VFD Output Motor Cable Number of Cores 5

[ThePunisher]

I observed that all VFD motor cables are 3core construction. I would like to understand why a 3 x 1c VFD output cable cannot be used.

I have a medium voltage VFD controlling a 1000kW, 4kV induction motor. I am insisting to use 3 core but our client is looking on single core arrangement because that is what their previous consultant recommended. In this regard your expert opinion or reference will bemush appreciated.

 

[dpc]

Current practice is to use shielded (screened) 3/C cable between motor and VFD to minimimze RFI coming from the cable.

3 x 1/c configuration will work fine to run the motor, but the radiated energy coming out of the feeder can wreak havoc on any nearby low-level circuits.

Tell him it's the standard of the industry. Check any VFD installation manual.



David Castor

http://www.cvoes.com

 

[ThePunisher]

David, if the 3 x single core cables are individually screened, will this mitigate the radiated RFI?

Just thinking out loud on what client may respond with.

Thanks you.

 

[Skogsgurra]

Screening single core cables has some inherent problems. The voltages and currents in a VFD system are usually not very 'nice'. Putting all phases within the same screen balances much of the effects from the VFD waveforms. Not completely, but to a large extent. MarkE recently published interesting measurements that he had done on different cable types that are used with VFDs. I shall look for the thread when I have closed this.

There are different types of MV VFDs. The Robicon (now Siemens) seems to be 'nicer' than most other, but I would still be very hesitant to use single core cables. Some very large rolling mill motors have to use single core cables simply because there are no three-phase cables available for those currents and voltages. But that is an exception.

For 1 MW, I would definitely go with three-phase cables.



Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[Skogsgurra]

This is Mark Empson's thread: thread237-303429

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[azamfaez77]

I will definitly to go with 3C cables reason for easy installation and cheap. But if client insisted to use 1C cable, i will follow them. But will remain them about the cost.

Technically, i did not see any different.I have experience to lay VFD to 800kW (6.6kV) induction motor.

 

[azamfaez77]

anyway, pls double confirm with your VFD supplier.

 

[waross]

uh azamfaez77; You may consider looking at the link provided by Skogsgurra.
EMC issues with VFDs
thread237-303429
You may decide that there is a difference.

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

 

[ThePunisher]

Thank you for link Gunnar. It was very informative.

The operating conditions that was based for testing did not mentioned any discrimination of 3 core or single core arrangment.

I am into the opinion that if I use singe core, 3.6/1kV, screened copper cables arranged in trefoil formation plus grounding cable and they are properly clamped together and their screens are bonded together at both terminations using EMC glands, they will offer the same like a 3 core screened cable.

I am not pretty sure though if single core screened, EPR cables are available at 3.6/1kV nominal? Any ideas?

 

[Skogsgurra]

There is a big difference between the two configurations. You have better believe it. But if your VFD has low HF emission, it may work with separate cables.

It would help if you told us what VFD you will be using. A lot depends on that.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[ThePunisher]

Gunnar,

Based on proposals, it would be more likely to be an ABB ACS1000.

I also have a strong belief that there is a significant difference. However, I am trying to get some technical information to explain it or better yet any technical literature that I could refer to understand and effectively explain and convince them to use 3 core, 3.6/6kV cables instead of the single core equivalent. I can't deal with the vendor at the moment because there is P.O. placed yet but we might pre-order the cables that is why I have some concerns.

 

[Skogsgurra]

The ACS 1000 has a sine filter built-in as standard. I would say that you can use either cable type. Use what is practical and economical and in line with your local practice.

I am afraid that wasn't much help

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[dpc]

It will help to put whatever you use inside of a steel conduit.

David Castor

http://www.cvoes.com

 

[electricpete]

I am by no means knowledgeable about vfd cable process. I would like to propose a small explanation for the benefit of 3/c transposed and overall shielded as opposed to 3 1/c's individually shielded. Please tell me if wrong or if obvious.

1 - The transposition reduces magnetic field radiation. It is similar benefit to twisted shielded pair for signal cables (except for signal cables the concern is acting as magnetic receiver vs transmitter)

2 -With 3 cables in one shield, the fields should cancel resulting in very little shield current. In contrast 3 individually shielded 1/c cables would each have current in their shields. That shield current itself can be a source of emi.

=====================================
(2B)+(2B)' ?

 

[Skogsgurra]

That's right, Pete.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[ThePunisher]

Thanks for reminding Pete. If the 3 x 1C, screened output cables are laid in trefoil formation, I would expect that the EMI issue is resolved and closely replicate that of a 1 x 3Core cable.

Am I right with this opinion? In a 3 core cable, the separation between the 3 conductors are dependent on the thickness of insulation that surrounds each copper conductor and the overall screen will tend to trap any stray voltages.

On the 1 x 3core on the other hand, this separation will be dependent on the thickness of insulation+jacket+air gap due to a probable imperfect installation (trefoil) in the field.

However, will this difference cause significant EMI to cause a problem?

 

[Skogsgurra]

I started making a picture showing different ways of arranging cables.
It is here:

http://gke.org/pub/files/Cable arrangements for VFD 1.jpg

Not at all complete.

Putting cores in different screens is a bad idea. High currents are capacitively coupled into the screens and that will cause lots of problems.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[ThePunisher]

Hi Gunnar, may request more information as to how this high currents which capacitively coupled to screens create more problems?

I am doing my further research to get a close answer to this but would help a lot if you throw me your thoughts further. Appeciate your comments much...

 

[Skogsgurra]

As I said before. You will not have any problems with the ACS 1000 since it has a sine filter built in.

For other inverters with PWM output, there are very good papers on the web. Just google +PWM +"motor cable" and variations thereof.

Those pictures were started so you can visualize the difference between a symmetrical three-core cable with a common screen and other configurations where current is coupled from the separate cores to the respective screens.

That current now has to flow along the cable screen to ground and will produce HF emission and HF voltage drop in the screen. That is not what you want. But, as I said, no probs with the ACS 1000.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[ThePunisher]

Thanks Gunnar, I got some pretty good technical papers to start with.

To all, thank you for your assistance. I'll take it from here.