need help with guitar amp project

[smuel]

guitar amp oscillation

Hi,

I've just about finished building a guitar amp, and im having strange problems with the power amp circuit.

everything works fine, it amplifies the signal, and sounds good, but the amp is drawing more current from the positive rail than it is from the negative. after a few minutes it starts drawing excessive current from the postive supply and the negative stays normal.

the funny thing is, it was working fine the other day, and i cant remember changing anything, then it starting doing this. with zero input signal the safety resistor ive installed on the positive supply is showing 15mA through it which rises with time, and the negative supply is sitting at 6mA.

i have replaced all the components, and it is still doing the weird thing.

any ideas?

Thanks heaps,

Sam

 

[bamdad]

guitar pcb project

Your power stage is doing something called the "Thermal Runaway", as the power BJTs heat up, they start to draw exessive current since the threshold voltage of the base-emitter diode will fall down, and the collector current will rise exponentially... what you should do is to place a small but high power resistor in series with the emitters of the output stage power transistors. another point is that if you have parallel output transistors, this series resistor must be placed before the common connection point of the emitters.

 

[smuel]

Hi, thanks for the advice. I've already got 5w .22ohm resistors in series with the power transistor emitters. i think i might have found the cause of this strange behaviour, i noticed earlier today that there is low amplitude intermittant high frequency oscillation at the output. and now i think this is probably due to the fact that i didn't put too much though into the pcb layout, so i might go back and make another one.

unless you guys have any tricks to prevent this kind of oscillation?

thanks,

Sam

 

[bamdad]

compensation with a zero (a series capacitor and resistor) might do some magic to prevent these oscillations.

 

[IanP]

You can also try to limit the band of the power amplifier in such a way that it will only operate within the standard bandwidth of say, 20kHz.
In most cases there is a resitor devider in the feedback that sets the gain of the power amplifier. Connect a cap in parallel with the resister which is connected between the output and input stage ..
Without schematic it is really difficult to estimate its (the capacitor, that is) capacitance, but it will be somewhere within picofarads (100pF+) range ..
And a RC serial network in parallel with speakers mentioned above is rather a "must" ..
Regards,
IanP

 

[bamdad]

The HF resonance you mentioned will cause the thermal runaway you observed in the output power stage, since it will heat up the base-emitter junction and the result is an increase in base-emitter current in fixed bias, and thus the collector current will increase too, and your devices will heat up...

 

[pinkmouse]

Two quick checks:

1) Have you checked for DC on the output?

2) Does it still oscillate with the input shorted to ground?

 

[smuel]

Thanks for all the advice guys. i tried a few of the tricks you mentioned, but still didn't have much luck. so i bit the bullet and went back and designed a new pcb. this time putting much more effort into it... and now it is working absolutely beautifully. no more annoying oscillations

so i guess the circuit design is ok, but my pcb design must have just been really bad (it definetely was) tracks were too long, inputs to close to outputs etc..

anyway, thanks for all the help. the project is almost complete now except for a problem with power supply switching noise, but ive just posted a question in the analog design forum about that, so hopefully ill have some luck.

thanks heaps,

Sam

 

[1_C_E_M_A_N]

Did you check the PCB for misplaced components, most amp builders depend on the PCB layout, and encountered the same problem, until they noticed that there
was a problem with the PCB, some having hair-line, thus resulting to overheating
of some component and thus drawing more current from the supply.

try checking the component placement on the board, it might help.