Star delta starter overvoltage 2

[Masbibe]

Hi guys,

While I was reading some other post one thing come on my mind.

There is induction motor connected for star delta start with classic scheme (three contactors: main, star and delta).

Firstly main contactor and star contactor close their contacts. Motor is accelerating and after some time star contactor open its contacts.
Then, before delta contactor close its contacts, as it is well known rotor current will continue to flow decreasing exponentialy down to zero.
Decreasing will depend on rotor time constant L/R.
This rotor current will induct some voltage on stator windings with frequency of rotor speed.

So this is the point:
In this situation motor terminals U1,V1,W1 will be on network potentials (because main contactor remain closed). On stator windings you will have in first moments voltage that is near to nominal motor voltage and this voltage will lagging in relation to network voltage. So in some next few moments (when angle between network and stator voltage reach 180 degrees) you will have approximatly double nominal voltage on motor terminals U2,V2,W2 (depending on rotor R/L and actual rotor slip).
This can be problem for motor, cable and contactors isolation.

I am interesting is this problem is well known in literature. Are there problems with isolation more often then usually on star delta starters.
Is this a thing that you must be careful or nobody care for this.

Milovan Milosevic

 

[amptramp]

I am not sure I understand your question, but maybe you should investigate "closed transition" Y-D starters. The closed transition starter uses resistors to mitigate the inrush currents with their associated electrical and mechanical stresses that occur during the transition to the Delta connection.

See if this link is relevant to your question:

http://www.mastercontrols.com/Articles/Wooddall/TA_RWood.htm

 

[Masbibe]

amptramp,

In other word this is what I meant to say. In transition time when you change connection from star to delta, to be more precize when main contactor is closed and star and delta are opened potentials of motor terminals U2,V2,W2 are:

Vu2 = Vu1 + Eu2_u1
Vv2 = Vv1 + Ev2_v1
Vw2 = Vw1 + Ew2_w1

Vu1,Vv1,Vw1 - potentials of motor terminals U1,V1,W1 - frequency=line_frequency, voltage=line_voltage (because main contactor is closed this potential are equal to line potential)

Eu2_u1,Ev2_v1,Ew2_w1 - EMF on stator windings (inducted because of rotor current) - frequency=(1-slip)*line_frequency, voltage~line_voltage

So when when Vu1 is in phase with Eu2_u1 (or oposite with Eu1_u2) you will have nearly double nominal voltage on terminal U2.
same thing is for other 2 stator windings.

I hope now is clear.

Thanks for link, I hope I will find something interesting.

Milovan Milosevic

 

[jraef]

MilovanSRB,
What you observed is true. You can indeed have what are called "transition spikes" in an open transition start-delta starting scheme. There are ways to mitigate it to a small extent, but that is the chief drawback to using that technology. Not only can you get voltage spikes, but they can and do then cause torque spikes, sometimes great enough to shear motor shafts.

Is it known? Yes. Is it discussed? Not widely (although we have discussed it several times in here). Should it be a concern? Yes. Do suppliers inform potential users of the risk? Not often enough.

There is a great white paper written about this that used to be available from a supplier of Fire Pump Controllers, used as a tool fr convincing people not to use open transition Wye-Delta starters on critical equipment such as fire pumps because of the risk of failure of the equipment. I'll see if I can resurrect that link, I have posted it here in the past.


"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]

Here is the thread with the whitepaper link (I think) that jraef mentioned:

http://www.eng-tips.com/viewthread.cfm?qid=29369

 

[Masbibe]

Thanks guys,

I already know for problems with current and torque peaks when delta contactor close its contacts I also find some papers about this. But I never heard about this voltage problem in any literature. I was find out this phenomen while I was thinking on some other post.
In one factory that my company was on commissioning we had lot of star delta starters (maybe about 20). All of this star delta starters had circuit breakers with only magnetic release (overload protection was in ABB reley UMC who also control opening and closing delta and star contactors with fixed transition time from opening star to closing delta I think maybe 50ms). And on many of them during transition from star to delta this breaker was triping (I cant remember exactly on which current this magnetic part was tripping but probably not below 7-8 * In). So they must always send one electrician guy to hold some spring with screwdriver to avoid trip during start (this was our automatic start with one man per motor). After that problems we replaced all magnetic breakers with calassic fusses and there was no problem any more (total energy impuls during transition is low but current peak is big).




Milovan Milosevic

 

[jraef]

I have had success in some of those situations by "rolling" the conductors on the main contactor (K1). Rolling means you shift the conductors one direction or the other, but maintain the same relative rotation. It creates a phase shift in the transition that can allow natural differences in motor winding resistance to dampen the spikes just a bit, which is often enough to mitigate the nuisance tripping. MCS, the people in that link who publish those papers, used to sell a little device called a "Leading Phase Indicator" that would tell you the best roll direction to make. But you have a 50-50 chance anyway, so just be prepared to do it again if it didn't work the first time.

Now I just talk users out of Star-Delta and into using soft starters. They are so much less expensive than they used to be that even though Star-Delta is still a little cheaper, the difference between them makes it not worth the risk.


"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

 

[Masbibe]

jraef,

Just to clarify one thing.
This is my observation.
You can connect windings of stator bassicly in two different ways in relation to star connection.
Phase voltages of delta connection can lag or lead for 30 degree in relation to phase voltages of star connection.

Current peak that occur when you switch on delta contactor depend on: line voltages in moment of oppening star contactor, line frequency, puase bettween opennig star and closing delta contactor, rotor speed, inertia and load.

So best is to use one of this two delta connecton which will give less angle between line voltages and inducted voltages on stator in moment of closing delta contactor because current peak will be smaller.

When motor is already connected for Y-D start you can change delta connection by "rolling" conductors on main contactor (but not together with conductors of delta contactor) or somewhere else (delta contactor, terminals U1,V1,W1 or terminals U2,V2,W2).
Rolling can be done in just one direction (depending on delta connection) because when you roll in other direction you will get that all windings are on zero voltage. In other words this would be potentials on the end of windings:
U1-Lx, U2-Lx, V1-Ly, V2-Ly, W1-Lz, W2-Lz (Lx,Ly,Lz are L1,L2,L3 respectivelly or not depend on delta connection).
So there will be no current (after rotor current fall to zero).

There are no possibility to know which of those two connection is better until you tray (or there are possibility but you must know a lot of things which you ussually dont know).

Of course you can affect on transient current peak by other means (time of transition, pause of transition)

Please for comment of anybody. Is there any mistake above.

Milovan Milosevic

 

[RajeevCell]

MilovanSRB, you said,
"After that problems we replaced all magnetic breakers with calassic fusses and there was no problem any more"

Replacing magnetic breakers with fuses is not a good idea for motors. There's risk of single-phasing.

Were your motors very small? Why didn't use DOL instead?

 

[Masbibe]

RejeevCell,

We had ABB reley UMC (Universal motor controller) that measure all three line current so we had phase loss protection and overload protection (UMC control star delta start and open contactors in case of overload or phase loss fault). Circuit breaker (later fuses) was used only for short circuit protection.

Thanks for response.

Milovan Milosevic