Electrostatic speakes - DC offset, clipping and amplifier oscilation

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

I've looked all over the net and haven't managed to find any relevant info on how amplifier malfunction can affect electrostatic speakers

couple of questions

1) DC offset - how does it affect the electrostatic speakers? will the membrane be off center when DC offset is present?

2) Clipping, what does clipping do to the membrane? am I correct that since the cooling is different for an electrostatic that nothing bad can occur?

3) What will amplifier osculation (sub and super sonic) do to the electrostatic speakers? This third question is something i'm really interested in

Thanks to anyone that can provide answers
 
Administrator
Joined 2007
Paid Member
Good questions. It will be interesting to see everyone's thoughts on this.

Now as far as I know, electrostatics as we think of them (large panels like the Quads) always use a transformer to step the audio signal from the amp up to a much higher voltage that can deflect the diaphragm. The transformer means that DC offset can not be passed to the diaphragm which means that a DC offset has no effect.

Edit... I'm referring to a normal steady state DC offset. An offset that appears abruptly such as an amplifier stage suddenly applying say 30 or 50 volts DC could well cause damage as that would pass through the transformer as a pulse and move the diaphragm abruptly.

Clipping I'm not sure about. What I think has to be avoided is excess movement of the diaphragm so that there is no possibility of flashover, which would happen if it moved far enough to touch the other fixed high voltage plate. So I would imagine that clipping in itself isn't a problem as long as the applied clipped signal didn't move the diaphragm to far.

High frequency oscillation I think is a non issue, and if the amplitude of the oscillation were high enough to cause problems then I think the amp would be smoking before the speaker suffered.

Genuine low frequency oscillation is rare in solid state gear but if present and of high enough amplitude then it could cause excess movement of the diaphragm.

Those are my best guesses as to what may or may not happen.
 
The Acoustat 121 interfaces have been seen here minny times......the low frc tranfourmer has a one ohm on the neg side of the input.....an two 50k before the panels as you can see.....
Thes partes well get vary hot befor the transfourmer fails.................but it can fail....
These parts can get hot with amps... with nom low DC offset on there outputs when driven hard.......so it would led me to think if these parts were not there....any transfourmer would Sharate....with clipping an higher DC on the outputs............
Also if panels bias is low....or if high humidity get to the panels...
the partes well get hot an can fail if driven hard....
but how dose this affect the panel sound .....roll of the topend an has less output........
Tranfourmers crossover at 200-300hz I never heard any diff in sound of the panels.....long as the bias is up.....ML work fine an any other ESl I have owned..........
all just one mans findings.....after owen 10 are more pr of Acoustats...........an 30 are more panels......
Now we all know to keep clipping high DC offset amps off any speaker....if we can...good luck
 

Attachments

  • 362325d1374834931-acoustat-2-2-c-mod-repair-mk-121c_xovr-5b1-5d-1-[1][1].gif
    362325d1374834931-acoustat-2-2-c-mod-repair-mk-121c_xovr-5b1-5d-1-[1][1].gif
    39.5 KB · Views: 242
Hi
Regarding dc-offset; audiotransformers don't like high amounts of low frequency; they will tend to saturate / distort. Any dc-offset will stress the transformer. Luckily most amps will have a very low (millivolt range) offset. If there is an input capacitor at the ESL, any DC voltage will be blocked anyway and is irrelevant. So, it won´t affect the mebrane, but the transformer.

The membrane of the ESL won´t suffer from clipping. In dynamic multiway systems, the tweeter has much lower power handling. If the amp distorts, large amounts of high freq. distortion products may be generated, blowing up the small tweeter. The same thing for oscillation.
 
Thank you very much for the three very detailed replies.

I am very glad to hear that typical amplifier failures cannot do any damage to the membrane. I guess the only thing that can damage electrostatics is too much power forcing the membrane to make contact with the charged plates but that can be easily avoided.
 
It's been a while since this thread died but I've just come across it so I've resurrected it....

I've run home made stats(originally from eraudio kit) since 2003 and gone through a couple of transformers since. I've always blamed clicks and pops but I'm sure there are other considerations....

the reason for this post though is on topic as I've thought about DC offset quite a bit in the past...I sometimes run a Pass F6 into the stats and this is a pain to bias here in the uk because my mains fluctuates alarmingly. With the bias there is also the dc offset to consider which varies with the bias changes.

I have noticed with the F6 there are good days and bad days and I'm sure they are related to variations in the bias and dc offset as the mains voltage wanders from the value when I set the bias.

So the bad days may well be a combination of non optimal bias and DC offset.
I'm going to try the F6 with a mains regenerator(constant 230v) and bias accordingly. This theoretically will mean no DC offset whatsoever.

Has anyone any further thoughts on this?
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.