Head-Fi.org › Forums › Summit-Fi (High-End Audio) › High-end Audio Forum › KOSS ESP-950 Thread
New Posts  All Forums:Forum Nav:

KOSS ESP-950 Thread - Page 34

post #496 of 1421
Quote:
Originally Posted by kevin gilmore View Post

 

>The 2sj79 drives the cathode.  This is grounded grid drive. Way more linear than common cathode with input signal at the grid.

>Output from the plate.

 

The characteristic curves change depending on which port the input signal is applied to?  News to me...

 

Common grid is used to avoid the so-called "Miller capacitance" where the interelectrode capacitance is amplified by the amplification factor.

 


Edited by sorabji - 10/1/12 at 5:54pm
post #497 of 1421
Thread Starter 
Quote:
Originally Posted by spritzer View Post

The T2 is deceptively simple at it's core.  The SRX circuit was clearly the inspiration as that's the best all tube electrostatic amp circuit I've ever seen.  Ultra stable, high gain so feedback isn't a problem and the cross coupled input stage is very clever indeed.  It's an all tube design so it has some major limitations.  It has an inherent DC-offset which needs to be countered with the PSU or a bias suppply, it's not fully DC coupled and resistors do not make for good current sources.  So they took that circuit and used SS parts to get rid of all the limitations they had to contend with in 1969.  Some of what they did is very, very clever... 

Yeah some of this is what I find a little difficult to read in the schematic. Things like CCS plate loads, etc, I follow in principle but looking at it on a schematic doesn't result in the kind of feel for things that I get from more pedestrian designs like plain old complementary-symmetry amps, or basic tube stages.  I understand in general what these solid state parts are doing to linear-ize the circuit, I just don't grok exactly how it gets done.  Like I said, maybe it's all explained in that other thread, I will have to just wade through that.

post #498 of 1421
Quote:
Originally Posted by sorabji View Post

My ESP-950's finally came in from headphones.com.

This is a very nice product, and it's too bad they don't get more due recognition. (due to audiopile hype-predjudice, imo)     They are much more neutral than my Stax Lambda Signatures and just as comfortable, perhaps more so.  The pads appear to breath a little better.

Going to order connectors so I can hook them up to my homebrew electrostatic amp that I built almost 25 years ago. +/-  750 volt supply with MOSFET buffers for each diff amp using 6SL7 and 6BL7 tubes.

Tried some minphase EQ as well.  Responds pretty well to it.  Better than the Stax, probably due to less high Q resonances @ high frequencies. (assuming purrin's plots are accurate)

I'm not sure if 750V is safe for the ESP/950. I remember reading about a chap blowing up the stators with 700V. redface.gif
post #499 of 1421

http://offer.ebay.com/ws/eBayISAPI.dll?ViewBids&_trksid=p2047675.l2565&rt=nc&item=300786744820

 

was thinking of bidding, if below 500. but its already over it . :O

 

i mean what are people thinking ? its clearly a very old pair ( look at the color of grills ).

 

am looking for another pair to wood-ify or alu-mod them, but not at this price.

post #500 of 1421
Quote:
Originally Posted by obobskivich View Post


I'm not sure if 750V is safe for the ESP/950. I remember reading about a chap blowing up the stators with 700V. redface.gif

Responsibility is in the hands of the user by the sensible use of the volume control.  Headphones can be repaired. Inner hair cell nerves cannot.

 

The reason the voltage is so high is b/c it just happens to be what the surplus transformer I decided to use produced.  A nice side benefit is higher gain per stage and better linearity from the use of a bigger load resistor.

 

The plates are at ~0V or pretty close to it most of the time with no signal. (if that's what your wondering about)

post #501 of 1421
Quote:
Originally Posted by singh View Post

http://offer.ebay.com/ws/eBayISAPI.dll?ViewBids&_trksid=p2047675.l2565&rt=nc&item=300786744820

was thinking of bidding, if below 500. but its already over it . :O

i mean what are people thinking ? its clearly a very old pair ( look at the color of grills ).

am looking for another pair to wood-ify or alu-mod them, but not at this price.


I think the color of the grills is just the camera flash. IIRC even the original Stereophile review images they were black, and that's from ~1992. I think it's sort of like how the T70 can appear purple in some images, and they are not.
Quote:
Originally Posted by sorabji View Post

Responsibility is in the hands of the user by the sensible use of the volume control.  Headphones can be repaired. Inner hair cell nerves cannot.

The reason the voltage is so high is b/c it just happens to be what the surplus transformer I decided to use produced.  A nice side benefit is higher gain per stage and better linearity from the use of a bigger load resistor.

The plates are at ~0V or pretty close to it most of the time with no signal. (if that's what your wondering about)

When you say 750V, do you mean bias, or signal? I'm talking bias, which is what I interpreted from your statement (perhaps wrongly - it's been a long weekend). If you're talking signal, they should be able to run that without blowing up just fine. But 700V bias will, as far as I know, kill them. redface.gif
post #502 of 1421

Bias is 600V

 

No, you don't want arcing!

post #503 of 1421
Quote:
Originally Posted by obobskivich View Post


I think the color of the grills is just the camera flash. IIRC even the original Stereophile review images they were black, and that's from ~1992. I think it's sort of like how the T70 can appear purple in some images, and they are not.

 

I meant, the color faded with time, not that it came in different flavor back then .

post #504 of 1421
Quote:
Originally Posted by obobskivich View Post

When you say 750V, do you mean bias, or signal? I'm talking bias, which is what I interpreted from your statement (perhaps wrongly - it's been a long weekend). If you're talking signal, they should be able to run that without blowing up just fine. But 700V bias will, as far as I know, kill them. redface.gif

Sorry, what I actually meant to say was the supply for the amplifier is +/- 750 volts.

 

Here's a question for those who've built electrostatic amps:  What is the effect of the series resistor just before the diaphragm?  I've used up to 1 gig ohm.  Thing is, it's hard to measure what the voltage on the diaphragm actually is with such a large R.  There must be some trickle current leakage just due to the atmosphere. 

post #505 of 1421
Quote:
Originally Posted by sorabji View Post

Bias is 600V

No, you don't want arcing!

Oh yeah, in that case you'll be fine. smily_headphones1.gif

singh,

I'm not so sure about that. I guess whoever buys them can post here and tell us though, lol.
post #506 of 1421
Quote:
Originally Posted by sorabji View Post

Sorry, what I actually meant to say was the supply for the amplifier is +/- 750 volts.

 

Here's a question for those who've built electrostatic amps:  What is the effect of the series resistor just before the diaphragm?  I've used up to 1 gig ohm.  Thing is, it's hard to measure what the voltage on the diaphragm actually is with such a large R.  There must be some trickle current leakage just due to the atmosphere. 

 

It changes the charge rate of the diaphragm and it's more to do with distortion than anything else.  We use 5M for most of the 'stats (Stax, Koss, Beyer) but 10M for the Sennheiser models. 

post #507 of 1421

Thanks for the memory jog on the effect of the series R.  Been a while since I thought about it much.

post #508 of 1421

the force driving the membrane is Q*V - if extra charge can move on/off the membrane at audio frequencies then the sensitiviy is modulating at audio frequency giving 2nd order distortion

 

ideally the diaphram itself has >100 MegaOhm/sq sheet resistance to keep the charge from moving around at audio frequencies as the diaphram moves

 

the series R to bias the membrane coating should still be very high too in case the ideal of high membrane sheet resistance isn't met the external RC time constant should be as low as practical to prevent nonliner distortion


Edited by jcx - 10/3/12 at 6:10pm
post #509 of 1421

Back down in price now on Amazon, ordered a pair earlier this afternoon.  

 

Edit: Cancelled my first order @ $750, was able to score them later in the evening @ $650!


Edited by sphinxvc - 10/4/12 at 5:01am
post #510 of 1421

I would hope so, and not a penny more.smile.gif

New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: High-end Audio Forum
Head-Fi.org › Forums › Summit-Fi (High-End Audio) › High-end Audio Forum › KOSS ESP-950 Thread