MCC134 Incorrect temperature value

[Francky]

Hello,

I have a Raspberry5 and 2 MCC hats: MCC134 and MCC128.

The MCC134 is connected with 4 K type thermocouple but it seems that all inputs are reading 9C higher than a multimeter would read from the same sensors.

When I had the MC134 alone, temperatures seem in the expected range so it seems it might have started after I installed the MCC128, but when removing the MCC128, result are still high.

Does the stock slope and offset makes sense here?

Found MCC134 at address  [HatInfo(address=0, id=323, version=1, product_name='MCC 134 Thermocouple Input HAT')]
Channel 0 set to Type  1 Coefficient MCC134CalInfo(slope=0.9995942395331735, offset=-5.424700045439295)
Channel 1 set to Type  1 Coefficient MCC134CalInfo(slope=0.9995924187283793, offset=-24.076377532285278)
Channel 2 set to Type  1 Coefficient MCC134CalInfo(slope=0.9995927307750806, offset=-27.509583343136)
Channel 3 set to Type  1 Coefficient MCC134CalInfo(slope=0.9995938139937812, offset=-22.98250636926832)
Logging Interval (sec) 60
Logging temperatures, Ctrl-C to exit.
Temperatures 11.14 10.67 28.06 23.03
CJC  26.7397420913332 26.367750118973106 25.348451403433273 24.073498502352948
ADC noscale nocal -67282.0 -67654.0 11855.0 -4457.0
Calibration Date  2023-12-27

Values from the Python script are the same than with the MCC utility.

Not sure what I could do at this point?

Thank you

 

 

Edited March 21 by Francky

[Fausto]

Hello @Francky.

The MCC DAQ HATS are calibrated at the factory.  The coefficients are as they leave the factory.

Did you read the MCC 134 calibration coefficients from the factory?

Did you modify the calibration coefficients?

Are the thermocouple tips in contact with a metallic surface? If they are, then use an epoxy or thin film on the ends.

23 hours ago, Francky said:

Temperatures 11.14 10.67 28.06 23.03

Are all four K-type thermocouples together or in different locations?

Please describe the testing application, materials, and environment.  (Images are always helpful.)

Are you using an active cooler on the RPi5?

What are the temperature values from the MCC DAQ HAT Manager utility application?

Do you see a difference if you test your code with a one second time interval?

Are there any pumps or motors near the RPi and/or thermocouples?

Regards,

Fausto

[Francky]

32 minutes ago, Fausto said:

Hello @Francky.

The MCC DAQ HATS are calibrated at the factory.  The coefficients are as they leave the factory.

Did you read the MCC 134 calibration coefficients from the factory?

Did you modify the calibration coefficients?

Are the thermocouple tips in contact with a metallic surface? If they are, then use an epoxy or thin film on the ends.

Are all four K-type thermocouples together or in different locations?

Please describe the testing application, materials, and environment.  (Images are always helpful.)

Are you using an active cooler on the RPi5?

What are the temperature values from the MCC DAQ HAT Manager utility application?

Do you see a difference if you test your code with a one second time interval?

Are there any pumps or motors near the RPi and/or thermocouples?

Regards,

Fausto

Hi @Fausto

I'll try to answer your question, in order below:

 

The calibration from the factory are the ones shown in my examples, I did not change them.

Temperatures 11.14 10.67 28.06 23.03:

• The 1st 2x Thermocouples are in the fridge, hanging, not in contact with any material (11.04C and 10.67C should read 3C and 2.5C)
• The 3rd Thermocouple is on the evaporator of a fridge, contact on metal (28.06 should read 20C)

I tried removing it from the metal, same result.

• The 4th Thermocouple is an ambient temperature, hanging not in contact with any material  (23.3C should read 16C)

==> So roughly the MC-134 reads 8C higher across all sensors

All four are K-Type and I can read them OK with a multimeter connected to them, in place of the MCC-134

I cannot use the active cooler on the RPi5 as it would not fit under the hat

Temperature from the DAQHat UI application are the same as read by the Python script

My polling is every minute, and I just tried every 1s, results are the similar

No pump or motor near the RPi

 

Note that I did try the MCC-134 alone, without the MCC-128 on top and had the same issue.

MCC-128 works fine.

 

 

 

 

 

 

 

Edited March 22 by Francky

[Fausto]

Hello @Francky.

Thank you for the reply and image.  When you test the thermocouples with the multimeter, do you disconnect the thermocouple at the MCC 134 screw terminal or the mini thermocouple connectors?  If you disconnect the thermocouple at the connector, then the wire section from the opposite mini connector to the MCC 134 screw terminal is an unknown.  Please relocate the MCC 134 to be on top (2nd board), connect K-type thermocouples directly to the MCC 134 screw terminal, and then retest.

 

 

Regards,

Fausto

[Francky]

1 hour ago, Fausto said:

Please relocate the MCC 134 to be on top (2nd board)

Hi @Fausto

I thought that was the issue, temperature started reporting fine but then crept up slowly without reasons.

I will try replacing my thermocouple junctions and will report back.

Found MCC134 at address  [HatInfo(address=1, id=323, version=1, product_name='MCC 134 Thermocouple Input HAT')]
Found MCC128 at address  [HatInfo(address=0, id=326, version=1, product_name='MCC 128 Voltage HAT')]
Channel 0 set to Type  1 Coefficient MCC134CalInfo(slope=0.9995942395331735, offset=-5.424700045439295)
Channel 1 set to Type  1 Coefficient MCC134CalInfo(slope=0.9995924187283793, offset=-24.076377532285278)
Channel 2 set to Type  1 Coefficient MCC134CalInfo(slope=0.9995927307750806, offset=-27.509583343136)
Channel 3 set to Type  1 Coefficient MCC134CalInfo(slope=0.9995938139937812, offset=-22.98250636926832)
Logging Interval (sec) 60
Logging temperatures, Ctrl-C to exit.
Temperatures: 3.78 4.35 25.03 21.08 Voltage: 0.198 4.296
Temperatures: 3.78 4.35 25.03 21.08 Voltage: 0.197 4.269
Temperatures: 4.24 4.55 25.07 21.17 Voltage: 0.196 4.307
Temperatures: 4.70 4.84 25.35 21.56 Voltage: 0.196 4.257
Temperatures: 5.24 5.27 25.68 21.88 Voltage: 0.196 4.222
Temperatures: 6.10 6.05 26.37 22.57 Voltage: 0.196 4.252
Temperatures: 6.68 6.51 26.68 22.91 Voltage: 0.196 4.332
Temperatures: 7.31 7.03 27.10 23.42 Voltage: 0.196 4.351
Temperatures: 7.52 7.29 27.35 23.52 Voltage: 0.196 4.234
Temperatures: 7.78 7.58 27.63 23.86 Voltage: 0.196 4.291
Temperatures: 7.86 7.65 27.65 23.81 Voltage: 0.200 4.388
Temperatures: 8.14 7.91 27.87 24.06 Voltage: 0.199 4.370
Temperatures: 8.27 8.03 28.27 24.53 Voltage: 0.199 4.343
Temperatures: 7.39 7.94 28.54 23.97 Voltage: 0.198 4.299
Temperatures: 7.73 8.10 28.60 24.23 Voltage: 0.198 4.310
Temperatures: 8.06 8.25 28.59 24.17 Voltage: 0.198 4.275
 

Edited March 22 by Francky

[Fausto]

Hello @Francky.

The MCC 134 Electrical Specifications mentions a 15 minute warm-up time.

Please verify the three CJC sensors are not damaged.

 

You initially reported the following CJC temperatures.  Have readings varied significantly?

CJC  26.7397420913332 26.367750118973106 25.348451403433273 24.073498502352948

Please retest the MCC 134 boards without the mini thermocouple connectors; direct connection to the screw terminals.

https://digilent.com/reference/daq-and-datalogging/documents/measuring-thermocouples-with-raspberry-pi

 

Regards,

Fausto

[Francky]

Hi @Fausto

Connecting the sensor directly to the MCC-134 was the solution.

The mini thermocouple connectors or the leads may be the culprit. I'll investigate more but thank you for your help.