Other Techniques to Measure Speed of an Induction Motor Besides Using Tachometer

Yes. Look on the nameplate.

If you don't want to use a tacho, which is the simplest, the other method will be , you make a mark on the shaft, when motor rotate, you look at the mark, you blink your eye fast , adjust the speed of your eye blinking , at the correct speed of blinking , you should see the mark stationary, then you count the number time of your eye blinking per second, multiply this by 60 , that will be the rotation rpm. I don't know of another cheaper method .

LynDoor™ Industries has recently patented the Blink-O-Meter Speed Freq. Version 2.456.1.23.

It mounts on your head and counts the number of eyelid blinks per second and records it on a strip chart recorder.

Available in limited quantities in hot pink colour, with a magnified heads-up-display.

Offer void where prohibited by good taste.

If you have an ammeter, the slip is (at least approximately) proportional to the current drawn. Compare this with the full load current and nameplate rpm (vs synchronous rpm).

Thanks!!... any other similar ideas you can think of?

We need to know more. Is it supplied by an inverter? How accurate do you want to measure the speed? Do you have an ammeter? If not they're not that expensive.

If it's not inverter-driven, and supply frequency is right for the motor, you can get a good estimate as Tornado said in #4.

If it is inverter-driven, the inverter should have an display of output frequency which you can use to work out synchronous speed. You can then adjust that for current, at least for frequencies below nominal. Maximum current is roughly constant at turndown, maximum power reduces proportional to frequency.

Codey

Oh yes sir!!!

Thank u for replying, but the problem is that the motor i'm using has a staggering rotation speed of around 2500 rpm!!! its quite not a possibility to blink that fast :(.

Is there not any electrical method or any other mechanical method to do it??

Strobe, fluorescent, and some other types of lights actually blink or flash according to the frequency of their electrical supply. You might be able to apply some reflective strips to a rotating shaft, flange, etc.; and do some calculations from that.

Is the motor coupled to a mechanical device? Say like a gearbox, or jack-screw? If so you could figure the number of revolutions required to cause a range of motion, then measure the time it takes for that number of revs. The motor speed can then be calculated as an inverse proportion relationship to the measured time.

Thank your everyone for your answers!!.. it really helped me a lot!! :)

take care everyone!!

you could use a photointerupter

How about an optical/rotary encoder.

Rotary encoder - Wikipedia, the free encyclopedia

There is no free lunch, Some money and some effort is always needed.

You can make your own stroboscope. Buy a torch bulb and connect it to the output of an adjustable frequency pulse generator which can be made with couple of 555 timer ICs and and few more passive components, Pulse ON Time has to be significantly low in compression of off time.

You can have a good quality leaner potentiometer with graduation on its bezel to control the frequency of circuit,calibrate output frequency with graduation on bezel, Connect the bulb on output of circuit and focus it toward shaft, put a Piece of tape on motor shaft, adjust the frequency with pot and you will observe shaft in stationary position at the circuit frequency that is equal to the RPS of motor.

But don't forget to connect stroboscope to power.

Use a strobe light with a known blink rate that can be adjusted. Put a mark on the shaft, turn off the lights, turn on the motor and the strobe. Adjust the speed of the strobe until the mark appears to be stationary. Whatever speed the strobe light is running will be your RPM's of your motor. This is how cars used to be timed, and I've used this method many times for various projects.

But sir wouldnt it be difficult to observe whether the stroboscope is at the exact frequency as the shaft itself??.. like i told earlier, the motor is rotating at a speed of around 2500 RPM and varying only slightly due to the different loads attached to it.

Is it possible to get the exact rotating speeds of a motor rotating at such high speeds??

Absolutely, When your mark stops moving, you'll have your speed in sync. Of course with anything, there is a margin of error. As long as the strobe can blink that fast you should be ok.... the light will need to blink approx. 42 times per second... to a strobe with a 50-60hz capability will do the job... it's not an exact science, but you are looking for a cheap way to do it, there it is. Otherwise, you'll need to get a position sensor/counter to measure the revolutions.

2500 rpm is not that fast, aside from stroboscope and visual speed measurement or a reflective tape and optical sensor, you can use a magnetic pick up if you have a gear in your system. A 60 tooth gear will produce a frequency equal to rpm at the output of the magnetic pick up.

It depends on how Rube Goldberg you wanna get.

You could make a siren disk. You should be able to get somewhere just over 1 kHz when you blow through a tube onto the disk. Use tuning forks and beat frequency to find the exact speed.

Make a crude capstan. Take the motor to the roof of a building. Use the kind of slack line across the capstan that they use in the Navy. I can't think of the name but perhaps someone can. When you tug on the free end, the capstan begins winding up the other end. Do this for 20 seconds and measure the length of line.

Attach an LED to the shaft. Let the light hit a photodiode and measure that on a scope.

Mount the motor vertically. Fasten a strong wire and weight to the shaft. Put a spring scale/load cell in the line. Measure the force and back-calculate velocity.

OK, I'll stop.

Photo tach. Prices range widely, but there are some cheap ones, such as this: http://www.amazon.com/gp/product/B001N4QY66/

Strobe disc. This is an almost-free method, using a fluorescent lamp. Here's a pretty good description, with ready to print disc images -- http://blog.zensunni.org/faceting-information/measuring-the-speed-of-lap-rotation-using-strobe-discs/

Stroboscope. These can be expensive, but there are some schematics around if you are electronically inclined. http://www.google.com/search?q=stroboscope

An ordinary incandescant lightglobe "blinks" at mains frequency, here it's 50 times per second/ 300 times per minute. Divide that into 2500 RPM and you get 8.33. Mark a length of tape that goes around the motor shaft just once ( exactly ) into 25 equal lines. Wrap the tape around the shaft and shine a lightglobe onto it. If it appears stationary it's doing 2500RPM if it goes backwards it's slower etc.

Do the maths and put rows of lines or dots corresponding to different speeds alongside each other.

Tip! Check your line frequency.

pantaz is correct. you would need something like a xenon strobe, but the optical tachometer is by far the best solution and very accurate.

Sounds like a school assignment to me. But I appreciated help when I was in school, too!

You say you don't have a tach to directly measure the speed. But you don't say if you have all sorts of other test equipment. You also don't say if you need to know the average motor speed or need to tell the variation in motor speed, which means you need to be a bit more sophisticated.

How about measuring vibration? Hang a weight on the motor shaft and measure the frequency of vibration.

Gyroscopic precession? Get the rotating mass of the motor armature and then put it on a turntable where you can measure the gyroscopic forces to predict the rotational speed of the motor.

Put a propellor on the motor (water or air type) and measure the forces generated in the appropriate medium. From known characteristics of the prop, calc the speed.

Static electricity? Arrange the motor to drive a Van de Graaf static electricity generator and measure the static generated. Compare with static generated with other motors of known speed.

I like the suggestions by TPV45, too.

Jon.

THANKS for your help.

No, i do not have all the equipment as yet. And the budget is very small too. so i wanted to have some unorthodox way of looking into it which would be cheap.

I have to monitor the motor speed at different loading conditions. so this wouldnt be the 'average' of it.

Thank you for providing me with a number of options but by having a look into it, it seems a little more complicated than i can help.

And yes, i second your opinion that TPV45 provided some good ideas with the idea of LED and photodiode being the best.

You still haven't told us much detail. Is it fixed frequency? If you just want to know speed vs load at fixed frequency it would be interesting to know why this is important and how accurate you need it. What is the driven equipment?

Slip at rated load is only about 3% (full-load speed will be on the nameplate) so if you only need to know speed +/- say 5%, which is good enough for many purposes, you're already there!

Codey

GA. I like the vibration one. That would work well. I'm betting you could use the rotor current (perhaps back emf?) and run it into the sound card (via a good divider) on your computer, into some free FFT software and measure the frequency for free. If not, go to a yard sale and pick up an old record player so you can swipe the needle cartridge and use that as a pickup.

Or, find an old variable air capacitor, attach it to the shaft, run it as part of a resonant circuit and use that as an oscillator driver. Measure the frequency and BINGO!

Or a paddle scoop that picks up a ping pong ball and throws it. Measure the distance . .

You can do it with a bicycle and a stopwatch. Measure the circumference of the bicycle wheel and mark one spot with masking tape. Measure or calculate the circumference of the motor shaft or a pulley attached to it. Push the bicycle wheel against the shaft or pulley and count the number of rotations in some fixed time interval. Do the math, and you will have RPM's. Of course, this will only work if you can push the bicycle wheel against the rotating part without losing a hand or some other useful body part.

Alternatively, you can just buy a mechanical tachometer. The cheap ones sell for less than $100.

http://www.tooltopia.com/kd-tools-2967.aspx?utm_source=googlebase&utm_medium=cse&utm_term=KDT2967&utm_campaign=googlebase_18u

Great advice sir!.. this one seems to be the most appropriate!! plus the suggestion by TPV45!!...

Thanks a lot

and if contact or access to the end of the shaft is a problem do the optical

from #17

Photo tach. Prices range widely, but there are some cheap ones, such as this: http://www.amazon.com/gp/product/B001N4QY66/ is only $14.16USD

or if you like to build stuff Google optical tachometer arduino

my guess is the Bill Of Materials (BOM) is < $25 USD.

I used inexpensive bicycle speedometers to calibrate the tachometers on 2 old "Enterprize" direct reversible slow turning Diesel engines. The little wrist watch size indicators could be programed for different sized wheels. So it was simple to put reflective tape on the huge flywheels which were marked all 360 degrees around. My crew was half Philipino and half Irianian. We towed 2, 300" bardges and a 87' Harbor Tug, 14,000 miles. The Tow was almost 1/2 mile long. Louisiana- Persian Gulf, 6 months at sea, one stop 2 week stop in Nambia, Average speed almost 3 kts. The Bicycle speedo's were very important. I had the only conversion. With Ramadan and Christmas approaching everyone was in a hurry. Meaning more RPM. I ran the engines from 160 to as high as 190 RPM. So with my little screwdriver I could turn up the real Tachometers 5 or 10 RPM and everyone was happy. I wasn't straining the gear or burning as much fuel. 40 or 50 RPM was almost 1,000 gals a day difference. It was quite an adventure. My vote is for the bicycle speedometers and a little math.

I would be very afraid of a bicycle wheel spinning at 2500RPM. I believe they are only designed for a fraction of that, and surly would not be balanced enough to hold together at that speed... Danger!

If the circumference of the bicycle wheel is 100 x greater than the circumference of the shaft the bicycle wheel will turn at approximately 25 RPM.

The shaft will turn at 42 Revolutions per second if the motor is spinning at 2500RPM.

Figuring a 28" bicycle tire spinning at 42 revolutions per second:

• We are looking at the equivalent of having it run at approx 88" of linear travel for each revolution (pi x 28"dia).
• 88" x 42 revolutions = 3696" of linear travel per second
• That makes out to approx 308ft/second
• Which turns out to be approx 18,480 Feet per minute
• Which is 1108800 Feet/hour
• Which is 210MPH

I don't know about you, but a bicycle tire and rim running at 210MPH would make me very nervous considering they were built with a top speed of maybe 40MPH for a typical professional racer.
I would wager any unbalance in the rim/tire will let itself be known if the form of catastrophic failure... but I dunno... Are there any bicycle rims/tires rated to 200+ MPH?

Did it again... Misread the post. I get it now... your not talking about coupling the tire to the shaft, but rubbing the tire against the shaft. Well then.

_{<Sneaks off to corner of the room to insert foot into mouth>}

I agree but... I think the shaft is envisioned as only perhaps 1" dia and at 2500 rpm = 3.325m/s. If that drives a 28" bicycle wheel it is only spinning at ~90RPM ~1.5Hz.

Try this: Treysit Sirometer available from local Tecumseh or Briggs and Stratton dealer for around $20 .

Instructions:

http://www.treysit.com/treysit-sirometer.pdf

Very accurate, rugged, dependable.Immune to voltage spikes,transients,EMP's,voltage sags,surges, lightning strikes (except direct hits by lightning or thermonuclear devices), dropping,stepping on, (except by stampeding animals, such as Bovines or Pachiderms).Accuracy around 2%.However, the accuracy can be affected by earthquakes,aftershocks,sonic booms,very loud music, or shaky hands.

No joke, check it out.

Also,there used to be a tuning fork method that resembled a comb, with different length teeth that would resonate at an exact frequency.Anyone remember these, or am I telling my age?

I would attach a small belt pulley to the shaft and run a belt to the largest pulley I could find. You may be able to get the speed down enough to observe a timing mark on the large pulley. Then calculate the speed based on the ratio of the pulleys.