Drive - motor cable length (again)

[thinker]

O.K., this is a well discussed subject. It is basically known how to calculate a critical cable length (standing or reflected waves issue). We recently got a note from one of major drive suppliers that the critical cable length must be significantly reduced when more than one cable run in parallel. Specifically, for 4 cables in parallel the critical length is reduced by the factor of 5. Is there any practical experience to agree with that? I realize that there is coupling between cables changing the equivalent distributed parameters of the whole run, but so significant implications?

 

[Marke]

Hello thinker

One of the major concerns, is the capacitance between cores and between cores and earth. The output of the inverter produces a high frequency wich is PWM modulated with the low frequency output envelope. With modern IGBTs, the transition from high to low (and low to high) is very fast due to the switching time ofr the output devices. This results in a high charging current to that capacitance and therefore the potential for significant power dissipation in the output devices during switching. If the capacitance is excessive, you will have a failure in the output devices.
As you increase the length of the cable, you increase the capacitance. If you run cables in parallel, you also increase the capacitance, but parralel cables provide effectively a higher capacitance than long cable due to the absence of the series inductance of the cable. i.e. the effective capacitance at 10mHz of a single 100M cable is less than the effective capacitance of four 25M of the same cable in parralel.

Best regards,

Mark Empson

http://www.lmphotonics.com

 

[thinker]

Marke, thanks for your input. We just used to think that the critical cable length is a limiting factor to motor terminals voltage level only. It appears that there is another limitation on drive switching losses. In this case drive suppliers should state not only the length, but the number of parallel cables, however they typically do not specify that. We got this info as a response to our request.

 

[Marke]

Hello thinker

You will find that it is common for drive manufacturers tospecify the maximum length of unscreened cable and a separate length for screened cable. The difference being the increased capacitance.

Best regards,

Mark Empson

http://www.lmphotonics.com

 

[Marke]

It is also common for drive manufacturers to quote "total length" of cable on drive output. This sort of covers the parallel cables and also the situation where one drive controls a number of motors such as the fans in a kiln.

Best regards,

Mark Empson

http://www.lmphotonics.com

 

[MMAK]

Hi,
Is there a comprehensive reference for RWT, or we just depend on the suppliers of VFDs. If there is a guide over the subject please let me know. I would like to work out some basics for the department, on Designing... and an appropriate justification for selecting Reflective Wave Traps.

Best Regards

 

[DickDV]

In my experience, this issue of inter-lead capacitance is meaningful only on higher voltage (400+) drives and only in the smaller horsepowers as in less than 5hp. Above that, the increased current thru the lead capacitance gets lost in the motor current.

I have used the following rule without problems:

If carrier frequency is 8khz or less (why would you want higher anyway?) and motor is Class F insulation with MG1 P31 endorsement

Then with 100hp and higher, leads can be up to 250 feet without any additional devices in the motor leads

And, with 10hp and less, leads can be up to 60 feet without any additional devices in the motor leads

Extrapolate linearly between these two points for other hp sizes

This gets around all the smoke, mirrors, and snake oil claims from the manufacturers and has a long history of being a good guideline to follow.

Of course, if someone chose a cheaper motor than spec'ed above, they're on their own or will need to develop their own rules, probably quite different from the above.