Guys,
Did you put the heat sink to GND?
If the problem still persist then try to increase the input capacitor to 220/330pF ... maybe your environment is full of EMI
I have just checked my amplifier with the input unconnected, taping with an 1m unisolated coper wire in the input RCAs. No problems here
Hope this help,
M
Did you put the heat sink to GND?
If the problem still persist then try to increase the input capacitor to 220/330pF ... maybe your environment is full of EMI
I have just checked my amplifier with the input unconnected, taping with an 1m unisolated coper wire in the input RCAs. No problems here
Hope this help,
M
OK. The difference in output transistors appears fundamental here. Other than a change in lag cap, can there be any other issue or fix. The change in R6/R41 was mentioned. Is this change common to all three here?
Also, can this output be sensitive to the capacitance of the transistor insulator pads? If so, which pads were used?
Also, can this output be sensitive to the capacitance of the transistor insulator pads? If so, which pads were used?
Added 100pf in parallel with the 100pf in place. Square wave shows no overshoot at all. Clean, but as expected more rounded leading and trailing edge, with the transition now around 36kHz.
However, if I plug and unplug the output from the signal generator, fuses go poof.
The insulators are some material I got at a local electronics surplus place as a sheet, so I don't know exactly what it is. By construction, it's a grey polymer reinforced with what looks like fine square woven glass, and with a self stick adhesive on one side. It's about 0.2mm thick.
Sheldon
However, if I plug and unplug the output from the signal generator, fuses go poof.
The insulators are some material I got at a local electronics surplus place as a sheet, so I don't know exactly what it is. By construction, it's a grey polymer reinforced with what looks like fine square woven glass, and with a self stick adhesive on one side. It's about 0.2mm thick.
Sheldon
Sheldon said:
Power up and slowly drive the amp into overload/clipping. Is it stable or does it start screaming blue murder, eventually blowing the fuse?
I've see such a behaviour in amps that have the nasty property to decrease the phase margin, eventually becoming unstable, with the input signal increase. Low beta transistors may create such problems.
roender said:
OK, this is pretty interesting. With the scope probe attached across the 10R input resistor, the amp seems stable. I can plug and unplug the generator, on or off, no problems. Can drive the amp to clipping with and 8R load and 1 kHz sine or square wave, no problem (can't ramp it too slowly or hold it there too long, as I've only a 20W resistor). It clips at about 25V, so that seems normal. Works the same, whether I have the probe at the input end or ground end of the resistor.
Then, to do the control, I disconnected the probe and tried to insult the input with a screwdriver and with the generator input. The amp no longer cares?!!!
I doubt this has any significance, but I had a bunch of extra UF4007 diodes, so I used those for the antiparallel diodes across the 10R. Could the capacitance of the safety diodes cause a problem?
Sheldon
roender said:
The diodes were always there. Nothing was changed from the last time I blew fuses by fiddling with the input RCA. But now I don't have a problem. Or, I should say, that the only problem is that I've no idea was was causing the original issue. Static, noise from some other source?
roender said:
It clips at 25vrms output. Should have made that clear. Supply is 34Vdc at the rail. Pooge caught me being sloppy on this issue a few posts ago. Sorry for not learning my lesson well.
Sheldon
roender said:
Strange
Did you smell some sulphur smoke in the air when the amp was unstable?
No, but that doesn't mean much. My sense of smell (which used to be very acute) is pretty much shot. Beginning to wonder about my sanity now too.
Also, the instability issue seemed instantaneous. The rail fuses would go in less than a second of contacting the wrong thing in the front end or tapping the RCA connector together. Not much time to heat anything up. No obvious signs of burnt components. Amp works fine, output looks fine.
Sheldon
