Remote Temp Sensor

[JohnC]

Good morning group,

I would like to know if it would be possible to monitor stator heat using a indoor outdoor remote temp. guage.

Not sure what component to remove. I was thinking again... scary at times, then maybe cast it in the stator. I'm sure one of you's electronic wizzards could let us know how to make the modifications.  ;~}  

THX,

John E. Cox

Ottawa, ON

Canada

[dinges]

Have never done this, but it should be doable, depending on the design of the remote temperature unit.

If it uses a discrete component for measuring the temperature (like a LM35 temp. sensor), it should be easy enough to remove it from the circuit board and mount it into the stator, with extension wires.

However it's hard to tell whether it's actually possible with the particular temp. unit you have. Got to see what's inside.

So the answer is yes, it's doable in theory, but exactly how depends

Do you have a temp. sensor that you intend to use? If so, open it up and post a good detailed picture of the parts of the PCB. Then we could start giving more to-the-point advice.

And I agree on the thinking bit. Can be scary, no, downright frightening.

[disaray1]

In agreement with Peter here, couple of things to consider first. An indoor/outdoor temp sensor may not have the temp range your looking for...does it go up to say 250 or 300F? Also consider the resistance in the sensor wire, from the mill, to the display....could be 200 to 300 feet(or more!) and that resistance would likely throw off your reading.

    Here's an idea. Get an engine temp sensor from an auto that has electronic fuel injection (a two terminal thermistor), hook that to one end of the wire length required, the other end to your ohm meter. Put the thermistor in your oven , along with a good baking thermometer, and start taking the ohm readings as the oven temp goes up. Go from room temp to 300 degF in small increments. Got your chart now. Sink the thermistor in your stator mold. Use your multimeter for the display.  You'll know by the ohm reading how hot (or cold) she is. Cheap and easy.

 disaray1  Good Day, Eh?

[dinges]

Maybe John could clarify, but I was assuming he was using a wireless remote temperature transmitter, the kind that usually works on 433MHz. So, the wires wouldn't need to be very long, a few meter at most.

Temp. range of the thing is ofcourse an issue.

[JohnC]

It's a remote sensor, no wires.

Have to wait till the wife is gone. :~} Will use her hair dryer to test the limit.

THX

John

[johnnythefox]

You may also want to see how well it well work close to a revolving magnetic field?

good luck

johnnythefox

[Ungrounded Lightning Rod]

If you do embed a sensor in the stator be sure that the wires are bundled very close togehter, and if it's a chip the chip is edge-on to the rotor's field.

Otherwise induced voltage from the moving magnetic field may totally swamp the temperature signal you're after.

Wound-wire thermisters may have enough turns to couple to the field big-time and should be avoided.  Ditto sensors with a metal case, which would experience extra heating from eddy currents.

If I were adding a temperature sensor to a radial-flux stator I'd put a hunk of about #10 wire (to serve as a heat pipe) radially into the area I wanted to measure (to minimize eddy current heating in the wire), bond the sensor to that  wire OUTSIDE the between-the-pole region (with lots of heatsink compound) and then bury the sensor and exposed part of the wire in some thermal-insulating foam so the sensor mostly sees the wire's temperature.

[Ungrounded Lightning Rod]

I might bury a second sensor inside the stator - then use the buried one (only!) to check the calibration of the one on the wire.  (I'd do that by running the mill until the temperature seen by the external sensor rose and stabilized, then stopping it and quickly reading the temperature shown by the buried sensor.)

[BigBreaker]

Why not embed the tiny thermistor into a tiny piece of brass or copper tubing?  Seems like that is a lot easier.  My thermistors are all about 2-3mm in diameter so the edge currents wouldn't be very high anyhow.  Twist the leads to help avoid interference or thread the leads through their own piece of tubing.  You'll need very little inductance to kill double digit hertz.

Also the solution to "wireless" sensing to use a whetstone bridge style circuit with the thermistor and a calibration pot.  That circuit can cancel the wire losses very simply so you can run cheap wire down the pole or all the way to the house.  Compared to a 10k ohm thermistor, that loss is quite low anyhow.

So either make the wire resitance low compared to the thermistor or use a whetstone bridge style voltage comparator that balances with no current running through the sensor.