Tiny giant amp

[whomeno]

Just bulit the tiny giant amp from the pcb. using a 12 volt 5 amp power supply. every thing works great till I get to 10:30 on the volume them it seems to go in and out on the low strings only.
if I back it down past 10:30 then it is fine. Any ideas what might be going on?
Thanks

[anotherjim]

Could be insufficient power supply capacitance. If the power supply isn't very close to the power amplifier, the 470uF provided for on the PCB probably won't be big enough. I would try 1000uF across the pcb power input.

[potul]

Also check that you don't have a heat issue. You need a good heatsink in this amp, if it gets too hot it will cut out.

[whomeno]

anotherjim
I have the power going from power jack to on/off switch then to pcb. Where would I put the 1000uf at? the pcb called for a 15volt 4amp power supply. since I used a 12 volt 5 amp supplly
I omitted the lm 338 the 120 R and the 1 k resistors. and jumped the to pins where the lm 338 would be.
Here is a pic of the pcb.

potul
I mounted this in a 125b enclosuer. used heat sink paste and bolted the tda 7240 to enclosuer. so heat should not be a problem.











Thanks guys

[anotherjim]

The way the input opamp U1a is biased injects power supply noise into the signal path - which gets amplified and causes supply ripple which gets back into the input... the harder the amp is worked, the more the supply voltage ripples and the bigger the problem gets. Removing the voltage regulator doesn't help, although I think the original reason for that was to use a laptop power supply that gave at least 18v.
The point of adding capacitance to the power supply is to help reduce any ripple in the voltage and it's added in the obvious way - across the +V and GND pcb connections. As there is already 470uF capacitance, it's going to take at least another 470uF to make a difference, and the extra capacitor might as well be 1000uF.

[whomeno]

Thanks I will add the 1000uf.

[whomeno]

anotherjim
Could not put the 1000uf on the pads so removed the 470uf and replaced it with a 1000uf,a little better but I have some 1500uf coming. going to try one of them next.
What I don't get is this one is built from the pcb. I bulit on on vero and replace the 470 uf with a 1000uf and it works fine. using the same power supply also. Looks like I will use the vero from now on ( just saying).

[anotherjim]

I think the real problem is the op-amp bias scheme. It depended on the voltage regulator keeping a stable voltage despite the power amplifiers varying current draw. Difference between PCB and Vero can be down to different "wire" thickness and routing.
It's possible that it wouldn't even need that 470uF changing if the op-amp and it's input bias had separate power filtering so that the power chips current usage had no effect back in the preamp.