VFD Cable Reflectance 3

[peebee]

I've lately been hearing that long motor cables can be a problem with VFD's due to reflectances, even on cables of roughly 100 feet. I don't doubt that this could be a problem on unusually long cables, but it seems to me that on most installations the cable would not be nearly long enough to be resonant.

Here's my reasoning:

Assume a drive switching frequency of approximately 12kHz, which is on the high side for most drives. Dividing this into the speed of light, 3x10^8 m/s, the free-air wavelength would be about 2500 meters. Let's say the wavelength in the cable is 1/3 that, or about 800 meters. To avoide resonance, let's keep our cable to 1/8 that length, or 100 meters. That's 328 feet.

So it's my contention that any modern drive should work with any cable up to about 330 feet with no standing wave reflectance, and due to the conservative nature of the calculations and assumptions above you should be able to go much further than that with most drives.

Any thoughts on this?

 

[electricpete]

I don't know the answer.

But one thing that might come into play is that the switching frequency of 12khz actually might resemble a voltage square wave with fundamental frequency of 12khz and harmonics at multiples of this (up to infinite frequency in the case of ideal square wave). In case of real world square wave with finite rise time, we need to know the rise time before we can really start to judge the frequency content.

 

[electricpete]

I am a little more confident in my answer after searching around a little bit.

The following article seems to provide a pretty detailed description. I only got to page 2 where it seems they indicated that the critical cable length is determined by the switching time (rise time), not switching frequency. There are 19 pages total.

http://www.ab.com/drives/techpapers/InstallationConsiderations01.pdf

 

[peebee]

Perfect. Thanks.

 

[nbucska]

I don't think the resonance is a problem -- the wideband
radiation due to the long cable and the many harmonics
will certainly exceed any acceptable or legal limit by
several magnitudes of order.

Don't even try to use more than a few feet cable and that
one, too, preferably shielded. If you make the drives,
you must assure that it meets the FCC rules.








<nbucska@pcperipherals.com>

 

[peebee]

That's an interesting argument, nbucska, but definitely contrary to industry practice. My experience, which seems to be validated by the article referenced by electricpete, is that regular cable in lengths of hundreds or thousands of feet has been the norm on VFD's for years, and that standing waves only recently became a problem due to the switch from bipolars to IGBT power transistors.

You do seem to have a point that radiation should also become more of an issue. But I've never heard of anyone having a problem with that before, and I've never heard of anyone using shielded cable to feed a low voltage motor.