Regen VFD as a load bank for testing

[jraef]

I am consulting for a new start-up company who makes a small alternative energy generation system (hey, aren't we all?) that uses a 25kW SCIM as the generator. For acceptance testing, they have been using a set of halogen lamps to load up the generator, staging them on and off by wattage as they put the system through its paces. The test is actually of the regulation system that is controlling the prime mover, so they want to be able to fully load the generator and shed load very quickly to determine the step change response capability.

I proposed they consider using a Regen VFD on the 25kW motor instead, with a plan to convert the "load" command to a "brake" command via a PID loop. For example they tell the PLC to "Load to 50%", which has the PLC send a commanded speed signal to the VFD that scales lower and lower until the regen output is 12.5kW, then holds there.

Anyone done something like this before? Anything I'm missing here? It just seems too simplistic!


"If I had eight hours to chop down a tree, I'd spend six sharpening my axe." -- Abraham Lincoln
For the best use of Eng-Tips, please click here -> faq731-376

 

[waross]

I would search the site, Jeff. I seem to remember some discussion of using regen VFDs for dyno applications. The details escape me but there may be some pertinent information.
This sounds like an interesting challenge. I hope that you are able to keep us posted on the progress.

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

 

[davidbeach]

But you have to do it while holding a pretty tight 60Hz; otherwise the anti-islanding detection will kick in and trip it off.

 

[waross]

For steady state loading the VFD should be great. For checking the throttle response to step load changes it may not do so well. With about 3% motor slip between no load speed and full load speed and 3% droop on the control scheme, you may find that the VFD has problems simulating real world loading. Your loading may drop off very fast as the prime mover speed is pulled down by the step loading. This would not be significant with resistive loading.

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

 

[itsmoked]

For sudden loading and shedding the VFD seems counter productive. Those lamps are much better at that. They even allow heavy hits as the tungsten heats up.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[DickDV]

As long as you have a good fast processor, you should be ok even with step changes. I've done several like this absorbing energy from hydraulic actuators (oil motors). All I've ever used is ABB ACS600's and 800's but the results were quite satisfactory.

Some of these actuators were running 12-14000rpm, too.

 

[hesselberth]

see attached

 

[jraef]

No attachment...


"If I had eight hours to chop down a tree, I'd spend six sharpening my axe." -- Abraham Lincoln
For the best use of Eng-Tips, please click here -> faq731-376

 

[hesselberth]

sorry forgot to add

 

[waross]

There is no question of the advantages of a regenerative drive for most dyno applications.
However this is a special case.
As I understand it, the transient response of the control gear of the prime mover is being tested. In such case the transient response of the VFD becomes part of the equation. The transient response of the VFD may interact with the control scheme under test in an unknown and non linear manner. This may render the test result meaningless. Much expense for nothing, and for transient testing very little returned energy. By all means use the regenerative drive for most of the testing, but go with a load bank to test the transient response of the prime mover control scheme.

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

 

[Skogsgurra]

My experience is that the torque loop in all VFDs that I have worked with is faster than 1 millisecond. It is also stable and well damped.

Compared to that, the AVR or speed controller or just about anything that needs to be tested is slow. So, I wouldn't hesitate at all. VFDs are good for all torque, load, speed testing that I have seen.

There is one case where water brakes or lamp banks are needed and that is when your grid isn't strong enough to 'swallow' the power your prime mover delivers. But, at 25 kW, that is probably not a problem.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[jraef]

Excellent info, thanks everyone.

When I mentioned the possibility of using a regen VFD it piqued their interest and although I have done this for motor test stands, I was not 100% sure on the issue of doing it for an induction generator. I know it's not much of a difference really, it's just my being overly cautious with something new. You have reinforced my confidence that the differences and limitations are no more than I thought they might be.

Bill, although I'm not comfortable revealing the exact nature of the system (my customer may be reading this), there is an inherent lag in the step change capability anyway.

May the power of the Forum be with you all!


"If I had eight hours to chop down a tree, I'd spend six sharpening my axe." -- Abraham Lincoln
For the best use of Eng-Tips, please click here -> faq731-376