This thread has been locked.
If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.
Whenever code involving the LEDs (including the Safety MCU Demo software and the two freeRTOS blinky demos) is run on my developer board, two of the LEDs, specifically LED2 aka NHET1(5) and LED1 aka NHET1(27) remain on at full intensity, completely independent of the flashed code. Though I don't believe it's causing any problems with the rest of the code's functionality, I would like to know what is causing this issue and how to prevent it.
Hi,
I have forwarded your query to our experts. They will get back to you.
Regards,
Praveen
Hello Ecosat,
Which board you are using: HD or USB stick? What is the revision? For HDK, revE has 8 LEDs, but revD has 6 white LEDs plus 2 tri-color LEDs. USB stick has 6 white LEDs and 2 tri-color LEDs.
Please also check the versionof the demo SW. Thanks
Regards,
QJ
We are using the HDK, with the revE (with 8 white LEDs) revision.
We are running Code Composer Studio 5.5, using the current version of freeRTOS. I can't seem to find any documentation of which version of the Safety MCU Demos we are running, other than that the software was taken directly from the CD that came with our HDK.
Hi Ecosat,
I run the safety demo on RevE HDK. After clicking "LED Light Show", the picture shows 6 white LEDs and 2 tri-color LEDs. If using the old rev demo GUI and demo code (for revC/D HDK) , the LED1 and LED2 will keep on whenever you check any box around LED1 and LED2.
Please download the latest demo GUI from http://www.ti.com/tool/hercules_safety_mcu_demos
The latest one will check the board ID and board rev number in EEPROM, and program the proper demo code to MCU flash. The new code works ok (but there is still a bug, I just found it). LED1 or LED2 will be ON when you uncheck the box, and all other demo LEDs will be ON when you check the boxes for them.
Regards,
QJ
So I ran the software and installed it on the board, and LED1 and 2 are both on as soon as the demo opens (not just when the LED light show is selected). Is this the expected behaviour then, due to the glitch?
Ecosat,
When you oepn the demo, all LED will be flashing for seconds, then turn off. The LED1 and LED2 should not be on.
I will do another test, and give you a separate demo code to try.
Regards,
QJ
Hi QJ,
We have been doing some additional testing with the Demo software.
LEDs 1 and 2 are on as soon as the demo software finishes installing on the board. However, by turning on the LED light show (which LED 1 and 2 do not follow, they remain on at full strength) and then stopping it, the two LEDs will turn off along with the cessation of the light show. At this point they become fully controllable.
We have also managed to replicate the error by testing the parity error, as soon as this error is run the two LEDs turn on once more and act as above.
Our main problem with this error currently is that the LEDs act in a similar manner when running FreeRTOS on the board, both LEDs remain on and are not affected by code that should change their state. We fear that this might be indicative of greater problems.
Hi Ecosat,
I checked the Demo source code (they are also in DVD), there are several bugs there. For turning off all the LEDs, the demo code writes 0x08110034 to HET DOUT register. This value is not correct since it set pin5(LED2) and pin27(LED1).
After you run hetInit(), you are not able to control the LEDs using HET DOUT registers. In hetInit(), the HET pins for those 8 LEDs are programmed to output PWM signals. Please stop PWM ( pwmStop(hetRAM1,pwm?) before control those pins using DOUT.
Regards,
QJ
Hi QJ,
If that's the case, it looks like our board is working correctly. It appears that the FreeRTOS code only initializes six of the LED pins as serving that purpose, so it was only that bug that was confusing us.
Thanks for your help.