TPS65994AE: USB Type-C UCM Disable Enable Test Fail

Marty Kuo

Prodigy 110 points

Part Number: TPS65994AE

Tool/software:

Dear Sir:

We have HLK test，At present, HLK has been tested and has come to an end. The Fail items are as follows:

if the attached log can identify the issue?Thanks~

USB Type-C UCM Disable Enable Test_0617.htm

1 month ago

0 Kevin Kosta 1 month ago

TI__Genius 13365 points

Hi,

Thanks for reaching out on E2E!

An expert will get back shortly.

Thank you,

Kevin

0 Zoe Ko 28 days ago in reply to Kevin Kosta

TI__Expert 3480 points

Hi Kevin,

Any update?

Raymond visited the customer before and should know the issue more.

Could you help check with him for the issue and reply soon?

Thank you very much!

0 Tommy Lin 28 days ago in reply to Zoe Ko

TI__Mastermind 20765 points

Hi,

There is another thread on e2e (https://e2e.ti.com/support/power-management-group/power-management/f/power-management-forum/1385321/tps65994ae-hlk-test-fail-pd-f-w-will-always-send-out-the-battery-charger-state-causing-ppm-fail)

Is the other thread the latest update of this issue?

Regards

0 Zoe Ko 27 days ago in reply to Tommy Lin

TI__Expert 3480 points

Hi Tommy,

The link page not found.

0 Tommy Lin 26 days ago in reply to Zoe Ko

TI__Mastermind 20765 points

Hi,

Seems like the thread got removed, but basically the other thread reported that PD is always sending some message related to battery and cause the ppm test to fail.

Can you ask them to provide the PD log and I2C1 log when the test failed?

Regards

0 Marty Kuo 26 days ago in reply to Tommy Lin

Prodigy 110 points

Hi Tommy,

PD log and I2C1 log as attach file，Thanks~

1104.Clevo_eGFX_PortA_PD_Logs.zip

0 Tommy Lin 25 days ago in reply to Marty Kuo

TI__Mastermind 20765 points

Hi,

I do not see the I2C1 log in the zip file, it only contains PD log. Can you provide the I2C1 log again?

In terms of the PD, I do not see much difference between the failed and pass case.

The error in the reported started with Partner is attached, expected to be detached. But in pass and failed log I did not see disconnect happening at the end of the log.

Regards

0 Marty Kuo 20 days ago in reply to Tommy Lin

Prodigy 110 points

Hi,Tommy,

I2C1 log，as attach file，thanks your support~

x37_I2C.csv

0 Tommy Lin 19 days ago in reply to Marty Kuo

TI__Mastermind 20765 points

Hi,

Can you detail the test procedure of this UCM disable test? From the I2C log it showed that the port partner is still connected to PD (physically) and PD did not receive any command from EC to disconnect.

Regards

0 Marty Kuo 16 days ago in reply to Tommy Lin

Prodigy 110 points

Hi Tommy,

The problem is that pd always returns the battery charging status through pd report, so why does pd fw always return chrger?
I would like to ask if there is a way to set GRT_CONNECTOR_STATUS 49:48 to 00b (No Charging)in the PJT of PD_FW, as shown in the following table

Our Pjt file as attach，Thanks~

8171.X370_TBT5_TPS65994BH_F909p12p15_Tiger_PDv3115_20240619.pjt

0 Tommy Lin 15 days ago in reply to Marty Kuo

TI__Mastermind 20765 points

Hi,

May know the reason why you want this to report as 00b?

Do you have the Battery Charging Status Change notification enabled or disabled in the UCSI interface?

Regards

0 Marty Kuo 15 days ago in reply to Tommy Lin

Prodigy 110 points

Dea Tommy,

This test is to plug in the PD device and perform the USB Type-C UCM Disable Enable Test. The OS will disable Type-C through the UCSI interface, but the PD f/w continues to send battery charge status, which means that the device is still there, so the test fails.

BR,

Marty

0 Tommy Lin 14 days ago in reply to Marty Kuo

TI__Mastermind 20765 points

Hi,

The problem here is although OS send the disable command to EC, PD never receive the command from EC to disable the port prior to the UCSI get connector status command, that is why the connection is still there and it is reporting charging. I would suggest you disabling the PD by writing to 0x28 port config register to disable the port when EC receive the disable type-C command from OS.

Regards

0 Marty Kuo 12 days ago in reply to Tommy Lin

Prodigy 110 points

Hi Tommy,

Please tell me, regarding BATTERY CHARGER STATE, why does PD F/W always reply to charge? It has nothing to do with PORT DISABLE/ENABLE at present, but PD replies CONNENT STATE error. In addition, UCSI does not need to download 0x28 REGISTER according to TI and MS SPEC. Please TI Provide UCSI CMD which group of OS PPM POLICY CMD needs to be DISABLE PORT.Thanks~

0 Tommy Lin 12 days ago in reply to Marty Kuo

TI__Mastermind 20765 points

Hi,

Please note that the PD is the LPM in the USCI scheme and does not directly support UCSI command from OS/OPM, hence why there is the need of driver in PPM that parse the UCSI command from OS/OPM and why in the device TRM the UCSI commands are in 4CC command format.

If PPM receive a port disable command from OPM then PPM should write to 0x28 register and disable port. Likewise if OPM needs to re-enable the port PPM needs to rewrite the 0x28 register again.

Regards

0 Marty Kuo 9 days ago in reply to Tommy Lin

Prodigy 110 points

Hi, Tommy,

According to the TI SPEC UCSI 4CC CMD PPM flow chart, there is no process for closing the 0.28 register.
1. Please confirm which PPM CMD you want to execute,
2. This still has no answer why TI keeps F/W REPORT BATTERY CCHARGING STATE. This has nothing to do with closing PORT. The current UCSI HLK is caused by TI returning to charging state.
Please reply to the above questions, thank you

Marty

0 Tommy Lin 8 days ago in reply to Marty Kuo

TI__Mastermind 20765 points

Hi,

The listed UCSI 4CC commands are the 4CC command that PPM can use when receiving the corresponding UCSI command from OPM. If the corresponding UCSI 4CC command is not listed then you will need PPM write to the relevant register, such as writing to 0x28 to disable the type-C port.

The PD reporting battery charging state is requested by Intel.

Regards

Power management

Power management forum

TPS65994AE: USB Type-C UCM Disable Enable Test Fail