Introduction: Spindle Speed Sensor From a Guitar Tuner
I just got a Pro membership a couple of days ago, thanks to Random_Canadian's generosity. (Thanks for the hookup!) I figured it was time to quit lurking and actually contribute, so here goes my first Instructable.
As a CNC technician, there have been a few occasions where I've wondered if a spindle or motor shaft was spinning as fast as the motor controller claimed. An old trick I learned a long time ago involved a digital guitar tuner. The trick was to hold a piece of stiff metal shim stock, around .020" or so thick, against a protrusion on the shaft. As the motor strikes the shim it should cause it to vibrate at a frequency that the tuner would turn into a note reading. If you were to look up the frequency for that particular note, some quick math would give you a reasonably accurate RPM.
The problem with this method is that most industrial machinery is not located in a concert hall. The ambient noises in a typical factory reduce even the most careful measurement right back into guesswork. So, if we generate an electrical signal instead of relying on sound, we can isolate and refine our measurement into something useful, and gain a tolerance of around 5 percent.
You'll need:
A small permanent magnet
Old headphones
Enameled magnet wire (pretty fine; mine feels around 0.010")
A digital guitar tuner with an input jack (you can use a good one - we won't be modifying it in any way)
A 1/4" male to 3.5mm female plug adapter
The patience to splice very thin, very fiddly wire
As a CNC technician, there have been a few occasions where I've wondered if a spindle or motor shaft was spinning as fast as the motor controller claimed. An old trick I learned a long time ago involved a digital guitar tuner. The trick was to hold a piece of stiff metal shim stock, around .020" or so thick, against a protrusion on the shaft. As the motor strikes the shim it should cause it to vibrate at a frequency that the tuner would turn into a note reading. If you were to look up the frequency for that particular note, some quick math would give you a reasonably accurate RPM.
The problem with this method is that most industrial machinery is not located in a concert hall. The ambient noises in a typical factory reduce even the most careful measurement right back into guesswork. So, if we generate an electrical signal instead of relying on sound, we can isolate and refine our measurement into something useful, and gain a tolerance of around 5 percent.
You'll need:
A small permanent magnet
Old headphones
Enameled magnet wire (pretty fine; mine feels around 0.010")
A digital guitar tuner with an input jack (you can use a good one - we won't be modifying it in any way)
A 1/4" male to 3.5mm female plug adapter
The patience to splice very thin, very fiddly wire
Step 1: The Sensor
The sensor is as cheap and easy as it gets. I picked up a package of 1/2" ceramic magnets from Radio Shack for around three bucks. (I buy these five-packs every time I need one magnet, and then the rest vanish. Apparently, they're biodegradable, or I have mice that are far more sophisticated than I would expect.) By wrapping some fine enamel-coated wire around the magnet (see photo) we have a magnetic pickup, like in an electric guitar. It's tough to see in the picture, but there's probably fifteen or twenty wraps of wire on there. I didn't worry about north or south poles, and I don't think that would really have an impact on function.
Step 2: Add Wiring
Lop the connector off of a set of old headphones, and splice on your magnetic pickup. I used a meter and Identified which point on the plug went to which wire - most headphones these days are stereo, with a discrete left, right, and common return - and wired the pickup onto the return, which should be the top (furthest from the wire) conductor on your plug, and either of the other two wires. Chop off the remaining wire and heatshrink everything together. You should now have something that resembles the picture.
Step 3: That's It. Go Measure Something.
So, now you have a magnet on a string. That doesn't help much, does it? How do you use this thing?
An object made of ferrous steel moving through the field generated by the magnet will affect the field. This, in turn, will induce an electrical current through our coil of wire. If the steel object were the head of a bolt attached to the spindle on a drill, every time the bolt head passed by, the guitar tuner would think it was a guitar string vibrating. By comparing a musical note reading from our tuner to a chart like this one,
http://www.phy.mtu.edu/~suits/notefreqs.html
we see that low C is 16.35 Hz, or 16.35 oscillations per second. That, times sixty seconds in a minute, gives us 981 RPM. If you had multiple cams (protrusions) on your shaft, you would divide your measurement by the number of cams. In the picture, you see that I used a paddle bit in my drill and read an A sharp at half trigger, and since it's a hand drill (max 1500RPM) it ought to be pretty slow, and this guitar tuner that I've got is only good for the first octave anyway.
A# = 29.14Hz x 60sec=1748 divided by 2 (since there's two impulses per rotation, one per side of my paddle bit) =874 rpm. Give or take.
During testing, I did discover that the non-grounded, non-shielded drill gave quite a bit of interference, so I put a three-foot headphone extender between the tuner and the sensor. That ought to allow you to hang this thing wherever you want, since the sensor itself is so small.
Also, check your tuner - some are good for multiple octaves.
Any questions, give me shout.
An object made of ferrous steel moving through the field generated by the magnet will affect the field. This, in turn, will induce an electrical current through our coil of wire. If the steel object were the head of a bolt attached to the spindle on a drill, every time the bolt head passed by, the guitar tuner would think it was a guitar string vibrating. By comparing a musical note reading from our tuner to a chart like this one,
http://www.phy.mtu.edu/~suits/notefreqs.html
we see that low C is 16.35 Hz, or 16.35 oscillations per second. That, times sixty seconds in a minute, gives us 981 RPM. If you had multiple cams (protrusions) on your shaft, you would divide your measurement by the number of cams. In the picture, you see that I used a paddle bit in my drill and read an A sharp at half trigger, and since it's a hand drill (max 1500RPM) it ought to be pretty slow, and this guitar tuner that I've got is only good for the first octave anyway.
A# = 29.14Hz x 60sec=1748 divided by 2 (since there's two impulses per rotation, one per side of my paddle bit) =874 rpm. Give or take.
During testing, I did discover that the non-grounded, non-shielded drill gave quite a bit of interference, so I put a three-foot headphone extender between the tuner and the sensor. That ought to allow you to hang this thing wherever you want, since the sensor itself is so small.
Also, check your tuner - some are good for multiple octaves.
Any questions, give me shout.