Unreliable temps with multiple TMP1101_1 phidgets

[pg_ammo]

Hi there,

I'm using multiple (3) TMP1101_1 phidgets as thermocouple readers with a single VINT hub (HUB0002_0). I've noticed that if I only have one of the TMP1101_1s plugged into the hub, the temperature readings I get are seemingly reliable (room-temp-ish readings, around 20C). The moment I plug in another TMP1101_1, the readings cut in half, to 10C, for both.

This seems like a bug in the software, but I'm unsure. I'm wondering if anyone else has seen this, or if what I'm doing is specifically unsupported for some reason.

Any help appreciated, thanks!
Patrick

[jdecoux]

When using a TMP1101, it is important to ensure the thermocouples you are using are electrically isolated from each other on the sensing end (not touching each other or connected metal surfaces)

Do you see this issue when using individual, separate thermocouples on each TMP1101?

[pg_ammo]

I'm using a multipoint thermocouple from Omega (like this: https://www.omega.com/en-us/temperature ... 3-PP6-PP10)

I would assume they are electrically isolated from each other, but they're also wrapped in a metal sheath, which is conductive, so maybe not?

I do know that using a different TC reader, like the Measurement Computing 8-channel TC DAQ (this one: https://digilent.com/shop/mcc-usb-temp- ... q-devices/) doesn't have this issue. (But we'd like to move away from that and towards the Phidget devices.)

[burley]

Do you have a multimeter? I'd be interested to see a continuity check between the different outputs of the probe. Is there an internal short or something? They COULD be isolated, but there could be damage, or of course (and this is more likely) they could also be grounded to the probe sheath which is not uncommon in which case they are definitely not isolated. The TMP1101 will not be happy with grounded thermocouples, you'd need to use the TMP1100 for that.

[pg_ammo]

Got it, thank you!