PGA460: Timeouts PGA460 GUI

Hi Pablo,

Just to be sure that I understand correctly:

Does the setup look like the following?

    PC <---> MSP430 <--> XBEE ............................ XBEE<-> PGA460

Thanks!

A.

Hi Pablo,

The PGA460 is limited to 115.2kBaud max when using UART, but the device does support SPI (USART) up to an 8MHz clock for much faster communication. The PGA460 UART and SPI commands are identical in terms of the data that is sent, including the use of the same RXD (MOSI) and TXD (MISO) pins. You'll only need to add a 8MHz clock to the SCLK pin, and drop the Start/Stop bits from the frame to use the SPI mode. See 7.3.6.2.1.8 USART Synchronous Mode of the PGA460 datasheet for details on this communication mode.

Just so I understand, are you still trying to use the PGA460-Q1 EVM GUI, but over a XBEE wireless connection rather than USB, or are you referring to a XBEE GUI? If the PGA460 GUI, what behavior are asking be updated in the GUI to support this? Alternatively, you can create a Serial COM port based GUI to extract the PGA460 time of flight data by using the PGA460 Energia Library and Code Example.

Hi Akeem,

We have performed some tests and found some interesting. We thought that our MCU were causing the issue so we removed it from the equation and prepared the next setup:

MSP --- XBee <---> XBee --- PGA460

With this, we get the same result. We can configure the PGA460 profile but we cannot get a good curve in the GUI. It is cut. Please, check the image:

Right after the bottom, it is set to 0 in the datapath even when the MSP is receiving all the curve (Checked with an FTDI):

40 bf ff ff ff ff ff ff ff ff ff ff ff ee d6 81
39 18 0a 03 02 01 01 01 01 05 1f 36 36 2d 18 0a
04 03 02 02 01 03 05 06 06 05 04 02 03 04 04 03
03 02 02 02 02 02 03 03 03 03 04 07 07 07 04 03
02 02 02 12 a4 ff ff ff bc 67 3c 40 3f 2e 1f 21
21 1f 1b 0d 08 06 0d 14 14 0d 0c 0c 0d 19 29 2e
2e 29 1c 15 4b 7a 7e 7e 7d 62 34 24 1a 29 42 47
42 30 2d 1d 0e 13 25 2b 2a 20 19 18 10 0b 19 27
29 2a
40 80 3f 40 00 bf 40 06 b9

Represented:

I believe that GUI's timeout when receiving message is to tight for this setup. Would it be possible to have a version with a more soft timeouts?

Best regards

Pablo,

It appears that the echo data dump (EDD) is being disrupted a little more than half way through the record cycle. This is usually due to UART read of the EDD being disrupted by a subsequent command to the PGA460. The delay built into the GUI is based on the given record length and UART baud rate plus margin, so in your case of 8.192ms, the GUI should always allow for enough delay time between the EDD read command and filling of the UART buffer with the 128 bytes of data. Are you sending any auxiliary commands from the master controller during the EDD capture? Also, what UART baud rate are you using? If not the maximum, are you able to increase the baud rate to 115.2kBaud? I have confirmed the EDD is still fully captured when I set the record length to the 8.192ms and the lowest GUI supported baud of 9.6kBaud.

The best way to determine this is by using a logic analyzer to see if the UART bus is reading all 128 bytes of EDD data. If so, then there could be a queuing issue in the GUI due to the overhead from the Xbee.

One thing you can try as a workaround is increasing the record length cycle to check if the disruption occurs at ~50% of the given record length, or if it is always halted just before ~4.9ms. If it is always after 50%, and you only need up to 8.192ms, double your record length to 16.384. Yes, the resolution will decrease, but you will be able to continue with the evaluation.