Go Down

Topic: 18B20 in strong electro magnetic fields? (Read 1 time) previous topic - next topic


Hi there!

I'm working on an axial flux alternator. As a safety feature, I would like to integrate temp sensors into the stator, as close as possible to where the magnets pass, so I can monitor internal temperature and trigger a shut down in case the alternator starts overheating during operation.

I have some 18B20 lying around, but I'm not sure if they will perform well inside a generator due to the presence of strong inductive forces, magnetic fields etc. Maybe it's just a question of having the right filters? The Alternator is slated to run at a steady 500-550 RPM, so EM noise should only occur in a distinct frequency range during normal operation.

The whole stator element, including sensors, will be pored in resin before operation. So, what ever I decide to do, once it's done, it's done :)

I'm open to any suggestions, ideas... Thanks!


First, any wires going to chip will generate a voltage when it crossing magnetic field, may be it wouldn't be a big problem with digital interface, yet all depends how strong field is.
Second, Si what chip is made from, would be subject to hall effect, and again a lot depends on strength of magnetic fields.


Aug 16, 2011, 05:50 pm Last Edit: Aug 16, 2011, 05:52 pm by retrolefty Reason: 1
I'm open to any suggestions, ideas... Thanks!

Whatever temp sensor you decide on you might consider adding a couple of spare ones before the "
The whole stator element, including sensors, will be pored in resin before operation"

At the refinery I worked in before retirement, the very large industrial motors (up to 5K HP!) would have RTD sensors embedded into the stator windings and they usually installed several spare sensors in case of sensor failure.

As far as magnetic flux effecting temp sensors, for the RTDs they were wired to analog differential op-amps so common noise appearing on both wires tended to cancel out. Not sure how a 18B20 would react to the rotating field.


I assume you can cast the temp sensor in a metallic shield something like - http://arduino-direct.com/sunshop/index.php?l=product_detail&p=151 -

magnetic fields will be blocked while heat can pass through at a slightly slower pace.

The proof will be in the pudding test I guess :)

(all disclaimers apply as I never tested it)
Rob Tillaart

Nederlandse sectie - http://arduino.cc/forum/index.php/board,77.0.html -
(Please do not PM for private consultancy)

Go Up