TPSF12C1QEVM: The AEF does not perform completely as expected

Tobias Ubostad

Part Number: TPSF12C1QEVM
Other Parts Discussed in Thread: TPSF12C1,

Tool/software:

Hello,

We are trying to implement the TPSF12C1 AEF on 24 VDC system. We have tried to assess its performance using the TPSF12C1QEVM on a simple breadboard with an AFG source to simulate the CM noise as illustrated in the datasheet of the eval. module with some modifications as shown in the illustration below.

The CM chokes used are 90 uH.

The quickstart calculator values are attached.

Our measurements are shown below.

First row is the measurement with a 100 kHz sq. wave CM signal. Left is with the AEF enabled, while right is with AEF disabled. CH1 is the INJ voltage, CH2 is the CM voltage signal and CH3 is the voltage measurement between L and chassis (PE) in the illustration above. Red (Math) is the FFT of CH3. It seems that the performance of the AEF is quite good. However, the next measurement is with a 250 kHz sq. wave and the performance is visibly degraded, even though it seems that it should be fine with the quickstart calculator values..

Why is the performance degraded? I'm not sure if I can see an instability in the INJ voltage.. Is there an issue with our approach/ measurement setup? We cannot do testing on the final product yet..

100 kHz

250 kHz

TPSF12Cx-Q1 - Copy.xlsx

Thanks!

Best regards,

over 1 year ago

0 Naresh Adepu over 1 year ago

TI__Mastermind 19820 points

Hello

I received your questions. I am looking at it - I will get back to you

0 Naresh Adepu over 1 year ago in reply to Naresh Adepu

TI__Mastermind 19820 points

Hello

Could you please confirm what is the value of X-caps that are being used for the measurements? Is the noise CM dominant or DM dominant? If there is prominent DM noise, then we recommend adding a small additional DM choke for measurement.
Could you please check the following link for circuit optimization and debug steps. I understand that you might have to tune some component values in the injection and damping stage to improve the performance at 250kHz. [FAQ] TPSF12C1 and TPSF12C3 power-supply filter IC FAQs - Power management forum - Power management - TI E2E support forums

Thanks and Regards,

0 Tobias Ubostad over 1 year ago in reply to Naresh Adepu

Prodigy 30 points

Hello Naresh,

Thank you for your response.

1. The value of the X-caps are 1 uF each. This is CM dominated noise. As suggested, we tried to add a small DM-choke, but that did nothing to aid the results. This suggest that what we see on the oscilloscope is CM noise.

2. I have looked through the FAQ and there are some good troubleshooting tips there, but most of them are already tried with no improvement..

So I am wondering if you maybe have some concrete tips that can help us? It is especially the noise around 250 kHz that is problematic. The setup is exactly as shown in the original post, and the Y-caps to PE are 2.2 nF on the left side and 1 nF or the right.

Best regards,

0 Naresh Adepu over 1 year ago in reply to Tobias Ubostad

TI__Mastermind 19820 points

Hi Tobias

I received your question. will get back to you.

0 Naresh Adepu over 1 year ago in reply to Naresh Adepu

TI__Mastermind 19820 points

Hi Tobias,

Is it possible to share the slides showing the measurements and the debug steps followed? please include SENSE and INJ pin waveforms for each debug step.

The solution steps that I propose would be very specific to your results and the setup used.

Thanks and Regards,

Power management

Power management forum

TPSF12C1QEVM: The AEF does not perform completely as expected