2
\$\begingroup\$

I am designing a product that uses a thermocouple to monitor temperature.

I am trying to decide what is the best way for users to attach the thermocouple to the PCB. I have found the connectors that Omega Engineering offers. The only problem is that they are a little pricey and hard to find through any of the typical component distributors (except Newark).

Does anyone know of any other standard thermocouple connectors? Do you think it would be bad to use a different style connector like an audio connector for example?

\$\endgroup\$
6
  • \$\begingroup\$ What's wrong with screw terminals? \$\endgroup\$ Jul 4, 2013 at 1:12
  • \$\begingroup\$ @IgnacioVazquez-Abrams Screw terminals could work for non-connectorized thermocouple. There also are industry-standard 2-prong thermocouple connectors. \$\endgroup\$ Jul 4, 2013 at 1:15
  • \$\begingroup\$ @PICyourBrain What temperature measurement accuracy are you trying to achieve? \$\endgroup\$ Jul 4, 2013 at 1:16
  • \$\begingroup\$ Within a degree is probably fine \$\endgroup\$ Jul 4, 2013 at 4:24
  • \$\begingroup\$ J type Thermocouples are typically +/- 1.5C for Class 1. It depends on the Type K,T,J etc. and Class 1 or Class 2. J type Class 2 is +/- 2.5C. \$\endgroup\$
    – Spoon
    Jul 4, 2013 at 11:20

2 Answers 2

2
\$\begingroup\$

the thing to consider with thermocouple interconnections is that each additional disparate metal type junction you include in the circuit may introduce error.

while looking for PCB connector options like you were, I came across this thread, which describes, for example, the benefit of K-type over J-type when connecting to copper PCB traces. have a read: http://www.nerdkits.com/forum/thread/2779/

any given connector you consider will have some combination of metals, & not just the core metal, but also the plating, if any (gold, tin, etc).

i also stumbled across these people selling PCB-mounting thermocouple connectors, with varieties for each thermocouple type: https://www.labfacility.com/pcb-mounting-thermocouple-socket-iec-p168-pg71/

Anthony.

\$\endgroup\$
1
\$\begingroup\$

The temperature sensor and the wires are one in the thermocouple. The temperature must be known where the thermocouple meets the circuit board. This spot is called cold junction.

The DIN EN 60584 standard specifies the thermal voltage to be expected in relation to 0 ° C. This is important to note because the thermal voltage caused by the Seebeck effect is not linear.

In most cases, an RTD (PT1000) is connected to the connector on the cold junction in a thermally conductive manner. The measurement inaccuracies from the thermocouple and from the RTD thus flow into the temperature measurement. For this reason alone, temperature measurement with thermocouples is complex and less accurate than just an e.g., PT 1000.

The length and the temperature gradient along the sensor cable also play a role. If you have many measuring points to monitor, the length of the extension cables should be the same to ensure better comparability. To ensure this, the individual measuring lines are lengthened to the required length with extension cables or compensating cables.

So yes, you can use audio connectors if you can take the temperature of the connector into account. Still, I don't think this should be done because these connectors don't evenly distribute heat well. Sockets for thermocouples have better properties in this regard and are also more suitable than anything else in terms of their mass.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.