Determine wheel position

**Futterama** · 12th October 2005 08:41 PM

Hello you guys,

I have this wheel (flywheel on a RC gas engine) and I want to measure the RPM's.
I have paintet the flywheel black, and put a bit of white paint on one spot on the wheel. See the drawing.
Now, my problem is, what kind of optical component could be used to sense if the black or white color is in front of the sensor?

I thought about drilling a hole (or 2 to balance it) in the flywheel (to use an optical interrupt switch), but it's hardened steel, and it won't give in for the drill bit.

Any suggestions?

Regards,
Futterama

**jbeng** · 12th October 2005 09:58 PM

A reflective-type opto coupler should do the trick. Basically, it's an led and phototransistor mounted side-by-side. Light (or IR) from the led is reflected back to the phototransistor only when the light-colored spot is in front of the pair.
JB

**Futterama** · 13th October 2005 05:18 AM

So what you are saying is that IR will only be reflected from a white (or bright color) surface, and not a black?

Do you know any part numbers for such a reflective-type opto coupler?

Is the QRB1114 (Phototransistor - Reflective Object Sensor) from Fairchild the type of device you're talking about?

Regards,
Futterama

**Someone Electro** · 13th October 2005 05:42 AM

Creful wen using IR becose some black sufaces also reflect IR.You cod meaby try it out whith a camera and an IR LED to see if it reflects IR.

**electronist** · 13th October 2005 11:03 AM

I think the white surface will work fine. But if in doubt how about gluing a small piece of mirror on the white surface. And if you cant drill through then use the edge to attach some small obsticle and use a simple led- ldr pair

**Nigel Goodwin** · 13th October 2005 11:25 AM

Originally Posted by electronist

I think the white surface will work fine. But if in doubt how about gluing a small piece of mirror on the white surface. And if you cant drill through then use the edge to attach some small obsticle and use a simple led- ldr pair

You really don't want to add any weight to it, or remove any by drilling holes, these engines rotate EXTREMELY quickly - unbalancing the flywheel will introduce serious vibrations.

**bryan1** · 13th October 2005 11:29 AM

When you say it's an R/C gas engine it's probably doing something like 10,000 rpm so you'll need a pretty fast acting sensor for it to pick up. Maybe a hall effect sensor will be able to do it. I'm not sure as I haven't had the chance to try it out yet. How big is the fly wheel :?: as you could remove it, anneal it, drill a hole for a magnet, re-balance it and harden it again. Then place a magnet in the hole and mount the hall effect sensor within the specs and use a pic to count the pulses.

Just a thought Bryan

**Futterama** · 13th October 2005 11:57 AM

bryan1, the rpm of the nitro engine is more like 30-40,000.

But still, thats below 1000 rpm per second.

The flywheel is approx. 50mm in diameter and approx. 5mm thick (haven't measured it though).

I have thought about the anneal thing, but I don't know how to do it (it's something about heat up and down, slow/fast ect.). If I do that, I'll drill 1 or two holes (two to balance it) and use an optical interrupt switch (don't know if there is an optical interrupt switch with 6-7mm gap out there). Since the flywheel is steel, it will respond to magnetics and then I'm not sure you can use a magnet in a hole.

And of course I will use a PIC to count pulses, that's not a problem :wink: The problem is the mechanical-to-electrical "conversion" :roll:

**eblc1388** · 13th October 2005 01:01 PM

Do not change the physical properties of the flywheel.

The opto relfective coupler method would work on a painted surface but you'll probably need some signal conditioning/shaping before feeding it into the PIC.

Another strange idea is to magnetise the flywheel. You can then use a pickup coil to pick up the induced voltage.

**Futterama** · 13th October 2005 05:54 PM

Originally Posted by eblc1388

Another strange idea is to magnetise the flywheel. You can then use a pickup coil to pick up the induced voltage.

And how can that determine the rotation speed?

**eblc1388** · 13th October 2005 08:01 PM

Originally Posted by Futterama

And how can that determine the rotation speed?

The frequency of the voltage is directly proportional to the rotating speed.

**Futterama** · 13th October 2005 08:20 PM

You will have to explain how you would magnetize the flywheel. Or is it like on my drawing?

**eblc1388** · 13th October 2005 08:29 PM

We were taught back in our school days how to magnetise a piece of steel using a strong magnet, moving it in a looping fashion.

Perhaps you can do the same on the steel flywheel.

**Futterama** · 13th October 2005 09:43 PM

Originally Posted by eblc1388

We were taught back in our school days how to magnetise a piece of steel using a strong magnet, moving it in a looping fashion.

Perhaps you can do the same on the steel flywheel.

I'll give it a try. I have a strong magnet from an old harddrive.

**jbeng** · 14th October 2005 09:17 PM

Originally Posted by Futterama

So what you are saying is that IR will only be reflected from a white (or bright color) surface, and not a black?

As Someone Electro said, the black will probably reflect some IR, but it's likely that the white will reflect more IR than the black will.

Originally Posted by Futterama

Is the QRB1114 (Phototransistor - Reflective Object Sensor) from Fairchild the type of device you're talking about?

Yes, that's the type of device I was referring to.
JB

Electronic Circuits Projects Diagrams Free

Thread: Determine wheel position

Thread Tools

Display

Determine wheel position

Tags for this Thread