Test motors under load 5

[EngRepair]

In our rewind shop we want to test repaired motors (up to 60 HP) under load.
It seems that such dynamo-meters from various manufacturers are extremely expensive and we would try to do something ourselves using brakes or generator or motor etc...
Is there any idea how to do it?
Probably, someone has already done it.
Any help will be valuable.

 

[waross]

Use an induction motor and a regenerative VFD.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[EngRepair]

Hi, Bill,
I'm not familiar enough with the VFD, can you explain in more detail?

 

[itsmoked]

A VFD or Variable Frequency Drive is a tool for controlling the speed of an AC induction motor. They are used widely in industry, so widely they have reached commodity status and so you can get spectacular functionality for your money with them.

You should google for VFD.

There is an elite version of VFDs that can return energy to the power supply from an over driven motor. You can use one of these and a motor to load another motor for testing.

After you have looked up VFDs and educated yourself on them a bit come back and check these threads relating to using VFDs for test setups. Don't bother looking at them until you understand VFDs a little better though.

thread237-344979
thread237-173814
thread237-137071
thread237-261800
thread237-169558



Keith Cress
kcress -

http://www.flaminsystems.com

 

[LionelHutz]

The cheapest, assuming you have some old "junk" motors lying around, might be using DC injection braking. If you wanted to do a bit of a science project then get an old motor, probably around 150hp or so, and re-wind or re-configure it to be a single winding to take DC. Then, feed it with a phase controlled rectifier. Otherwise, you could just use a bigger motor, likely around 250hp or so, and just feed it DC and it would work as a brake capable of continuous operation. A smaller motor would also work, but your tests would be time limited due to the DC current level being applied to the load motor.

We used to use a 25hp motor to load a 10hp motor for testing. We basically use one of our soft-starters with the SCR connections reconfigured to create a current controlled rectifier.

We have since switched to using VFD's to load the motor since we now have larger test motors. I was hoping that the VFD would create a load capable of acting at a constant torque but that never worked well. So, I set them up so the VFD is energized at 60Hz once the motor is running and then a pot lowers the VFD operating frequency which puts a load the other motor.

 

[EngRepair]

Hi, LionelHutz,
Thanks for your suggestion.
That was my first thought because it seems to me, that's the simplest.
I have some experiance in use of DC current for braking of AC motors.
Can you say something about the calibration ie. how to determine braking torque/power knowing the voltage and current of the DC supply?

 

[LionelHutz]

I doubt you'd find a simple way to correlate the DC current to a braking torque. If you need a good accurate torque and power measurement, it'd be best to instrument the load and measure the values directly. Use a load cell x distance to get torque and a toothed wheel and pickup for rpm.

I suppose if you instrumented it then you might be able to correlate the DC current to a torque but once you setup the instrumentation you might as well just keep using it.

 

[iop95]

A DC generator with a variable resistive load may be a solution; with some aproximation may get/calculate torque from field and load current. Resistive load heat may be good in winter...

 

[mikekilroy]

If you use another motor on a vfd as the load, simply wire its dc bus to the drive running the motor and circulate the current; then you need no fancy regen; your 60hp test stand will pull around 2 amps from the wall to supply the losses only (the "load" generator motor will supply the juice for the motor under test). We do it every few years with customers.

http://www.KilroyWasHere<dot>com

 

[waross]

The choice would then be either two "Plain Jane" VFDs or one expensive VFD with regeneration capability.
How do the prices of basic VFDs compare with Regen VFDs?
Either way, if you are going to do many hours of testing, the energy savings will add up in your favor.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[mikekilroy]

Yep, and ideally, buy really capable vfd's, not super generic if you want lots of potential capability.

We did a couple high performance 200KW vfd's with high speed fiber optic link between for A Nasa shuttle project once; we could sync down to the PWM pulses between the two drives. The motor was actually a redundant motor being proposed for the hydraulics pump on the shuttle; pancake style: 180 degrees was one complete motor, other 180 degrees the 2nd one for backup. we could use one as a motor, other as generator load to fully test the whole capabilities. We even synced the PWM pulses to not have any extra shaking or eddy current losses or anything. These same drives are only about 30% more expensive than generic ones, so the moral is buy good like Waross suggested.

Keeping in mind that the motor under test will be taking MORE power than the generator puts back on the DC link, there is no need for regen drive - just a normal old DB circuit - in only one of the 2 drives - for stopping fast.

http://www.KilroyWasHere<dot>com

 

[itsmoked]

Excellent point Mike! I've now seen that work too.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[LionelHutz]

A regen drive will also recirculate the energy. In fact, if the motor under test is operated across the line the system would likely be more efficient and even save more energy than using 2 VFD's with the DC buses connected.

The question is do you want to be locked to running the motor under test on a VFD or not? We setup our regen drives as loads so we can apply a load to whatever is being tested.

 

[Woodyjumps]

Short of having a dyno:

Best way by far is the regen VFD.
For it to be accurate you MUST perform the autotuning function, otherwised you will not be able to trust the torque and power values.

The poor man's way, but still very accurate:

Build a pony brake. (Search "pony brake dynometer" on google, and look at the images.)
Long arm style, use ~ 20" flat or multiple grooved pulley with an assortment of tapper lock bushings.
Oil soaked maple blocks for the friction parts.
Use a digital scale / load sensor.
Use a prox sensor on the shaft or pulley for speed pulse.
Get data aquisition hardware for your laptop from DATAQ.com so you can trend rpm, amperage, and the scale output all at the same time.
Viola - You have a very accurate dynometer for about 10% of the cost of a 60hp regenative VFD. The only caveot is that you have to manually tighten the screws.

 

[LionelHutz]

45kW is a lot of power to be dissipating into some wooden blocks. I can't see that braking for very long before you get smoke and fire.