Single Stator Electrostatic Headphone - issues? - Page 5 - diyAudio
Go Back   Home > Forums > Amplifiers > Headphone Systems

Headphone Systems Everything to do with Headphones

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 10th December 2012, 09:21 AM   #41
diyAudio Member
 
Join Date: Oct 2007
Location: Aarhus, Denmark
Hi Kevin

& thanks for sharing your thoughts and ideas on electrostatic amplifiers & headphones. I've decided to buy a couple of MPSA42/92's to try out but otherwise will be using the KSC3503 and KSA1381 from Fairchild. Both are 300 VDC devices, have higher power handling capacity, and lower current rating + slightly better data.

I will also be using only one signal stator and the other will be grounded - this has been discussed earlier in the thread. The idea is to keep the amplifying circuitry as simple as possible and single-ended. FYI I have attached an image of the amplifier schematic. I've used a very similar circuitry before - careful selection of components is essential for sound-quality but otherwise it IMHO is remarkably fluid, transparent & not least musical sounding.

Best regards,

Jesper
Attached Files
File Type: pdf HTamp.pdf (10.9 KB, 19 views)
__________________
... Being mindful about sustainability also in my audio endeavors matters to me ... paying CO2 - and a quite fair pollution compensation helps create a good listening "energy" ...
  Reply With Quote
Old 10th December 2012, 05:06 PM   #42
Struth is offline Struth  Canada
diyAudio Member
 
Join Date: Aug 2012
Location: books at londonpower.com
Hi Guys

The MPAA42/92 are quite linear over the current range required for a stator amp, so don't discount them too quickly.

What I have read suggests that the dual stator system is far superior to a single stator on pretty much every count: distortion is lower, drive power can be reduced for a given loudness, maximum loudness is increased.

The circuits for a dual-stator system are dead simple. A very decent single-input balanced output can be attained with just one low-voltage BJT and two mosfets. Complexity is relative and in most cases improves performance.

You can go to a three-block approach with identical stator drivers made with a single BJT and single mosfet, plus a single BJT/jfet/mosfet balanced driver. All three blocks can be easily improved by adding a single BJT to the stator drivers and two BJTs to the balance circuit.

In your circuit there are split rails, so why not eliminate the bias divider for Q1's base and tie it to ground through a base leak? You can apply feedback from the output back to Q1's emitter to have a very linear circuit. If Q1 was turned into a diff amp, then you could stabilise the DC at the output quite easily.

Note also that the 24k Rs used to load the EF stage and the VAS will be dissipating over 1W each at idle. Both restrict the negative swing and signal symmetry. You could split the VAS load and bootstrap it to the output for better symmetry. The 820R in the EF collector is not needed.

Have fun
Kevin O'Connor
__________________
London Power / Power Press Publishing
www.londonpower.com

Last edited by Struth; 10th December 2012 at 05:12 PM.
  Reply With Quote
Old 11th December 2012, 02:21 PM   #43
diyAudio Member
 
Join Date: Oct 2007
Location: Aarhus, Denmark
Hey Kevin,

Thanks again for commenting & sharing your thoughts - might I ask you a couple of questions on this ... ?

Quote:
The MPAA42/92 are quite linear over the current range required for a stator amp, so don't discount them too quickly.

What I have read suggests that the dual stator system is far superior to a single stator on pretty much every count: distortion is lower, drive power can be reduced for a given loudness, maximum loudness is increased.
I am not discounting the MPSA42/92s but would like to listen to them to hear what their characteristics are. They look interesting.

About the double stator headphone, yes, I've also learned that it's a superior principle. However, I am also building a double stator headphone with the slight twist that I ground one of the stators. Same advantages as normal double stators (as I understand it), yet can be a single-ended amplifier circuitry (my preference) with a 6 dB (one stator) loss of sensitivity. Which is ok with me.

Quote:
In your circuit there are split rails, so why not eliminate the bias divider for Q1's base and tie it to ground through a base leak?
Do you have an example of this (maybe a link to a schematic)? I haven't heard of it before ...

Quote:
Note also that the 24k Rs used to load the EF stage and the VAS will be dissipating over 1W each at idle. Both restrict the negative swing and signal symmetry.
I'd rather keep it simple without feedback or bootstrapping (listening tests from someone I know & trust on this - high-end audio manufacturer & myself). Would it be beneficial to use a PNP transistor for the EF stage to give more balance of loads & symmetry?

Greetings,

Jesper
__________________
... Being mindful about sustainability also in my audio endeavors matters to me ... paying CO2 - and a quite fair pollution compensation helps create a good listening "energy" ...
  Reply With Quote
Old 11th December 2012, 03:44 PM   #44
Struth is offline Struth  Canada
diyAudio Member
 
Join Date: Aug 2012
Location: books at londonpower.com
Hi Guys

Note it is MPSA42 and MPSA92 - one 'A' not two, a typo.

I think it is safe to say that people who do not understand how circuits function believe that only simple circuits can sound good. They should keep in mind what Einstein said about simplicity: Things should be as simple as needed but no simpler.

Selecting BJTs by how they sound is quite ridiculous in my view, as is a fear of feedback or bootstrapping. There is absolutely no such circuit as a "zero feedback" one - it simply does not exist in this universe. The inherent imperfections of manmade devices and physics as we understand it makes it impossible for there not to be feedback in circuits where there is no "designed in" feedback. Your sim uses local feedback, apart from intrinsic feedback within active elements. So, what you mean to say is that you do not want to use global feedback. Should we also assume you are afraid of coupling capacitors?

The purpose of a playback amplifier is to not corrupt the sound; if you can hear what the circuit is doing then that condition is not being met.

Dual supplies provide advantages inasmuch as an output and input cap can be eliminated. Your circuit uses a critical resistor choice to eliminate both, when a simple circuit change does this effortlessly without critical values. DC coupling the input makes no sense whatsoever. DC offsets from preceding circuits will upset the output of the next block, and so on. Besides, subsonic filtering is an asset when driving speakers or headphones.

If you want the simplest design, then go to a single rail. Two BJTs will provide the gain needed along with DC stability and low THD. You could go to a single BJT but THD will be higher. In both cases, coupling caps are needed for input and output. At the output side, having the cap in place allows the greatest output from the ESL without any critical matching, where not having the cap places a DC voltage on each stator that effectively reduces the bias value. Without output caps, the DC condition at idle must be accurately balanced.

Clinging to single-ended restricts potential for performance and musical enjoyment. Everything is more critical when you live in that camp and you get a skewed view of most aspects of circuit design, component specs, sensitivity to parameter spreads, etc. SE has a pretense of simplicity but extracting acceptable performance from it is much more complicated than with push-pull.

Grounding one stator is definitely NOT the same as push-pull drive. The only way it could be made equivalent would be to modulate the plate with a half-size signal and apply a double-size signal to the driven plate. This requires a much higher voltage drive circuit which is more likely to exhibit higher distortion than a lower-voltage circuit. The driver could modulate the plate bias through a voltage divider, but why not do things in a less expensive way that provides better performance?

Everyone chooses their own path. The path can change and so do the choices. This does not mean we have "compromised our principles", rather that we have learned and adapted.

Have fun
Kevin O'Connor
__________________
London Power / Power Press Publishing
www.londonpower.com

Last edited by Struth; 11th December 2012 at 03:53 PM.
  Reply With Quote
Old 11th December 2012, 10:33 PM   #45
Struth is offline Struth  Canada
diyAudio Member
 
Join Date: Aug 2012
Location: books at londonpower.com
Hi Guys

Here is a simple two-transistor driver. Q2 is a mosfet to handle the dissipation and voltage. Q1 does not have to be high voltage, so can be high gain instead.

R2 is not strictly needed. All the caps are low values so can be film types.

Have fun
Kevin O'Connor
Attached Files
File Type: pdf ESL-d1cfp.pdf (39.1 KB, 32 views)
__________________
London Power / Power Press Publishing
www.londonpower.com
  Reply With Quote
Old 12th December 2012, 05:40 PM   #46
Struth is offline Struth  Canada
diyAudio Member
 
Join Date: Aug 2012
Location: books at londonpower.com
Hi Guys

Upon further reflection, there is a simpler way to salvage the grounded-stator dual-stator ESL driven by an SE circuit.

Most such drivers use resistive loading, just like the one I posted above. This resistor can be split into two equal portions to generate the half-signal required to modulate the plate. A cap ties the signal to the plate.

Note that the plate is usually tied to the bias supply via a high-value resistor, most often about 10M. This provides a "soft" point to connect the modulating signal.

The ESL Headphone Driver note in Designer Notes vol.1 (out Feb.2013) explains in detail why the grounded-stator is nearly as bad an idea as the single-stator ESL element.

Jesper, I did not mean to offend you with my previous post and apologise for any hurt feelings. It is just that certain paradigms grow wearisome after many years of seeing them so many people be crippled by them

Have fun
Kevin O'Connor
__________________
London Power / Power Press Publishing
www.londonpower.com
  Reply With Quote
Old 17th December 2012, 06:49 AM   #47
diyAudio Member
 
Join Date: Oct 2007
Location: Aarhus, Denmark
Hi Kevin - good morning.

Now the weekend has passed and energies are up a bit so I have more time to answer your posts. First: thanks for considering & answering ;-)

When I read what you write I reckon that our experiences in audio may have been different. I worked for many years in an audio store which sold both consumer audio system as well as higher end/high end audio components e.g. B&W Nautilus (the Nautilus), 800, 801 etc. The amplifiers we sold were less high end (the best probably were Holfi and then Nakamichi), although from time to time we had the chance of listening to e.g. VTL, Mark Levinson, Lyngdorf Audio (TACT), Harman Kardon, etc. - due to being in the field. We also carried Audioquest and TARA Labs cables including the top models from TARA Labs.

During this period I personally experimented with component mods & amplifiers and built quite some - e.g. inspired by L'Audiophile's Le Monstre & 30 watts class A and Holfi's designs (that were quite well regarded in this part of Europe). I experimented with and/or listened to complementary outputs, differential inputs, single-ended inputs & complementary outputs and eventually all single-ended circuitry with no capacitors in the direct sound path. To my ears these topologies - with BJT transistors - sounded better by a very clear margin. However, I am aware that other paths are possible and may give satisfactory results to other people - given their preferences.

Hope this helps in clarifying my choices. Please also observe that I'm looking at what you have suggested with interest, yet from the perspective that my previous experiences have given me. To me that is fine.

Regarding your comments on grounding one of the stators for a single-ended ESHP I can't quite understand what you write... In case you are interested in sharing a schematic I'd be pleased to see it - otherwise I may get back to this in February when your (?) magazine is out. Can I ask you what this magazine will otherwise cover? And is there a website?

Best regards,

Jesper Mønsted
__________________
... Being mindful about sustainability also in my audio endeavors matters to me ... paying CO2 - and a quite fair pollution compensation helps create a good listening "energy" ...
  Reply With Quote
Old 28th January 2013, 12:34 AM   #48
Struth is offline Struth  Canada
diyAudio Member
 
Join Date: Aug 2012
Location: books at londonpower.com
Hi Guys

Upon further further reflection - a double reflection but hopefully not a standing wave - driving the plate won't really work.

Since the diaphragm is supposed to have a "high resistance" surface inasmuch as the charge distributed over it is supposed to stay in place, superimposing an AC voltage on the DC voltage is likely to cause distortion.

Grounding one stator of a dual-stator element is distinctly NOT the same as driving it push-pull. The idle condition for both is the same. With a signal, there is a constant tug towards the grounded stator with a variable push and pull from the driven stator. This is really no different than having only one stator except that the idle position of the diaphragm is correct.

You can still use SE circuits with the dual-stator element if that is your desire - just build two per cup and drive one out of phase.

Stax's diff amp with diff feedback is elegant but a bit compromised inasmuch as half the signal is by way of passive pull-up. But... it was easy to do back in the day. Modern devices allow simple true push-pull circuits that can direct drive the stators. These can still be diff amps with diff feedback, or circuits using common-mode feedback. In either case, it is a single input and dual outputs. The three-block system with one inverter and two drivers is simpler conceptually.

Have fun
Kevin O'Connor
__________________
London Power / Power Press Publishing
www.londonpower.com
  Reply With Quote
Old 29th January 2013, 09:36 AM   #49
diyAudio Member
 
Join Date: Oct 2007
Location: Aarhus, Denmark
Hi Kevin,

... so this electrostatic headphone project still simmers in your thoughts ...

Thanks again for commenting and suggesting .... maybe I can ask you a question to this end? When considering distortion levels for a single stator (grounded stator) version vs. a push-pull version of an electrostatic headphone does the playback level of the music then matter? I.e. might it be so that the lower the playback level the less the distortion difference will be between a grounded stator & a push-pull headphone? I play very low so it would even out the differences between the two approaches.

Hope things are well

Jesper
__________________
... Being mindful about sustainability also in my audio endeavors matters to me ... paying CO2 - and a quite fair pollution compensation helps create a good listening "energy" ...
  Reply With Quote
Old 29th January 2013, 03:40 PM   #50
Struth is offline Struth  Canada
diyAudio Member
 
Join Date: Aug 2012
Location: books at londonpower.com
Hi Guys

Jesper, if you look at the drive circuits used for most ESL headphones, you see that the individual outputs are almost always single-ended. All the tube stuff and all the Stax stuff at least up to the mini driver uses a single active element working against a resistive load - so an SE output.

This could be bad hum-wise were it not for the fact that there are two such outputs per cup driving the two stators with an identical noise signal. So just as a push-pull tube amp achieves hum canceling electromagnetically in the output transformer, the push-pull ESL drivers achieve hum balance electrostatically in the ESL element itself. This allows a slightly noisy supply to be used effectively, keeping overall cost and size down.

The asymmetric drive of having just a single stator or a grounded stator will cause distortion in the ESL element at all listening levels.There should be no harm in trying it. At least with the dual-stator element you will have the option of push-pull drive.

ESL drivers are as close to perfect as one can get in a speaker. Push-pull drive is the best form of this concept and many would say that it is an essential aspect in attaining performance called "high fidelity". The balanced driver circuits provide an easy means to eliminate and/or greatly reduce many distortions from the electronics itself, and in the case of PP vs SE, of the ESL as well.

Have fun
Kevin O'Connor
__________________
London Power / Power Press Publishing
www.londonpower.com
  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
Single stator electrostatic headphone. gentlevoice Headphone Systems 16 2nd February 2010 11:33 AM
Single stator design yrmeli Planars & Exotics 6 11th November 2005 07:42 AM
single stator esl. tade Planars & Exotics 2 4th July 2005 05:00 PM
Electrostatic stator design martinv Planars & Exotics 16 6th September 2003 06:46 PM
DIY Electrostatic Stator insulation schpeltor Planars & Exotics 7 13th February 2002 07:11 PM


New To Site? Need Help?

All times are GMT. The time now is 12:16 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright ©1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2