Supplying two temp controllers with a single element thermocouple

Question

I am installing a process temperature controller and an overtemp controller for a heater bundle. Normally I would just install two thermocouples, one for each controller. But space is limited and I can only fit one probe. So if I were to have two separate T/C wires coming out of one probe, would that affect the temp readings? Has anyone tried this before? I am aware of a dual element probe but that is something I would have to purchase, whereas I have the single element probe in hand. Your thoughts?

Answer 1

The Overtemp / Limit controller must have independence from the Process Control T/C. By Law: NFPA 86.

Answer 2

It would work but affect the accuracy in two ways.

Depending on how you hook it up, it would throw off the cold junction compensation. If the split happens outside of the controllers. A good controller will monitor the temperature of the TC terminals for the compensation.

It would also affect the loading. Typically the controller is compensated for its internal load on the TC. If you put two controllers in parallel the load is effectively double. This will cause inaccuracy in the reading.

Depending on your controller it may be possible to calibrate out both of those effects by swinging the TC through its range in an oil bath and then storing those points in the controller. Most likely the controller just accepts a specific type of TC and is precompensated for it.

Answer 3

With normal industrial temperature controllers, this will work fine, absolutely no problems at all. Hobbyist toys will probably not work properly, and if the unit won't tolerate either floating or grounded junctions, it is suspect.

The only effect is that the small current that is passed through the thermocouple to detect a broken sensor will be doubled so there will be a slight additional error as a result of that. As long as your leadwires are not too high resistance that's going to be negligible. Typically up to 100\$\Omega\$ is tolerated within the accuracy specs. This is not really a serious concern in most cases, as 50\$\Omega\$ represents a lot of wire (for example, 84 double feet / 25 double meters of AWG 20 /0.8mm type K (Chromel-Alumel wire).

---

However, note that you're losing some redundancy with this setup, so, for example if the leadwires get pinched and shorted away from the hot zone the controller will call for excessive heating and the overtemperature limit will agree. In the case of many other failures, such as a failure in the controller itself or a stuck contactor, the problem will likely be detected. There's also a small chance that either the controller or the overtemp could fail in such a way as to make both think the temperature is lower than it should be.