What's new
What's new

Load Reactor / Filter Questions

ptsmith

Cast Iron
Joined
Feb 15, 2018
I have a 10A 3% impedance TCI line reactor on the output side of my VFD to try to reduce the motor noise mentioned in the thread linked below. It makes it tolerable but I'd like to reduce it more to protect the motor.

I'd like to add capacitors and possibly resistors. I've found info on how to wire them but nothing on what values to use.

And some load filters have just caps and others have caps and resistors. Any advice on what would be needed for my application? I have a 240V single phase in 3HP VFD outputting 240V 3 phase.

Thanks!

Link: https://www.practicalmachinist.com/vb/transformers-phase-converters-and-vfd/fuji-frenic-mini-setup-359036/
 
The usual plan is to put the cutoff of the filter below the annoying noise frequency. The noise itself does not damage a properly made motor, although some motors may fail from wire flexing fatigue if they are not tied and varnish dipped correctly. That may not always be due to the VFD, or the noise.

The resistors associated with capacitors may be for discharge, OR they may be low value larger resistors used to lower the "Q" of the filter, and prevent large currents from flowing at the resonant frequency.

Filter design is a specialty. That does not mean it cannot be done, just that there are all sorts of tricks and gotchas that you want to look out for.

If you put in just the reactor, I do not think damaging spikes are likely, and I would let it alone, actually. You are asking to make a sine filter, essentially. In that case, you want the filter frequency to be between the highest sine output frequency and the lowest switching frequency.

That is going to depend on the way the VFD operates.... if it goes to plain square wave modulation at any point, and other details.

Another consideration is that the filter will reduce output power somewhat. The reactor will do that by itself.
 
you don't need resistors, just add delta connected capacitors after the reactor.

10uf per hp at 4Khz or higher frequency is more than enough for a 5% line load reactor at 240v (assuming the line/load reactor is matched for the vfd and the motor hp)

in your case you have a 3% reactor so i would set the pwm to at least 6Khz and verify the reactor doesn't heat up after the capacitors are added.

Some of the "line" reactors have thicker laminations and may overheat if used as a load reactor. the "line/load" reactors are rated for both.

You can use Y connected capacitors, but if you do so, you cannot ground the "neutral" point of the capacitors, if you use an oscope to measure the voltage at that point, you'll see a square wave at the pwm frequency of the vfd (say 4Khz) at +170, -170 vdc. it is this voltage that kills the bearings in the motors. It can't be filtered out with a common 3 leg line/load reactor.
 
The only difference between a "line" reactor and a "load" reactor is the current rating. Load current on a VFD is higher than line current. So you can ALWAYS put a "load" reactor on the line side, but not vice versa.

Reactor + caps and resistors (RC snubber) = dV/dt filter. You can buy them already made.
 
Thanks for the responses. You guys are too helpful. That's more answers than I can keep up with. Let me add to this and you can correct me where needed.

I've been trying to understand this for quite a while and did figure out a few things.

Dv/dt filters is were I got the idea to add caps and resistors. Now that I think about it, the only load reactor I saw with caps-only was a drawing. So I guess that's not the norm.

Looking at dv/dt filters on ebay, you can often read the model of the reactor. They all say "line reactor" on them. I've looked up many of them and as far as I can tell they are standard line reactors. Another thing I noticed is that they are always low impedance, 1.5% @480V. And I just found this. It appears *all* TCI load reactors are 1.5%: https://transcoil.com/wp-content/uploads/2018/09/Reactor-Filter-Size-Chart-29995.pdf

That chart list different model numbers for line and load reactors, which seems to confirm what Jraef said. Galco list line reactors as "Use on drive input or output" so that's confusing.

I think I should just buy a dv/dt filter as I'm getting more and more confused. So last question:

This 12A TCI is rated 10HP at 600V and 7.5HP at 480V. I read on Galco somewhere (although I can't find it now) that would rate it at 3HP at 240V, what I'm running. Does that sound right? V1K12A00 - TCI - Motor Protection | Galco Industrial Electronics

PS: I'm mildly dyslexic and a long post like that with a lot of google searches involved requires many rewrites. It was a major ass whipping and I'm sure it's still all screwed up!
 
If your primary desire is to reduce audible noise, you are headed in the right direction and since the filters are rated based on current, then holding current constant, as your drive / motor voltage ratings go down so does the power rating. So not at all surprising that a 12A rated unit would be adequate for a 10hp motor @ 600V and only 3hp at 240V.

Now if you have increased the switching frequency and the audible noise is now tolerable, and now your primary concern is protection of the windings . . . you probably have nothing to worry about unless it is a very old motor as most motors manufactured in the last few decades can operate fine on a 230V VFD.
 
The only difference between a "line" reactor and a "load" reactor is the current rating. Load current on a VFD is higher than line current. So you can ALWAYS put a "load" reactor on the line side, but not vice versa.
......


Hmmmm... You KNOW the maximum current on the load side, the VFD has a current limit. So if the reactor is good for 2x (200%) of full load current, you should be good. Many VFDs are rated for a short term 200% current.

But on the input side, you have the charging current for the capacitors in the VFD, which is unlimited during normal operation, aside from the mains impedance, and the effect of the reactor. So I'd have said it was generally the opposite.... A line side reactor would generally work for the load side, but the reverse might not be true.....
 
If your primary desire is to reduce audible noise, you are headed in the right direction and since the filters are rated based on current, then holding current constant, as your drive / motor voltage ratings go down so does the power rating. So not at all surprising that a 12A rated unit would be adequate for a 10hp motor @ 600V and only 3hp at 240V.

Now if you have increased the switching frequency and the audible noise is now tolerable, and now your primary concern is protection of the windings . . . you probably have nothing to worry about unless it is a very old motor as most motors manufactured in the last few decades can operate fine on a 230V VFD.
I have a Lagun FTV-2 mill which is circa '80s? I'm not sure of it's age. I believe the motor insulation is class B. The VFD carrier frequency is set to 16kHz.

The reactor made the noise tolerable but I'd like to protect the motor as much as possible. I should be able to find a used TCI load filter on ebay for ~ $50, maybe less. It's surprising how low the resale value of these is.
 
I set up a similar Lagun almost 10 years ago and still running it every day on a 480V VFD . . . no problems so far and no load reactor either.

Shared album - Ken Brown - Google Photos

I put a prox on the spindle gear and display spindle speed in RPM . . . doesn’t matter which set of pulleys are being used as it reads the spindle speed directly. Has worked great for nearly a decade.
 
I like your control layout. I may copy it.

It looks like you changed directions without stopping first. I thought you couldn't do that with a VFD.
 
I like your control layout. I may copy it.

It looks like you changed directions without stopping first. I thought you couldn't do that with a VFD.

ou can, the VFD will slow to stop, then reverse, in general. It is not a "plug reverse" situation, which would overcurrent the VFD in most cases.
Y
 
OK, thanks. I actually saw a youtube video where the guy was having trouble doing that. Wanted to power tap. Ask if anyone knew of a way. He got several comments, but no solution.

So that's why I thought you didn't want to do that. I'll give it a try when I get home.
 
I just paid $30 shipped for this. Pretty grungy. Hopefully it just needs a little cleaning and is in good working condition.

s-l1600.jpg
 
OK, thanks. I actually saw a youtube video where the guy was having trouble doing that. Wanted to power tap. Ask if anyone knew of a way. He got several comments, but no solution.

So that's why I thought you didn't want to do that. I'll give it a try when I get home.

It depends on the accel/decel settings. If too fast, then yes it may be an issue. On any reasonable VFD tyou can set those times to suit. With a high inertia load, long is not always better than medium, though.... You can get in trouble with long and with short, a medium time often works where long fails.
 
Ok thanks.

I got my dv/dt output filter hooked up. There is no audible improvement. Hopefully it's at least protecting the motor more.

I was surprised by it's size. The 12A reactor in the filter is half the size of the 10A it replaced. Puzzling!

Pic #1 shows the 10A on the right.

Reactors.jpg

10A-Reactor.jpg

12A-Reactor.jpg
 








 
Back
Top