TPS25750: Dead Battery mode

Good Morning Chen,

I will give the code a look today and update you by Monday EOB, at the latest.

Are you receiving any errors, or do you have any logs that show the behavior of the code?

Kind regards,

Conner Gillette

Hello Conner:

The code didn't report error, but it didn't help the battery recover from Dead Battery Mode. It always required a power cycle on the USB-C terminals to set the battery into normal charging mode.

Based on my understanding, the battery should enter normal charging mode immediately when the flag was cleared. For example, if the battery was fully discharged and triggered CUV protection, it should wake up when USB-C plugged in, and the MCU can clear the Dead Battery flag, and then the battery can charge normally. But based on our test with this code, the battery only charged at 1A, until the USB-C removed and plugged-in again.

Do you have any update on this? 

Thank you!

Chen

Can you provide me the values stored in PowerPathStatus and 0x26 when being called by I2Cs_GetsData function? Is there any way to provide a debug log of the values of each variable as you step through each line of this code?

Kind regards,

Conner Gillette

Hello Conner:

We will try to collect the data as you requested above.

At the meantime, we are trying to test 9V and 15V charging function, so I generated 9V/27W only and 15V/45W only BIN files. If we are using an universal USB-PD power supply, which can cover 5-20V, how can we configure the BIN to get 9V/15V out of it? Based on our test so far, it will automatically output the highest power(voltage), even the BIN file dedicates to 15V/45W. Could you please advise? I believe the TPS25750 should be able to negotiate to get the required voltage. 

Thank you!

Chen

Would you mind clarifying: You want a single bin file that is capable of 9V/3A and 15V/3A, and always selects the higher voltage when available?

Regards,

Conner

Hello Conner:

Yes, that is correct. We want to test 9V and 12V charging behavior, including EMI/EMC, so I create two BIN files, one with 9V/27W only, and the other one 15V/45W only. But when we tested on the universal USB-PD power supply which claimed from 5V to 20V/60W, the power supply always outputs 20V. Also we tried a 5-12V power supply with the 9V/27W BIN, 12V was automatically selected.

Thanks.

Chen

BTW, I used 7.0.3 GUI to generate the BIN, but I think it should be functional on this simple configuration.

Could I have your settings file (JSON format) from the GUI? And could you let me know if your device is behaving as a source, sink, or dual role?

Hello Conner:

It is sink only. I still have the issue to upload the file here. The settings are simple:

1. Sink only with TPS25750D

2. N/A

3. 27W (9V)

4. N/A

5. No USB data

6. N/A

7. No

8. No

9. BQ25790 or BQ25792

10. 12.6V

11. 2A

12. 0.2A

13. 0.2A

Thanks.

Chen

What are the sink PDOs in the advanced configuration? (register 0x33)

Hello Conner:

I used to have 2 PDOs in one BIN file in the advanced config: 5V/3A and 9V/2A in one; and 5V/3A and 12V/2A in the other one. But it has the phenomenon as mentioned above.

Now I just use the basic config, without any advanced config, so just 9V/3A and 12V/3A.

Thanks.

Chen 

Please select 'Advanced Configuration' > 'Enable Advanced Configuration' > the 'Configuration' tab > 'Transmit Sink Capabilities (0x33)' and tell me the value of each listed Sink PDO.

Hello Conner:

It has 2 PDOs: 3A/5V, and 3A/9V

Thanks.

Chen

You had previously said "But when we tested on the universal USB-PD power supply which claimed from 5V to 20V/60W, the power supply always outputs 20V".

I assume that you have a 20V PDO in your sink capabilities, but now that we've discussed multiple bin files, I don't know if we're talking about the same bin file.

The problem is negotiating 20V instead of 15V, correct?

Hello Conner:

Sorry for the confusion. Yes, there are several BIN files under testing, and we should just focus on one, the 15V/2A BIN file. 

It has 2 PDOs, 5V/3A and 15V/2A. But based on our test, the input voltage was still 20V.

Thanks.

Chen

I sent you a message with my email address requesting the settings file.

Regards,

Conner

Hello Conner:

For the Dead-battery issue, I just got the data of PowerPathStatus, and it is 006b; and I confirmed with FW engineer, even it was in "dead battery mode" and 1A charging, the value was still not 10b. We are using "state = PowerPathStatus[5] & 0b11000000", so whenever PowerPathStatus = 10b, it will return state = 80. However, we never saw this value during our test.

Thanks!

Chen

Can I have a full dump of the entire register's contents? (0x26) It should be 5 bytes long.

I've received the json and bin files. The json files look good, so tomorrow I will test the bin files on an EVM.

If that doesn't yield results, I'll ask you to email me a copy of your schematic.

Kind regards,

Conner Gillette

For both bin files, I was only able to negotiate the 15V or the 9V (for their respective bin file). Could you go ahead and send a copy of your schematic and/or your setup that is showing you 20V passing through?

Kind regards,

Conner

Hello Conner, the root cause was the EEPROM writing issue. Now this is fixed. Thank you!.

That's great to hear!

Kind regards,

Conner