Hello Troy -
Cutting the leads won't affect the accuracy but introducing new
junctions of dissimilar metals will. Here's a great snip from the
Aeroelectric list that addresses the issue:
"I have posted at least two discussions about thermocouple wire and how
it works - look at the archives.
That said, here is the short story. Thermocouples work due to the
Seebeck effect. That is, if you connect two wires made of dissimilar
metals they will generate a voltage at the junction that is proportional
to the temperature at the junction (and dependent on the types of
metals). It is important to remember that it is ANY junction of
dissimilar metals. This includes the junctions where the thermocouples
are connected to whatever you are using to measure the temperature. The
designers of these instruments always provide some way (internal to the
instrument) of compensating for the voltage generated at this junction
so the resultant temperature readout is that at the other end of the
thermocouple wires.
So... If we interpose another set of junctions in a length of
thermocouple wire - such as a connector - that could cause problems or
maybe not. I say could for two reasons.
First, if the temperature at each additional junction is the same then
the voltages generated will all cancel and you will still get the same
result on the readout instrument as you would without the additional
junctions. This is very likely the case if you pass the thermocouple
wires through a hole in the firewall (using the techniques that Bob
recommends for providing a fire resistant pass through) and use a
connector wholly within the engine compartment since both ends of the
connector will be in close proximity and at essentially the same
temperature. If you use a "barrier" connector with one side mounted on
the firewall and the other in the engine compartment, the situation is a
little worse since the two sides are obviously at different
temperatures. On the other hand if you use some sort of heat insulation
around one or both sides of the connector, since the pins are intimately
connected (both electrically and thermally), they are likely to be very
close in temperature and will cause a relatively small error.
Second, you are trying to measure relatively high temperatures and a
small temperature difference error generated by the connector may be
irrelevant to your actual usage. If you are using the EGT for leaning,
the actual temperature is only vaguely important - the relative peaks
are what you will look for. An error of 10 or 20 degrees would be
immaterial in this case. With the CHT the numbers are lower but the
errors are probably still not detrimental to your usage.
<<snip>>
Bottom line: If you are careful in what you do, you can use any
connector you want with minimal if any errors. If you are not careful
and end up with the two sides of the connector at significantly
different temperatures you will get an error that is approximately the
difference between the temperatures on the two sides of the connector."
Good stuff.
D
-------
Troy Maynor wrote:
>
> Hi All,
>
> I wanted to cut and route my CHT and EGT thru some cannon plugs at the
> firewall but I was told long ago that this would affect the accuracy of the
> readings if leads were cut. I since had planned to run those things thru a
> grommet with firestop material. Am I confused? Have others cut thiers with
> success?
> Troy Maynor
> N120EU Europa Monowheel Classic
> Left to finish:
> Paint,interior,engine install, wiring.
> Weaverville, NC USA
|