Understanding Effects of Coiled Feeder Cables on Motor Starting.

[IBRCAN]

I am a technician in the on-site power generation industry. I do a lot of work in quarries, and every now and then I run into a crusher motor starting issue.

I understand that, as with any electrical installation, feeder cables from the generator to the mcc should be kept as short as possible to reduce the effects of voltage drop. One of my co-workers in particular is adamant that cables should never be coiled to prevent voltage drop.

Am trying to understand and quantify the effects of the feeder cables being coiled. Obviously the coils cause undesirable heat build-up for one thing. The coils will introduce a small amount of inductive reactance into the circuit as well.

Case in point--200 hp, soft started jaw crusher motor, 320 kw prime rated diesel generator with PMG. This generator had been used to start the same and similar sized loads before. During start up of the crusher, soft start was faulting on under voltage. I connected power analyzers at both the generator and soft start. There appears to be at least 100 V lost between the generator and soft start during motor start-up, with in rush current of approx. 700 A at 600 V. Distance between generator connection box and mcc was probably 25', however there is over 100' of 4/0 copper cable run per phase, with the excess coiled up and hung on the side of the generator trailer. There were perhaps 10 coils per phase, 3 ft in diameter. Calculated voltage drop using cable diameter and length indicates a negligible voltage drop, only a few volts. I also calculated the inductance of the coils, which seems to be in microhenries and would not cause significant losses.

I know there are a lot of other variables at play when it comes to generator/motor starting, but I am looking for guidance as to the effects of coiled cable in particular. Thanks.

PS, The generator AVR was equipped with a line loss compensation mode, which I enabled and seems to have solved the motor starting issue.

 

[Skogsgurra]

Coiling or no coiling - is that the question?

Your calculation seems to be right. The coiling is of no importance. We are talking 50 or 60 Hz and the effect that a few microhenries have is negligible compared to the internal voltage drop in the generator.

Coiling can add heat because cooling is less efficient but the heat that develops during the few seconds it takes to start the machine is also negligible. Heard "amateur physists" talk about things like this many times. They usually wer funny hats, also.

Gunnar Englund

http://www.gke.org

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

 

[edison123]

I have seen significant voltage drop and cable heat due to coiling especially when the motor current is in-rush mode thus leading to longer start times and even speed crawling some times in our motor repair shop. I avoided the problem by arranging the cables in zig-zag manner (like the overhead crane trailing cables) so that the cable is not looped.

Muthu

http://www.edison.co.in

 

[waross]

Single conductor or three conductor cables?
For three conductor cables the effect of coiling is negligible but for the voltage drop, you should consult the manufacturer's voltage drop tables due to the cable self inductance. Be aware that when reactance is included the phase angle of the load becomes significant. Due to the high ratio of reactive current in the starting current the voltage drop may be more than the current indicated in the tables.
Single conductor cables will have more self inductance than trefoil cables and this will add to the inductance created by coiling.
Remember that to accurately calculate the voltage drops, you must consider the X:R ratio of the cable impedance and the power factor of the load current.

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

 

[jraef]

I have to say that in my years of doing soft starter commissioning, I came across this issue 4 times and there was significant voltage drop resulting from coiling the cables. In all 4 cases I had them uncoil the cables and lay them out zig-zag, the problem immediately went away. I no longer wear a foil hat, but there is something to this, despite the purely mathematical insistence that it should not be so.

To Bill's point, the situation in these portable crusher installations is that the cables have ALWAYS been single conductors, in my case always left on the spools. The reason is, they use DLO (Deisel LOcomotive) cable because it is extra flexible. But it is also extremely expensive and because the equipment moves around, they are loathe to cut the cable every time. So they spool it up with a long pigtail out of the center of the spool to make the connection. They are often very tight coils and in my cases, steel spools. Iron core reactor? That's what I attributed it to and even on a wood spool or no spool, the air core reactor effect seems equally plausible. The fact that the problem was immediately solved by unspooling the cables was proof enough for me.

"Will work for (the memory of) salami"

 

[Skogsgurra]

We have to consider the numbers. Of course, there is no doubt that a long piece of cable, when coiled, adds voltage drop. But the short length and the few turns mentioned in the OP does not constitute what I consider a significant impedance increase. The fact that fixing the AVR solved the problem is a hint that the coiling didn't casue any problems in this case.

Gunnar Englund

http://www.gke.org

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

 

[2dye4]

If the phases are coiled separately there will be much more inductance than if the bundle is coiled as one unit.

An online inductance calculator gave

152 uH for a coil 36 inches in diameter and 5 inches long with 10 turns.

http://www.daycounter.com/Calculators/Air-Core-Inductor-Calculator.phtml

ohms = 377*152e-6 = 0.05 ohms.

At 700 amps this drop is 40 volts due to reactance.

It just might be so that the coil is less desirable.

 

[wroggent]

Skogsgurra, are you sure we're talking about only 50 or 60 Hz? He mentioned a 'soft start' which to me might mean harmonics and accordingly higher reactance for those components. Also, I would think (I'm not really familiar with generator controls) the AVR would increase reactive power output of the generator to maintain a voltage. Seems like that would essentially improve the power factor which would compensate for the reactive losses due to coiling.

 

[iceworm]

Similar to jreaf:
Years back I was involved (as very junior engineer) with a temporary power, 2MW, 480V generation feeding a permanent installed switchboard with a DOL 400HP motor. The temp feeder was made up of multiple 500kcmil single cables per phase.

When the motor started the temp cables looked like a bunch of snakes and the voltage at the switchboard dipped way low.

The chief engineer has us lay the cables out - no coils, group up the cables in threes (one of each phase) and tie them up securely. Snake motion went away and most of the voltage dip went away.

His comment: "Keep the magnetic field contained - keeps the reactance down."

That was 40 years ago - I still can't calculate the VD caused from having the seperated/coiled phase conductors, but, it's there.

Yes, I do still wear a foil hat on occasion. Last time was for our End of the Mayan Calendar party

ice

Harmless flakes working together can unleash an avalanche of destruction

 

[IBRCAN]

In this case they were single conductor cables with no spool, so air core reactors. Most of our customers don't usually coil their cable...but when they do and everything works ok, they figure there must be no problem with doing so (worked just fine before!). Add in a couple other factors such as cold crusher oil etc. and coiled cables could mean the difference between a successful start and a frantic call for service.

Skogs: I put the AVR in "line loss compensation mode", which acts to boost generator terminal voltage with increasing load in order to compensate for feeder cable voltage drop when there is a particularly great distance between the generator and load. The fact that enabling this mode significantly increased the voltage at the soft start suggests that there was a large voltage drop across the cables.

wroggent: in a stand alone generator set-up such as this, the AVR changes field excitation to maintain generator terminal voltage.

In my experience, the starting pf for a crusher motor is typically .3 to .5, so the large amount of reactive current is only making matters worse. Good to see that others have witnessed this effect and it's not just an urban legend. Thanks to all for the info.

 

[jraef]

You know come to think of it, all of my experiences involved soft starters (because that's who was paying me at the time). So there may be something to the idea that even a small amount of added inductance can mean the difference between getting enough accelerating torque and not getting it, given a necessary current limit as controlled by the soft starter. In other words we often install a soft starter mainly for current limiting, especially on portable equipment powered by generators or in far flung locations with pre-existing voltage drop issues. So adding that last straw of coiled cables may be the back breaker in some instances, but not all.

"Will work for (the memory of) salami"

 

[Skogsgurra]

This could easily turn into one of those eternal threads.

wroggent: The harmonic contants doesn't matter at all. It is always the fundamental component that produces torque.

Re coiling: I think that the OP mentioned microhenries. That seems correct. And with microhenries, there should not be a problem. But the 152 microhenries are definitely a problem.

Cable "excess coiled up and hung on the side of the generator trailer". That implies (to me) that the coils were more or less concentric and that the inductive effect therefore should cancel. The plot tightens.

But, hey! What is this? "hung on the side of the generator trailer" - that may mean that one of the coils actually had ateel in its flux path and that may have increased the impedance of that coil.

And. To conclude. The inductance of an air coil with ten turns and three feet diameter seems to be around 200 microhenries. And that is a lot more than I thought. So, coiling may be bad after all. But should not be if you keep the coils concentric. If you do not hang them on a steel plate. Perhaps...



Gunnar Englund

http://www.gke.org

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

 

[waross]

I imagine that there are three inclined bars or pipes, one for each phase. Each phase cable will be hung in loops on one of the supports.
This is similar to welding leads hung on the side of a welding generator trailer. There may be little interaction or cancelling between phases and while there may be some iron effect, there will not be ferrous encirclement.
Is that about right Jeff?



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

 

[living2learn]

[152 uH for a coil 36 inches in diameter and 5 inches long with 10 turns.]

Just wondering where does the 5" long came from? The 36" diameter makes perfect sense as well as the number of turns. If the length corrosponds to the length of the wire from the beginning of the coil to the end, then the inductance would be way smaller.

 

[waross]

That is from a formula for form wound induction coils. Consider a coil wound on a 36" diameter form. The length of the coil on the form is 5", or the coil forms a cylinder 5" high by 36" in diameter.

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

 

[jraef]

Maybe "thick" would be a better term than "high".

"Will work for (the memory of) salami"

 

[waross]

I agree with you in the case of coiled cables Jeff. I was referring to the formula that hams use when winding single level, spaced, induction coils on a coil form. In the formula the length is not the length of the wire but the length along the form at right angles to the conductors.

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

 

[7anoter4]

Following waross idea, it is incredible how a single loop can double the reactance of 300 ft. single core cable of 600 V, 500 mcm. If we take Radius of 18"[36" diameter] of a loop using the approximate formula L[inductance]=miuo*N^2*Area/h miuo=4*pi()/10^7 H/m Area=pi()*(18*2.54)^2=6566.929 cm^2 and h=1.01*2.54=2.565 cm
L=3.21725E-05 H and Xloop=2*pi()*60*L=0.012128747 ohm
The reactance of entire 300 ft. of 3 phases , 500 MCM in trefoil arrangement it is only 0.009887112 ohm.(!).