Hello,

I have a Class D amplified speaker I am trying to fix (circuit attached).

The fault at the moment is that when the speaker is powered on, and with no input signal, one of the speaker channels half-bridge configured FET's begin to immediately overheat. Hot enough that it turns the solder on the Drain tab molten.

To prevent damage to the speaker while diagnosing the fault, I have disconnected the HV + and - rails and just have the lower voltage IC's and pre-amps working.

Now when I feed a sine wave into the input of the circuit, I can trace it all the way up to [IN HI] and [IN LOW] inputs on pin 13 of U9 and U10 (audio drivers on page 3 of attached pdf). The problem is, this signal is still just a clipped sinewave and not a PWM signal.

So as per the title, can anyone tell me where the PWM from running a Saw Tooth wave and input signal into a comparator in the circuit is occurring ?

This still may not be related to the actual fault but I am first just trying to get my head around the circuit.

 

It’s all done in the ADAU1701.

The sawtooth / comparator is ancient technology.

This is all done with ADCs and a DSP. That chip takes in the analog signal and outputs the PWM signals.

 

And the technology just keeps marching forward. Newer chips put everything, including the MOSFETs in a single chip.

I just bought a Bluetooth amp that outputs 50+50 Watts in a case that is about 2x2x.75 inches. The external 24V power supply is twice as large as the amp! And it works great. For $36.

 

It’s all done in the ADAU1701.

The sawtooth / comparator is ancient technology.

This is all done with ADCs and a DSP. That chip takes in the analog signal and outputs the PWM signals.

Ah of course, that explains a lot. Thanks for that.

So the output from the ADAU1701 chip should already PWM.

When I measure the output on an oscilloscope from [DAC0] & [DAC1] (PIN 46 & 45), I only see an in phase sine-wave (the input signal) and not a PWM signal. These outputs then connect directly into the inverting and non-inverting inputs of op-amp U6A on page 1.

Does this mean ADAU1701 is faulty, or am I missing something here?

Cheers.

 

According to the datasheet the DAC output pins unsurprisingly supply analogue outputs (0.9V rms full scale).

 

Look at the input and outputs of the IRS20957’s. That will be the PWM signal

 

And the technology just keeps marching forward. Newer chips put everything, including the MOSFETs in a single chip.

They also have amazing efficiency, 97+%.
Which for all practical purposes, it eliminates the requirements for heat sinks.