TPA3255 project using ZK-3002 issues

[dandunham]

Hello all, this is my first post and actually my first experiment with any sort of digital audio amp. I purchased a ZK-3002 no bluetooth, and I am having some strange issues. I will go over the basic setup first. I am using a switching power supply at 40 V DC just to keep things well within spec before pushing it hard. I configured it for the fan on at all times and for stereo. I have made sure grounding is all tied together including my input from an old receiver. I was able to trace a solid audio signal from my older receiver to the input of the of the new board with no issues. The problem I have is there is no sounds output and I just want to verify I haven't missed something in the setup. when I put my scope on the output i am seeing 20VDC with about 2 V of noise on the top. one thing I haven't done is to connect speaker negative to power supply ground. I have included a pic of he scope with the the output on the top AC coupled so it just shows the noise and the bottom trace has audio input. The noise on top of the DC seems to be between 500Khz and 1 Mhz. I am hoping I just did something silly so if you have experience with these I'd love to hear your input.

 

[AnalogSteph]

Amplifiers with a single power supply and capless output are BTL (bridged) affairs, so seeing half the supply on both + and - outputs is expected. Neither should be shorted to ground. On a scope you'll generally have to use differential probing using two channels. This board appears to be less than ideal when it comes to output filter design, so seeing some switching frequency escaping is not unexpected.

 

[DVDdoug]

one thing I haven't done is to connect speaker negative to power supply ground.

Good! It's a bridge amplifier. There should be a speaker (+) and a speaker (-). Bridge amplifiers don't have a speaker ground. They should both be at about 20VDC (no differential DC voltage).

The high frequency noise is probably normal without a load, since it's a class-D (switching) amplifier. A low-pass pass filter normally needs a "load". Even then then no filter is perfect, but a small bit of radio frequency into the speaker shouldn't hurt anything.

I don't know why you're not getting a signal...

 

[NTK]

These TPA3255 based amplifiers usually (99+% ?) have an LC low pass filter at the output to filter out the switching noise. The cutoff frequency of these LC filters is usually at 1 decade below (= 1/10) the switching frequency. Since a second order filter has a slope of -40 dB/decade, you'd expect the switching noise voltage at the output to be ~1/100 (= -40 dB) of the PWM waveform voltage before the LC filter. Therefore, a 2 V measured switching noise at the speaker terminal may be a bit high but should still be OK.

Note that most speakers will have very high impedance due to driver inductance at Class-D amp switching frequencies, so switching noise of several volts is usually harmless.

 

[dandunham]

Thank you all for the knowledge, as soon as I get a chance I will do some more testing with this thing and see what's going on. I will update asap.

 

[dandunham]

Just an update. I had a little time to play with it and I decided to bypass the digital volume control chip and i was able to get sound but it was very distorted. I think i will pull that chip out and see what happens.

 

[dandunham]

I pulled it and did a simple jump from input to output pins and no distortion at all. It now sounds fantastic.

 

[trungdtmc]

This board uses 1kohm input gnd pin resistor which is too low