• State Resolved
  • Locked Locked
  • Replies 9 replies
  • Subscribers 63 subscribers
  • Views 1026 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.

TPS65994AE: Debug accessory mode while powered only from VBUS

Mikko Koutonen
Prodigy 20 points

Part Number: TPS65994AE
Other Parts Discussed in Thread: TPS65994QFNEVM, TPS65994

Hi,

I'm trying to use the debug accessory mode while the PD controller is powered only from VBUS (in dead battery mode).

I have measured that LDO_3V3 (and LDO_1V5) are present on the board and that the config is loaded from the external EEPROM (traffic seen on the I2C3m bus). I have configured Debug Accessory Mode event to one of the GPIOs. I would expect that the PD controller would now detect the debug accessory cable attached and change the state of the GPIO accordingly. But this does not happen.

For reference, the debug accessory mode event is working well while the PD controller is powered from VIN_3V3.

Is there something that should be done before the debug accessory mode detection starts to work while only powered from VBUS?

Thank you!

Mikko

  • 0
    TI__Genius 15195 points

    Hi Mikko,

    Let me confirm this with the team but I suspect that the PD is not able to enter debug accessory mode while still in dead battery mode. When booting in dead battery mode (no VIN_3V3 initially), you need to manually clear the dead battery flag by issuing a 'DBFg' 4CC command to the PD controller via I2C. The PD controller does not automatically clear the dead battery flag and will remain in dead battery mode until the flag is cleared. 

    I will confirm whether or not debug accessory mode can be entered in dead battery mode and get back to you.

    Best,

    Alex

  • 0
    TI__Genius 15195 points

    Hi Mikko,

    You should actually be able to enter debug accessory mode in dead battery. 

    Can you confirm whether the PD is unable to detect debug accessory and enter debug accessory mode, or is only unable to trigger the GPIO event? Please use the Application Customization GUI or an MCU to read 0x5F Data Status register and 0x69 TypeC State register to see if the PD enters debug accessory mode while in dead battery. Please send register contents or a screenshot from the GUI.

    Best,

    Alex

  • 0 in reply to Alex Liu
    Prodigy 20 points

    Hi Alex,

    Thank you for the reply! I have attached a zip file with screenshots from the GUI showing registers 0x5F and 0x69 in both dead battery mode (VBUS) and in normal mode (VIN_3V3).

    To me 0x69 seems that in dead battery mode the PD controller is unable to detect the Rd resistors on CC lines (we have 5K1 on both). Also I find it strange that 0x5F shows Debug Alternate Mode=False even while powered from VIN_3V3 when the debug mode is actually working. Also USB2 Connection shows a connection there; we actually mux UART to USB2 data pair while in debug mode.

    I was also able to reproduce the 0x69 register state (no Rd resistors detected) on TI EVM board (TPS65994QFNEVM).

    Br,

    Mikko

    Debug accessory issue.zip

  • 0 in reply to Mikko Koutonen
    TI__Genius 15195 points

    Hi Mikko,

    From your screenshots, it looks like the PD is not entering debug accessory mode at all when in dead battery. The PD is not detecting the debug accessory device, as seen in the 0x69 TypeC State register. Debug alternate mode is something different.

    Can you send me your .pjt GUI project file and the firmware base image number you are using? I can try and replicate it on my side.

    For clarification, what debug accessory device are you connecting at the far-end? Is the debug accessory device providing VBUS to the PD controller?

    Best,

    Alex

  • 0 in reply to Alex Liu
    Prodigy 20 points

    Hi Alex,

    Please find attached .pjt file. Firmware base image is TPS65994_F609_06_10.bin.

    We have in-house made debug cable. As I mentioned, in debug mode DUT muxes UART to USB2 data pair instead of USB data. At far-end we have UART-to-USB bridge which connects to test PC to view the UART traffic. VBUS is provided via the same cable to PD controller.

    Br,

    Mikko

    0005.config.pjt

  • 0 in reply to Mikko Koutonen
    TI__Genius 15195 points

    Hi Mikko,

    I will take a look at the project file and try to do some testing on my side. I will update you in 1-2 days.

    Best,

    Alex

  • +1 in reply to Alex Liu
    TI__Genius 15195 points

    Hi Mikko,

    I took a look at your project file and all looks good. One thing I want to check is the debug accessory detection process.

    I thing something is going wrong when the PD does cable/far-end device detection. Can you capture the analog signals for CC1 and CC2 so I can see what voltage level they are at? If you are using the EVM, there are test points for CC1/CC2 on each port. 

    • When the PD is not in dead battery mode, the PD is the source and the debug accessory device is the sink. The debug accessory device must provide Rd/Rd pull-down on both CC lines for the PD to detect a debug accessory connection. This I believe is fine as you successfully enter debug accessory mode when not in dead battery.
    • When the PD is in dead battery mode, the PD is the sink and the debug accessory device is the source. The debug accessory device must provide Rp/Rp pull-up on both CC lines for the PD to detect a debug accessory connection. It is possible that the Rp being presented to the PD in dead battery mode is not there or is too weak for detection.

    Best,

    Alex

  • 0 in reply to Alex Liu
    Prodigy 20 points

    Hi Alex,

    Before going to the lab to capture the waveform I'd like to point out the second bullet from your message.

    You said: "The debug accessory device must provide Rp/Rp pull-up on both CC lines for the PD to detect a debug accessory connection."

    Our debug cable has physical Rd pull-down resistors on CC1 and CC2. From your answer I understand that in dead battery mode we should have pull-ups instead of the pull-downs. Could that be the root cause?

    Br,

    Mikko

  • 0 in reply to Mikko Koutonen
    Prodigy 20 points

    Hi Alex,

    I confirmed with 56K pull-ups on CC1 and CC2 that this was the root cause. With pull-ups the debug accessory mode is now working in dead battery mode.

    Actually, it seems that the cable with the pull-ups is working also in normal mode where PD controller is powered via VIN_3V3 if I feed VBUS from the cable to PD. With VBUS disconnected it only works with pull-downs, which I think makes sense.

    Register 0x69 shown below in condition where debug is working with 56K pull-ups and PD controller is in dead battery mode:

    Thank you for the help on solving this issue.

    Br,

    Mikko