IWR1843: RX ADC DC Offset Calibration Failure on Boot-time Calibration

Hello,

I'm very sorry for rushing you, but could you please provide a response?
Even if it's just whatever information you can prepare at this point,
I would appreciate receiving your answer.

Best regards,
Yasuaki

Hi Yasuaki,

1. Are you using these samples on a customer board? Have you checked the supplies and clock on the board properly?

2. Can you disable RX ADC DC offset calibration in boot cal config API and check if rest of calibration work properly? 

3.If it is one custom board, can you swap a unit on that board and check if a working sample also flags failure on the board?

4.There is no time limit from the reset assertion for running boot cal. Typically boot cal is completed within 20-30ms

Regards

Also, If you have a TI EVM, is it possible to place this sample on TI EVM and check?

Regards

Hello,

Thank you for your response.
Since the units experiencing this issue are in their fully assembled product state, we will proceed with the investigation based on your questions from this point forward.
The same phenomenon has occurred in 2 out of 100 units.

The main factors we have considered from multiple perspectives at this point are as follows:
1. IC internal manufacturing variations (lot dependency, concentration in specific serial numbers, etc.)
2. Effects of external circuit and power supply startup waveforms
3. Variations in boot-time calibration execution timing
4. Data integrity and write timing issues
5. Software/firmware sequence control errors

We would appreciate your guidance and opinions on the following points:

[Questions]

1. Regarding the boot-time calibration function in this IC, are there any similar past cases or potential risks you are aware of?
2. Concerning the possibility of calibration process failure, are there any trends observed from past data regarding manufacturing lots or internal test conditions?
3. For identifying the root cause of the failures, what type of isolation testing do you recommend we proceed with - at the IC level or mounted board level?
4. Please advise on any specific points of attention or recommended measurement items regarding IC control signals, initialization sequences, power startup, etc.
5. If you have any other potential causes your company considers, additional investigation perspectives to explore, or opinions regarding sample return procedures, please share your insights.

Best regards,
Yasuaki

Hi

Sorry for the delayed response.

We have not seen failures occurring due to RX ADC DC offset calibration. This failure generally indicates problem with the power supply. There is no know issue or history associated with this issue

Were you able to check disabling RX ADC DC offset calibration in boot cal config API and check if rest of calibration work properly?

Regards

Also, can you check if the power supplies and clock are as per the spec required by IWR1843?

Regards

Can you please let me know the version of the software used? SDK version and Demo version?

Regards

Hello, 

Thank you for your response.

Regarding power supply, we have confirmed four types of voltages and verified that there are no problems.
Also, we have confirmed the voltage waveforms of the ADC DC Offset Calibration PASS units and NG units and verified that there are no significant differences.

The SDK version and Demo version is "mmWave-SDK 03.05.00.04".

"Can you disable RX ADC DC offset calibration in boot cal config API and check if rest of calibration work properly?"
This investigation has not been implemented yet.
Please tell us what can be understood from this investigation.
Currently, other boot-time  calibrations performed simultaneously with RX ADC DC Offset CAL are PASS, and we do not understand the intention of confirming other CAL results in a state where RX ADC DC Offset CAL is disabled.

As shown in the figure, when we replaced the MMIC (IWR1843 A) from the circuit board (PCB A) that had CAL FAIL onto the circuit board (PCB B) that had CAL PASS, we found that CAL became FAIL.
Also, when we replaced the new MMIC (IWR1843 C) onto the circuit board (PCB A) that had CAL FAIL, we found that CAL became PASS.
From these results, we suspect that there are no problems with the circuit board or electronic component mounting, and that it is an IC defect.

Best regards,
Yasuaki

Hello, 

"Can you disable RX ADC DC offset calibration in boot cal config API and check if rest of calibration work properly?"
I conducted this investigation.
I confirmed that the calibration status of ADC DC Offset Calibration was FAIL, while all other statuses were PASS.

Regards

Hi

Can you use Low Power ADC mode? When the IF bandwidth is less than or equal to 7.5MHz bandwidth, low power ADC mode is recommended

Regards

You can config in Low power mode using the below function

rlSetLowPowerModeConfig

rlReturnVal_t rlSetLowPowerModeConfig(rlUInt8_t deviceMap, rlLowPowerModeCfg_t* data) in in rl_sensor.c

Regards

Hello, 

We tested boot-time calibration in Low Power ADC mode.
We confirmed that all calibrations, including RX ADC DC offset calibration, PASSED.
However, our application uses sampling rates of 10MHz or higher.
Therefore, we cannot operate in Low Power ADC mode for our product.
Are there any constraints or inconveniences with Regular Power ADC mode?

Regards