• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Subscribers 33 subscribers
  • Views 700 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

TPS65982: TPS65982 using the external PATH at < 20V.

WST
Expert 1025 points

Part Number: TPS65982

I was experimenting with a TPS65982-EVM, and also my own board with a TPS65982 part on it. I wanted to see how warm the TPS65982 on the EVM layout gets when sinking 20V/5A and 12V/3A. I used a 20V brick supply with the EVM and set the switch to 6, to provide 5V, 12V, or 20V. I then set my board to ask for 5V and 12V. The EVM sets up the 12V contract without a problem. I then sink 3A from the EVM and it handles this, but the TPS part on the EVM board gets hot to the touch. I can repeat this with 20V @ the 5A limit of the spec using a marked cable. With 20V the external FETs on the EVM are used and the TPS part doesn't get hot and the external FETs on the EVM board seem to stay cooler than the TPS part with 3A.
So, I wanted to setup the EVM so that the PP_EXT path would be used even with the 12V contract so that the external FETs would carry the current, not the TPS part. The spec sheet says that I can even go down to 5V using the FETs on the EVM board. I was planning on doing this on my board also with higher current contracts. So, I connected an Aardvark to the EVM board and changed the Source capabilities register so that only the 5V and 12V contracts were present. I set my board to have a 5V and 12V sink contract at 3A. I then changed the EVM Source capabilities register to use the PP_EXT path. When I connect the boards the contract is made but no PSRDY happens. I see a source capabilities message and a Request for the 12V and an Accept, but no PSRDY. It then repeats with a SrcCap again with the lights on the EVM repeatedly blinking on and off. If I switch back to the PP_HV path, everything works fine. I removed my board and used a second EVM set to SW=7. I again edited the SrcCap register on the EVM set to 6 to use the PP_EXT path and I had the same problem, the PP_EXT path can't be used with 12V.

Is there something else that needs to be changed to allow the PP_EXT path to be used for the 12V contract? I would rather do this as I said, because then the TPS part wouldn’t carry the 3A contract current. I also tried changing the System configuration register so that the PP_HV path was disabled, but this didn't help.T

  • 0
    Expert 1025 points
    A follow up. I realized that even though the EVM board had the FETs on it, that in the normal configuration the FETs were connected to J6 which is always 20V (or whatever the power brick is). So, that would explain why when I tried to use the PP_EXT path as the 12V source it wouldn't work. However, I then removed jumper J6 and connected the HV_SOURCE (12V) from J5 pin 1 over to J6 pin 2 to provide 12V to the other side of the FETs. I changed the internal switch to PP_EXT for the 12V source contract as before. This time it made the PD contract at 12V using the PP_EXT path. However, when I try to load this at 3A I see hard resets on the PD traffic. It won't take any load at all. Is there some current limit programming I'm missing that is causing this?
  • 0 in reply to WST
    Expert 1025 points

    There was another problem, I found that the load I was using had a voltage limit that was keeping the voltage from reaching 12V. Therefore the EVM was doing its job and shutting off when it couldn't reach the contract voltage. Once I raised that limit to be above 12V I found that it made the contract using the PP_EXT path with no problem and I could load it with 3A.

    I also tried the external path at 5V on the EVM and this worked also when I changed the jumpers to use J5p2 as the source supply (the 5V regulator).

  • 0 in reply to WST
    TI__Mastermind 36756 points
    Hello,

    Thank you for keeping us up to date with this issue/solution. I will now close this thread.

    Thank you,
    Eric