TPS65987D: PD negotiation not consistent for 20V

Hi,

Could you please provide PD logs that I could take a look at? Also, what are your ADC values?

Thank you,

Hari

I assume you mean the PD State Machine Trace from the GUI, right?   This is from the last connection.

Port 1 Module 31 0x1ff = UNKNOWN

Port 0 Protocol 0xcb = PRState_RECEIVED_GOODCRC

Port 0 Protocol 0xc0 = PRState_SQUELCH_ACTIVE

Port 0 Protocol 0xc1 = PRState_SQUELCH_IDLE

Port 0 Protocol 0xc5 = PRState_RX_BUF_RDY

Port 0 PD 0x28 = PEState_Sink_TransitionSink

Port 0 Protocol 0xc0 = PRState_SQUELCH_ACTIVE

Port 0 Protocol 0xc1 = PRState_SQUELCH_IDLE

Port 0 Protocol 0xc5 = PRState_RX_BUF_RDY

Port 0 PD 0x29 = PEState_Sink_TransitionSink_PS_RDY

Port 0 PD 0x2a = PEState_Sink_Ready

Hi,

No sorry, I meant the PD logs showing the CC communications from a PD analyzer such as the one from Total Phase. 

Thank you,

Hari

Ah ok. I have a scope that can decode PD messages.  I think I can capture the entire negotiation but I'm not sure how to relay the information.  I'm seeing a lot of Rp-3.0 messages. Would you prefer the straight binary?  Sorry I'm not familiar with the encoding

Hi,

I am not familiar with reading the PD messages this way as I'm not sure how to decode the binary.

If you do not have a PD analyzer, would you be able to test this on an TPS65987D EVM? I can not this issue on my EVM. What are your ADC values from the board?

Thank you,

Hari

I don't have a tps65987d EVM, only the tps25750EVM unfortunately.  My adc values:

ADC1 R1 = 10k

ADC1 R2 = 20k

ADC2 R1 = 8.2k

ADC2 R2 = 10k

I'll see if I can capture more information on my scope.

Hi, I will take a look from your ADC values and see if you may need to change the configuration.

Thank you,

Hari

ok, so I've got data.  Here is the sequence messages:

PwrRole/Type: 1 MsgType: SourceCapab

PwrRole/Type: 1 MsgType: SourceCapab

PwrRole/Type: 1 MsgType: SourceCapab

PwrRole/Type: 1 MsgType: SourceCapab

PwrRole/Type: 1 MsgType: SourceCapab

[break for 180 ms]

PwrRole/Type: 1 MsgType: SourceCapab

PwrRole/Type: 0 MsgType: GoodCRC

PwrRole/Type: 0 MsgType: Request

PwrRole/Type: 1 MsgType: GoodCRC

PwrRole/Type: 1 MsgType: Accept

PwrRole/Type: 0 MsgType: GoodCRC

PwrRole/Type: 1 MsgType: PS_RDY

PwrRole/Type: 0 MsgType: GoodCRC

And then nothing. CC stays high at ~1.6V and I'm stuck at 5V at 1A

Hi,

What's your POCI value? It looks like your ADCIN1 divider value is about 0.67 which will not correspond to the correct default configuration which will allow 20V sink. For this, you will need to adjust it so that the divider will select configuration 3 or 5.

Thank you,

Hari

I'm sorry, I have the values swapped above. 

ADC2 R1 = 10k

ADC2 R2 = 20k

ADC1 R1 = 8.2k

ADC1 R2 = 10k

Also to note - why is the chip showing me a PDO for 20V in the status registers but only negotiating to 5V?

Are there other things I can check?  I really need this to come up as 20V and I'm banging my head against a wall here.

by POCI I think you mean the PD chip to the SPI chip?  That is hard tied low.

I just had a realization.  

Which PP paths are being opened for configuration 3?  

PP_HV1 is set for a SINK path

PP_HV2 is set for SOURCE path on my pcb.

Hi,

Based on your ADC values that you provided, your ADCIN1 divider value is 0.55 and you mentioned your POCI is always low, which corresponds to this configuration:

Therefore, the PD controller is attempting to set the external power path to sink power in dead battery mode and the default configuration is set to Infinite Wait. Please take a look at Table 8-7 for more details on the different configurations. It looks like you will need to change your ADCIN1 resistors to match with either Configuration 3 or 5 which will enable 20V sinking.

Thank you,

Hari

Alright I'm tracking with you, but I'm still seeing some weird behavior.  I adjusted my resistor values and still not seeing 20V.  I checked for everything I can think of.  If I supply power to the rest of the system all my power supplies come up and everything behaves as expected.  It's just this chip not doing what I'm expecting it to do.

Alright.  I have the total phase PD analyzer.  What do you want to see?

Hi,

The other issue I could see might be the external power path that you have configured. Could you show how your external power path is set? Based on your other configuration, I assume you are using your own external power path, correct?

Thank you,

Hari

I attached a screenshot of the schematic.  I am using PP_HV1 as my sink path and PP_HV2 as my 5V source path.  My 5V source path works if I power up the chip from 3.3V instead of VBUS I write the configuration registers and TX PDO etc and that seems to work.  However, if I write 20V into the registers as a PDO, it never goes to try to negotiate for 20V even if I set the registers using the GUI after it's been booted, not using VBUS to power it.

Also, here are the screen shots of the PD messages.  This tool is nice!

The chip is requesting garbage it looks like.  

ok more information.  If I supply 3.3V, export my project settings into device RAM, then plug in my USB power supply - I get what I'm expecting mostly.  It negotiates for 15V instead of 20V.  I used the default Project from the GUI "tps65987ddh -> DRP prefers data host".  I didn't change any other settings so this is stock from the GUI.

Why is configuration 3 not sinking power correctly?

something that bothers me here too is my system load is easily 12W .... why is the PD chip not even presenting the resistor on the data lines to get the 3A?  It current limits to 1A.

Hi,

It looks like you will be using the internal power path of the PD controller, therefore, if you look at the datasheet configuration 5, it says it will be for an external switch so you will have to change your setting to use Configuration 3 since you do not have any external FETs set up based on your schematic. Have you tried Configuration 3 yet?

Also, you will not be able to configure any registers of the PD controller in dead-battery mode and without loading a proper patch configuration via EEPROM or EC.

I would recommend testing your design first on an EVM and taking a look at the schematic shown in the User's Guide.

Thank you,

Hari

Ok, so per my diagram above and the datasheet, I'm confident it is supposed to be in "Configuration 3".  Why does it not negotiate correctly?

Is there a way we can get in direct contact?  I believe this is a fault in your firmware and I want to get this resolved quickly.

Hi,

Is your schematic older then? It looks like you have configuration 5 on there, shown in blue.

Could you please use an EVM to test this? I will take a look at your schematic again to see if there's anything missing.

Thank you,

Hari

Understood.  I've changed my schematic so that the resistor configuration is set to 3.  I have a screen shot above from 7 days ago

this is now my resistor divider configuration

Hi,

Could you please confirm you have LDO_3V3 up when the other device plugs in and provides VBUS to the PD controller?

Otherwise, for some reason the other device may not be offering 20V/3A on the source PDO.

Thank you,

Hari