How important is the bias supply to the performance of ESLs? I'm building a headphone amp and trying to decide between a higher voltage multipler straight from the wall, and a 115v to 230v transformer and a smaller multiplier. Also, how important are the capacitor types (polypro, film-foil, teflon) and diode types (i.e. fast recovery, schottky, avalanche etc.) in these supplies? Also, what about ripple supression in these supplies?
Max bias I'll need is 600v...
Thanks!!
Max bias I'll need is 600v...
Thanks!!
Hi,
Q: how important?
A: quite unimportant?
Q: use Tranni?
A: functionally it is even, but I´d prefer a trannie because of galvanic decoupling to the power line. In Germany and many other nations it is not allowed to take the voltage right off of the wall.
Q: importance of Caps
A: unimportant. Use the cheapest that fullfill the voltage and capacity requirements. You don´t need Foils for this.
Q: diodes?
A: unimportant, just fulfill the voltage requirements. Use cheap standard highvoltage types like 1N4007 or series connections of those.
Q: ripple rejection
A. the usually used and always positive(e.g. for safety) high ohmic value resistor between cascade and the transducer itself reduce ripple ( RC-Filtering). Adding a blinker-circuit reduces influences further by completely intermediate decoupling the ESL from the supply. As a bonus You get a very sensitive control device.
jauu
Calvin
Q: how important?
A: quite unimportant?
Q: use Tranni?
A: functionally it is even, but I´d prefer a trannie because of galvanic decoupling to the power line. In Germany and many other nations it is not allowed to take the voltage right off of the wall.
Q: importance of Caps
A: unimportant. Use the cheapest that fullfill the voltage and capacity requirements. You don´t need Foils for this.
Q: diodes?
A: unimportant, just fulfill the voltage requirements. Use cheap standard highvoltage types like 1N4007 or series connections of those.
Q: ripple rejection
A. the usually used and always positive(e.g. for safety) high ohmic value resistor between cascade and the transducer itself reduce ripple ( RC-Filtering). Adding a blinker-circuit reduces influences further by completely intermediate decoupling the ESL from the supply. As a bonus You get a very sensitive control device.
jauu
Calvin
In addition:
The hot side of the HV-supply can be filtered with a high value resistor or a neon oscillator ('blinker') and the resistance of the coating as well.
But what about the side of the HV-supply which is connected tom the trannie. Do we need something there to avoid noise etc.? It is directly from the mains.
Any suggestions?
The hot side of the HV-supply can be filtered with a high value resistor or a neon oscillator ('blinker') and the resistance of the coating as well.
But what about the side of the HV-supply which is connected tom the trannie. Do we need something there to avoid noise etc.? It is directly from the mains.
Any suggestions?
Hi,
I´d use whats standard
- a good switch with small cap over the contacts
- a fast fuse
- a MP-cap
- Varistor
-additional filtering is not necessary, but if it makes You feel better, I´d suggest a complete input block (mains connector, switch, fuse-holder and filter as one component)
jauu
Calvin
I´d use whats standard
- a good switch with small cap over the contacts
- a fast fuse
- a MP-cap
- Varistor
-additional filtering is not necessary, but if it makes You feel better, I´d suggest a complete input block (mains connector, switch, fuse-holder and filter as one component)
jauu
Calvin
Calvin:
thank you for your help. What do you think about something like this, which would be battery-powered:
http://www.emcohighvoltage.com/CASeries.pdf
also, as for the blinker-circuit, do you have a schematic I could look at?
thank you for your help. What do you think about something like this, which would be battery-powered:
http://www.emcohighvoltage.com/CASeries.pdf
also, as for the blinker-circuit, do you have a schematic I could look at?
I talked to Emco today, they won't sell in small quantites. However, I found a circuit that uses a 555, perhaps this is what you mean by a blinker:
http://www.quadesl.com/speaker/diyesl/hv_supply.pdf
http://www.quadesl.com/speaker/diyesl/hv_supply.pdf
Hi,
no, MJ and I don´t talk about that supply....there´s no component blinking in that circuit!
The circuit we mean looks like this:
jauu
Calvin
no, MJ and I don´t talk about that supply....there´s no component blinking in that circuit!
The circuit we mean looks like this:
An externally hosted image should be here but it was not working when we last tested it.
jauu
Calvin
anyways, thanks for the clarification. I assume that this is a 90v DC breakdown voltage that the schematic refers to? I see a lot that are 65 AC / 90 DC breakdown voltage. How does the neon bulb act as a control device?
The neon lamp monitors the voltage on the capacitor. If the voltage rises above about 90V, the lamp turns on. The voltage will rise when the charge on the diaphragm of the speaker leaks away into the air or through other paths. When the lamp turns on the diaphragm will recharge. This will lower the voltage across the capacitor turning the lamp off again.
I bought modules directly from Emco about 15 years ago. I didn't have any trouble at that time, but maybe the fact that I was an engineer working at HP helped. Try Pico, they have some very small HV DC-DC converter modules and may be more willing to sell small quantities or have distributors who will.
I_F
I bought modules directly from Emco about 15 years ago. I didn't have any trouble at that time, but maybe the fact that I was an engineer working at HP helped. Try Pico, they have some very small HV DC-DC converter modules and may be more willing to sell small quantities or have distributors who will.
I_F
I_Forgot said:
thanks for the explanation. So, may I ask why this "cycling" of the bias voltage is desired?
As for the DC-DC power supplies, do you feel that it is better than a simple voltage multiplier? Remember, I only need maximum of 600 V bias, as this is for a headphone amplifier. How would the HV DC-DC supply compare to say using a Hammond 229A230 which is 6 VA and would be wired for 230V CT @ 25 ma and then a simple voltage doubler circuit? Like this:
http://photo.head-fi.org/data/500/alwayson.gif
Maybe I could add the neon to this supply, if it makes sense... just looking for a nice solid recommendation and then I'll go from there and experiment!
Calvin:
as for your recommendations of:
- a good switch with small cap over the contacts
- a fast fuse
- a MP-cap
- Varistor
I'd love to see a schematic if you have one tucked away somewhere. I'm not sure I'll use the switch, but it would be nice to know how to size the fuse, and where to place the MP-cap (I'm assuming something like 0.022 at line voltage across the primaries) and what to do with the varistor (probably a heat indicator as in power amps?)
Thanks!
Hi,
basically the Blinker isolates the panel from the supply while ´off´.
Too the blinkrate gives You an idea about the leakage losses. One blink every 10secs is ok. Quickly blinking and there is a problem somewhere.
A dedicated HV-supply is easier to work with, comes with proof marks (like CE) and is (when completely capsuled) the safer way. Functional there´s no difference to a standard tranny-cascade-design.
have a look at Schaffner, item FN280, Power entry module. (the 2xL and 2xCy are not necessary for Your application, but filtering the supply voltage isn´t a bad thing either, especially when the line voltage is transformed upwards.
Add a varistor between contacts N and P. This will limit the voltage going into the tranny/cascade to a safe value for the components.
jauu
Calvin
basically the Blinker isolates the panel from the supply while ´off´.
Too the blinkrate gives You an idea about the leakage losses. One blink every 10secs is ok. Quickly blinking and there is a problem somewhere.
A dedicated HV-supply is easier to work with, comes with proof marks (like CE) and is (when completely capsuled) the safer way. Functional there´s no difference to a standard tranny-cascade-design.
have a look at Schaffner, item FN280, Power entry module. (the 2xL and 2xCy are not necessary for Your application, but filtering the supply voltage isn´t a bad thing either, especially when the line voltage is transformed upwards.
Add a varistor between contacts N and P. This will limit the voltage going into the tranny/cascade to a safe value for the components.
jauu
Calvin
hey guys, I'm still having some problems with my electrostatic headphone bias circuit. I have need for an adjustable bias of 330-600 VDC. I have room on my project for a switch (i.e for a high and low range bias, if I cannot get away with just a pot for the entire range needed) and a panel mount pot, as well as two tip jacks for using a DMM to set the voltage without removing the case cover. See attachment for my first try:
Problem is, I don't think my multiplier is working. I have 528 VDC right before the 100k resistor. I am getting 278 VAC out of the transformer. My pot is a simple carbon single turn Alpha pot.
Can anyone comment on the circuit? I'm betting I'm missing something pretty obvious.
I took my multiplier circuit from this link:
http://www.play-hookey.com/ac_theory/ps_v_multipliers.html
Thanks!
An externally hosted image should be here but it was not working when we last tested it.
Problem is, I don't think my multiplier is working. I have 528 VDC right before the 100k resistor. I am getting 278 VAC out of the transformer. My pot is a simple carbon single turn Alpha pot.
Can anyone comment on the circuit? I'm betting I'm missing something pretty obvious.
I took my multiplier circuit from this link:
http://www.play-hookey.com/ac_theory/ps_v_multipliers.html
Thanks!
The pot is shorting the output of the supply to ground. Voltage multipliers usually can't supply much current, so almost any resistance across the output will kill it completely. If you have 528V and you're grounding it through 1M, you're attempting to draw 528 uA from the supply (assuming you've got the pot set to 1M, more if the pot is set lower). Don't expect to get more than about 10 uA from this sort of supply without pulling down the output.
The pot will not last long with 500+ V across it.
Remove the pot and the supply will probably work OK.
I_F
The pot will not last long with 500+ V across it.
Remove the pot and the supply will probably work OK.
I_F
Your voltage multiplier circuit is a tripler that will provide about 1kV out (triple applies to the amplitude of the input voltage, not the rms value). If you want 300-600V, you should start with a voltage doubler, but as you have discovered, such supplies are not easily made adjustable.
If you really need it to be adjustable, get one of the CCFL power supplies from the surplus dealers and hang a diode and small capacitor from it. Those supplies are usually very crude switchers whose AC output will vary somewhat with the input voltage. They are also safer to work with than line powered supplies.
I_F
If you really need it to be adjustable, get one of the CCFL power supplies from the surplus dealers and hang a diode and small capacitor from it. Those supplies are usually very crude switchers whose AC output will vary somewhat with the input voltage. They are also safer to work with than line powered supplies.
I_F
Here's a 5VDC in 1000VAC out supply that is cheap ($5), small, and safe:
http://www.goldmine-elec-products.com/prodinfo.asp?number=G16565
You can probably swing the output voltage over a pretty wide range by adjusting the input voltage. Any dope who knows which end of a soldering iron to hold can make a 5V adjustable supply. It is a lot harder and more expensive to make the adjustments on the HV side.
I_F
http://www.goldmine-elec-products.com/prodinfo.asp?number=G16565
You can probably swing the output voltage over a pretty wide range by adjusting the input voltage. Any dope who knows which end of a soldering iron to hold can make a 5V adjustable supply. It is a lot harder and more expensive to make the adjustments on the HV side.
I_F
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.