induction motor type of connection

[aletric]

dear every body
regards
suppose two induction motors both(22kw) , one have star connection at 380 vac , and the other have delta connection at 380 vac .
the question is which one can consume high current at start up at same mechanical load , or what is the expected results
thanks for reply

 

[jraef]

We can't answer that without knowing if the motors were DESIGNED to be connected in Star or Delta...



"Dear future generations: Please accept our apologies. We were rolling drunk on petroleum."
— Kilgore Trout (via Kurt Vonnegut)
For the best use of Eng-Tips, please click here -> faq731-376

 

[aletric]

Dear mr. jraef
thanks for your reply
the two motors in my question, one is designed to be connected in star at 380VAC-60 HZ
and the other motor is designed to be connected in delta at 380VAC -60 HZ
my question is , at same mechanical load , which motor will consume high current in start up
in other words , are there any relation between the type of connection of induction motors and the starting up
current at same conditions.
thanks,,

 

[waross]

I don't believe that there is any difference. Delta connected NEMA motors were once more common in North America than they are now. When trouble shooting we never noticed any difference between delta wound motors and star wound motors. Note that trouble shooting almost always included monitoring the starting current.

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

 

[zlatkodo]

Generally speaking, if both motors are designed in the same way (with the same values ​​of flux densities in the air-gap, back-iron and teeth), then there is no any differences at the same load.
Differences may occur only if the selected winding has a different harmonic content in "wye" and "delta".
Zlatkodo

 

[zlatkodo]

My previous answer is related to symmetrical three-phase windings.
In the case of asymmetrical three-phase windings, currents in delta connection are much higher (due to circulatory currents) and this is why the "wye-connection" should be preferred.
Zlatkodo

 

[ScottyUK]

Hypothesis:

The potentials applied to the windings of a delta-connected machine are defined by the supply characteristics. In a star-connected machine the star point is an internal connection and is not connected to system neutral; the potential applied to each winding of a star-connected machine is therefore not solely defined by the supply but also by the impedances of the interacting magnetic circuits of the stator windings and the rotor. I expect that the neutral potential will move around a fair bit while the stator magnetic field reaches a stable state.

In both cases the point on the voltage wave at which the supply was energised will have an effect on the behaviour at initial switch on, and there will be saturation effects due to transient DC components. It seems fairly likely that there will be measurable differences in behaviour between a star connection and a delta connection, although whether they are of any practical significance is an entirely different question.

There are better analytical minds than mine who might care to model this; magnetic circuits on the edge of saturation or possibly in saturation are one step too far for my Sunday evening.

 

[electricpete]

In a star-connected machine the star point is an internal connection and is not connected to system neutral; the potential applied to each winding of a star-connected machine is therefore not solely defined by the supply but also by the impedances of the interacting magnetic circuits of the stator windings and the rotor. I expect that the neutral potential will move around a fair bit while the stator magnetic field reaches a stable state.

For simulating motor starting, Krause (who seems to present a standard... most others seem to followo his approach) assumes that wye motor neutral voltage remains the same as supply system neutral voltage.

It can be proven correct (transient or steady state) using Laplace analysis under assumption of linear behavior
For real motors with non-linear behavior during start, it's an unknown to me.

Was discussed a little here:

http://www.eng-tips.com/viewthread.cfm?qid=251952

On the original question, I agree with other posters fwiw... If there is any difference attributable to connection, it would be related to dc and harmonic components. If the interest of original question is simply the fundamental steady locked rotor currrent (excluding dc), these differences wouldn't matter.


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

 

[jraef]

Yes, I suppose it will depend on the definition of "at startup" in the original query. Is it the first few cycles, or the first few seconds, or the first few minutes? Cycles, maybe, but anything longer, not likely.

Aletric, are you asking this because of a specific problem you are having, such as a breaker tripping on startup?

"Dear future generations: Please accept our apologies. We were rolling drunk on petroleum."
— Kilgore Trout (via Kurt Vonnegut)
For the best use of Eng-Tips, please click here -> faq731-376

 

[ScottyUK]

Hi Pete,

Yes, it is the non-linear behaviour which I was thinking of when the winding see the initial DC offset from switching. I am sure it can be modelled, but only with a fairly sophisticated model of the magnetic circuits. I have so far drawn a blank in my brief attempt to find any detailed investigation of what happens, perhaps because it is of little interest other than purely academic curiousity.

 

[electricpete]

There was some related discussion here:
thread237-285699
If we consider only the magnetizing flux linkage (which must sum to zero among three phases), then the instantaneous neutral voltage (including during transient) would be the average of the three phase voltages (using any voltage refernece for all 4 measurements... take your pick).

But the leakage flux linkage does not have the same requirement (sum to zero among three phases), so certainly seems to introduce potential for the neutral voltage to vary if the leakage reactance behaves non-linearly.

I have gone some distance toward trying to simulate this in the past, but never completed. It's not as easy as you'd think but I'm sure it can be done.

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