What to look at when debugging Amp1-B

[jmjohnson]

It seems like most of my posts are me asking for advice, hopefully I can start to contribute a bit more in the other direction soon.

Anyway My AMP1-B has been built for a while and the soldering looks good, a small bit of practice and the right tools go a long way. I just got the PSU going over the weekend and have 24, -24 and 5v. I’m not sure if this is important but the 5v and +- 24v are from two separate PSUs. I am not sharing the ground directly between them.

When I power up the Amp it does not work. Voltage polarities are fine, the voltage from the PSUs doesn’t change once the Amp is up either muted or awake. The Amp is completely dead though. When awake there is no led and no relay click. I am reading voltages around the board and to the chip but I don’t know what I should be looking at to help zone in on the problem. When powered up for a little while the chip gets slightly warm to the touch.

All thoughts and suggestion appreciated,
Thanks,
James

[firestorm]

James,
I think you need to have a common earth reference for all 3 power supplies. So try to join the 0v ref for all 3 power supplies together, at one point.

Regards

[jmjohnson]

Thanks. I have tried it but it makes no difference.

[Jan]

Hi,

Did you check that you did not mix up the transistors in the relay/balancing section? There are two types. It could also be that one of these transistors or the comparator chip is damaged.

[jmjohnson]

Going through the threads it seems that a number of people locky, kenk, deepspace, plopla, bearman, myself and possibly others either have or have had some difficulties in getting the Amp1 up and running.

Amp3 seemed to have much more information on how to test but in looking through the Amp1 threads there seems to be less information and the small few ideas posted seem to trail off with resolution. It isn’t clear if any of the people who had posted problem questions found a resolution.

I’d love to get this amp up and running but short of desoldering and starting over (not something I am keen to do) I’m not sure where to go with it from here.

One thought I have had is for Jan to open a troubleshooting thread for each amp so that people have a single place to go to look for this information and can add to it in terms of additional information and hopefully success stories/lessons learned.

Regards,
James

[locky]

Sounds like a good idea. I second that...

[simingx]

I would check the ground connections to the SMD components. The ground plane is really effective as a heatsink (I had to set my soldering station to quite a high temperature!) so you may not be getting contact on the PCB itself even though the solder is on the component.

[bearman]

I got my second Amp1 working well. The first one, I made the mistake of putting the amp chip all the way down to the board and got a small solder bridge which fried the amp. The second amp try at first didn't work but figured out that the 5v neg. was not connecting the pin. I also used two separate PSU's. I also have the AMP 1 not the 1-B.

It seems that most of the problems are of the Amp 1-b and I would suspect it would be in the relay section.

[Mike Andrew]

I'm having fun with AMP-1B, and the more I listen, the more I like it.
However, I don't think I'm getting anywhere near 2x60W out (into 8 Ohms) - it's plenty loud enough for my listening habits, but I like to be able to annoy the kids, so I'd like to try optimizing the stage gains.

I'm driving the amp from the headphone output of my CD player (It has a volume control) and I see about 250mv p-p (hard to tell exactly) at the input when the mute comes on. This is using +/-35Vdc rails. A couple of questions:

1) I'm trying to 'scope the outputs of the input stage op amps (OAOUT1 and OAOUT2, pins 22, 25 to AGND). I understand this should be about +/-2V, but I see nothing more than a few mV. Shouldn't this be an inverted replica of the input (i.e. a gain of -1)? Am I making the measurement correctly?

2) According to the Tripath info, the maximum modulator feedback should be about 4Vp-p. How can I measure this? Do I need to make a differential measurements between FBKOUT1/2 and FBKGND1/2? Same for OUT1/2 and OUT1/2 Ground?

Hope these aren't too dumb. I still like the sound of the AMP - regardless of what the gain(s) might be.

Thanks,

Mike Andrew

[Jan]

Hi,

Mike, It seems your amp is closing down prematurely. With only 0.252V peak to peak, it is running at about 10% of maximum power.

It could be that with 35V rails, you are close to the over voltage limits. When power increases, there is always some transients up and down. I'd suggest you try put an additional diode in series with each supply, to lower voltage and see if this changes (increases) the maximum output.

The modulator input = input section output = input signal times input gain. If your input section has a -1 gain then the modulator input isn the same as your input signal, (but inverted). Therefore, you do not need to measure the modulator input.