DLP5531Q1EVM: Inconsistent Boot Up

Hello Daniel,

Thank you for reaching out and using the E2E forums! An expert from our team will review the thread and get back to you shortly.

Regards,

Tristan Bottone

Hi Dan,

Thanks for the detailed description. Could you try the following test?

With SW3 disabled to enable normal operation into main app, power on your EVM with the SPI Cheetah adapter connecting the EVM to your host computer.

When PROJ_ON is toggled to the ON position, what do you see next to Mode (highlighted below)?

Does it show "Communication Error," "Standby," or "Display?"

1. Communication Error:
   1. Maybe DMD is seated incorrectly?
   2. Normally indicates SPI communication error or cable not working, but since you can see bootloader mode, the cable is confirmed good.
2. Display:
   1. System is likely working as intended, but check your Error History and refresh the list
      
3. Standby:
   1. There is a serious enough issue that the system must leave main app or never reaches it
   2. Please check Error history

I will be out of office this Thursday/Friday for a national holiday, by the way. Please allow us extra time to respond if this is not resolved tomorrow.

Adding more to this as an edit:

Daniel Guzi said:

I can verify the flash memory, which passes. Afterwards if I restart the system, some of the time the system successfully boots into the main application. If it doesn't, I can repeat those steps and it eventually works.

Are you doing this with SW3 enabled for BOOT_HOLD? This sounds more like a hardware issue to me if it fails here as well.
I would check the TPS99000 Datasheet for Power On Sequence. Let's first look at Error history, though.

Thank you and regards,
Michael Ly

Hi Michael,

When I bootup with SW3 disabled, the mode shows as "communication error." I have also taken a look at the DMD and it appears to be seated well. Nothing appears in the error history.

Yes, I am verifying the flash memory with SW3 enabled for BOOT_HOLD. When SW3 is enabled, I am always able to get to the bootloader application. After I've verified the flash, I disable SW3 to boot into the main app.

Thank you,

Dan

Hi Daniel,

Thanks for the prompt response.
Are you saying that you have removed the DMD and reseated it? Is it pressed down well into the socket/not loose?
I am asking because there are times where the thermal paste on the back of the DMD can get onto the pins and then into the DMD socket located on the controller board. This can cause improper connection.

Have you attempted to reprogram the flash file while in boot hold mode yet? I would try that out with SW3 BOOT_HOLD enabled and then power cycle (with BOOT_HOLD disabled between powering off and then back on).

While we are on the topic of firmware, though, are you using headlight firmware? Do you know if your project is the headlight project with GPIO9 pulled high? There's a note in the EVM schematic for this. I will send you a friend request just in case. I may or may not be able to get back to you before the end of the week, but would like you to know I will be on holiday starting this evening for a national holiday. I will return Monday morning Dallas time (GMT-6).

Regards,
Michael Ly

Hi Michael,

I have not reseated the DMD, but it appears to be well seated in its socket.

I have reprogrammed the flash and that seemed to fix this issue. Another issue did emerge after the reprogramming. I believe your colleague has included you in that thread offline.

I am using headlight firmware. It looks as though GPIO9 is pulled high.

Thank you,

Dan

Dan,

Let's continue this conversation offline. I will put this thread as "waiting for customer" until we come to a conclusion--this is just to prevent TI's system from thinking this thread has been unattended. When we reach a conclusion on this, I'll post the answer here (if it is not sensitive) so other users can view in the future.

I've sent a new reply since we appear to getting an Error 674:

Thank you and regards,
Michael Ly