Bending thermocouple wire

[bret]

I cut all mine just after the SS braided area, it is just normal Tefzel wire beyond that with an outer jacket joining the two wires, I made my own custom harness with like size Tefzel wire to a Doitch? gold pin connector and OHM tested every sensor before and after the harness construction, same same. you need to extend that wire anyway to reach the EMS ya?

[bret]

Originally Posted by wjb

Me, too. Thermocouples generate a voltage difference across a dissimilar metal junction that's a function of the temperature of the junction, but don't (and can't) generate any appreciable current, something that the IR (current*resistance) drop would affect. There's no issue cutting the wires to the correct length.

They are, yes, usually a single solid wire (not a fine gauge stranded set), so minimizing tight radius bending will contribute to their longevity.

Feel free to leave them long, or cut them as you see fit.

Checkout the Omega Engineering FAQ for more detail (http://www.omega.com/prodinfo/thermocouples.html). No association with Omega other than being a very happy customer over the years. (and, they probably make your thermocouple)

on the Dynon harness, the brown wire harness IS multi strand Tfzel wire, and solid wire at the junction from inside the SS braid to the sensor. on mine anyway ;-)

[wjb]

Originally Posted by 1001001

I don't have a reference handy (I'll try to find one) but in industrial control systems great care is taken to properly calibrate the resistance of thermocouple wires. If wires need to be shortened or lengthened the sensor must be recalibrated. In addition, only special "thermocouple wire" is used to extend thermocouples. In industrial controls, typically it is now simpler to run a short thermocouple wire to a local transmitter than to run thermocouple wire all the way back to the control panels as was done commonly in the "olden days." I work in an industry in which we have equipment that has been running since the 1940s alongside brand new stuff. It's interesting to see the changes.

This is true for RTDs (Resistance Temperature Detectors), which can be affected by lead length if used in "two wire" configuration (a 4 wire configuration uses a separate current source to drive the sensor, leaving the remaining two wires for a pure voltage (zero current) measurement). Perhaps someone was calling it by a wrong name?

[Mich48041]

Maybe decades ago before the advent of FET transistors, thermocouples with an exact wire length were matched to an instrument.
Although copper wire has been used successfully in some cases to extend thermocouple wire, it is not recommended. When two dissimilar wires are spliced together, a new thermocouple is formed. A voltage is generated which will add or subtract from the voltage output of the intended thermocouple. If both thermocouple wires are spliced to copper wire at the same location and exposed to the same temperature, then the affects of the two splices might cancel each other. However, it is not good practice to use any type of wire other than thermocouple wire that is made for the type of thermocouple used. As with any aircraft wiring, fewer splices results in fewer potential trouble spots.

[mburch]

There is a somewhat related kernel of truth here that shouldn't get missed:

When cutting and splicing thermocouple wire using standard copper or gold-plated connectors and pins, best practice is to make sure that the splices for the "high" and "low" wires are located in close physical proximity to one another. This ensures that both halves of each joint are exposed to generally the same ambient temperature. That's desirable because each of those connections is actually a little dissimilar metal junction all by itself, but the effects cancel each other out if the high and low sides are at the same temperature.

This is not normally a problem for our application, since the thermocouple extension wire we use typically comes as a conjoined wire pair, and we generally put the high and low connections together out of installation necessity. But it could pose a problem if for some reason a guy ended up putting some extra splices into, say, the yellow wire but not the red.

(Note to engineers: Yes, the above is an oversimplification, but for the kind of accuracy we are shooting for, I think it's a good enough rule of thumb.)

Edit: Hah, Joe posted the same advice while I was typing this!

[Taildrgr]

From the Dynon installation manual:

"Wire length of CHT/EGT harnesses supplied by Dynon is not critical. It is acceptable to cut the thermocouple (brown) wire of the CHT/EGT harness to shorten the appropriate length for your installation. It is recommended to not cut the ?armored? portion of the cable near the sensor end of the CHT or EGT sensors.
The EGT wire may be extended provided that: 1) Type K thermocouple is used for the extension wire. (This is different than CHT wire.) 2) There is no temperature change across the connection between the junction that is created. For example, do not create a junction across the firewall. If either of the above guidelines are not followed, your EGT readings will be incorrect."

Note: this is from Dynon, may not apply to other systems.

[RVDan]

When a thermocouple is interfaced to a pyrometer, which is an instrument powered by the thermocouple, the total harness resistance is critical to typically +- .05 ohms. Older turbine EGT indicators were all pyrometers and had a trimming resistor to adjust for changes in harnesses when the engine was changed. A pyrometer input impedance is typically something like 8 ohms.

Since the advent of high impedence instruments thermocouple lead/harness resistance hasn't been critical.

Hope that helps the understanding of where that thought came from.

[dynonsupport]

It's been covered here, but to simplify it to why there is some historical data around cutting themocouples:

If you have an analog gauge, you can't cut thermocouple wire. These indicators are current devices and are calibrated to the resistance of the thermocouple circuit. It was really only 15 years ago that this was all airplanes.

If you have a digital gauge, these can be cut. These are voltage devices with basically zero current in the system and the change in resistance from the length change is not a factor.

[drone_pilot]

I understand the Cutting/Extending of thermocouple wiring not being an issue for SkyView according to the Dynon instructions. My question was only about bending the wires into a bundle forward of the firewall! Geesh you guys! )

Made for very interesting reading though! Lots of smart and talented folks building/flying the RVs!

[Mich48041]

Coiling excess thermocouple wire will not affect its function as long as the insulation is not damaged by sharp bends. A coil of excess wire does not look good and adds weight and might be more vulnerable to damage from vibration. It is helpful to have a small service loop of extra wire at the EMS.