Battery powered (or passive) RPM sensor

Thread Starter

valer85

Joined Feb 18, 2021
18
Hi Everybody,

I need to measure the rotation speed of a rotating shaft, but with a wireless sensor.

This is a similar application:

I'm interested in the speed of the plastic dispensers. The most basic solution would be to use an inductive sensor + a phonic wheel. BUT, since the unit is mounted on a rotating arm, I can't power the sensor (rotary joint is not an option).

I searched on google but I was not able to find anything. Do you know if there's any kind of wireless inductive sensors without battery? or at least with a battery that lasts months. The sensor would transmit about 8 hours per day, so it's not low duty cycle. The rpm of the shaft is nothing special, about 30 rpm, so the sensor does not need a high frequency input.

very important: I need an off-the-shelf product, not something DIY style.

In my mind I was thinking about a sensor which could be powered magnetically with the shaft rotation itself. So a kind of passive sensor which transmits to a receiver positioned in the static part of the machine. But since it sounds very fancy, also a battery one could be taken into consideration.

Any idea of a suitable sensor for this application?

thanks
 

KeithWalker

Joined Jul 10, 2017
1,789
Is this requirement for a machine which is already assembled, or are you still in the design stage?
The simplest solution would be to rotate the object being wrapped, keeping the dispensing arms still. Then no remote transmission link is required.
 

ericgibbs

Joined Jan 29, 2010
12,942
hi 85,
You could mount a small magnet/s on the rotating shaft assembly and mount a Hall Effect sensor on the on the body of the wrapper, so that the magnet/s would pass close to the Hall device.
The pulse from the Hall device would drive the analog input of low cost MCU, say a Arduino Nano and LCD display.

E
 

shortbus

Joined Sep 30, 2009
8,591
The simplest solution would be to rotate the object being wrapped, keeping the dispensing arms still.
You may not know what that video showed. It was a "round baler", that makes round bales of hay or other animal feed(corn stalks are popular around here). The bale was being rotated while it was being wrapped ,by rollers under it that are part of the baling process. The bails are 100 or more pounds in weight.
 

shortbus

Joined Sep 30, 2009
8,591
You could mount a small magnet/s on the rotating shaft assembly and mount a Hall Effect sensor on the on the body of the wrapper, so that the magnet/s would pass close to the Hall device.

That could be done by an off the shelf bicycle computer, a cadence computer. They work that way, a magnet on a pedal arm and Hall sensor on the frame tube.
 

Thread Starter

valer85

Joined Feb 18, 2021
18
The simplest solution would be to rotate the object being wrapped, keeping the dispensing arms still. Then no remote transmission link is required.
I can't change the design, and the concept is the same one as in the video, so of course I can't rotate the object to be wrapped, otherwise there wouldn't be any problem :)

hi 85,
You could mount a small magnet/s on the rotating shaft assembly and mount a Hall Effect sensor on the on the body of the wrapper, so that the magnet/s would pass close to the Hall device.
The pulse from the Hall device would drive the analog input of low cost MCU, say a Arduino Nano and LCD display.

E
uhm... I'm sorry but don't understand how this could work. I need to measure the speed of the rotating shaft (of the plastic dispenser itself), not of the arm which holds the plastic dispenser. But if you put a magnet there, how could it be useful for this purpose? once it passes close to the hall sensor in the static part, assuming it's on the right side of the shaft, how could that give a measurement of the speed?
 

Thread Starter

valer85

Joined Feb 18, 2021
18
As I said in the other post I have to measure the rotation speed of the plastic dispenser itself, not the rotation speed of the arm.
 

djsfantasi

Joined Apr 11, 2010
7,693
In your video, it looks like there may be a fraction of an inch below the plastic roll. Or at the bottom of the idler roller.

Could you mount a magnet in either location? Then, a nearby Hall effect sensor could count revolutions. Then a tachometer could measure the RPM. Like the one suggested above.

As a matter of fact, that device contains the Hall effect sensor and IS an off-the-shelf product. You just need to mount a magnet on one of the rollers. I suggest the idler roller is the best location.

00AD9642-0526-4652-A9E1-C28C003E4863.jpeg
 
Last edited:

KeithWalker

Joined Jul 10, 2017
1,789
I assume you are trying to measure how much wrapping material has been used.
How accurate must your result be? Depending on a realistic answer to that question, there may be other ways to measure it.
 

Thread Starter

valer85

Joined Feb 18, 2021
18
I apologize, maybe I wasn't clear enough or I'm not understanding your ideas.
I retry.

the final goal is to measure if that plastic roller is rotating or not (so if the plastic is not being applied to the object) so even measuring a binary condition would be enough (rotating - not rotating).
As Keith stated correctly, an even more advanced measurement would be to measure the amount of applied plastic, but that's only an option and not required at the moment.
Having said so, not much accuracy is needed, just a binary output. I hope this ease the problem.

@djsfantasi: I'm sorry I'm missing how you transfer the tachometer reading to the static part. I can mount magnets wherever we want, but I don't understand this part.

@sghioto, I hope I answered above.

thanks for your effort guys, really appreciated
 

djsfantasi

Joined Apr 11, 2010
7,693
I'm sorry I'm missing how you transfer the tachometer reading to the static part. I can mount magnets wherever we want, but I don't understand this part.
The Hall effect sensor and the tachometer is mounted on the static part. The magnet is mounted on the roller. When the magnet passes the Hall effect sensor, as the roller rotates, it registers a pulse. This pulse is passed to the tachometer which performs the “calculations” that transform the pulses to RPM.
 

Thread Starter

valer85

Joined Feb 18, 2021
18
I apologize if I don't understand.
how can you detect (or measure) the rotation of the roller with a magnet on the roller and a sensor on the static part??
The roller, as you can see in the video, rotates, but also the arm which holds the roller rotates.
So when the arm passes in front of the static part, once per revolution, I don't understand how the magnet could be useful to measure the roller rotation.

Let's say that, for any reason, the roller is rotating but the magnet is never on the sensor when the arm passes in front of it. in this case the sensor will always detect nothing.
I don't know if you see what I mean..
 

shortbus

Joined Sep 30, 2009
8,591
Like I said in post #5, a bicycle cadence sensor is the fastest and easiest way to do this. That is what they are designed for, to count the cadence/rotation of the pedals as you ride. Mine is an old one that uses a mgnet and sensor, they are held on by zip ties to the pedal arm/crank and the frame. But the new ones are wireless bluetooth. https://www.wahoofitness.com/devices/bike-sensors/wahoo-rpm-cadence-sensor They're water proof, vibration proof inexpensive and off the shelf. Plus they will do exactly what you want.
 

LesJones

Joined Jan 8, 2017
3,189
Am I correct in assuming that it is the rotation of the two rolls of plastic film that you are trying to detect ? If so to get round the problem of using a battery to power the system which you want to last for several months I think using a small PM motor as a generator on each roller could solve the problem. The motor could provide enough power to drive a small transmitter that could encode the voltage output from the motor (Which would be proportional to its speed.) You could could then have a a receiver mounted on a static part of the machine that would display the speed. ( Or just indicate if the speed was above some preset level.)

Les.
 

Thread Starter

valer85

Joined Feb 18, 2021
18
Like I said in post #5, a bicycle cadence sensor is the fastest and easiest way to do this. That is what they are designed for, to count the cadence/rotation of the pedals as you ride. Mine is an old one that uses a mgnet and sensor, they are held on by zip ties to the pedal arm/crank and the frame. But the new ones are wireless bluetooth. https://www.wahoofitness.com/devices/bike-sensors/wahoo-rpm-cadence-sensor They're water proof, vibration proof inexpensive and off the shelf. Plus they will do exactly what you want.
this one looks nice, I need to check if they have also a non-bluetooth version, with the receiver that can be interfaced with my ECU as an input.


Am I correct in assuming that it is the rotation of the two rolls of plastic film that you are trying to detect ? If so to get round the problem of using a battery to power the system which you want to last for several months I think using a small PM motor as a generator on each roller could solve the problem. ........

Les.
this is what I wrote at the end of my first post. But nothing off-the-shelf exists (to my knowledge)


Then all you need is a hall sensor a LED and a battery.
this is not an off-the-shelf solution and I need to detect the rotation with an ECU, not by looking at it! Otherwise looking a the plastic dispenser would have been enough to detect a problem :)
 
Top