TPS23861: Issue with TPS23861PWR PSE Power supply and PoE Camera Connectivity

Hi Poovarasan,

Can you confirm your probe points? Are you provbing between Vpwr and Drain? Ensure you probe between Vpwr and Drain. 

How can you confirm you have power at the device? Is the only thing powering the device PoE? Or is there an auxillary input?

Regards,

Brandon

Hi Brandon,

Please refer to the probing points highlighted in the image below.

My end application is a camera. Even though the waveforms look like this, I can still get data from the device, and the streaming works well for a long time (few times it's stopping).

Also, our device doesn’t have auxiliary power; the power is provided only through magnetics.



I hope this message clears you're doubt. 

TPS23861 IEEE 802.3at Quad Port Power-over-Ethernet PSE Controller datasheet (Rev. I)_Application curves.pdf


Hi Brandon,

I have completed the probing based on the application curve provided in the datasheet, and I’ve attached the PDF for your reference. Please review the scope images, and let me know if everything looks correct.

I’ve also followed your instructions for additional probing, and the scope images are attached for you to review. These images were captured while the camera was in streaming mode. Please confirm if the PSE is functioning correctly.



Thanks,
Poovarasan S

Hi Poovarasan,

Thanks for your test waveform. Sorry that I may still need know more information to address the potential issue:

1. About the probing point: May I assume there is only one probe is putted between "P1_Drain" and "GND"? The other not used channels are either open or connect their grounds wire to the same "GND". This way you can ensure all passive probes share the same ground. 

2. In the first new waveform, does the camera enter the streaming mode at the "a" cursor point? And your camera keeps on before and after entering the streaming mode?  

3. You mentioned before that " in some cases, the camera stops streaming, although the device itself remains powered on.", if the camera just cannot streaming but it is still on (assuming no other power sources beside the PSE), this means the power delivery from PSE to the camera was always exist. In your measurement, if it is "P1_Drain" - "GND" voltage, 400ms >20V is beyond the overcurrent protection time and should already cut the power before that. So far it is more like the PSE waveform is a measurement issues for now. But would like to get clarified with more information. 

Best regards,

Diang

Hi Diang,

Thank you for your response.

1. Regarding the first point, as you mentioned, I used only a single probe for probing, and the other channels are not connected to my board's ground.
2. For the second point, the cursor "a" indicates that I can access the device IP only after that specific time period. Until then, the device remains off. Once the pulse begins, I am able to ping the device and start streaming.
3. As for the third point, I didn't fully understand it. Could you please elaborate in more detail? If you need any further information, kindly let me know.
   Note: I don't have other power source beside PSE for my camera. 

 
Best regards,
Poovarasan S

Hi Poovarasan,

Thanks for your reply!

Since you do not have other power source beside PSE for my camera, can you check if the camera is still on when the streaming function stops? Just want to double check if the streaming stop is caused by camera losing power or just the camera streaming function get disturbed. Could you measure the current flow through your camera by using a current probe on the Ethernet wire, so we can also view the power interface current to double confirm? 

Since from the waveform aspect, the timing (400ms or longer) could be too long to be considered for a PSE foldback protection & power cut off. Besides, if you can dump the PSE register, it will be very helpful to check if the PSE detects any faults.

Best regards,

Diang