Electric motor rotor balancer?

[DennisCNC]

Any one here have experience with motor rotor balancing machines? I would like to know some basics how they work, what is used to sense the vibration (piezo material or other) does the machine tell you how much it is out of balance or is it guess work?

If you know of a site that has more details please share the link!

Thanks

[mc-motorsports]

http://www.schenck-ind.com/balance1.html
I was told Schench is the industy leader, I know for a fact they are expensive. They use "vibration velocity transducers" or accelorometers. (I know I didn't spell that right) Initially its guess work, how the machine knows where the weight needs to be added or removed I have no clue, after adding or subtracting test weight, the machine is dead on. And you just use magnetic sensors and a laser to track RPM by drawing a white or black line on the shaft or rotating assembly and make sure they the laser and sensors are 90* apart and the balancing equiptment is conversational.

[neilw20]

If you can feel it vibrating, you already have a sensor. Your hand.

Better still to get some measure of the amount of un-balance is to set up a dial indicator against the motor in the radial plane where you wish to measure and see how much the pointer wobbles ,as long as the motor is not really hard mounted.

Given that they are close together, 2 worm drive hose clips with the heavy parts 180 degrees opposed mounted on the shaft should give no correction.
Use felt tip pen to mark a line on the shaft at 90 degrees to the big parts of the clamps, and 2 more at 120 degrees apart in a different color.
Now move the clamps around the shaft towards the mark. You have 'added' weight near the the line. Put a mark across both the clips so you can duplicate their angular position EXACTLY.
Run the motor and see how much it vibrates. Note how much wobble.
Move both the weights 120 degrees around to next color. Run it again. How much?
Move both the weights around 120 degrees to next color. Run it again. How much?
Now you might have some idea in which 120 degrees the weight needs to be added.
On a circular graph you can draw a vector from the center for each reading
at 120 degrees.
Now join the ends of the triangles and find the center of the triangle.
This gives the radial direction from the center of the relative amount of un-balance, and should give some idea of the angular position required to add weight. Now mark the heavy/light spot and fiddle with the clips to correct for the vibration.

Now you have 'added' weigh to fix the problem.
On the opposite side to where you added the weight you can remove weight by drilling/grinding to remove weight.
Remove the clips, start drilling and measuring and you will balance the motor. This will only be correct for this end of the motor. The other end needs doing too, and this is dynamic balancing. The correction at the other end may be totally different and at a different angle.

The 3 weight method is good because you can't screw up a triangle and you must get a vector result. If you want to practice before fixing it for real put an old pulley or piece of round stock with a hole on the end of the motor and practice on the stock. Even a piece of hardwood will work.

If you get inconsistent angular results and keep chasing your tail do each of the 3 measurements both clockwise and CCW, then draw the triangles for each direction, and average the resultant vector. Sometimes phase errors like this can result from the mountings interfering with the wobble measurement. Make sure the dial indicator stays in the same place for each of the test runs.

When getting close to balanced you will probably detect that the amount all of a sudden moves 90 degrees around the shaft. This is because the position you have selected to correct is off by a few degrees. At small un-balance amounts the angular position becomes quite sensitive.

I have balance tail shafts on cars using a similar method, but just using a piece of chalk to mark the heavy spot. Make sure the car is mounted safely!

Any comments...?

Neil

[vger]

Dennis

I balance rotating equipment as a business. Neil has made very good comments! One of the oldest names in balancing machines is IRD, later called Entec-IRD. I still have an IRD 350 analyzer balancer that is used very rarely when I have a large very low rpm (under 150 rpm) job that is critical.

Machines for balancing motor rotors are fairly simple. One such machine called a "soft bearing" consists of a variable speed drive motor mounted low to the floor with a crowned pully. A belt runs from that to a swing-arm mounted, weighted, crowned idler roller under the balancing sling. The belt then runs up and over the armature and returns to a second fixed idler and back to the drive motor. This allows the belt to be lifted up and over the end of the rotor shaft and placed centered on the armature. The balancing sling consists of two shaft supports that are slung hung so they can move easily. A pair of bearings are set in the swinging part to support the shaft. A sensor is attached to the swinging part to measure the horizontal displacement. See crude attached drawing of one end of sling.

The electronics part is a bit more complicated. Most motors are balanced in at least 2 planes such that each end of the motor is balanced. With both ends of a motor out of balance at different vectors, you can balance one end and smooth it out, but when you balance the other end, the first end will be out again. There are programs for doing 2 (or more) plane balancing such as one by IOTECH called eZ-Balance. (30 day free trial) http://www.iotech.com/catalog/daq/ezbalance.html

IOTECH has data aquistition hardware to connect to various pickups.

Accelerometers by wilcoxon at:
http://www.wilcoxon.com/vi_index.cfm?CatM_ID=1

Accelerometers by Vibrametrics at:
http://www.vibrametrics.com/index.as...lerometers.htm

Does the machine tell you how much it is out of balance?
Yes and no. They can tell you to a great deal of accuracy either the displacement, velocity, or acceleration that is being measured. To know how much weight must be added or removed, and where, requires at least one trial run with a known amount of "trial weight". From the amount of change with the trial weight added, and knowing the "phase" of the unbalance, one can calculate the balance weight and location.

The smallest trial weight I have used was 0.5 grams, the largest was about 25 pounds. The size, weight, and speed of the rotor all come into play when determining an appropriate trial weight. Remember, Inertia (cenfrifugal forces) can be your friend or a deadly enemy!

Give me an private message if you like.

Hope I've helped a bit
Steve

[vger]

Neil,
I use those hose clamps quite often when balancing drive shafts on cooling towers. The ones I use are stainless steel (cooling towers are really nasty corrosive due to water treatment) and a cut piece of stainless as a final balance wieght clamped to the shaft. I use the 4 run method for balancing all the time. I use the vector chart with circles method. Although I have a computer program for doing the drawing and calculating it can be done on paper with a compass. If you are interested in the details drop me a private message.

Steve

[DennisCNC]

Thank you both for the tips, I will think over the info and let you know if I have more questions.

[vulcom1]

Dennis,
Not sure how large a balancer you need but I have been working on a home built one for my small turbine. I have the schematic and also the other instructions that I downloaded a while back. It is not very big but will do the job for me when working. If you are interested let me know and I will send or place the files on here.
John

[kulk_sd]

Originally Posted by neilw20 If you can feel it vibrating, you already have a sensor. Your hand. Better still to get some measure of the amount of un-balance is to set up a dial indicator against the motor in the radial plane where you wish to measure and see how much the pointer wobbles ,as long as the motor is not really hard mounted. Given that they are close together, 2 worm drive hose clips with the heavy parts 180 degrees opposed mounted on the shaft should give no correction. Use felt tip pen to mark a line on the shaft at 90 degrees to the big parts of the clamps, and 2 more at 120 degrees apart in a different color. Now move the clamps around the shaft towards the mark. You have 'added' weight near the the line. Put a mark across both the clips so you can duplicate their angular position EXACTLY. Run the motor and see how much it vibrates. Note how much wobble. Move both the weights 120 degrees around to next color. Run it again. How much? Move both the weights around 120 degrees to next color. Run it again. How much? Now you might have some idea in which 120 degrees the weight needs to be added. On a circular graph you can draw a vector from the center for each reading at 120 degrees. Now join the ends of the triangles and find the center of the triangle. This gives the radial direction from the center of the relative amount of un-balance, and should give some idea of the angular position required to add weight. Now mark the heavy/light spot and fiddle with the clips to correct for the vibration. Now you have 'added' weigh to fix the problem. On the opposite side to where you added the weight you can remove weight by drilling/grinding to remove weight. Remove the clips, start drilling and measuring and you will balance the motor. This will only be correct for this end of the motor. The other end needs doing too, and this is dynamic balancing. The correction at the other end may be totally different and at a different angle. The 3 weight method is good because you can't screw up a triangle and you must get a vector result. If you want to practice before fixing it for real put an old pulley or piece of round stock with a hole on the end of the motor and practice on the stock. Even a piece of hardwood will work. If you get inconsistent angular results and keep chasing your tail do each of the 3 measurements both clockwise and CCW, then draw the triangles for each direction, and average the resultant vector. Sometimes phase errors like this can result from the mountings interfering with the wobble measurement. Make sure the dial indicator stays in the same place for each of the test runs. When getting close to balanced you will probably detect that the amount all of a sudden moves 90 degrees around the shaft. This is because the position you have selected to correct is off by a few degrees. At small un-balance amounts the angular position becomes quite sensitive. I have balance tail shafts on cars using a similar method, but just using a piece of chalk to mark the heavy spot. Make sure the car is mounted safely! Any comments...? Neil

Please let us know more on instrumentation part of balancing machines

[kulk_sd]

Please let us know more on the instrumentation part of the balancing machines. So that we can run some experiments on it.

[Herbertkabi]

Soundcard Based Dynamic Balancing Machine
by
Miklós T. KONCZ, HUNGARY, (kmiklos@vnet.hu)

There you can understand the principle,
and build it,
and later to go to accelerometers,
then you need to buy a new soft, costs few kilos,
cheers,
herbert

[RomanLini]

Buy a 3 axis accelerometer and a datalogger, then plug it into your PC.