TPS65131: Noise issue on dual DC/DC converter (Load current waveform)

Hi Salvatore,

Is the TPS65131 supplying a display? If so, the current load might be related to the display pattern.

BR

Patrick

Hi Patrick,

The TPS65131 does not power a display but a PGA855 (VCC and VEE) as shown in the following image.

In any case, I can't explain the current flow for VEE, which is probably the cause of the noise.

I look forward to hearing from you.

Regards,

Salvatore

Hi Salvatore,

I am not so familiar with the FDA, but maybe you could check whether the IN-, IN+ and VOCM of the PGA855 are stable or not first.

And for the INA823, what are the voltage of IN- and IN+? And is the output of INA823 stable or not?

Since TPS65131 is the power supply to the FDA, so the pulse current load is very likely related to the amplifier operation. If you have already captured any waveforms of the amplifier circuit, please let us also have a look, thanks.

BR

Patrick

Hi Patrick,

IN- and IN+ are the signal inputs from the BNC. A 50-ohm cap was used for testing. I can confirm that the VOCM is stable at 1.75V. The INA823 output is always 0V because, as mentioned, the input is 0V.

Meanwhile, I ran some tests:
- I powered the TPS65131 with an external 5V supply (across C163) -> no change with noise.
- I powered the circuit after the PGA855 with external +10V/-10V VCC/VEE (removing R118 and R117) -> it works correctly as per the PGA855EVM evaluation board, where each branch draws less than 20mA.
- I reinstated the internal power supply, removed the PGA855, and connected a 120-ohm linear resistive load -> no whine.
- From the previous condition, I limited the current of the 5V input to 500mA -> the whine reappeared.
- From the previous condition, I removed the current limit and resoldered the PGA855 -> the whine is still present.
- I connected the +10V/-10V output from my board to the PGA855 evaluation board -> the whistling noise is present, behaving like my board.
- I removed the ferrites between the power supply and the load, but the noise/interference on the current has increased.

So I don't think that the problem is in the downstream circuit, since I have no problems with an external power supply.

Regards,

Salvatore

Hi Salvatore,

When you did the "I powered the circuit after the PGA855 with external +10V/-10V VCC/VEE (removing R118 and R117) -> it works correctly as per the PGA855EVM evaluation board, where each branch draws less than 20mA.", have you captured the current waveform of +/-10V?

BR

Patrick

Hi Patrick,

yes, I captured it with the oscilloscope: the current waveform is stable in this condition without noise.

Regards,

Salvatore

I ran further tests, of which I'm attaching a screenshot from the oscilloscope.

I replaced the PGA855 with an ITECH IT8511A variable load and noticed that the current behavior is "strange" for low values: the positive branch has a linear current trend after 130mA, while the negative branch has a linear current trend after -50mA. For lower values, however, noise is present on both the oscilloscope and the whistling sound.

Regards,

Salvatore

Hi Salvatore,

I will give you feedback later.

BR

Patrick

HI Patrick,

any news about that?

Regards,

Salvatore

I will give the feedback tomorrow

Sorry Patrick, but did you see my original post? There's the electric diagram with connections.

This is a very important project for me and we can't find a solution for this problem. Why the component produced by Texas Instruments is not working properly?

Regards,

Salvatore

Hi Salvatore,

Based on your waveforms, I suspect the audible noise is caused by the light load in power save mode. In this mode, IC will skip some switching cycles to reduce the power loss, but it may have chance to cause the converter works in audible frequency range. Is it able for you to add more loading to make the IC quit the power save mode?

BR

Patrick

Hi Patrick

After some tests and carefully reviewing the schematic, your suggestion solved the problem.

Thank you very much for your time.

Regards,
Salvatore