Vin vs 3V3 input MAX31855

[Steve1445]

I inadvertently plugged my 3V3 power from Raspberry Pi W into the Vin pin on the thermocouple chip. However, I was receiving "thermocouple short to ground" and "thermocouple short to Vcc" error messages. I troubleshot through connections, cold solder, etc to discover the misplacement. The errors stopped immediately. The upshot is the thermocouple now does not read temp at all, it's stuck at 32 degrees F.

tldr: When I was in error, the temp tracked in test very well, but now corrected to input on the correct pin, no reading at all.

[adafruit_support_bill]

Connecting 3.3v to the VIN should not have caused any damage. The VIN pin is just the input to the on-board 3.3v LDO regulator.

[Steve1445]

So I can live with the error messages then? since it operates ok on the Vin input.

[adafruit_support_bill]

Those messages typically indicate a problem with the thermocouple probe itself. If you post some photos showing your soldering and connections we can take a look.

[Steve1445]

Thanks, I will concentrate on the probe, I was considering switching out to another one.

combo.jpg (202.41 KiB) Viewed 389 times

0226191314.jpg (579.53 KiB) Viewed 389 times

[adafruit_support_bill]

Those pullup resistors should not be necessary, since there are pullups on the board. You can wire directly for either hardware or software SPI as shown here:
https://learn.adafruit.com/max31855-the ... y/hardware

For the probe connections, are you using type-K thermocouple extension wire?

[Steve1445]

Yes on the wire

[adafruit_support_bill]

Do you get the error messages all the time - or only when actually heating the kiln? It is possible that you are picking up some electrical interference. Heating elements and temperature control/switching systems are a likely source.

[Steve1445]

All the time. I read some other posts here in the forum and I did my best to isolate the power supply and turn off any interference (ballast lights, etc). I put the power supply on a filtered surge suppressor and powered off anything else in the garage. So the only thing going was the controller.

When the kiln is running the errors continue, but like I said earlier, the temp tracks very well and the test I ran performed just like I wanted it to; its just the constant error messages.

I removed the pull up resistors, no change in messages.

Wire

wire.jpg (229.97 KiB) Viewed 372 times

[adafruit_support_bill]

How long is your extension wire? Is it shielded?

[Steve1445]

not shielded and about a meter

[adafruit_support_bill]

A meter is long enough to make a decent noise antenna. Although it does look like twisted-pair which should help with the noise rejection.

If you have access to an oscilloscope, a look at the actual thermocouple signal might be helpful in tracking down the problem.

[Steve1445]

No o-scope access unfortunately; will have to see about another thermocouple to see if it persists. Would placing a cap across the Gnd and Vin (I think) help filter out any noise. I think I saw that in a forum post, something on the range of .01µF?

[adafruit_support_bill]

It might help - and certainly would not hurt anything to add a cap.

[Steve1445]

Well I have been back and forth over the 31855 board, checked again all the connections, re-soldered where I thought I could have done a bit of a better job and removed the 1 meter lead and thermocouple from the input.

In its stead I put in directly 1.4 Vdc (roughly equivalent to a higher temperature reading) to see if removing the hardware on the way to the kiln would clear the error message. It does not. Unless I am assuming something wrong here or missing something, I think the idea of noise creating the errors or from a bad thermocouple and lead can be set aside.

At this point the only thing left is to replace the 31855 board itself.