Gallien Kreuger 1001RB mk II turn on thump

[dmeek]

When I switch the amp on, I get a loud thump after
the turn-on mute delay. I can also make it thump when
I short emitter and base of Q20. Other than that the
amp is OK. I checked Q2A and Q2B on the curve tracer
for matching. Also checked Q4,Q5,D2 and D3.
Anyone heard of this problem?

Doug


1001RB 206-0260-B.pdf

[Enzo]

Turn-on thump? yes, we hear of that all the time. Usually it is not really a problem.

Scope your output, what does the thump look like? Is there any DC offset? CHeck with NO speaker load.

This is a complex amp, it runs on +/-43v, but also has extra rails at +/-101v. Are all those supplies in the right ballpark, similar voltages on each polarity and CLEAN?


Q4 Q5 are just a switch that turns V- on to the differential pair.

[MarkusBass]

There is a known issue regarding thump in this amp but it was related rather to turn off and to the preamp. You may check whether you get the same problem when signal from preamp is disconnected. Or it maybe easier just to short the input of the power amp. Do you get thump in this case too?

Mark

[dmeek]

Checking the output with the scope, at turn-on initially I have -400mV w/load,
-900mV w/no load. After the thump, O volts either way.
The thump or maybe POP is a better description happens at the end of the power-on muting delay.
Since the amp has power-on muting I would expect it to be quiet. The pop is a fast rise slow
decay pulse. No thump on power-off.
+/- 43 and +/-101 are OK.

The amplitude of the pulse is +12v with load, +25v no load. I don't see any effect on
the power supply rails. As I said, I can cause it by forcing the mute on and off while
the amp is on.
I've never seen a muting scheme like this one. Why couldn't they use a
FET switch like everyone else?!

Doug

[Enzo]

A JFET circuit would not have the built in added effect of testing the V- rail. This circuit also shuts down the power amp if the fault is triggered, a JFET merely shunts the signal input. Look at a lot of SWR power amps. Upper left corner instead of lower left, but same idea.

[dmeek]

Problem solved. It turned out to be a small amount of leakage in Q11.
The evidence was a reading of 30mV across R26 when in forced mute mode.
Mute mode is forced on by shorting R41, idle current is then turned off.
Therefore there should be no voltage across R24,R31,R33,R22,R20 and R26.

Doug