TPS65983B: GRL for PD test fail

Hi Rock,

What is the GRL test number associated with this fail?

Thanks,

Emma

Hi Emma,

The customer went to GRL for testing, no matter whether the PD is set to 96W or 90W, its OPP will exceed 100W and the protection will be activated at 106W.
Could it be that the Firmware Tool "TPS65983 Thunderbolt Application Customization 4_14" cannot manually set the PDO over 85W?
Maybe it is manually changed to 96W, but some protection mechanisms have not been changed. Please help confirm whether PD 96W can be modified through this version of TOOL?
Can TI provide another firmware?

Regards

Rock

Hi Rock,

Could you please follow up on a few questions for me so that I can have a clear understanding of this issue?

1. What is your GRL test number associated with this fail?

2. What is the reference design you are following?

3. Have you made any changes or updates to the pjt file?

Thanks,

Emma

Hi Emma

1. What is your GRL test number associated with this fail?
wait for customer feedback

2. What is the reference design you are following?
Tapex Creek

3. Have you made any changes or updates to the pjt file?
Please check the following pic and pjt file, the customer only changes the port power capability to 96W

Thunderbolt Pro Dock V6.67-96W-15W.pjt

Regards

Rock

Hi Rock, 

Can you send me a picture or the project file of what their highest source cap looks like? They also need to change that.

Thanks,

Emma

Hi,
I would like to be friends with you on the TI E2E Support forums. Thank you!

Hi Rock,

I accepted your friend request. Please share the project with me there. 

Thanks,

Emma

Hi Emma
I have shared the PJT & FW for you in a personal message, please help to check this issue, thanks.

Hi Rock,

Thanks for sharing the project. I will take a look and follow up with you in a few days.

Thanks,

Emma

Hi Emma,
Do you have any updates about this issue? thanks.

Hi Rock,

Just to confirm you are seeing an issue with the OPP not turning on at 96W, however the customer is able to source 96W. Is that correct?

Thanks,

Emma

Hi Emma,

YES， It can provide 96W, but OPP will not start when the pumping load exceeds 100W.

Hello,

I understand. Let me follow up with the firmware team to see if this is a bug we can reproduce on our side. I will follow up with you on Monday.

Thanks,

Emma

Hi Emma:

Our new product EV TEST only has OPP test Fail. We look forward to your good news.

Hello,

Our firmware switches have default settings. With an output of 96W, you will have 4.8A of current at 20V. This will select PP_EXT current limit (detailed on page 15 of the datasheet). This should select setting 14 for your current limit. Setting 14 has a max of 5.529A, so it is expected behavior for this current limit to exceed 100W at its high end. 

In order to check which current limit is being programmed, I can check the switch setting. Can you read the register 0x68 before you apply the short and let me know the result? 

Thanks,

Emma

Hi Emma:

1、Please tell me how to read the value of register 0x68?

2、Thunderbolt 3 certification must test OPP (the specification does not exceed 100W, and the output must be stopped within 5 seconds when it exceeds 100W) Can TPS65983B do it?

3、The results of GRL test 90W and 96W are the same Fail.

Hello,

You can use the GUI tool in debug mode to do a register read and then follow up with me with the data. The 90W test has the potential to have a max current allowed of up to 98.3W, depending on the setting. Please check this register and let me know the result. 

Thanks,

Emma

Hi Emma:

I can't get the GUI tool to read the data in the register, maybe because it can't be selected through the Raiden interface, so I clicked "Read" and it was disconnected.

In addition, I cannot find the location of the 0x68 register. Which page is this?

For the 90W setting, we tested it in GRL and the result was Fail! The output stops until 106W.

Hello,

Please use the command tab and use the Raw Register Read field, as shown below. 

Thanks,

Emma

Hi Emma:

Do I need to connect to the Device? How do I set the GUI Tool to connect to the Device I connected?

Hello,

You will have to do an I2C sweep using the type of connector you have (Tiva, Aardvark, etc). It should find the proper address. Then enter debug mode. https://training.ti.com/tps6598x-gui-tool-displayport-alternate-mode-debugging?context=1140344-1140208 this video explains more about debug mode.

Thanks,

Emma

HI Emma:

1、I cannot link to the device here. I will ask Rock Shao for help.

2、Can you tell me why I want to read the contents of register 0x68?

      Does this help me with the problem I am responding to?

Hello,

You are reading the contents so that we can determine the current limit setting to see if what we are seeing is expected behavior or do we have an issue with our current limiting.

Thanks,

Emma 

Hi Emma:

Why can't I read anything?

Hello,

It looks like you aren't connected to anything. Which gui version are you trying?

Thanks,

Emma

HI Emma:

This is the version of the gui tool I used.

Hello,

I will check in with the firmware team and see if you should be seeing an overcurrent behavior at 90W. Can you try some other powers and let me know the results?

Thanks,

Emma

HI Emma:

ROCK helps me read the value of register 0x68, please refer to the figure below.

Hello,

Was this for a setting of 96W or 90W?

Also, please read 20 bytes of the register and let me know the result. The relevant bytes are 9-12. 

One other question, what value for Rsense are you using? 

Thanks,

Emma

Hi,

That is setting for 96W.

Does TPS65983B have the tolerance at the OCP function? For our PD FAE say, there has a tolerance problem in the OCP, so suggest design under 90W to guaranteed to be less than 100W at OCP.

HI Emma:

Rsense : 5m ohm

Hi Rock,

Can you read all 20 bytes of the register?

And yes you are correct, this is a tolerance issue with the OCP function. Your thought is right here, to lower the power to lower the OCP. Can you try lowering the power offering and let me know your result?

Thanks,

Emma

Hi Emma,

Please see the result

without load

with load

Re check (no plug)

Hi Rock,

Emma will be able to review your results next week as she is currently out of office this week.

Thank you,

Hari

Hi Emma:

Are there any new developments?

Hello,

The firmware team is currently evaluating the register values. Their other feedback is below: 

     For 90W the current limit (w/o any margin) would be 4.5A. This would mean PHV_EXT ILIM code of 14. This code allows ILIM between 4.523A to 5.529A i.e., 90.46W to 110.6W. This could possibly explain the 106W that customer reported.

Can you please try a power setting of 80W to ensure that the setting is setting 12, which has a max power possible of 98.3W?

Thanks,

Emma

HI Emma:

When we set PDO=85W, the tested data is 89.6W. (85W products have passed Thunderbolt certification)
There is only 4.6W more, but when the PDO is set to 90W or 96W, the measured data is 106W.
Maybe this is within the error range that you recognize, but because the Thunderbolt certification test does not allow more than 100W, does it mean that I can't use TPS65983B to set the PDO to more than 90W? because the test will exceed 100W.

What we want to confirm is this problem, whether TPS65983B can set the PDO to 90W or 96W, but it can be limited to 100W during the OPP test, if not, please let us know.

Hello,

What can we do next? Our product PDO wants to be changed to 96W,
Is it possible to modify the firmware to PDO = 96W, and the OPP test can also be within 100W?

or can those parts of HW be modified?

Hello,

The current limit ranges cannot be changed, as they are intrinsic to how the part operates and current limits. The TPS65987DDH has adjustable current limiting but this configurability is not offered with the TPS65983B.

Thanks,

Emma