But now that you have brought it up…haha

In case others are curious about the vent tube, attached is a description and pics pulled from B&K literature.
Attached Images
File Type: png BKmic_backchamber_vent.png (564.6 KB, 163 views)
  Reply With Quote
Old 12th June 2015, 06:44 PM   #10
diyAudio Member
 
Join Date: Mar 2011
Yes, I did use a backplate, but since the diaphragm was already electrostaticly offset I didn't consider it to be stator as such. It was operated as a dipole radiator.

I'd like to do a 'reset' here and bring us back to what the project was all about. I wanted to see how well the new tube design worked and needed to have an ESL to test it with. The speaker project was a default design for that purpose. I simply used what was around the lab and saw the microphone paradigm as a place to start. The real project was to design a DIY HV triode that could replace the transformers typically used with ESLs.

I was hoping that someone here might be interested in that aspect of the experiment. I didn't anticipate that the thread would be shunted into a protracted discussion about the theories and opinions of panel designs.
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
ER Audio ESL III Kit, and ESL questions in general wixy Planars & Exotics 27 29th July 2014 11:14 PM
Driving ESL with valves/tubes Darth Vader Planars & Exotics 98 9th June 2013 07:41 PM
How is a ESL driven directly by tubes? ak_47_boy Planars & Exotics 3 6th November 2006 07:44 PM
Tubes for direct-driving ESL Prune Tubes / Valves 32 14th March 2004 02:59 PM
ESL amp: tubes vs solid state: where is the winner? Milo Tubes / Valves 43 28th April 2003 09:51 AM


New To Site? Need Help?

All times are GMT. The time now is 10:37 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2015 DragonByte Technologies Ltd.
Copyright ©1999-2015 diyAudio

Content Relevant URLs by vBSEO 3.3.2