Tube amps and electrostatic speakers

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Yes, but not easily.

"Normal" tubes can be used for swings of a few hundred volts (good enough for e-stat headphones), but ESLs typically require a few thousand volts. So, you can use high voltage tubes and high-power high-voltage supplies. Dangerous stuff, expensive, difficult to stabilize, and very difficult to make reliable.
 
even before "Speaker Builder" (1980?) was published there were several articles in Audio Amateur concerning ESL's and ESL amplifiers -- here's a good reason to go to the audioXpress website, order the CD-ROM and keep Ed Dell in business , issue 3/1977 (geez, my oldest son was born that year!) has a hybrid amplifier using a pair of 8068's as the finals. 600 Volts.

I would take issue with the author's use of a backwards filament transformer to derive the HV -- most transformers don't have the insulation to take the high voltage.
 
Seems like a zillion years ago, but I used TV horizontal output tubes (6DN5's, 35LR6's or one from that club) class A @ 3kv to directly modulate a DC arc (creating the worlds most impractical tweeter). Quite certain it would work for ESLs...it's been done several times.

Be EXCEPTIONALLY careful...a second chance with a high current high voltage supply is highly unlikely. "Normal" ESL power supplies, while dangerous, are fleas by comparison. Build a shorting interlock on the power supply, work with one hand in your pocket, life is more important, etc etc.
Paul
 
Acoustat has made the X, which is a 3 panel system with a direct drive OTL power amp.

The last stage is a SRPP made with 6HB5 (TV sweep tube) tubes.

This is worthwhile to copy, I've done it myself.

But let me warn you again: this is not a beginner project.
The amp and or powersupply can easily kill you !

Dick.
 
I currently designing such kind of amplifier

Be aware, the only 4 diy otl amp I seen on the web use choke load.

A capacitive load+choke load= very bad ringing issue :bawling:

the acoustat X use a SRPP topologie with minimal current trough (about 6mA, not enough!!!) Also, srpp are srpp and in preamp, this configuration don`t have the sonic qualitie that I like:bawling:

The only way to bulid a good otl amp is to use a pure classA and resistor coupling or tube ccs.

That will require twice the voltage and the effiency will be VERY low but that seem to be the only way.

For example, my amp will deliver about 2100VP-P, will run at 2500V and will consume about 1400Watts at idle (like a house heating plate(1500watts))

Martin
 
etalon90 said:
I currently designing such kind of amplifier

Be aware, the only 4 diy otl amp I seen on the web use choke load.

A capacitive load+choke load= very bad ringing issue :bawling:

the acoustat X use a SRPP topologie with minimal current trough (about 6mA, not enough!!!) Also, srpp are srpp and in preamp, this configuration don`t have the sonic qualitie that I like:bawling:

The only way to bulid a good otl amp is to use a pure classA and resistor coupling or tube ccs.

That will require twice the voltage and the effiency will be VERY low but that seem to be the only way.

For example, my amp will deliver about 2100VP-P, will run at 2500V and will consume about 1400Watts at idle (like a house heating plate(1500watts))

Martin


Martin, why do you think that 6ma is not enough??

Also what tubes are you intending to use for your project??

_-_-bear :Pawprint:
 
It is ok to drive els with tubes. I have years ago build a 40W se amp. 2 parralled Svetlana 811-10 transformerdriven from 300B and 6SN7 srpp driver. The amp is driven at 1200V dc. The amp is not transformerless. The outputtubes drives a 1-2.5 centertapped transformer. To give the differentialdrive to the speakers. It works better than differential drive from the tubes. Less ringing and lower outputimpedance to get a better drive tin the high frequencies. My els is 40cm X 200cm and the capacitance is 1200pF.
There is no ringing on the drivesignal and the amp i capable to give out about 500V at 12kHz. I should be enough.
If you use tubes forget all about full output at 20 KHZ.

Before the se project i had an 6550 pp amp with 1 - 7 outputtransformer it had far to high outputimpedance. I modified it to 2-6SA7 regulator/outputtubes kl A driven at 250 V it works ok but much has much less power than the 811 amp.

I use the electrostatics to 300Hz, below that @-6dB a 10" speaker in closed boxes q= 0,5 and below 60-70 Hz @-24dB adjustable cornerhorn with 15" speakers. This systen is not rocking good at rockmusic but very good at symphonic orchestra and jazz.
 
An ESL need a lot of current if you want to acheive 20khz.

My esl (acoustat 2+2) are 1800pf per stator. At 20Khz, that represent 4.4kohm...

Esl need around 1000Vpeak-peak per stator for an appropriate listening level.

So the current needed for chargind the capacitance=1000/4400=227mA

How can you charge the panel if you have only 6ma avaible????

So the good thing is that in music, you never find a 20khz equal level compared to the rest of the spectrum... So you can divide this number by 2 or 3.

My amplifier will use 8 845 (4 parallel single ended, one for each stator). The current will be 160mA per output section (80ma per tube).

from calculation, I can hope -2Db at 18Khz at full output (2100Vp-p)

My output impedance will be approx 1Kohm


the stage will be like that. A 100% otl!!!

417A(rc)-->EQ-->10Y(lc)--> parafeed centertaped choke for phasesplit-->vt62(LC)-->845(rc).

I curently in the stage of buying the parts
 
An ESL need a lot of current if you want to acheive 20khz.
So the current needed for chargind the capacitance=1000/4400=227mA
How can you charge the panel if you have only 6ma avaible????

Martin, you are right, this is not possible.

So the good thing is that in music, you never find a 20khz equal level compared to the rest of the spectrum... So you can divide this number by 2 or 3.

This is also true, but i'm very shure that the division number is much larger than 3. Don't know howmuch, but it must be because when listening to an Acoustat X with the OTLamp there is nothing wrong with the sound. (on the original unit swap out the opamp for a TL074, which is better than the original opamp).

And the Acoustat X is made of 3 panels in parallel !
I would suggest to listen to one before judging.

In normal music the spectrum above 12kHz is besides hardly used, also much less in amplitude.

Hence, testing with test tones will not work with the mentioned SRPP, but music is no problem.

Ofcourse the described class A amp from Martin would work and sound very well, but it's a bit overkill.

Dick.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.