Source for a contact tachometer?

It may be outside of the realm of what alot of you guys do but I am always suprised with the knowledge that I find on this site.

I am trying to instrument some previously designed gearbox test stands to meet my own test and measurement needs, and I am looking for something where I can hopefully place a contact type tachometer against a shaft and get highly accurate measurements of RPM with out modifying the test stand at all.

Typically speaking when we use tachometers in this line of work we use standard hall effect magnetic pickup sensors, where usually we have an existing gear on a shaft and each gear tooth that goes buy creates a electrical spike. Basically you have the gear functioning as an encoder wheel.

Unfortunately it is looking like it will be very difficult to modify a few of the test stands to put any type of gear on them, thus what I am wondering if anyone has heard of an off the shelf contact tachometer that can be mounted semipermanently against the shaft and output me hopefully a signal containing multiple pulses per revolution that I can record and use to monitor the gearbox.

What ever I would be using would have to be very accurate as in many cases I am trying to determine shaft speeds and gearmesh frequencies of that are several times higher than the shaft that I could measure speed from. Thus any error would quickly multiply.

I have done some searching on Google and yet all I seem to find are a bunch of these hand held combination tachometers that have a wheel on one end that look like this.


However I really need something more permanant and something that will also direct output its pulses to me. I am envisioning that there has to be someone who makes something that works similar to the old travadials that mount on a lathe carriage, where you just press this box with an encoder wheel hard against the shaft and then have a working tachometer. Has anyone ever seen such a thing or have an idea where to look for one?

I know some of the groups have printed out mylar sheets index lines and than wrap them around the shaft and use a reflective type optical encoder, but I would like to avoid that if at all possible, as they can be very prone to dirt and other problems.

Just wondering if anyone has ever seen such a thing?

Thanks,

Adam

did you try automation direct? Omron? Measurement computing, or how about National Instrument/Honeywell. There are "cheap" data aqusition modules and sensors. you can go several ways, you can glue a magnet to the shaft and use essentially bicycle spedometer (but of course there are industrial versions)

or you can put a tone ring on and use an inductive or a hall sensor.

measurement computing, cheaper then N.I. option, included software.
USB Counter Timer Solutions from Measurement Computing

Honeywell Sensing and Control

anyway, I use inductive non contact and put a steel tone ring (looks like a crude gear) around the shaft. and a national instrument USB compact daq NiDaqMX

epoxying on a tiny magnet would be an easy option for you.

Starret 104
http://i72.photobucket.com/albums/i163/miltons_stuff/Starrett-104.jpg

dsergion,

I tried searching with a few of those websites and didn't find anything close to what I am looking for. We actually are using very similar tachometers to what Honeywell sells in the hall effect sensor department but the key issue is affordably rigging something up that goes onto the test stand. In the end we will also be processing/recording the signal wth some NI cards.

One thing I am thinking may need to be done at the end of the day is to have a mock gear cut either waterjet or wire EDM and try to find a coupling to put it on at. Any idea on the cost of having a 127tooth 9" or so gear made up that way?

None the less since this is not our group's test stand and this is a huge company it is going to take some work to actually get that on the stand.

What I did notice you mentioned was using a steel tone ring, do you know of any commercial manufacturers of tone rings preferably one that could bolt on? I did a quick google search on "Tone Ring" and got nothing. Sure I know this is a machinist site and making a crude ring shouldn't be much of an issue to many but the headache of getting any nonproduction work through the shops here at work make that no easy task! In the end of the day we may still need to rig something up but in efforts to "save" money corprate management have pretty much have gutted any shop capable of doing non standard development type work, and as a result you have to spend 10X to get anything done around here. Hence an off the shelf solution would be my #1 preferance.

Winfield, that would be the concept I am thinking of, of course on a slightly more permanant and modern scale.

Thanks,

Adam

Adam, you mentioned 127 teeth. This is a common number of teeth in Imperial to Metric change wheel, where a 120 - 127 pair is used to give metric pitches. Should be available on Ebone.

Edit. I was thinking of the one on my 11" Cochester and I can't find one of those tonight, but this is a smaller "Non Genuine Myford" one RDGTOOLS 127T CHANGE GEAR FOR MYFORD LATHE on eBay (end time 03-Mar-11 14:12:18 GMT) This firm is an hour away from me further north in Yorkshire, they are quite reputable. A lot of the stuff is Indian in origin, but that wouldn't matter for your purpose.

Steve

tone ring is a trivial thing. I have had them waterjet. just as easy to buy a stock gear of rather coarse pitch and cut out the center.

if you are using a daq it is simple to tune the thing to handle whatever your signal is, so you dont need a perfect square tooth

good rule of thumb for an inductive 2 wire sensor is put 60 teeth on about a 3" disk.
teh inductive is simple. gives sine wave output. cant tell direction. to read direction you need dual hall effect and signal processing.

the teeth can be water, laser, or even drilled holes

see graphic. for your use pretend that is about a 3" disk tehn it's close to actual scale and 30 teeth is good. 60 is more convenient because HZ then= RPM, but they are getting a bit fine pitch and more finicky.

http://www.brakeandfrontend.com/article/79736/wheel_speed_sensors_abs_and_tcs_signal_use.aspx

I guess the correct term for the sensors I use is "variable reluctance" I just know it as a 2wire speed sensor at Caterpillar.

can also glue magnets or even metal chunks on and use the same sensor perhaps.
or if you have some splines on your shafts you can image them. or a cardon joint yoke.
the daq processing is typically adjustable enough to hande whatever you can get.

I'm in the business of designing and operating lower powertrain test stands.

if you need more specific info, just ask. or email. or call.

adammil1 said:

dsergion,

Winfield, that would be the concept I am thinking of, of course on a slightly more permanant and modern scale.

Click to expand...

The Starrett was just kicking around in my tool box and was posted for fun. I was thinking of a more modern and simple way using a photodiode/ laser/ LED non-contact sensor wired to a counter. Modern simple and no contacting wearing parts. You could just use a piece of reflective tape on a shaft. Or a magnet and pickup. Or a eddycurrent sensor off of a shaft. Why complicate a tachometer with gears and wearing parts?

Still not what you are looking for but more modern than Winfield's, Stewart Warner 82682 good to 4000 rpm .Bill

I have an Extech model 461895. I believe I got it thru MSC or some tooling supplier.

It works creat for verifying spindle RPM and table speeds using the contact part of it.

It does have an optical eye which you then take reflective tape which is supplied for checking surface speed. I don't use that part of it too often.

Adam, what speed is the shaft you will be measuring? Unless it is low speed, you will not need a gear on it, a couple points is more then enough accuracy for vibration measurements. How many lines of resolution does your FFT have? That might be more limiting then a single reference point tach.

You can do a simple hub with cap screws on the OD as your sensor targets, or take a bearing locknut and screw that into a hub. The lockring has grooves already cut into the OD for tightening it and are suitable as targets for sensors.

For mounting the sensor, us a magnetic base you would use for a dial indicator, that is what I use on our machines for temporary installations.