Thermocouple Meter

[pwdrive]

Mmm..maybe that's one for the solid sate gurus.

(Edit) I wonder how non linear it is..logarithmic? something anti logarithmic, I bet Dr Mark H or one of the other boffins might have a suggestion/answer.

Lawrence.

[pwdrive]

Hi Al, I'm wondering then if old "blown" thermocouples can be replicated, there seems to be a fair selection of modern thermocouples available with various characteristics so a match might be possible, I would have thought that it's the heating element that fails, if that is so then it appears that the element wire/strip is still available? Maganin comes to mind.
In all this I am thinking say of someone who has a duff thermocouple on an old piece of test gear or an old WW2 transmitter then maybe a thermocouple assembly could be made up as a replacement to drive the meter, assuming that the meter movement is still ok.
The other thing that comes to mind is that it could form the basis for a Q meter assuming that a signal source of low enough impedance can be found or made, in that respect I wonder what the output impedance of a standard signal generator is before it hits the attenuator, if sufficiently low enough then maybe that output could be used, it could be the basis of an accurate homebrew Q meter?

Lawrence.

[Skywave]

When I find 5 minutes, I'll make a plot of current through the meter versus meter scale indication. That should me give a starting point to discovering the required law to linearise the scale. Having said that, I can think of applications where a non-linear scale would not be a hindrance - such as this meter's original application, or even as an S-meter.
All hinges on what the law turns out to be.

Al.

[pwdrive]

Might be sods law Al, lets hope not.

Lawrence.

[Skywave]

Sod's or anyone else's I know not, but it is certainly a strange law.

Scale reading: over the scale range 0 to 80, it looks like an inverse-square law: at scale = 70, coil current = 1.4 mA. But at scale = 80, the curve suddenly straightens out (well, nearly straightens) and continues in that fashion up to the scale end = 100, where coil current = 1.8 mA.

Inside the meter itself, hand-written on the magnet: fsd = 1.8 mA. Well, that's re-assuring to some degree. Inside the case, there is a small resistor in series with the coil - presumably to bring the fsd voltage up to some convenient numerical figure: I haven't measured any resistances yet.

As for the current versus scale reading, I am beginning to wonder if the magnet has lost some of its magnetism. The movement itself is also substantially under-damped: a shunt R will easily fix that.

Al.

[pwdrive]

Al, I don't know too much about meter design, maybe it's designed to match the thermocouples characteristics over it's heating range whatever that may be, good luck with it anyway.

Lawrence.

[camallison]

The thermocouple is possibly a type "K" Cr/Al and the table pages are here: http://instrumentation-central.com/TechN...TableC.pdf

If possible, measure the mV coming from the thermocouple over the full range and see if it tallies with this table. Thermocouples are indeed non-linear and follow a distinct inverse-square law.

Colin