Floating or grounded neutrals on motors and supplies 1

[Younsi]

Hello
I am interested in learning how industrial motors (1000HP to 50,000 HP MV to HV)and their supplies are connected. More specifically I would like to know if the neutral of the motors and the newtral of the supply side are grounded. Also how does the neutral connection/grounding approaches differ from Asia to Europe and North/South America ? Your help and opinion is much appreciated
Best regards
Karim

 

[waross]

Grounding of motor neutrals is a bad idea unless you want to sacrifice motor thermal capacity in an attempt to balance unbalanced supply voltages.
Supply grounding and grounding methods depend both on protection philosophy and the type of supply.
Wye or delta/ primary or secondary/ solid ground, impedance ground, corner ground, no ground, zig-zag grounding or wye:delta grounding.

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

 

[electricpete]

I agree with Bill. Within our plant auxiliary power distribution system our plant (US) we have a variety of supply systems: solidly grounded supply systems (480vac), ungrounded supply systems (480vac), high resistance grounded systems (4kv) and low-resistance grounded systems (13kv). In all cases, the neutral point of a motor wye connection is left floating. Bill points out one good reason, system unbalances would result in much higher motor current unbalance since the motor attempts to generate a balanced three phase voltage.

An interesting thing about motor dynamic analysis (such as Krause), it seems to assume the motor neutral is connected back to the supply system. This seems a very wide assumption for motor simulation but I haven't ever heard of matching practice.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[Younsi]

Many thanks to Bill and Pete. Your answers are most helpful. If one could develop a 1000 times more sensitive ground fault or stator insulation leakage current differential CT, on what motors could one use it ? I was askd this question at a recent IEEE on Motors and Gens Conf.
Thanks
Karim

 

[electricpete]

I think it would be very useful as an alarm feature for critical motors (or better yet if continuous leakage value is available, use it as a trendable parameter). That would be a useful warning that some investigation is required.

On the other hand, using a more sensitive trip function wouldn't benefit most applications - certainly where loss of motor impacts the process severely. Also with existing ground fault trip the damage is limited to the winding, not the core. The improvement I guess would be to be able to save the winding...I'm not sure if this benefit would offset the disadvantage of more trips.

Going back to the benefit of early warning, we had two instances of intermittent ground fault trips, after which motor tested fine and ran fine. These were warnings that something was going on.

One turned out to be a quarter inch hole in the groundwall insulation at the very bottom of the knuckle on bottom end of vertical motor. It never went hard ground and didn't show up on insulation tests, but when we opened it up we saw it. There had been water in the motor at some time and apparently dripping from knuckle to ground plane below created a fault that blew open that portion.

Another turned out to be a cable splice fault on 15kv cable for 13.8kv system. Actually the EPR cable tested 30 megaohms at 10kv, which we have come to accept as normal for these runs of cable 2000', sometimes wet. But between 10kv and 15kv it broke over to zero ohms.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[rockman7892]

Waross

Very intersting response, I was not aware of the two points you mentioned.

Can you quickly explain how grounding neutral point of motor helps balance unbalanced line voltages and how doing this incerases thermal heat in the motor?

 

[waross]

The back EMF of the motor is balanced both in voltage and phase angles.
If the line voltage is unbalanced the mismatch will cause excess current. This crrent will be limited by the system impedance and the motor impedance. This is why a small voltage unbalance causes a large current unbalance.
The unbalanced current is in a direction to correct the voltage unbalance however the system impedance is often so low in relation to the motor impedance that there is little actual correction of the applied voltages.
In an instance where a large motor is supplied by a transformer with not much more capacity than the motor requires and with a relatively high impedance the correction of a voltage unbalance may be noticeable.
My comment concerning an "attempt" to correct unbalanced voltages was somewhat tongue in cheek. The attempt will usually yield so little effect as to be futile. The extra current will result in more motor heating and rotor heating may be be an issue also.
I have seen a "soft" distribution line with a unequal currents, voltages and the resulting displaced neutral corrected to a large degree by the action of a wye:delta transformer bank with the primary neutral connected. The KVA size of the transformer bank was significant compared to the KVA of downstream unbalance. The action has some similarities to the action of a motor fed by unequal voltages. The relative sizes of motors compared to transformer banks results in the corrective action of motors to usually be much less than the possible corrective action of a large transformer bank.
Again, yes there is some correction, but no, don't plan on it, it is usually not enough to be useful.

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

 

[Younsi]

I understand then that we have a whole range of scenarios as far as neutral grounding on the supply side. With a floating one the differential current sensor shoudl not work as there is no return path of the leakage current to the supply.
Thanks for the info
Regards
Karim

 

[electricpete]

I'm sure you know, a standard motor differential can work with floating neutral - it just compares current flowing into line terminal of A phase with current flowing out of neutral of A phase (into the other two phases). If there is a ground fault within the motor or phase-to-phase fault within the motor, a difference is seen.

Also I am pretty sure the grounded motor neutral is very rare if it exists at all. I wonder what NEC would say about it? (I don't know).

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[electricpete]

Clarification in bold:
"Also I am pretty sure the grounded motor neutral is very rare if it exists at all. "
should read
"Also I am pretty sure the externally-connected motor neutral is very rare if it exists at all. "

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[electricpete]

I guess delta-wye starter with external running leads contradicts that. Really what I meant to say is I haven't ever heard of connecting motor wye point to either ground or to any system neutral.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[waross]

I would let both statements stand, Pete, even with a star delta starter.
If a motor is large enough that differential protection is desired, it will be ordered with all winding ends accessible and suitable for installation of CTs.

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

 

[electricpete]

I voted for you in the IEEE DEIS election Karim. Maybe you can bring some more focus to rotating electrical machines (seems like most of the interest is in transformers and smartgrid).

Anyone else in IEEE DEIS... you've got until 9/29 to vote.

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

 

[waross]

Once the differential CTs are sensitive enough to pick up the capacitive leakage current, there may be no point in further improvement in sensitivity.

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

 

[electricpete]

I thought the basis for the minimum differential setting is CT accuracy, rather than capacitive leakage current.

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

 

[waross]

Accuracy is good but when your setting must be above the level of the capacitive current, how many orders of magnitude more of resolution is really useful, or am I missing something?

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

 

[electricpete]

All I was saying was I was under the impression errors you have to assume for the CT’s (including steady state and transient) are typically much higher than the cap current. Maybe if you have very very long power cables inside the zone that might not be true(?).... but for our differential-protected motors the CT’s are in the motor term box.

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

 

[electricpete]

Correction:
"cap current"
should have been
"capacitive current"
(I am assuming no capacitors connected to ground of course)

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

 

[waross]

Hi Pete;
I was responding to this particular post:

Many thanks to Bill and Pete. Your answers are most helpful. If one could develop a 1000 times more sensitive ground fault or stator insulation leakage current differential CT, on what motors could one use it ?

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

 

[electricpete]

Aha. Sorry I had no idea of the context of your comments. Now it makes perfect sense. My apologies.

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