PGA460PSM-EVM: Device unresponsive over UART on TXD and RXD pins

Hello Mark,

Thank you for your question. The PGA460PSM-EVM does have several communication interfaces so it could be possible that you may have inadvertently changed the communication preferences. I am not aware of a factory reset over UART, but I will look into it to see if that is a possibility.

How do you have the SCLK pin wired on the EVM? 

Do you have any scope captures of the communication with the TXD and RXD pins when you were writing TEST_MUX to the unresponsive device?

Best,

Isaac

Hi Isaac,

Thank you for your reply.

I am using the PGA460PSM-EVMs in asynchronous UART mode, and I do not have the SCLK pin connected, only TXD and RXD.

I have attached some scope pictures below comparing the unresponsive PGA460PSM-EVM to a responsive device.

As the writing of the TEST_MUX register is a no-response command (PGA460-Q1 datasheet page 38) I've also included an attempted reading of the DEV-STAT0 register, which is successful with one device but not the other.

You can see on the unresponsive PGA460PSM-EVM what appears to be a pull of the TXD line high at the expected time and for the expected duration (judging by the reply on the responsive PGA460PSM-EVM), however it then does not pull the line low as would be expected or show any further signs of activity on the TXD line.

This initial pull-up does not occur if the wrong UART address is used.

I am using a 10K pull-up resistor to 3V3 on the TXD line as I'm using multiple PGA460PSM-EVMs on the same UART bus as per this thread here: https://e2e.ti.com/support/sensors/f/sensors-forum/829633/pga460-q1-how-to-drive-multiple-devices-by-uart. This has worked successfully with the other three sensors. I do not use a pull-up resistor when interfacing with the MSP-EXP430F5529LP, as I only every use the MSP-EXP430F5529LP with one sensor at a time.

In the scope plots the sensor under test is the only device connected to the UART bus.

Many thanks,

Mark

########

Green = RXD (from PGA460's perspective)

Yellow = TXD (from PGA460's perspective)

UART: baud = 2400, parity = None, stop bits = 2

Both devices have UART address = 0

########

Write TEST_MUX = 00; responsive PGA460PSM-EVM:

Write TEST_MUX = 00; unresponsive PGA460PSM-EVM:

Read DEV_STAT0; responsive PGA460PSM-EVM:

Read DEV_STAT0; unresponsive PGA460PSM-EVM:

Highlight on the initial TXD pull-up in the response; responsive PGA460PSM-EVM:

As above, + measurements:

Highlight on what appears to be an attempt at the same initial TXD pull-up of a response; unresponsive PGA460PSM-EVM:

Wider angle on this initial TXD pull-up after a DEV_STAT0 read, correct UART address = 0 used; unresponsive PGA460PSM-EVM:

Comparison showing this initial TXD pull-up after a DEV_STAT0 is not present when an incorrect UART address = 1 used; unresponsive PGA460PSM-EVM:

Hey Mark,

Thanks for the scope captures!

It looks like there is a way to return the device to factory settings using UART. It  not sure if it would work since you are having some communication issues, but I think it is worth a shot. This info can be found in the PGA460 data sheet under section 7.62

This thread also contains some more info: https://e2e.ti.com/support/sensors-group/sensors/f/sensors-forum/814314/pga460-programming-problem-of-eeprom

Let me know if this works for you Mark.

Best,

Isaac

Hi Isaac,

Thank you for your suggestions.

I've practised programming the EEPROM. Using the responsive sensors I've been able to verify that I've been successful by writing a different value of a USER_DATA register, power cycling the sensor, then reading the registers and verifying the change has persisted.

Unfortunately, writing and non-volatile programming the default values of the EEPROM registers as defined in the datasheet to the unresponsive sensor has not changed its behaviour, and it remains non-responsive.

Many thanks,

Mark

Hey Mark,

Thanks for giving that a shot, I figured there was a possibility that it might not work due to the communication errors. Has the communication on this device always been a problem or did it suddenly stop working? 

I believe the best course of action here is if the PGA460 can be replaced on your EVM. This would help verify if this device was damaged. Not sure if you have an extra that you can try out on the board.

Best,

Isaac

Thank you Isaac,

We can order some. As a sanity check, I believe this is correct type and package?

https://www.mouser.co.uk/ProductDetail/Texas-Instruments/PGA460TPWRQ1?qs=Mv7BduZupUjgt2GrI95CBA%3D%3D

Many thanks,

Mark