vfd to motor 100 ft? 6

[jdenison]

What kind of problems should I expect with installing a vfd 100 ft from my motor that is 25hp 480v 33.7a w/ an Allen Bradley power flex 70 vfd. The conductor size I came up with is 3 - #6 thw-2 and a #10 grnd. Any suggestions on better condutors? I also came up w/ a 90a inverse time trip breaker ahead of the vfd any other opinions?

 

[jraef]

100 feet is worthy of considering at least a load reactor to attenuate the potential risk of motor winding insulation damage from "standing wave" pulse generation. If you are not aware of this phenomenon, do a web search on that term, in quotes, and do some reading on it. Better yet would be a complete load filter package. It not only protects the motor from the effects of the VFD output, but can protect the VFD tput transistors from damage caused by problems in the motor r conductors. Cheap insurance.

As to the sizing of the circuit breaker, make sure you read and understand the NEC sections specific to this, somewhere around 430.52 (I don't have my version here with me). The breaker must be sized according to the VFD input current, NOT the motor current. 90A seems a tad large for that size drive, unless your VFD is a lot larger than necessary for a 25HP motor.


"If I had eight hours to chop down a tree, I'd spend six sharpening my axe." -- Abraham Lincoln
For the best use of Eng-Tips, please click here -> faq731-376

 

[amptramp]

As jraef indicated, motor insulation failure and winding blowout were a big problem in the early days of IGBT drives; however, this problem has been well studied and most drives and motor manufactures have recommendations for your use.

A good paper on the subject is:

http://literature.rockwellautomation.com/idc/groups/literature/documents/wp/drives-wp019_-en-p.pdf

Quote from Allen-Bradley Publication 20A-UM001M-EN-P – May, 2009 which is the Powerflex 70 User Manual:

Motor Cable Lengths:
Typically, motor lead lengths less than 30 meters (approximately 100 feet) are acceptable. However, if your application dictates longer lengths, refer to "Wiring and Grounding Guidelines for Pulse Width Modulated
(PWM) AC Drives" for details.

Per the following Allen-Bradley document "Wiring and Grounding Guidelines for Pulse Width Modulated
(PWM) AC Drives":

http://literature.rockwellautomation.com/idc/groups/literature/documents/in/drives-in001_-en-p.pdf

Page A-9:

A 25HP inverter duty motor (1600V insulation rating) is good for 600 ft without additional termination devices. However, this spec also requires the use of an appropriate cable, i.e., a shielded/unshielded cable designed for inverter use (an improperly designed cable can have capacitive charging currents that can cause nuisance overcurrent trips at long motor lead lengths).

I would recommend an inverter duty motor and a VFD rated shiedled cable and you should be fine.

With respect to the drive overcurrent protection:

For a 25HP Standard Duty Powerflex 70 drive (Model 20AD034), I would use a Bulletin 140M (part number 140M-F8E-C45) motor protection device. The devices are listed for use with the Powerflex drive and are typically cheaper than a UL489 breaker.

The previous information can be found in the following document (Powerflex 70 User Manual):

http://literature.rockwellautomation.com/idc/groups/literature/documents/um/20a-um001_-en-p.pdf

Page A-17

As jraef stated the circuit conductors are sized at 125% of drive rated input per NEC 430.122(A).
(FYI: NFPA standards including NFPA 70 2008 (the NEC) can be viewed online at NFPA.org)
Per Allen-Bradley a 25HP 480V Normal Duty drive has an input rating of 31.2A so 31.2A * 1.25 = 39A. Based on Table 310-16 a #8AWG @ 75 degC conductor is good for 50A so unless your branch circuit supply is far from the drive (voltage drop) a #6AWG is a bit of overkill. A #10AWG ground is ok.

 

[itsmoked]

Nice amptramp.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[DickDV]

I agree with amptramp except for that statement "A 25hp inverter duty motor (1600V insulation rating) is good for 600 feet without additional termination device".

I don't know whose recommendation that is but it is totally out of sync with general industry practice and experience. At 460V, a 25hp MG1P31 (inverter duty) motor starts being at risk right around 100 feet, just as jraef suggests.

Actually, the need for reactors, etc. is dependent not only upon motor lead length but also motor hp. Everyone seems to do the derate a little different but, the rule I use is "at 10hp or less lead length can be up to 60 feet, at 250hp or more the lead length can be up to 250 feet, and use linear extrapolation between those points".

But 600 feet at 25 hp, no. Brace yourself for trouble at that rate!

 

[edison123]

A lps for amptramp.

Muthu

http://www.edison.co.in

 

[amptramp]

I agree that 600 ft long motor leads sound extreme but after reviewing Allen-Bradley's literature this spec stands.

See Rockwell Knowledgebase Answer ID 34253.

A quote from the following drive system sample specification:

http://literature.rockwellautomation.com/idc/groups/literature/documents/sr/20a-sr001_-en-e.pdf

REFLECTED WAVE
A software algorithm to limit the reflected wave due to long cable lengths to a maximum of
2.25 times the bus voltage or 1600V, whichever is less, up to cable lengths of 600 ft (183m).
Hardware designs also limit peak voltages on the motor.

Please note the motor must have 1600V insulation and the drive output cable must be VFD rated for this spec to stand.

I would contact my local Allen-Bradley rep for a consultation.

 

[DickDV]

Ho, Boy!! If Allen Bradley says it then it must be true!

You may think that if you wish!

 

[amptramp]

Although I can understand DickDV's cynicism towards Allen-Bradley specifications, Toshiba also rates their G9 series drives for 600ft at 460V and <5khz carrier freq. But, of course, they specify the use of a "NEMA MG-1-1998 Section IV Part 31 compliant Motor".

http://www.toshiba.com/ind/data/tag_files/G9_Installation_and_Operation_Manual_PN_58681-002_3101.pdf

Page 20

Also, Marathon Motors specifies their Max Guard insulation system (which includes the Blue Max and Black Max series of motors) for use on PWM Inverters at any distance. Quote from Page 10 of the following document:

http://www.marathonelectric.com/motors/docs/manuals/SB371.pdf

"Motors that employ the MAX GUARD® system can be operated at
any cable length and/or carrier frequency. MaxGuard® surpasses the requirements of NEMA MG1-2003, Part 31, Section 4.4.2."

Is it possible our rules of thumb need updating? Is it possible a properly specified drive, carrier freq, motor cable, and motor can withstand the standing wave overvoltage and CIV (corona inception voltage) problems at 600ft for an acceptable mean time before failure?

On a related note, when installing 600ft motor leads, I would think voltage drop should be considered.

 

[DickDV]

The drive business has always been known as a huge generator of "snake oil" claims and on this subject it literally flows in the streets!

Anyone else care to comment of what constitutes a good rule for protecting MG1P31 motors on drives?

 

[MAGTiger]

It could get really complicated if you factored in such issues as process critacality, motor age, drive/ motor hp unity. But my very genral rule of thumb is upt to 25 ft, no only leads. 25 to 100ft a 1.5% reactor. 100 to 1000ft, a dvdt filter and beyond 2500 ft a sinewave filter. You may have noticed the gray area between 1000 and 2500 ft. In that range, I'd want to consider drive switching speed, and cable spec carefully. Just going to a sinewave filter can be a pricey choice. One thing to be careful of is lead sizing. Go with the bare minumum wire size that gives tolerable voltage drop. Too much over size adds unnecessary capacitance, very bad.

Neil

 

[LionelHutz]

DickDV - The rule could be that some days you just have to get lucky?

I was at one site and there was about 10' of cable. The 3% motor reactor caused a lot of extra ring-up which disappeared when it was removed. There was a case posted here where a sinewave type output filter caused neutral and/or ground current which then caused voltage problems. So, reactors or filters aren't always the solution.

 

[ozmosis]

In Europe and Australia/NZ there is another factor and that is EMC. To comply with the specific (legal) requirements of EMI (ElectroMagnetic Interference)then the VFD manufacturer has to stipulate the motor cable length to comply with the specific level of compatibility (C1, C2, C3..). This motor cable length is not the maximum length the VFD can operate a motor, it is the maximum cable length legally required. This varies hugely depending on the compatibility level. Some manufactures state anything between 150M (450ft) and 5M (15ft). It depends on the quality of the filter used and how it has been tested. It is all down to the VFD manufacturer to state their motor cable length.
This is the law.
Where it is not the law, it doesn't mean you do not need to worry about it!

 

[DickDV]

Lionel, you have hinted at what might not be the rule. Would you care to suggest what would be a good rule?

So far, we have widely diverse claims which lack specific application.

Or, is my claim of snake oil being verified here?

I, for one, think 600ft without protection is wildly risky especially for small motors.

 

[LionelHutz]

No, no good rule. Just saying a filter of some sort is probably a good idea but don't expect it to just work without still having problems some times.

It'd be interesting to hear from people who actually qualify every installation or who have tested different combinations.

 

[MAGTiger]

Problem is this is a complex problem. Parasitic elements of capacitance and inductance in the cable have not been given much concern on the input of the drive where the harmonics of significance are not much more than 1000hz. But when the drive output is pulsing at 2-16khz with voltage rise times in the hundreds of nanoseconds, these parasitic components become very significant (read, harder to predict the effects). And capacitive coupling comes into play too. So, the potential drive output problems to be dealt with require an entirely different thought process than do the potential input problems. Not to mention that the output issues are largely the result of voltage wave shape while the input problems are due to current waveshape.

Neil

 

[ozmosis]

However, the original poster jdenison is talking about 100ft. This should be ok. RA say it in their literature, and if it was our VFD then it would be no problem.
The cable size is ok and as MAGtiger points out, inductance and capacitance are factors of any load from a VFD but at this cable length and cable size then you should still be driving the motor and not just the cable.
Assuming the motor insulation is suitable for VFD use, this should be ok.
I wouldn't have a problem as long as there are no other factors not mentioned; i.e. this is not 100ft down a borehole running a submersible pump etc.

 

[amptramp]

The following snake oil discusses a technology being utilized to address the issue of long motor leads as well as bearing current problems:

http://www.yaskawa.com/site/dmdrive.nsf/link2/MHAL-6C3MQ2/$file/AN.G7.01.pdf

http://www.controleng.com/article/2...inverter_improves_motor_drive_performance.php

(Please copy and paste the links as the TGML was rendering the first link incorrectly so I turned TGML processing off. Thanks!)

 

[LionelHutz]

Interesting. A 3 or 5 level neutral point clamped inverter is quite common for medium voltage applications such as 2.4kV or 5kV. The Yaskawa drive should do as they claim, the main downside is that it must cost more than a typical H-bridge type of VFD.