Author Topic: Test load for motor controller  (Read 1562 times)

0 Members and 1 Guest are viewing this topic.

Offline tobiTopic starter

  • Newbie
  • Posts: 2
  • Country: de
Test load for motor controller
« on: July 09, 2019, 01:53:14 pm »
Hello,

I have motor controller design. I have to write software and during this time I don't want to have a real motor standing next to me (36V/40A/2mH). So my idea is to use a passive RL load.

My question:
- What kind of inductor do I have to use (air coil, ferrit)?
- Do I have to be careful with something like staturation, L over I characterisics?
- Where can I get an inductor like this?

Thanks.
 

Offline MagicSmoker

  • Super Contributor
  • ***
  • Posts: 1408
  • Country: us
Re: Test load for motor controller
« Reply #1 on: July 09, 2019, 02:18:04 pm »
I have motor controller design. I have to write software and during this time I don't want to have a real motor standing next to me (36V/40A/2mH). So my idea is to use a passive RL load.

My question:
- What kind of inductor do I have to use (air coil, ferrit)?
- Do I have to be careful with something like staturation, L over I characterisics?
- Where can I get an inductor like this?

Real motors experience saturation to some extent during normal operation so a gapped core inductor would be best in that regard, not to mention you would need several hundred turns to get to 2mH with an air-core solenoid (even more for an air-core toroid). However, it's going to require a really large core to handle 1.6J of energy storage... Also, an RL load basically simulates a motor when it is not spinning - no back EMF, of course - which means your motor controller is going to be operating over a very limited duty cycle range.

You will almost certainly have to wind this yourself, because a 2mH/40A inductor is not an off-the-shelf item.

As for making it an air-core, that's going to take a few hundred turns... like 400+. Yikes.

EDIT - finished sentence that I left dangling earlier.
« Last Edit: July 09, 2019, 03:11:09 pm by MagicSmoker »
 

Offline capt bullshot

  • Super Contributor
  • ***
  • Posts: 3033
  • Country: de
    • Mostly useless stuff, but nice to have: wunderkis.de
Re: Test load for motor controller
« Reply #2 on: July 09, 2019, 02:30:23 pm »
You will almost certainly have to wind this yourself, because a 2mH/40A inductor is not an off-the-shelf item.

Not a big deal, (at work) we use such stuff on a regular basis, and one can buy them from transformer manufacturers (the typical three-leg inductor / transformer for three-phase current).
Normally these things are used as an additional feeding reactor to somewhat larger VFDs, but they also nicely double as dummy loads for the VFD. One can push the nominal output current into these without burning all the real power into resistors, usually this is sufficient for load testing / thermal cycle testing of the VFDs, but not for testing the control algorithm. You want a real motor here (for SW development purposes).
Safety devices hinder evolution
 

Offline MagicSmoker

  • Super Contributor
  • ***
  • Posts: 1408
  • Country: us
Re: Test load for motor controller
« Reply #3 on: July 09, 2019, 03:18:41 pm »
You will almost certainly have to wind this yourself, because a 2mH/40A inductor is not an off-the-shelf item.

Not a big deal, (at work) we use such stuff on a regular basis, and one can buy them from transformer manufacturers (the typical three-leg inductor / transformer for three-phase current).

Yes, good point - there just might be a line/load reactor available off-the-shelf that meets the OP's requirements. Not the usual 3%/5% impedance types, though, as they are only available in discrete amounts of inductance inversely proportional to current rating.

 

Offline patrick1

  • Contributor
  • Posts: 49
  • Country: au
Re: Test load for motor controller
« Reply #4 on: July 09, 2019, 04:43:06 pm »
with respect, - go too the local scrap metal guys,  and stick a motor between your legs.   hold on tight @@
 

Offline H.O

  • Frequent Contributor
  • **
  • Posts: 864
  • Country: se
Re: Test load for motor controller
« Reply #5 on: July 09, 2019, 05:45:47 pm »
Old SCR-based servo drives used to use large inductors in series with the motor to limit dI/dt. For example, the old Indramat drive I ripped out of my milling machine had 5mH, 22A rated inductors fitted. Shouldn't be TOO hard to find something used and if you have the budget I'm sure just about any transformer manufacturer will make one for you.
 

Offline max_torque

  • Super Contributor
  • ***
  • Posts: 1327
  • Country: gb
    • bitdynamics
Re: Test load for motor controller
« Reply #6 on: July 09, 2019, 08:41:53 pm »
Get a cheap second hand (large) car alternator, and rip off the diode pack and field driver and you can usually get access to the 3 phases of the stator, which makes a perfect dummy 3ph load for no money at all.  If you connect the rotor's field windings to a bench supply, rather than use the voltage controller, you can make the motor turn when you need to as well!



 
The following users thanked this post: JPortici

Offline jbb

  • Super Contributor
  • ***
  • Posts: 1265
  • Country: nz
Re: Test load for motor controller
« Reply #7 on: July 09, 2019, 08:50:56 pm »
One thing to be aware of: the LR load is in many ways quite different to a motor load.  Obviously you'd start with inductance equal to the motor inductance.  But the resistance is tricky...

For equivalent power, you can use a larger value resistor to dissipate the appropriate power (V*I).  However, this arrangement has a high R to L ratio and will therefore be well damped and well behaved.  It's OK for power testing and very crude current loop testing.  It offers no Back-EMF (BEMF) signal.

To get realistic current dynamics, the L and R values should roughly equal the motor windings. This might well be covered by the inductors' internal resistance.  This arrangement should have a more realistic current dynamic, but will be totally wrong in terms of average voltage and power dissipation.  It offers no back EMF.  On the plus side, it won't dissipate a lot of power.

If you have money to spend, you can get power simulator electronics systems which have a power converter (looks much like your motor driver) with special controls that pretend to be a motor.  This will offer pretty good current dynamics, pretty fair power flows, some kind of back EMF waveform and some nice features like to ability to model a mechanical load.

The real acid test is of course to use a real motor connected to some kind of dynamometer (traditionally a separately-excited DC machine).  This offers the true behaviour: saturation, back EMF, resistance change with temperature etc.  It's a bit irksome to set up (needs stiff base, mechanics, guards, some kind of load controller for the dynamometer etc.) but can be very effective.
 

Offline tobiTopic starter

  • Newbie
  • Posts: 2
  • Country: de
Re: Test load for motor controller
« Reply #8 on: July 10, 2019, 06:40:36 am »

Not a big deal, (at work) we use such stuff on a regular basis, and one can buy them from transformer manufacturers (the typical three-leg inductor / transformer for three-phase current).
Normally these things are used as an additional feeding reactor to somewhat larger VFDs, but they also nicely double as dummy loads for the VFD.

Thanks for your suggestions.

Could a use Schaffner RD8137-36-1M5 + resistor (cable)?

Inductance is not quite right, but in principal?
 

Offline JPortici

  • Super Contributor
  • ***
  • Posts: 3573
  • Country: it
Re: Test load for motor controller
« Reply #9 on: July 10, 2019, 07:11:21 am »
Get a cheap second hand (large) car alternator, and rip off the diode pack and field driver and you can usually get access to the 3 phases of the stator, which makes a perfect dummy 3ph load for no money at all.  If you connect the rotor's field windings to a bench supply, rather than use the voltage controller, you can make the motor turn when you need to as well!





That's exactly what i did :)
 

Offline capt bullshot

  • Super Contributor
  • ***
  • Posts: 3033
  • Country: de
    • Mostly useless stuff, but nice to have: wunderkis.de
Re: Test load for motor controller
« Reply #10 on: July 10, 2019, 07:14:59 am »
Could a use Schaffner RD8137-36-1M5 + resistor (cable)?

No, since this one is a common mode choke. It will saturate at a very low non-common mode current.
Look for something like that: https://www.reo.de/de/produkte/drosseln/prufdrosseln.html
or https://www.reo.de/de/produkte/drosseln/motordrosseln.html
or https://www.reo.de/de/produkte/drosseln/netzdrosseln.html
(no recommendation, just the first seach hit for the term "Dreischenkeldrossel", there are several other makes)
Safety devices hinder evolution
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf