MCC 118 DAQ HAT Incorrect Response

[James7]

Hi, I've come across a problem where an MCC 118 hat will always fail to start a scan, returning with an 'Incorrect response' error. This occurs when both using my own code and running the example single_value_read.py. daqhats_read_eeproms detects the board's address correctly, as does daqhats_list_boards, however it does list both the Firmware and Bootloader version as 0.00. 

Up until recently, it's been working perfectly fine, however now I'm completely unable to read anything from it. Any troubleshooting guidance would be appreciated, however the raspberry pi in question is remote, and I don't currently have physical access to it. 

Edited October 21, 2022 by James7
Adding further images

[DAQman]

Try: 

sudo daqhats_read_eeproms

followed by:

daqhats_list_boards

Upgrade the firmware with the following commands:

mcc172_firmware_update 0 ~/daqhats/tools/MCC_172.fw

mcc172_firmware_update 1 ~/daqhats/tools/MCC_172.fw

mcc118_firmware_update 2 ~/daqhats/tools/MCC_118.hex

after, test each board with the MCC DAQ HAT Manager utility. (located in your Raspberry PI->Accessories.

best regards,

John

[James7]

Hi, thanks for the response. sudo daqhats_read_eeproms and daqhats_list_boards both returned as shown in the screenshots attached. daqhats_read_eeproms seems to return as normal, and daqhats_list_boards recognises the MCC 118, however the versions are listed as 0.00. Finally, the firmware updates on the two 172s returned as expected, however for the 118, it failed with an 'Error entering the bootloader' as shown in the screenshot attached.

As for the MCC DAQ HAT Manager utility, the PI is running the lite version of Raspbian, so only command line available sadly. Is there any way to access this from the command line?

[DAQman]

Hello,

I don't believe this is the problem but better safe than sorry. Jump into Raspberry PI->Preferences->RaspBerry PI Configuration. Click the Interfaces tab and make SPI and I2C are enabled. Otherwise, if the board is less than a year old and it is in the United States, I can issue an RMA number for a replacement. [requires the return of the old board]

Best regards,

John

[James7]

Hi again, I can confirm that both SPI and I2C are enabled. Is there really nothing else that can be done at this stage, or is it well and truly dead to the world? If so, is there any way of know what might have caused it?

Thanks,

James

[DAQman]

The microprocessor is not responding, that why you get a 0.00 for the firmware version. If you had the system in your hands, you could reduce it to just the MCC 118 at address 0 to see if one of the other boards is stepping on it. 

Best regards,

John

[James7]

Ah I see what you mean. Pretty sure I can guess the answer, but I'm guessing there's no way of programmatically reinitialising the hat, or I suppose forcing a software update? But yeah, I suppose there's not much I can do until I can physically check the Pi itself.

On a separate note, do you know what might have caused something like this? I've got another pi in a similar setup, so I'd like to take some precautions on that one if I can. 

Thanks again,

James

Edited October 24, 2022 by James7
Added second paragraph question

[DAQman]

Hi James,

It's only speculation, but static electricity (EDS) is usually the culprit. If the board had been working for sometime and it suddenly stopped, then more likely the chip went bad.

Best regards,

John

[James7]

Ah that makes sense, I had some similar suspicions. I'd been seeing some strange voltage spikes on the MCC 118's sensors, so makes sense that that would be the culprit. 

[James7]

Hi John,

I've had another thought on this subject. Is it possible that the code used to read from the MCC 118's sensor could have killed it? I suspect this is not the case given what you've said, but the code was recently modified to increase the scan rate of the sensor, as well as how often it is read by (from every 15 mins to every 1 min). I still suspect the voltage spikes over this, but could it have affected things?

Thanks,

James

[DAQman]

Hey James,

Program code will not damage the board. However, the 118 micro could have corrupted firmware if they ran the firmware updater with the wrong file. If this happened, power off the RPi and short pins 3 & 4 on the 5-pin W3 connector and fire it back up. [There's no W3 connector per say, just the trace holes in the PCB.] This may force it into the bootloader. Update the firmware, remove power, remove the short and fire it back up. Read the EEPROMs and list the devices. If it's your lucky day, it will show up functioning normally.

Good luck,
John

[James7]

There's not been any recent firmware updates sadly, aside from the one I did for these tests. I'll try and give it a go when I get the chance though, no harm in it.

[aremund]

Hello,

I have exactly the same problem as @James7, have any of you made any progress in solving the problem?

I would be very grateful for any help or expertise.

Best,

aremund

[aremund]

Hello,

I have tried both approaches, I reduced to a board and tried to fire with the shortage, but it did not fire up. Is now the only option to buy a new mcc118?

Best,

aremund

[DAQman]

Hello,

If SPI and I2C are enabled, and you have run sudo daqhats_read_eeproms, and daqhats_list_boards returns firmware version 0.00, you have a broken board. 

Best regards,

John

[aremund]

Hello,

Now I wanted to check the firmware version after trying the two troubleshooting options, and now I get a firmware of 1.03 and a bootloader version of 1.01, so it worked, but I can't tell how and why. At least now it works.

Best,

Anna