Thermocouple input gauge

[kattz]

Has anyone here had any luck with using thermocouple EGT gauges?  I have a beauty from Weston...  If you have, please PM me and give me some insight.  Much appreciated.

Kevin

[kattz]

BTW, on my cheap Chinese bench power supply, with the volt pot just cracked open and showing 0.00 and the current pot slowly increased from zero just slightly, the gage will snap from zero to 400°, and I can control it up to the gage max of 900° by moving the current pot ever so slightly.  Both displays show 0.00 for voltage and current, so my guess is millivolt power.

Maybe if I use something with analog out and put a resistor in front of the gage input...

Maybe an Arduino with PWM outputs?  Not sure about that... hmm. 

BTW, I know how thermocouples work... just not sure how to synthesize that signal.

[ame]

If your gauge is responding directly to the thermocouple output then it might just be a simple moving-coil ammeter.

Thermocouples generate a tiny voltage (as you know). Based on a casual internet search I decided that the instrument would use a type K thermocouple, which generates 41uV/C, so 900 degrees C (you didn't specify C/F) would be 36.9mV.

You might be able to do it with an Arduino. The Arduino can generate a PWM signal, which you could feed into a low-pass filter and voltage divider. You might not need a low-pass filter because the meter itself has some mechanical damping. You might need a buffer amplifier between the output of the Arduino and the input of the meter, in which case you could set the gain of the amplifier to take the place of the voltage divider.

If I can help I'd rather do it on the forum than by PM, since other people can read it and either:
a) learn about the topic
b) provide cromulent suggestions
or
c) avoid my stupid mistakes

[kattz]

That would be awesome, because I am fairly deficient in the knowledge business when it comes here to the Arduino. Would love to learn and I know I've driven some members of this form crazy asking questions!

I think it would be great if it was an open discussion on the form simply because other people could learn from it as well.

Let those who would be cromulent be cromulent.

Thanks! 

Kev

[ame]

Ok, well, what do you know?

I'm not really sure where to start. Is the gauge really just a moving-coil (or similar) meter? If not, what is it?

Do you have a sensitive digital multimeter? Can you repeat your test with the power supply set to almost zero and measure the output voltage in mV? As you increase the current limit (without destroying the meter!), does the voltage change? Can you plot a few samples of voltage (mV) against gauge reading?

Have you programmed an Arduino before? Do you have an Arduino to play with?

A program to generate PWM of a specified value is quite simple. Hooking up an Arduino to a moving-coil meter and generating a signal to move the needle is pretty simple too. If your gauge is a moving-coil meter then putting two-and-two together should be straightforward. That's a big IF- I don't know what's inside the gauge.

But, if you can show a relationship between mV and gauge reading, and if you can construct a circuit that converts the Arduino PWM output into a mV-range signal, then it ought to work without damage.

...is my guess.

[ame]

Actually, I just realised, if it's a voltmeter you only need a multiplier resistor in series. The multiplier drops most of the voltage, and limits the current, then the bit left over drives the meter movement.

[kattz]

As you said, the K type thermocouple puts out millivolts, so that mV signal would have to be generated.

I have an Arduino but lack the discrete components as of the moment to make it work... Tuesday, I'll have "stuff" and a breadboard.

[ame]

Ok. You probably need only one high-value resistor.