Working on a pair of Altas amps, I found shorted SGSD100 BJTs on the negative side on one of the amps. I have on order, 2STW100 to replace all in this amp.
In the meantime, I moved one of the SGSD100s from the positive side to the negative side for further testing. I then placed a TIP102 on top and cut the trace to the gate of the SGSD100 to help increase drive, as I have another Atlas that was already modified like this. The modification doesn't help though.
The main rectifiers have either +-41vDC, or +-64vDC on center legs. With no signal, I get +-41vDC on rails of the output BJTs, but then when I drive signal into the amp, clipping, I see +63.5vDC, and -51vDC. So, something on the negative side is not driving rail voltage to full -64vDC.
Then on the second Atlas VII amp still in the sync that I have, its a slightly similar issue, but both rail voltages go only to +-55vDC when driving full input. I would have expected closer to +-64v. from both amps.
Anybody have a suggestion for these older Vari powersupplies? Both amps play sound otherwise.
In the meantime, I moved one of the SGSD100s from the positive side to the negative side for further testing. I then placed a TIP102 on top and cut the trace to the gate of the SGSD100 to help increase drive, as I have another Atlas that was already modified like this. The modification doesn't help though.
The main rectifiers have either +-41vDC, or +-64vDC on center legs. With no signal, I get +-41vDC on rails of the output BJTs, but then when I drive signal into the amp, clipping, I see +63.5vDC, and -51vDC. So, something on the negative side is not driving rail voltage to full -64vDC.
Then on the second Atlas VII amp still in the sync that I have, its a slightly similar issue, but both rail voltages go only to +-55vDC when driving full input. I would have expected closer to +-64v. from both amps.
Anybody have a suggestion for these older Vari powersupplies? Both amps play sound otherwise.
Last edited:
OK the more I play with this, the more I think everything is actually alright. (Minus the failed/missing BJTs)
If I raise the input frequency to about 10khz sine wave, I DO then get +-63vDC on both amps for the rails. So, the variable rail voltage is is very closely following not only the amplitude but also the frequency of the inputted signal.
Weird. Cool. Seems a little temperamental also as I do see some artifacts on the scope during change/transistion. That might come out as distortion under operation. Anybody ever bypass the variable voltage and just run straight +-FULL rail voltage from the high side rectifiers, by-passing the vari drive? Not going to do that here. Just wondering thats all.
If I raise the input frequency to about 10khz sine wave, I DO then get +-63vDC on both amps for the rails. So, the variable rail voltage is is very closely following not only the amplitude but also the frequency of the inputted signal.
Weird. Cool. Seems a little temperamental also as I do see some artifacts on the scope during change/transistion. That might come out as distortion under operation. Anybody ever bypass the variable voltage and just run straight +-FULL rail voltage from the high side rectifiers, by-passing the vari drive? Not going to do that here. Just wondering thats all.
Last edited:
The heatsink design requires the variable rail voltage to help prevent overheatng.
The forum member MOER could answer any questions you have definitively.
The forum member MOER could answer any questions you have definitively.