Have looked through a few threads for advice on how to figure a power supply bleeder resistor, but haven't found a simple rule of thumb. Looked in Art of Electronics, too, but found nothing helpful. NP's A-75 amp used a 1.6K resistor of about a watt or two, but that amp has a lower unreg. supply voltage than the ZenV4, so I figure the R for the Zen has to be greater than for the A-75, right? Can anyone suggest an optimal value for the ZenV4, given its 50V unregulated supply. Or, (teach a man to fish and you feed him for life) can anyone supply a calculational (not definitional) formula. My algebra is no good. Have only a foggy notion of what a time constant is.
I assume from looking at the schematic that there is already .47microfarads of film cap in parallel with the big filter caps, sitting on the circuit board, right? So, no advantage to further film caps in parallel. Didn't see anything on the board that would act like a bleeder resistor, however, so want to make sure its safe to stick my mitts into the amp to make adjustments.
BTW, still not sure why the design didn't include DC rail fuses, like, say, the Leach amp. Would the fuse mess up impedance in the regulated supply or something??
Larry Wright
Seattle area
I assume from looking at the schematic that there is already .47microfarads of film cap in parallel with the big filter caps, sitting on the circuit board, right? So, no advantage to further film caps in parallel. Didn't see anything on the board that would act like a bleeder resistor, however, so want to make sure its safe to stick my mitts into the amp to make adjustments.
BTW, still not sure why the design didn't include DC rail fuses, like, say, the Leach amp. Would the fuse mess up impedance in the regulated supply or something??
Larry Wright
Seattle area
Zapped said:Larry,
The unregulated power supply or raw supply does need to be bled off. It can be done several ways. A 5K resistor from supply to ground would do fine.
In mine I used a string of 2.5K resistor, blue diode, and 2.5K resistor. This was to buffer any noise from the diode and keep the current down to 10 ma for the diode.
Works great, I do not know if buffering each side of the diode make any difference in injected noise. But it sounds like a good idea.
The power rating can be calculated by voltage squared, divided by resistance. Try a resistor about twice the dissapation as needed for long trouble free life.
Good Luck!
George
Have looked through a few threads for advice on how to figure a power supply bleeder resistor, but haven't found a simple rule of thumb. Looked in Art of Electronics, too, but found nothing helpful. NP's A-75 amp used a 1.6K resistor of about a watt or two, but that amp has a lower unreg. supply voltage than the ZenV4, so I figure the R for the Zen has to be greater than for the A-75, right? Can anyone suggest an optimal value for the ZenV4, given its 50V unregulated supply. Or, (teach a man to fish and you feed him for life) can anyone supply a calculational (not definitional) formula. My algebra is no good. Have only a foggy notion of what a time constant is.
I assume from looking at the schematic that there is already .47microfarads of film cap in parallel with the big filter caps, sitting on the circuit board, right? So, no advantage to further film caps in parallel. Didn't see anything on the board that would act like a bleeder resistor, however, so want to make sure its safe to stick my mitts into the amp to make adjustments.
BTW, still not sure why the design didn't include DC rail fuses, like, say, the Leach amp. Would the fuse mess up impedance in the regulated supply or something??
Larry Wright
Seattle area
Zapped said:Thanks, Panelhead. That is a cool idea. Discharge the caps and power the pilot light at the same time. Ingenious!
Larry Wright
Seattle area
Larry,
I cannot take credit for the idea. It came from Nelson.
The bypass caps you asked about are most likely not needed. After three months of daily use I added 2.2 ufd 100v flat film caps to the last filter cap. At first it seemed to add some detail and life. But after 4-5 hours the sound was about the same as before. My filter stage is a 10K 63V, 2.7 mH Choke, and 10K cap. The rails are running at 53.5 volts under load.
The rail fuses also are more trouble than help. If the input fuse is sized correctly and the diode bridge included on the ground leg the power fuse should smoke when there is a major failure. But with the constant current draw of the circuit, it should be easy on rail fuses.
George
Zapped, you can calculate the time(t) your supply caps can be bled by using the formula t=RC; where t=seconds, R=resistance in ohms, C=capacitance in Farads. Or the bleeder resistance for that matter. So if you have a 25,000 uF of cap and you would like 15 seconds of discharge time, your R=t/C. I don't have a calculator right now (at a public library). Power(P) dissipation of the resistor will be P=V^2/R. V is the unreg supply voltage. Double P value of the resistor...this will get hot.
You can use this resistor before working the innards of your amp so you can bleed off stored energy and avoid a nasty spark or shock.
You can use this resistor before working the innards of your amp so you can bleed off stored energy and avoid a nasty spark or shock.Having an LED is nice, of course, but I wouldn't lose a lot of sleep over bleeding down the power supply. The circuit will do that quickly enough on its own (any circuit Nelson's had a hand in will draw electrons out as fast as you can stuff them in).
I don't even use bleeders on tube circuits. Solid state voltages aren't high enough to worry about.
Grey
I don't even use bleeders on tube circuits. Solid state voltages aren't high enough to worry about.
Grey
Well lets see. A 100w amp will have 100v dc between the rails and a 200w amp will have 140v between rails........I would say you would have something to worry about.
And if the amp had 100,00uf of capacitance the stored energy would be ...1/2 CVsquared (Joules).
I would say that if you got zapped , you would have a lot to write home about.
Regards,
Jam
And if the amp had 100,00uf of capacitance the stored energy would be ...1/2 CVsquared (Joules).
I would say that if you got zapped , you would have a lot to write home about.
Regards,
Jam
With 5K ohm at 2 watt across the filter caps (23,400microfarads each) I also found it took a heck of a long time for the caps to discharge. And the Blue LED was almost heliarc bright. Experimenting last week, I dropped the bleeder resistance to 2 K at 5 watt (sandstone type) and separately (in parallel with the bleeder resisters) tried a bunch of increasing R's-in-series-with-the-LED to get the blue light down to a low glow. At about 50K ohm (!) in series with the LED, it is a nice mellow glow, amazingly still very visible across the room. The caps discharge quite a bit quicker with the 2 K ohm R's, which only get a couple of degrees above body temperature when doing their bleeder duty. Don't know if I will be wasting current capability this way when I connect the supply to the audio circuitry, but I feel safe now anyway.
Thanks for the advice and especially concern for avoiding unsafe practices. I am already Zapped; always want to avoid becoming more zapped.
Larry Wright
Seattle area
Thanks for the advice and especially concern for avoiding unsafe practices. I am already Zapped; always want to avoid becoming more zapped.
Larry Wright
Seattle area
jam said:
Big PA amplifiers can run + and - 120V rails.
Going across these rails gives up to 240V DC.
DC is rather more dangerous than AC because DC can cause muscle spasm and 'lock-on'.
That said 240V AC can cause lock-on too - been there, done that.
The old rule about working with hazardous voltages is to keep one hand in a pocket, and perform measurements with the other hand only.
Grey, you now know that solid state voltages CAN be high enough to be hazardous.
Eric.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.