now as I've thought about it, of course!!!!
F=Q*E, P~F so Q must to be constant to have a linear relationship between E and P.
Now I also get why there are Resistors between the membrane and the hv supply
I quickly ran a calculation for a 10Hz high-pass with 1nF capacitance for membrane
I get ~16Meg for the resistor. That is not to much for a diy low creep HV-Resistor.
is this why ESL's are sensitive to humidity and am I better off with a high R in the HV-Path or is a high R membrane still better for some arcane reason?
F=Q*E, P~F so Q must to be constant to have a linear relationship between E and P.
Now I also get why there are Resistors between the membrane and the hv supply
I quickly ran a calculation for a 10Hz high-pass with 1nF capacitance for membrane
I get ~16Meg for the resistor. That is not to much for a diy low creep HV-Resistor.
is this why ESL's are sensitive to humidity and am I better off with a high R in the HV-Path or is a high R membrane still better for some arcane reason?
on the fun side of this project,
I've already whacked together the said space blanked beween two sheets of perforated aluminium with a hv-supply from an old monitor lamp and a "way underpowered " line transformer and, I cant help it but,
underneath the distrotion, besides the crap-resonant nonstiff mount of it all, it feels like I've rubed of some dust from an old window, and I can see the outside a slightest more clearly now. Of course, now I intend to get the detergent and clean the window propperly.
kind regards
I've already whacked together the said space blanked beween two sheets of perforated aluminium with a hv-supply from an old monitor lamp and a "way underpowered " line transformer and, I cant help it but,
underneath the distrotion, besides the crap-resonant nonstiff mount of it all, it feels like I've rubed of some dust from an old window, and I can see the outside a slightest more clearly now. Of course, now I intend to get the detergent and clean the window propperly.
kind regards
That I don't get,
I mean the ESL is basically a capacitor (as far as I understand it now) that I want to keep a constant charge on one plate, depite the other plate seeing an ac-voltage.
A resistor will do exactly that, provided that the RC cuttoff Frequency is much lower than the frequency of interest.
Am I wrong in assuming I can model the esl as a capacitor ?
I mean the ESL is basically a capacitor (as far as I understand it now) that I want to keep a constant charge on one plate, depite the other plate seeing an ac-voltage.
A resistor will do exactly that, provided that the RC cuttoff Frequency is much lower than the frequency of interest.
Am I wrong in assuming I can model the esl as a capacitor ?
Your calculation doesn't just apply to the entire membrane, but also to each part of it. That is, for minimum distortion, you don't want the charge to move from one part of the membrane to another as the membrane moves. Hence, you need a high-resistance membrane. Also having a large resistor (or preferably several of them) right at the output of the high voltage supply is a nice safety feature.
By the way, another disadvantage of low-ohmic membranes is the sound produced if you should get a short circuit due to dust getting in somewhere.
By the way 2: you can find a lot of information here: Elektrostatic Loudspeakers
Last edited:
From a theoretical standpoint, the resistance should be large enough so the charging time constant is 10x that of a half period of the lowest frequency to be reproduced (from Wagner's Electrostatic Loudspeaker Design and Construction). That gives a far higher value for the resistance. 10 Hz is an excessively low frequency target though.
Practically, over 100 megohms per square is a decent target for diaphragm resistance on full-range units. You typically wind up with spots that are higher and lower, so the higher target helps minimize issues when it wanders. On midrange/tweeter drivers I've seen much lower resistance used on commercial units (500 k).
Most electrostatics use both a high resistance diaphragm coating and a high value resistor in series with the bias supply (10-100 megohms typical). The bias resistor makes the supply safer for humans and speakers.
A high resistance diaphragm coating is still preferred because it prevents charge migration across the diaphragm while it is in motion. It also limits arc current during fault conditions (over-driving, dust in the speaker, etc.).
Yes, high humidity can cause output to decrease since more of the voltage is dropped across the resistance in series with the bias supply instead of showing up across the speaker's gap. A high bias voltage helps minimize this effect, which is why proper insulation is also useful, but that's another conversation. An adjustable bias supply can also be used to try to rebalance things.
Practically, over 100 megohms per square is a decent target for diaphragm resistance on full-range units. You typically wind up with spots that are higher and lower, so the higher target helps minimize issues when it wanders. On midrange/tweeter drivers I've seen much lower resistance used on commercial units (500 k).
Most electrostatics use both a high resistance diaphragm coating and a high value resistor in series with the bias supply (10-100 megohms typical). The bias resistor makes the supply safer for humans and speakers.
A high resistance diaphragm coating is still preferred because it prevents charge migration across the diaphragm while it is in motion. It also limits arc current during fault conditions (over-driving, dust in the speaker, etc.).
Yes, high humidity can cause output to decrease since more of the voltage is dropped across the resistance in series with the bias supply instead of showing up across the speaker's gap. A high bias voltage helps minimize this effect, which is why proper insulation is also useful, but that's another conversation. An adjustable bias supply can also be used to try to rebalance things.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.