TMS320F28069M: IO issues when connected to Windows10

Alexis Fetet

Part Number: TMS320F28069M
Other Parts Discussed in Thread: CCSTUDIO, CONTROLSUITE

Hello,

After installing the CCStudio 12.3.0 and ControleSUITE v3.4.9, I am using the Simulink support package to program a LaunchPadXL TMS320F28069M.

I have 2 boards, 1 rev 1.1 and 1 rev 1.2. When I plug in the rev 1.1 card, I have 2 COM ports opened on my computer, and only 1, as expected with the rev 1.2. Is there a known issue regarding that revision and this issue, or an issue with the actual board itself? The drivers seem to be working just fine on the other board, both had JP6 OFF and JP7 ON, boot 1,2,3 ON.

Regardless of the revision and the number of ports, I am facing severe IO interference (mouse unusable, keyboard unusable, device manager not responding) when using the COM port of the cards with the computer in my program, occasionally. I am using SCI module A on GPIO Rx 28 and Tx 29, with data header 'S' and terminator 'E'. The code is generated by Simulink from models and consists of 3 int16 values. Do you have any guess on what is the issue here?

Immediately opening (one of) the COM port of the board seems to prevent the interference, I managed to upload a simple blinking sketch after closing Putty (thus the COM port), so with no SCI blocks in Simulink and I am getting no interference.

This is a follow up on issue: https://e2e.ti.com/support/microcontrollers/c2000-microcontrollers-group/c2000/f/c2000-microcontrollers-forum/1200855/launchxl-f28069m-mouse-control-interference/

To answer last question, USB-C adaptor is a HAMA 00200101, and I can confirm one of the boards is activating 2 COM ports while the other is activating 1 COM port.

Best regards,

Alexis

over 3 years ago

0 Vince over 3 years ago

TI__Mastermind 19795 points

Hi Alexis,

Thanks for your question. I will reach out to our board design expert to find the issue on the difference between rev 1.1 and 1.2.

One thing to keep in mind: the GPIO 28/29 are configured using jumpers. I will let the expert assist with this board:

Regards,

Vince

0 Kevin Allen18 over 3 years ago in reply to Vince

TI__Mastermind 45606 points

Hi Alexis,

I'm not certain of the Rev 1.1 vs 1.2 differences (if any) and will need to get back to you. In the mean time, can you provide screenshots of the COM ports in the device manager for the two different boards? If you were to re-program both boards XDS100 following the directions below, I would think they'd be the same.

https://e2e.ti.com/support/microcontrollers/c2000-microcontrollers-group/c2000/f/c2000-microcontrollers-forum/670519/faq-sticky-enable-xds100-usb-serial-port-on-launchxl-f28069m

Best,

Kevin

0 Alexis Fetet over 3 years ago in reply to Kevin Allen18

Prodigy 20 points

Hi,

The boards have been flashed with the same procedure and they still behave differently. Here is a screenshot of the available COM port when the only devices plugged on the computer are the 2 Launchpads, COM 8 and 9 are the same board, COM 11 is the rev 1.2 board.

2 boards connected, 3 COM ports open

0 Alexis Fetet over 3 years ago in reply to Alexis Fetet

Prodigy 20 points

Here is the driver version I got from downloading CCStudio 12.3.0 and ControlSUITE 3.4.9, unsure which one provides the drivers.

0 Kevin Allen18 over 3 years ago in reply to Alexis Fetet

TI__Mastermind 45606 points

Hi Alexis,

Can you check if the labeling on the XDS100 (U6) device is different on the two boards? I just remembered we may have had to use slightly different FTDI devices for certain builds due to supply constraints. I'm not certain if this would affect what you're seeing however.

Another thought is if both Side A and B are configured for Virtual COM Port when the EEPROM was programmed. These devices have dual UART capabilities.

Best,

Kevin

0 Alexis Fetet over 3 years ago in reply to Kevin Allen18

Prodigy 20 points

Hi Kevin,

I rev 1.1 has FTDI 1621-C D9MT8.1 FT2232HQ and rev 1.2 has FTDI 2140-C DKMW5C1 FT2232HQ

For the dual UART, I can not see any related parameter in my Simulink so I assume this would have been set beforehand using this procedure?

Regrds,

Alexis

0 Alexis Fetet over 3 years ago in reply to Alexis Fetet

Prodigy 20 points

Using the linked software, I can see that both sides are using virtual COM, setting only side B as virtual COM and side A as D2XX Direct only opens 1 COM port on Windows.

0 Alexis Fetet over 3 years ago in reply to Alexis Fetet

Prodigy 20 points

Regarding the I/O interference, we were running the COM connection on side A which was using the D2XX Direct drivers. Using side B with pins 15/58 and jumpers 6/7 ON/OFF seems to fix our issue so far.

0 Kevin Allen18 over 3 years ago in reply to Alexis Fetet

TI__Mastermind 45606 points

Hi Alexis,

Alexis Fetet said:
Using the linked software, I can see that both sides are using virtual COM, setting only side B as virtual COM and side A as D2XX Direct only opens 1 COM port on Windows.

Good, glad to hear this fixed that issue. It seems the older LaunchPads (rev 1.1) FTDI devices weren't programmed properly for just enabling 1x COM Port.

Alexis Fetet said:
Regarding the I/O interference, we were running the COM connection on side A which was using the D2XX Direct drivers. Using side B with pins 15/58 and jumpers 6/7 ON/OFF seems to fix our issue so far.

Could you not get GPIOs 28/29 to work when JP6/7 are OFF/ON? It should be able to be made to work. The only difference is the F2806x SCI module being used for the pins (A vs. B), which would require SW changes to be made.

Best,

Kevin

0 Alexis Fetet over 3 years ago in reply to Kevin Allen18

Prodigy 20 points

Hi Kevin,

Using GPIOs 28/29 resulted in issues with the computer's I/O, regardless of the driver used on side A. Using only side B did not produce the issue with the same program.

Best regards,

Alexis

C2000™︎ microcontrollers

C2000 microcontrollers forum

TMS320F28069M: IO issues when connected to Windows10