Sensing on rotating parts.

[vini_i]

I need to sense a temperature threshold on a component that rotates. If the component gets too got the bimetallic switch opens.

My idea for doing this is with inductive sensing. Have two coils that act as loosely coupled inductors. The coil on the rotating side would be connected to a resistor and the thermal switch with a resistor. The idea is that there would be four distinct states.

• Switch closed
• Switch open
• Open in the circuit (Fault)
• Short in the circuit (Fault)

These states would be sensed by the current consumption of the stationary coil.

Does a device exist that can both excite the coil and have built-in current sensing and thresholding?
If it was in an industrial DIN rail mounted package that would be great but I'm more than willing to work around other form factors.
My issue is that I'm not even sure how to search for a device like this.



Thanks

[duak]

Two ideas come to mind:

A rotary transformer could do what you want: https://en.wikipedia.org/wiki/Rotary_transformer  Google gives a number of hits.

The second would have two coils mounted on opposite sides of the rotating part connected electrically by the thermal switch.  The coils would pass over a stationary magnet .  Opposite the magnet would be a stationary pickup coil with a sensitive amplifier.  The idea would be to have a voltage pulse induced in one coil by passing by the magnet that would be passed to the other coil if the switch were closed.  If the rotating part turns too slowly, the induced voltaged would be small.  However, the magnet could be replaced by another coil driven by an AC source.  Please note that the signal amplitudes would be quite small and it would take a bit of electromagnetic engineering to work out.

[cur8xgo]

I need to sense a temperature threshold on a component that rotates. If the component gets too got the bimetallic switch opens.

My idea for doing this is with inductive sensing. Have two coils that act as loosely coupled inductors. The coil on the rotating side would be connected to a resistor and the thermal switch with a resistor. The idea is that there would be four distinct states.

• Switch closed
• Switch open
• Open in the circuit (Fault)
• Short in the circuit (Fault)

These states would be sensed by the current consumption of the stationary coil.

Does a device exist that can both excite the coil and have built-in current sensing and thresholding?
If it was in an industrial DIN rail mounted package that would be great but I'm more than willing to work around other form factors.
My issue is that I'm not even sure how to search for a device like this.



Thanks

Drive the stationary coil with AC through a fixed appropriate series R

Use something to monitor that AC signal

The impedance of the rotating circuit will be reflected and you will see changes in your AC signal depending on what is open/closed etc.. on the rotating circuit

Can try this with a function generator and scope

Then can implement it with a microcontroller (entirely) or a few comparators/RC filters and a 555

As an alternative, have you considered using a thermopile non-contact sensor to scan the surface of the rotating object as it spins? Not sure if thats what you are measuring or not. But they have fast enough time constants and great sensitivity at close ranges to where you could possibly use that. Not expensive either

[m98]

If you aren't looking to make it yourself, this is exactly the product you're looking for: https://mesa-systemtechnik.de/en/inpud-produkteigenschaften/

[vini_i]

If you aren't looking to make it yourself, this is exactly the product you're looking for: https://mesa-systemtechnik.de/en/inpud-produkteigenschaften/

I was not looking to make it myself. That looks like exactly what I wanted. Even more so than what I wanted. Thanks

[Renate]

I don't know your application, but sometimes the best solution to a specific problem is fixing the general problem.

Years ago, the Ampex VPR-3 video tape recorder had an innovative capstan motor.
It used a "printed armature" or printed circuit or pancake motor.
The rotor looked like a copper web that had been squashed to a disk.
You can see some photos here (and of other motors): http://www.et.byu.edu/~chasek/ME%20472/Motors/electric_motors-USoW.pdf
The advantage was that it had very low inertia for high acceleration.
Because this motor was so powerful and the rotor (or armature) so light weight it could overheat easily some thermostatic protection needed to be included.
Instead of actually measuring the temperature, they decided to make a circuit to model the thermal characteristics.
And that's how they implemented thermal protection.

[vini_i]

I don't know your application, but sometimes the best solution to a specific problem is fixing the general problem.

Years ago, the Ampex VPR-3 video tape recorder had an innovative capstan motor.
It used a "printed armature" or printed circuit or pancake motor.
The rotor looked like a copper web that had been squashed to a disk.
You can see some photos here (and of other motors): http://www.et.byu.edu/~chasek/ME%20472/Motors/electric_motors-USoW.pdf
The advantage was that it had very low inertia for high acceleration.
Because this motor was so powerful and the rotor (or armature) so light weight it could overheat easily some thermostatic protection needed to be included.
Instead of actually measuring the temperature, they decided to make a circuit to model the thermal characteristics.
And that's how they implemented thermal protection.

The application is a wire reel. It's a large portable piece of equipment with a built-in extension cord. The reel holds about 3 layers of wire. At full load, the equipment consumes 100A of 3 phase 480VAC. At full load, the reel needs to be unspooled down to the bottom layer. To keep the cable from overheating. At light load, all 3 layers can be on the reel.

We could use an absolute position encoder to check how far the reel is spooled out and combine that with a loading model to predict the cable temperature. That sounds too much like work and doesn't account for all conditions like burst loading. I want to just have a bimetallic switch on the inside of the reel up against the inner row of wire. When it gets too hot the switch opens and the equipment shuts down.

[wizard69]

slip rings would be dead simple and would work with the temperature switch already there.   This would also work at zero revolutions per minute which might be an issue if this something that dwells in one place for long.   I don't see how transformer coupling would work here.  Also slip rings can be made to be very rugged which sounds like an important feature if you are winding up a 100 amps extension cord on this drum.

As for anything industrial you will have to implement debounce in some manner or form as the hardware will likely get knocked about a lot.   Also I wouldn't run anything less than 24 VDC across the rings.

[jbb]

Texas Instruments does 'Inductance to Digital' converters which might be able to sense the change in impedance when the switch opens.  But they may not be suited to an industrial electrical noise environment.

[duak]

If the reel turns only three or four times per actuation and returns, I might consider a high flex cable inside the hub if there is room.  These cables are used in CNC machines and are rated for hundreds of thousands of actuations.