Unable to program DSP and Boot EEPROM on ADAU1452
Thread Summary
The user encountered issues programming the ADAU1452WBCPZ Sigma DSP and external W25Q80DVSNIG EEPROM using Sigma Studio and a PE Micro Multilink Universal programming probe. The problem was resolved by configuring the MCU's SPI Controller bus to high impedance, allowing the ADUSB2EBZ-BOARD programming adapter to drive the SPI interface successfully.AI Generated Content
We have created a new custom PCB design that uses the ADAU1452WBCPZ Sigma DSP. We are unable to program the DSP internally or program the external Boot EEPROM. We are using Sigma Studio, Ver 4.7 with a PE MIcro Multilink Universal programming / debug probe. We can see the Target SPI port is receiving data from the programming probe. There is no signal output on the Controller SPI port that is connected to the external EEPROM. The EEPROM is a WINBOND W25Q80DVSNIG.
I have reviewed the hardware and suspect that DSP is not clocking internally. Could this be related to the PLL not being locked? Also, we are using and external 12.288 MHz signal to drive the MCLK pin. Do we need to configure the PLL per Table 28 in the datasheet? Do you have recommendations to address this issue?
Thank you, John
Audio Signal Processors
adau1452
programming
software
Sigma Studio Ver 4.7
sigmaStudio 4.7
SigmaDSP Audio Processors
ADAU1452WBCPZ
Show More
Hello John,
What happens to program the EEPROM when you use the command "Program the EEPROM Through the DSP" is that SigmaStudio loads in a small program that acts as a data pass through program. It takes in pages of data and then writes it to the EEPROM using the master SPI port. So it does not sound like you are loading in this program.
Since you mentioned that you cannot program the DSP itself then programming the EEPROM will not work. In addition, what does not make sense is what are you trying to program to the EEPROM if you cannot write a program to the DSP? Anyhow, the issue is you are not talking to the DSP properly.
You mentioned you see the data going to the DSP so that is a good start. There are several commands that SigmaStudio will use that are to the USBi interface and not to the DSP. Things like delays to wait for reset or to wait for the PLL to lock, to wait for the hibernate command. So you really need to use a USBi interface and not a different USB to SPI interface. That seems to me to be the issue.
Do you have an evaluation board?
Dave T
Hello Dave,
Thank you for your detailed response. At the end of the day yesterday, we figured out what this issue was being caused by. The DSP Target SPI interface is connected to a 10-pin programming header. This was duplicated on our custom PCB design as it is implemented on the EVAL-ADAU1452 Evaluation Board. This SPI serial interface is also connected to an MCU with a Controller SPI interface.
The issue was found to be the MCU Controller SPI interface had been configured to static low and high output levels, which prevented the ADUSB2EBZ-BOARD programming adapter from being able to drive the SPI serial interface. (Note: I mentioned in my original post we were using a PE Micro Multilink Universal programming pod. This was a mistake.)
We changed the driven signals on the MCU SPI Controller bus to be high impedance. The ADUSB2EBZ-BOARD programming adapter was then able to program the DSP and EEPROM successfully with this change.
I think we can call this support case Resolved. Thanks for your prompt response and support.
John
Hello John,
Thanks so much for taking the effort to report back. You are correct that having two SPI masters on a bus does not work well unless one of them can be made to tri-sate and get out of the way. This is why we have the 1K resistors in series with the SPI lines on the USBi interface. (ADUSB2EZB-BOARD). The resistors keep it from being damaged if thrown into a multi-master situation. Good find on your part. I hope the rest of the project goes well.
Dave T
Please log in to reply.