VFD motor leads

[cyoung120]

Hi all. Was told today that it is no longer acceptable to use thhn from drive to motor disconnect in a damp location. Can anyone confirm this? Also there will be three sets of motor leads in conduit. Not sure if that applies but was also told that anymore than three also unacceptable.

 

[DavidA]

THHN/THWN is, as far as I know, still listed for use in damp locations. As for number of conductors in a conduit: Unless it falls under engineering specs or the conductor size is too large for the conduit, I don't see any problem with it.

 

[jim dungar]

Manufacturers generally discourage, if not actually advise against, putting multiple VFD output circuits in the same conduit.

 

[rcwilson]

VFD's create high frequency harmonics that can lead to voltage surges that damage motors. One means of mitigating the surges is to use VFD-to-motor cables with balanced capacitance and shielding such as a three conductor armored power cable with three ground wires in the spaces between the phase wires. Single conductor cables in conduit do not perform as well.

Search for VFD cables on the internet for better explanations.

 

[Jraef]

VFD's create high frequency harmonics that can lead to voltage surges that damage motors. One means of mitigating the surges is to use VFD-to-motor cables with balanced capacitance and shielding such as a three conductor armored power cable with three ground wires in the spaces between the phase wires. Single conductor cables in conduit do not perform as well.

Search for VFD cables on the internet for better explanations.

And this is ESPECIALLY relevant to multiple VFD outputs in the same raceway if that cannot be avoided.

The other issue is that the VFD cables tend to use RHW with XPLE insulation that is rated for 1000V for the individual conductors because of the higher level of voltage in the DC pulses that make up the PWM output. You wouldn't have that with THHN/THWN.

So although I don't think it's "illegal" from a code standpoint to use separate THHN / THWN conductors, it's just a really bad idea.

 

[don_resqcapt19]

I keep reading all of this stuff, but based on my experiences it seems to be mostly manufacturer CYA information.

They have a few hundred VFDs at the plant I often work at, all of them using single conductor THWN in rigid conduit, most with conductors between the drive and motor at 150+' and no load side reactors and we just don't see these issues. We don't see motor or conductor failures like I would expect reading the information that is out there.

In general the only conductors in the raceway are the motor conductors and until recently the 120 volt control conductors. We have had issues with the control circuits in the newer drives (the induced current from the motor leads into the control wires would sometimes not let the motor respond to a stop command) and now we only the motor leads in one conduit and the control in a second conduit.

 

[Besoeker]

And this is ESPECIALLY relevant to multiple VFD outputs in the same raceway if that cannot be avoided.

I think it can always be avoided.
Conduit isn't used very much here. The cable from drive to VFD is (almost*) without exception a steel wire armoured cable. Per drive. The question of running multiple VFD conductors in the same raceway not only doesn't arise it doesn't even enter the thought process.

*I'm sure somebody somewhere can find an exception or two.

 

[kwired]

I keep reading all of this stuff, but based on my experiences it seems to be mostly manufacturer CYA information.

They have a few hundred VFDs at the plant I often work at, all of them using single conductor THWN in rigid conduit, most with conductors between the drive and motor at 150+' and no load side reactors and we just don't see these issues. We don't see motor or conductor failures like I would expect reading the information that is out there.

In general the only conductors in the raceway are the motor conductors and until recently the 120 volt control conductors. We have had issues with the control circuits in the newer drives (the induced current from the motor leads into the control wires would sometimes not let the motor respond to a stop command) and now we only the motor leads in one conduit and the control in a second conduit.

I pretty much agree and have observed about the same things. Did have some small fan motors that would fail from time to time but have had no problem since installing line reactors. Have had RF interference problems in other equipment, when output conductors were installed in non metallic raceways, but never metal raceways.

 

[Besoeker]

I keep reading all of this stuff, but based on my experiences it seems to be mostly manufacturer CYA information.

I've seen it happen a number of times. I posted in another thread of a site where there was repeated failures of 750 kW motors. It needed a redesign of the motors to fix the problem.
On another site with submersible motors there was degradation of the motor insulation followed by eventual failure. At intervals of several weeks for each motor. Excessive dv/dt was the problem and solved with output chokes.

 

[jim dungar]

I've seen it happen a number of times.

Not to pick on you, but "a number of times" does not impress me.

What would you say is the % of failures for pumps/fans <25HP (20kW), how about <100HP (75kW)?

Literally thousands of small drives are installed daily using single THHN in conduit, how many do you expect will experience failures?

I am in the group that believes around 99%(?) of general VFD installations <=100HP (75kW) do not need special output cable. Of course this means there are situations where it is paramount.

 

[Besoeker]

Not to pick on you, but "a number of times" does not impress me.

The cost of having to rewind four 750kW motors didn't impress the motor manufacturer either.

Literally thousands of small drives are installed daily using single THHN in conduit, how many do you expect will experience failures?

I am in the group that believes around 99%(?) of general VFD installations <=100HP (75kW) do not need special output cable. Of course this means there are situations where it is paramount.

It isn't just the cable. The problem stems from the very high di/dt of the inverter output pulses. It's something we generally look at, particularly on larger drives. If we think it might need an output choke, we fit one.
If you use single core conductors in conduit the spacing between the conductors will introduce some inductance.

I don't know what failure rates are. Failures may be after long periods following installation and possible intermittent duty - the failure mode is cumulative. And maybe the failures just don't get investigated being put down to just one of those things that happen.

On one site with our local water company it took a number of failures in fairly short order before they decided to invite me to investigate the cause of failure.

And nowadays, motors are more likely to be manufactured to be suitable for PWM waveforms.

 

[kwired]

Dual voltage motors will take more of a hit from the voltage spikes if connected to the lower rated voltage.

 

[don_resqcapt19]

Dual voltage motors will take more of a hit from the voltage spikes if connected to the lower rated voltage.

Can you expand on that?

 

[jim dungar]

... particularly on larger drives....

This is the key part of the discussion.

Yes, I have heard of small drive (<25HP, 20kW) issues but they are extremely rare based on the quantity of these drives installed daily.

Yes, I would not install a drive system >100HP without taking precautions that address >99% of the potential failures.

 

[jdsmith]

.....

I am in the group that believes around 99%(?) of general VFD installations <=100HP (75kW) do not need special output cable. Of course this means there are situations where it is paramount.

This all depends on how long you want the cable to last and if the conductors will get wet. For many industrial installations I agree with you. The installations are often not in service long enough to cause the cable to fail.

I'm in the petrochemical business where we design for 50 years of service. In normal non-VFD service with proper surge arrestors, snubbers on MV vacuum breakers, etc. cables are still one of the lowest reliability components in the plant. VFD service longer than 200 ft (meaning 2x reflected wave exists) causes cables to fail sooner than they would in normal 480V service, so it's natural that we focus on it from a reliability perspective. There was a paper published at the 1999 PCIC conference that lists predicted life for different conductor insulation types and lengths that is a good reference on this topic.

I understand that my design basis does not apply to most people, but there is a petrochemical contingent on the forum that should not be lulled into standard cable for VFD service given the cost of downtime, years of service expected, and typical long cable lengths. Some of the typical petrochemical cables are fine for VFD service and should last 40+ years, but other common cable constructions will likely last 10-20 years.

 

[weressl]

I keep reading all of this stuff, but based on my experiences it seems to be mostly manufacturer CYA information.

..and not every smoker dies of lung cancer either.

Do you argue that the voltage rise is not a problem and the impedance imbalance of single conductor cables don't 'confuse' the modern drives and provide less than optimal control? You also failed to address the question of what insulation those conductors use?! On a case-by-case basis judgments can be made, but in a litigious business environment one can't expect the manufacturers to commit themselves to a wide variety of applications where most of the drive applications installed without any specific concerns,in other wiords nothing more than would be given to a FVNR starter.

 

[weressl]

This all depends on how long you want the cable to last and if the conductors will get wet. For many industrial installations I agree with you. The installations are often not in service long enough to cause the cable to fail.

I'm in the petrochemical business where we design for 50 years of service. In normal non-VFD service with proper surge arrestors, snubbers on MV vacuum breakers, etc. cables are still one of the lowest reliability components in the plant. VFD service longer than 200 ft (meaning 2x reflected wave exists) causes cables to fail sooner than they would in normal 480V service, so it's natural that we focus on it from a reliability perspective. There was a paper published at the 1999 PCIC conference that lists predicted life for different conductor insulation types and lengths that is a good reference on this topic.

I understand that my design basis does not apply to most people, but there is a petrochemical contingent on the forum that should not be lulled into standard cable for VFD service given the cost of downtime, years of service expected, and typical long cable lengths. Some of the typical petrochemical cables are fine for VFD service and should last 40+ years, but other common cable constructions will likely last 10-20 years.

I have installed drives as replacement for existing FVNR starters where the motors weren't necessarily ASD compliant - just severe duty Mill-Chem - and the cables were TC with THHN conductors. Most of these applications were 1/6 max turndown variable torque applications. However I have done this with the warning that be prepared for premature failure potential both for the motor and cable and to install rated cable and motor as replacement should it occur. So far I have not heard of a failure with the exception of one bearing got trashed,but that was explained away, even though the bearing had clearly shown circulating current damage. On the top of it, it was a small - 5 or 7.5 HP motor - with a late variant of Siemens drive. All others were/are ABB/AB and Toshiba.

Would NOT install a new application with ordinary cable and other than IEE841 motor.

 

[don_resqcapt19]

..and not every smoker dies of lung cancer either.

Do you argue that the voltage rise is not a problem and the impedance imbalance of single conductor cables don't 'confuse' the modern drives and provide less than optimal control? You also failed to address the question of what insulation those conductors use?! On a case-by-case basis judgments can be made, but in a litigious business environment one can't expect the manufacturers to commit themselves to a wide variety of applications where most of the drive applications installed without any specific concerns,in other wiords nothing more than would be given to a FVNR starter.

I specified that we use THWN in rigid conduit and that in most cases there are also 120 volt conductors for the HOA in the same raceway as the motor conductors. I understand the manufacturers position and why they says what they say. I just stated that in my experience we have not seen those issues. And, yes in general the smaller drives are installed without any more than what we would use for a FVNR starter.

I will say that most of the drives are 25 hp or less and that drives for motors over 100 hp do have load side reactors. However even with the large motors the conductors are, with a couple of exceptions, THWN in rigid conduit. For the larger motors there are no control conductors in the raceway with the motor leads as the plant standard does not permit #14 conductors in the same raceway with conductors larger than #4.

 

[Besoeker]

A correction ti something I wrote in post #11/

The problem stems from the very high di/dt of the inverter output pulses.

It should, of course, been the very high dv/dt.

 

[kwired]

Dual voltage motors will take more of a hit from the voltage spikes if connected to the lower rated voltage.

Can you expand on that?

If you have a 230/460 volt rated motor and connect it to a 460 volt drive the peaks of the output voltage can be in the 750 volt range. Connect the same motor to 230 volt drive and it will be half of that. Winding insulation is what the issue is here. Line reactors will smooth out the peaks but now the reactor first in line subject to the peaks. They are probably better designed to take it than some motor windings are.

For smaller motors, if you have a choice for voltage readily available it may be worth consideration.