Conductors for VFDs

Status
Not open for further replies.
boater bill said:
The IGBT's of the VFD switching at 2-10KHz tend to put harmonic noise back on the power lines and other devices will start to trip or behave erratically unless an input reactor is installed.
Click to expand...

We had one site where this vfd radiated frequency shut down all the patient monitoring systems.
 
ghostbuster,
Many VFD's have a "skip frequency" parameter. Thats means if you are having issues with high frequency noise, you can eliminate the particular frequency in the programing parameters.

Example: No problems at 60hz but you do have problems at 61hz. You program out 61hz and the drive will never settle at that frequency.
You can also skip a band of frequencies if needed.
This function can be customized to your specific application. This function was designed for the problems that were occuring in your situation.
Note: this is a standard parameter available on most all VFD's.

The "Carrier frequency" can be adjusted also. This is the speed at which the IGBT's (Transistors) switch on and off. This adjustment affects the switching speed, it does not have any influence on the speed adjustments. This is where I would start. Set this parameter as low as possible.
 
John Valdes said:
ghostbuster,
Many VFD's have a "skip frequency" parameter. Thats means if you are having issues with high frequency noise, you can eliminate the particular frequency in the programing parameters.

Example: No problems at 60hz but you do have problems at 61hz. You program out 61hz and the drive will never settle at that frequency.
You can also skip a band of frequencies if needed.
This function can be customized to your specific application. This function was designed for the problems that were occuring in your situation.
Note: this is a standard parameter available on most all VFD's.

The "Carrier frequency" can be adjusted also. This is the speed at which the IGBT's (Transistors) switch on and off. This adjustment affects the switching speed, it does not have any influence on the speed adjustments. This is where I would start. Set this parameter as low as possible.
Click to expand...

John

The patient monitoring system was "looking" for a signal in the frequency band 50-100 khz.The drive people were unable to attenuate all their frequencies within this band.They tried for several months.

It was a big "mess" with everybody suing each other(engineers,contractors,drive people,hospital admin. etc.)In the end ,the monitoring system was replaced ,because it was deemed "too sensitive":)
 
ghostbuster,
So the VFD won. I would have thought the VFD would have been eliminated. What is the drive manufacturer? Just curious. Thanks
 
John Valdes said:
ghostbuster,
So the VFD won. I would have thought the VFD would have been eliminated. What is the drive manufacturer? Just curious. Thanks
Click to expand...

John

The electronic light ballast system (base frequency approx. 27 khz. : along with higher harmonics of this base frequency) was also causing malfunctions for this patient monitoring system (installed and working perfectly until now for approx.10 years).

That is an additional reason for replacing this monitoring system.

The drive mfr. is one of the big "5".
 
Charlie: Belden makes a sheilded VFD cable.
I have a white paper from AB that discusses insulation, I will try and find it, but I recall it recommends XLP XHHW insulation.
Our standards call for any wire in a wet location to be XHHW for 8 and larger. The smaller sizes are normally for lighting and THHWN is fine for that
One simple change is to adjust the carrier frequency on the drive to 2 KHZ the factory setting is normally 8 khz and the lower frequency is easier on the motor,but louder.
 
charlie b said:
Anybody know of either a ?good design practice? or an actual NEC requirement related to the conductors between a Variable Frequency Drive (VFD) and the motor it serves? Specifically, do you need to use a conductor of a higher rated insulation system (e.g., a 1000V insulation, instead of the 600V used upstream of the VFD)?

This came up in a design discussion with a client who thinks the high level of harmonics generated by the VFD would damage the insulation, if the insulation system did not have a higher rating. He did not think this applied to the conductors from the branch panel to the VFD.

Can anyone confirm or refute the existence of such a requirement?
Click to expand...

After being burned by VFD noise hosing up every sensor within two counties (well, not that bad) I use specially made VFD cable. While I haven't seen anything in the code about needing to use better made cable, this cable isn't much more than regular and your motor system will last longer.

1. The layered insulation types clamp noise at all frequencies so your sensors don't get hosed up. It has to be properly installed and bonded for this to work right. Usually you have an insulated flex conduit with foil and braid and then a separate grounding conductor. You have paths home for all noise bands.

2. The insulation is thicker so it won't get eaten up as quickly.

3. With some filtering on the VFD (newer and name brand VFDs will have this built in) you won't be making those 2000V spikes on your 480V nominal motor.

All the major cable makers are selling their own VFD cable. You can get conduited or just plan old multiconductor cable. Read their literature. Check out their specifications. Give it a try.

I will never run discrete wires to a VFD again, no matter how close it is to the motor.

And do keep your VFD to motor run as short as possible as this helps cut down noise spikes too.

Matt
 
Matt,
After being burned by VFD noise hosing up every sensor within two counties (well, not that bad) I use specially made VFD cable.
Click to expand...
Were the cables that caused you problems in ferrous conduit? All of our wiring, both power and sensors are in rigid steel conduits and we have no issues between our sensors and VFD wiring.
Don
 
megloff11x said:
I will never run discrete wires to a VFD again, no matter how close it is to the motor.
Matt
Click to expand...


I have never used any of these special cables before and I have installed numerous drives without problems. I am certain they have there place. I guess I never had a situation like you mentioned Matt.

The rules that I follow may not be the same as yours.

1) Always install a line reactor.
2) Never switch the load side of a drive (contactor) Bypass not included.
3) Always run control and signal wiring seperately from input and output of drive in a seperate conduit. Signal wiring sometimes in it's own seperate conduit.
4) Always run motor leads and EGC in seperate conduit.
5) Always install a load reactor if the distance from drive to motor is over 100 feet. Or if special considerations are needed.
6) Never parrallel control circuits over power conductors in enclosures.
7) Set "Carrier Frequency" to lowest acceptable KH.
 
I would add

8) use shielded cable, with ground glands covering 100% of the shield circumference, for all control/analog signals.
9) use electrostatic shielded transformers, with proper in/out separation, to generate the control signals.
10) reread the OEM installation instructions.
 
This is an interesting thread. I just installed 5 VFD's and am having problems with 2 of them. 2 units blow the fuses in the VFD imediately on start up. 1 unit load side is over 100 ft to the motor (the furthest) and the other is within 10 ft but in PVC while all the others are in IMC.
Motor was replaced and filtering added to solve the problem for the furthest. The other problem has not been solved as of yet.
 
petersonra said:
Seems like that might be a code violation.
Click to expand...

I agree, however I would think what is meant is to run the motor conductors and EGC in the same conduit, separate from any other wiring that might be in the system.
 
cjnickjr said:
This is an interesting thread. I just installed 5 VFD's and am having problems with 2 of them. 2 units blow the fuses in the VFD imediately on start up. 1 unit load side is over 100 ft to the motor (the furthest) and the other is within 10 ft but in PVC while all the others are in IMC.
Motor was replaced and filtering added to solve the problem for the furthest. The other problem has not been solved as of yet.
Click to expand...

Remove the load conductors from the VFD. See if it starts and ramps up, using the LED readout on the control. If the motors have feedback (encoder, tach, ect...) you may not be able to run the drive without the load leads connected. In this situation set the input from these devices to zero.

Are you having this issue when you apply power to the motor or when you apply power to the drive? What fuses are you refering to?

If you are having the issue when you apply power to the drive, it sounds like the input diode bridge is shorted and will need replacement. Make sure that you use the inputs or keypad to start the motors. DO NOT use line power for motor starting. (like jumping out the start I/O) Install a line reactor regardless!!

If you are having this issue when you are trying to start the motor via drive inputs, try testing the IGBT's (transistors) with a meter with the diode check function. Meg motor and load conductors. Remove ground conductor and try to start it up. If it starts, it is the motor or conductors. Hook the EGC back up. Since this is the "closest" drive, a load reactor is not indicated.

But the simplest way to know where to look is by removing the load conductors, and try to dry run the VFD. If the drive works with out the motor. Well you get where I am going.

Feel free to pm or email me. I will do my best to assist you. jvaldes@windstream.net :)
 
Charlie, Inverter output conductors are reported to be a source of motor degradation and shortened service life. Increasingly drive manufacturers are recommending shielded cable w/concentric ground construction. Perelli PROTFLEX, Rockbestos GARDEX, AmerCable GEXOL and Helukable TOPFLEX are some examples of products that I've had good experience with.
 
John thanks for the input. The line side lands on 100A fuses in the VFD. We can get power to the unit but when applying power to the motor fuses blow. No feedback. The VFD does have a bypass. Motor was removed, sent out and returned, load conductors megged. Decision was made to install line reactor. Will keep you posted. Tks
 
Status
Not open for further replies.
Top