Ideas for Load Testing a Motor

First idea, couple the motor to a small generator with a variable resistance. As the resistance is decrease the generator produces more load that requires more power from the motor.

Second idea, couple the motor to a fan with a shut off damper. Start the motor with damper closed. Since the fan is not moving air, minimum load on the motor. Slowly open the damper and the increase in airflow will load up the motor.

Is load your desired measurement or torque.

Third idea. Connect the motor to some mechanical arrangement that can apply a variable brake to the shaft, large enough to dissipate the heat.

Torque and current are (almost) linearly related.

"Torque and current are (almost) linearly related"

Yea I know this...and why they want me to be able to measure both is beyond me. But this is pretty par for the course here.

"some mechanical arrangement" This is in essence my problem.

I like the brake Idea. Actually all of them are good ideas. But how can I automate this and be able to measure the torque being applied. I thought about some kind of torque wrench or meter of some sort. But I am not familar enough to even know what is available.

bill12780

Hello,

If you can afford it, get a dynamometer. If you want home-made, build a Prony brake.

Luck.

Find some small load cells to mount either under the load or the driver and multiply the force measurement by the moment arm for torque.

How long will the test be run? How many motors will be tested?

Is this for production motors or for engineering development?

Do some google or wikipedia searches for the terms you are not familiar with. I have a small bit of background with large motor testing, and I have seen DC generators and fan blades used to load an engines for testing and comparison. At EAA one year a vendor had a portable dynamometer cobbled together from scrap jet engine parts to test P-51 Allison engines up to 1000 HP.

Bottom line, there are multiple ways to solve your design requirement. Keep asking questions, researching, and some idea or method will arise that makes the most sense for your application. That's what makes design work fun.

Well, if you measure torque and current, you can generate a curve showing the relationship between the two. As PWSlack said, "Torque and current are (almost) linearly related."

Look up Dynamometer on wikipedia, or go to http://en.wikipedia.org/wiki/Dynamometer

Mount your motor in gimbals, attach a torque arm and a load cell.

For the "load" a dc variable speed drive will be able to provide the variable torque required, speed feedback measurement, and tie into a PLC such as AB ControLogix or other platform to automate the procedure.

It is not a complicated procedure, and once you have torque and speed you have power.

Complete the setup with good input power measurement of volts, amps, power factor, real and apparent power, harmonic analysis and you have a first class setup.

You could also include temperature feedback and thermography too see how hot things get, and verify cooling requirements.

inductance motors.[ Induction?]

What I need to do is be able slowly increase load on the motor to a predetermined toruqe and measure the overall(Single phase. So overall is redundant) current draw. It would also be nice if I could do this automatically with a ucontroller.{You mean controller?}

If you are Electronics kinda-sorry to disappoint you!

Ordinary idea is to do this:

Get hold of a heavy truck wheel+Brake Drum -complete with wheel braking Cylinder+piping.

Rig up to foot pedal+master Cylinder. You operate the foot pedal

Couple your 1HP motor to Wheel, nicely.-Take help from an Auto mechanic or a racer friend.

Run motor with free wheel as no-load. Current=I₀

Then in your brain+leg+foot Control Brake Light>>>>>>>>hard.

Holding your BRAIN+LEG+FOOT steady note various I_N=>>>

Show is over!

p.s. Why Large Truck Wheel?--"causeYou electronics man might burn the shoe liners.

I think the first writer had one of the best suggestions - out of which I like the generator idea.

Since you only talking about a 1hp motor a larger vehicle generator (500W+) could easily offer enough resistance to torque test a size of motor like that by varying its exiter current in its rotor and put a heavy load on the generator output (in test benches they used that generated power for charging backup batteries but you can just dissipate it in a similar low resistance load).

I do not remember the exiter current needed but I believe it is well below 5A that any microcontrolled PCB circuit can handle and no mechanical movements required.

The only thing is that a lot of generators may still have their regulator fit in them rather than being separate but if you know what to do it is not a problem anyway.

The fan idea is also a good one, as well as the break idea, but they all need mechanical intervention and I assume you want preferably electrical, do you?

Comon guys give him some more suggestion!

It is a bit more expensive, but if you use a Yaskawa servo motor and drive, you can apply a torque at a given speed to you motor under test.

In the drive you can program a speed and/or a torque. In this way it is possible lets say by a speed op 500 RPM, to change the torque from -20 Nm (breaking the motor under test) to + 20Nm (accelerating the motor under test).

Good luck

I also agree with the generator idea. You might load the generator with light bulbs that the uController can switch on with relays or triacs.

A small motorcycle or scooter brake, heck, use the front fork as the lever and a spring balance to read load.

Ok you did say 'automate'.

Simply use a magnetic tape strip in the output end of the shaft and one on the free end of shaft. attach a pick-up strip so when the strips pass the pickups, they read on a readout or ocilliscope. Use a torque wrench and measure the deflection of the shaft at each end in relation to each other. Then you will be able to measure the torque on the shaft as it rotates. Apply a load with a friction device or generator and measure the amps and torque simultaneously.

Don Kingery

If some level of accuracy (said better than 5%) is required, your testing station should have ability of calibration.

To calibrate any of "breaking type motor load" you may use one of well known (Its current I as a function of a torque T - e.g. data from its manufacturer) standard motor in the range of the motors needed to be tested. So:

1. Build your mechanical or electromechanical (generator + rheostat) load

2. Find the standard motor. Be sure have its I = f(T) drawn or in a table.

3. Couple the standard motor and the load, run reverse test/calibration, i.e. draw LoadTorque = function of (StandardMotorCurrent). Be sure use constant (or regulated) applied voltage, as well note the test ambient temperature.

4. Mark settings for required test measuring points on the load.

5. Prepare easy couplings to MUT (Motors Under Test).

This should create you testing stations. In Instrumentation & Process Control practices there are used 5-point, 3-point fast methods for these type of tests. Of course you may use your own design of the procedures.

PLC could be your controller. AS a good electronics designer you may build your own controller using analog as well as digital circuits with chips, also programmable ones.

Use a prony brake. Google it or try wikipedia for ideas on how to construct it. It is very simple to build, very inexpensive and the only device you might have to purchase is a spring scale with the proper load range.

I just did a bunch of research on this. I assume you are looking for overall efficiency. Whatever the case, load the motor with something..pump...fan...brake, but couple it with a torque meter. They are extremely accurate (.1%), and not very expensive. I found PCB to have a great selection and competitively priced. Just google PCB torque transducer.

Look into Powermag's eddy current adjustable dynamometer -- designed for testing motors from 0 to 100% torque load.

What about a V.F.D.

Pump

Valve

Two pressure meters

Flow meter

The water will get hot though, so maybe a radiator and fan.

maybe add two temperature sensors for better accuracy.

Advantages:

Things wont burn

No fumes

E(dP,mF,dT,Cp)