DLPC3433: HOST_IRQ stays high! Problem with firmware? DMD need to be connected?

Morgan,

Welcome to the E2E forums and thanks for your interest in our DLP technology.

The DMD does indeed need to be connected for the system to boot. Otherwise you will be stuck in initialization stage (HOST_IRQ).

Best Regards,

Philippe

Thank you for your quick response Philippe. 

I have now plugged in the DMD but HOST_IRQ still stays high.

I will work on getting oscilloscope readings. 

Morgan,

Please do. You may want to also verify that the flash firmware was indeed written properly. I recommend to follow the startup timing diagram in the DLPC3433 datasheet to see if the powerup signals are operating as expected (see section 9.2): https://www.ti.com/lit/ds/symlink/dlpc3433.pdf 

I hope this helps.

Regards,

Philippe

This is how I wrote the firmware:

truncate -s 2097152 FWSel_DLPC3433_DLPA2005_pm1_i2c0x36_v7p4p2.img

Sudo flashrom -p linux_spi:dev=/dev/spidev0.0,spispeed=1000

    - It prints “Found Windbond Flash Chip…”

Sudo flashrom -p linux_spi:dev=/dev/spidev0.0,spispeed=1000 -w FWSel_DLPC3433_DLPA2005_pm1_i2c0x36_v7p4p2.img

   - It prints "writing ... Done, verified... Done"

The powerup is operating as I mentioned earlier: 

I turn PROJ_ON high 

    - RESETZ goes high

    - INTZ goes high

    - Power rails (VOFS, VBIAS) go to ~5V

    - VCC18 is 1.8V

HOST_IRQ stays high which means the DLPC hasn't finished initializing.

What signals should I focus on getting a oscope shot of?

Morgan,

Are you sure that you are using pin mapping option 1 for your design? Do you still fail to boot using pin mapping option 2? Details on pin mapping for the DMD can be found in the DLPC3433 datasheet.

Regards,

Philippe Dollo

I am using the pinout option which is used in the DLP LightCrafter 3010-G2 evaluation module (EVM)

Still fails to boot with option 2. 

Morgan,

What are the actual voltage values for VOFS, VBIAS, and VRST? Per the DLP3010 DMD datasheet, these rails should be at 11V, 19V, and -15V respectively. If they are all at 5V then you may have a DMD problem.

If the above is correct, do you see SPI activity between the DLPC3433 and the EEPROM device?

Regards,

Philippe

Correct, the rails stay at ~5V and the DMD_DEN_ARSTZ signal is low so I don't think it gets to the DMD step. 

I don't see any SPI activity between the DLPC and EEPROM.

Morgan,

If there is no SPI activity with the EEPROM then the firmware build you are using may not be relevant at this point. You must have some kind of hardware issue preventing the DLPC3433 from reaching the bootloader stage where it reads from the SPI flash.

Do you have a schematic/layout you can share for review?

Regards,

Philippe

Yes, if you would send me your email, I can send the file.