TPS65220: TPS65220 PMIC not generating power supply on some boards

Hi Mohabbat,

Thank you for reaching out on e2e and providing all the details.

Looks like Buck3 RV (Residual Voltage) is causing the PMIC to not power up. It re-tries 2 more times and shuts down.

Before trying the next power up, can you check the Buck3 output voltage and try to completely discharge and try powering up again.

Sathish

Hi Sathish,

Thanks for feedback.

We have checked the waveform for the below supplies rails:

BUCK 3:

BUCK 2:

LDO 1:

LDO 3:

LDO 4:

We are not observing any output on the LDO 2(VDDAR_CORE) and Buck 1 (VDD_CORE).

Can you let us know which fault is causing this issue based on the PMIC register dump we have shared and also let us know how to fix this issue. This is the very critical issue and we need to resolve it soon.

Regards,

Mohabbat Singh

Hi Mohabbat,

Thank you for the waveforms. All the voltage levels are good except Buck3, it is not able to reach the full 1.1V and it could be because of several reasons like bad inductor, leakage from MCU, board manufacturing issues etc..

Did you check Buck3 voltage when PMIC is OFF?

Sathish

Hi Sathish,

This issue is not resolved fully, as we are seeing PMIC is unable to turn ON each time and during that time we are observing the same buck 3 waveform that i have shared.

For example, if are trying to turn ON PMIC 10 times then its able to turn ON properly 6~7 times only and for 3~4 times its failing due to buck 3. We have increased the capacitor value to 66uF but still facing same issue.

Looking for the fix that completely resolve this issue.

Regards,

Mohabbat Singh

Hi Mohabbat,

Increase all the Buck output capacitance to >80uF and check.

You can also plot all the output rails as shown in TPS65220 TRM power up sequence diagram and check if any other rails are causing the PMIC to not power up.

Sathish 

Hi Sathish,

We are observing this issue again on same board, even after mounting the 80uF capacitor and we need the soultion which can resolve this issue fully.

Regards,

Mohabbat Singh

Hi Mohabbat,

Is the error related to Buck3 (register 0x32=04) or any new issue?

When you tried repeated ON and OFF, did you make sure all the rails are completely discharged before starting the new power up? PMIC will engage the discharges before the rails are enabled but if a rail is not fully discharged then PMIC aborts the power up sequence.

Sathish

Hi Sathish,

Yes, same error we are getting in same board with both cases (working or non working). Below are the log that we have captured with the same board that have 80uF capacitor mounted on it.

PMIC log when issue occurred:

root@am64xx-bidi-proto-1:~# i2cdump 0 0x30
No size specified (using byte-data access)
WARNING! This program can confuse your I2C bus, cause data loss and worse!
I will probe file /dev/i2c-0, address 0x30, mode byte
Continue? [Y/n] y
     0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f    0123456789abcdef
00: 80 53 7f 0b a6 98 05 76 94 a4 86 21 21 50 21 30    ?S?????v???!!P!0
10: 21 41 80 60 02 60 a9 5d c0 00 f0 00 00 00 02 41    !A?`?`?]?.?...?A
20: c6 7f 02 00 00 e0 30 00 02 00 7f 01 00 00 00 00    ???..?0.?.??....
30: 00 00 04 04 00 0e 00 00 00 00 00 00 00 00 00 00    ..??.?..........
40: 03 40 38 22 07 63 cd 2e 81 80 15 15 16 0e 0f 3f    ?@8"?c?.????????
50: 47 00 00 17 03 00 88 0d 24 a0 26 36 63 44 6c 1f    G..??.??$?&6cDl?
60: 24 26 20 24 80 17 36 63 55 68 20 23 24 20 24 60    $& $??6cUh #$ $`
70: 30 36 63 67 74 1e 24 2a 1f e4 e4 e4 24 24 24 e4    06cgt?$*????$$$?
80: e4 e4 0f d6 7b 15 27 26 8d 13 27 1a 66 1c 1e 1f    ????{?'&??'?f???
90: 08 1d 26 21 08 26 00 00 00 00 00 00 00 00 00 00    ??&!?&..........
a0: 00 26 08 00 00 00 00 00 00 00 00 00 00 00 00 00    .&?.............
b0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00    ................
c0: 00 00 00 00 00 08 00 00 04 00 00 00 3f 87 08 00    .....?..?...???.
d0: 80 53 7f 0b 00 1f 00 1e 00 1e 52 00 52 2c 00 2c    ?S??.?.?.?R.R,.,
e0: 18 00 00 00 05 00 00 00 18 00 34 00 ec 00 00 00    ?...?...?.4.?...
f0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00    ................

PMIC log when issue not occurred:

root@am64xx-bidi-proto-1:~# i2cdump 0 0x30
No size specified (using byte-data access)
WARNING! This program can confuse your I2C bus, cause data loss and worse!
I will probe file /dev/i2c-0, address 0x30, mode byte
Continue? [Y/n] y
     0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f    0123456789abcdef
00: 80 53 7f 0b a6 98 05 76 94 a4 86 21 21 50 21 30    ?S?????v???!!P!0
10: 21 41 80 60 02 60 a9 5d c0 00 f0 00 00 00 02 41    !A?`?`?]?.?...?A
20: c6 7f 02 00 00 e0 30 00 02 00 7f 01 00 00 00 00    ???..?0.?.??....
30: 00 00 04 04 00 18 00 00 00 00 00 00 00 00 00 00    ..??.?..........
40: 03 40 38 22 07 63 cd 2e 81 80 15 15 16 0e 0f 3f    ?@8"?c?.????????
50: 47 00 00 17 03 00 88 0d 24 a0 26 36 63 44 6c 1f    G..??.??$?&6cDl?
60: 24 26 20 24 80 17 36 63 55 68 20 23 24 20 24 60    $& $??6cUh #$ $`
70: 30 36 63 67 74 1e 24 2a 1f e4 e4 e4 24 24 24 e4    06cgt?$*????$$$?
80: e4 e4 0f d6 7b 15 27 26 8d 13 27 1a 66 1c 1e 1f    ????{?'&??'?f???
90: 08 1d 26 21 08 26 00 00 00 00 00 00 00 00 00 00    ??&!?&..........
a0: 00 26 08 00 00 00 00 00 00 00 00 00 00 00 00 00    .&?.............
b0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00    ................
c0: 00 00 00 00 00 08 00 00 08 00 00 00 3f 87 08 00    .....?..?...???.
d0: 80 53 7f 0b 00 1f 00 1e 56 1e 52 56 52 2c 56 2c    ?S??.?.?V?RVR,V,
e0: 36 04 4a 4a 05 04 11 11 18 2e 34 2e ec 00 00 00    6?JJ?????.4.?...
f0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00    ................
root@am64xx-bidi-proto-1:~#

We are allowing the enough time to PMIC to discharge.

Regards,

Mohabbat Singh

Hi Mohabbat, 

Sathish is currently out of office, let me try to answer your questions in the meantime. 

The I2C register readings are still indicating that a shutdown occurred due to RV/DISCHARGE_TIMEOUT on BUCK3 occurred (0x32= 04)
However, No RV event is flagged in register 31.
And no timeout event is flagged in bit 7 in register 32. 

If you write '1' to clear bit 2 of 0x32 and read again, is this bit still reading as 1 in both cases?

Both logs are showing these faults, but the device where the issue is observed is moving to INITIALIZE state after shutdown event,
while the device with no issue is able to move to ACTIVE state.

As Sathish mentioned earlier, are you able to measure voltage on Buck3 while PMIC is OFF to ensure there is absolutely no residual voltage?

Best Regards, 
Sarah

Hi Sarah,

We did not observed any residual voltage on buck3.

Regards,

Mohabbat Singh

Hi Mohbbat, 

Thank you for the confirmation.

Any comment regarding re-reading 0x32 of the boards with no issue detected?

Sarah Lao said:

If you write '1' to clear bit 2 of 0x32 and read again, is this bit still reading as 1 in both cases?

Best Regards, 
Sarah