why rotor rub more likely on starting 2

[electricpete]

I read that the rotor is more likely to rub the stator due to unbalanced pull during motor starting than during motor running. Can anyone explain why that is?
Thx.

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

 

[electricpete]

The reason I ask the question is that I usually think of unbalanced magnetic pull as associated with the attractive force from air-gap field, independent of the currents. Since air gap field is lower during starting, no reason to suspect higher force.

One factor that seems obvious at first is higher current. Perhaps interacting with circumferential cross-slot component of flux. But it seems like that would be equal and opposite on opposite sides of the rotor.

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

 

[jbartos]

Suggestion: The motor shaft load vibration during the starting, being bigger, should not be ruled out. Concentrating on the motor itself, the high motor inrush current in conjunction with the motor rotor manufacturing tolerances causes higher vibrations.

 

[edison123]

Air gap field is constant from the moment rated voltage is applied. I have seen more rotor rubs during running (thermal bow etc.) than during actual starting. Probably, in sleeve bearing motors, due to lower air gap at the bottom during start up, one could see some rub but then that would be due to improper design or assembly.

 

[RalphChristie]

Pete

Just thinking of a thing, probably I'm totally wrong..

One factor that seems obvious at first is higher current. Perhaps interacting with circumferential cross-slot component of flux. But it seems like that would be equal and opposite on opposite sides of the rotor.

Due to the small difference in the rotor winding impedances, and also the stator winding impedances (they will never be precisely the same) the current through each winding will also differ with a small amount. I think how higher the current, how bigger will the imbalance be.

Therefore the forces on the opposite sides will not be the same, with the biggest difference when there is a high current, like an inrush current, during start-up.

But, like I said, maybe I'm totally wrong...

Ralph

 

[SteveKW]

I agree with Edison. Common causes I see,
Bad Bearings.
Bad bearing fit on shaft and/or endbell.
Loose rotor.
Loose stator.

 

[jraef]

Injected conjecture...

Staor windings twist ever so slightly due to magnetic stresses during across-the-line starting. If the motor is random-wound, it might make a difference in the magnetic force application, contributing an inequality of the aforementioned flux(?)

Seems a stretch but I have seen visible movement in lage motor windings on startup...

Quando Omni Flunkus Moritati

 

[vanstoja]

M.Bradford,1968,"Unbalanced Magnetic Pull in a 6-Pole, 10kW, Induction Motor with a Series-Connected Staor Winding", Electrical Research Assn. Report No.5216 (summarized in Proc.IEE,Vol.115,1968,pp.1619-1627) reports tests of wound rotor and squirrel cage rotors with 10% rotor eccentricity at 80,100 and 125% rated voltage in which he finds that starting transient UMP forces are higher than no-load UMPs by ratios of 1.57, 1.35 and 1.34 for wound rotor and 9.0, 6.7 and 4.1 for the cage rotor at the three voltage ratios. He says that the main differences between the magnitude of transient UMP in cage rotors and wound rotors is the the presence of multiple parallel paths in the circuit of the cage rotor which allow circulation of equalizing currents set up by the nonuniform airgap flux density distribution. The equalizing currents, by Lenz's Law, are in such a direction as to maintain a more uniform distribution and hence reduce the magnitude of UMP. In the cage rotor motor , the equalizing action of the cage is almost negligible during starting giving transient values of UMP nearly equal to the higher absolute values of the wound rotor motor.
The 10% rotor eccentricity condition at motor startup would not be at all unusual for a vertically oriented motor with fluid film radial and thrust bearings because, on coastdown to a dead stop on the thrust bearing, the rotor axis could have any eccentricity up to 100% with plain sleeve radial bearings or even 120% plus in pivoted pad bearings depending on the number of pads. With rolling element bearings, vertical rotor offsets would be controlled more by shaft deflection than by eccentricity in the bearings and 10% might be abnormally high.

 

[aolalde]

vanstoja:

I am confused reading this part of your report:

“He says that the main differences between the magnitude of transient UMP in cage rotors and wound rotors is the the presence of multiple parallel paths in the circuit of the cage rotor which allow circulation of equalizing currents set up by the nonuniform airgap flux density distribution. The equalizing currents, by Lenz's Law, are in such a direction as to maintain a more uniform distribution and hence reduce the magnitude of UMP. In the cage rotor motor , the equalizing action of the cage is almost negligible during starting giving transient values of UMP nearly equal to the higher absolute values of the wound rotor motor."


Shouldn’t this equalizing action be in the circuit of the wound rotor, instead of cage rotor?

I have read that increasing the number of parallel paths on the stator winding reduces the UMP.

 

[electricpete]

I have also heard that in general multiple parallel paths in either rotor or stator decrease ump.

It sounds like for some reason the squirrel cage rotor (multiple-parallel paths) provides less effective damping action during starting. I don’t know what that reason would is.

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

 

[jbartos]

Suggestion: Visit

http://www.ecmweb.com/ar/electric_watch_burnouts_hardtostart/

for: a bad bearing permitting the rotor to rub on the stator,...

http://www.pdma.com/Rotortest.html

for:
When this insulation has been damaged, high currents can flow through the iron, doing no real work but creating excessive heat. These high temperatures lead to abnormal thermal growth creating imbalances, and worst case, rotor / stator rubbing.
etc. for more info

 

[vanstoja]

Aolalde, electricpete
I'm not qualified to resolve the issues of damping current effects on transient UMP in cage and wound rotor circuits. However Bradford says in another discussion ("Unbalanced Magnetic Push", Proc. IEE, Vol.115, No.10, Oct.1968, pg.1522) of a letter addressing magnetic levitation effects on UMP (Freeman,E.M.Laithwaite, "Unbalanced Magnetic Push", Proc.IEE,Vol.115,No.4, April 1968,pg.538) that "The wound rotor motor was entirely series connected and thus the measured force is not reduced by damping currents." Further on he says "In the cage rotor, damping currents in the cage winding reduce the magnitude of the eccentricity fields. Damping currents do not exist in the stator winding." He
also finds that the calculated levitation force of <25N for his tested wound rotor motor is insignificant compared with the measured value of 1190N during startinf and thus can be ignored in the calculation of unbalanced force in usual induction motors.

 

[jbartos]

Suggestion: The following reference includes the motor eccentricity, torque cogging, etc. in terms of motor vibration.
IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 37, NO. 6, NOVEMBER/DECEMBER 2001 1727
Comparison of Vibration Sources Between Symmetric and Asymmetric HDD Spindle Motors With Rotor Eccentricity

Sang-Moon Hwang, Member, IEEE, Kyung-Tae Kim, Weui-Bong Jeong, Yoong-Ho Jung, and Beom-Soo Kang

 

[vanstoja]

From several papers on UMP, it appears that parallel circuits in either stator circuits or rotor circuits (squirrel cages)reduce UMP loading and have additive effects. Papers by Tenhunen of Helsinki UTech at

http://lib.hut.fi/Discuss/2003/isbn9512266830/article2.pdf

and the same except /article4 discuss UMP with eccentric rotors and provide further insights into current equalizing effects. These were found in an aol/google search of &quot;Unbalanced magnetic pull&quot;.
I've seen discussions indicating that startup suppression of current equalizing effects is related to reluctance effects.
Pertinent to UMP induced rotor-stator impacts at startup,Bradford's ERA report has an appendix II on &quot;Calculation of Transient Vibration of the Rotor&quot; which is not mentioned in his IEE paper. He determines UMP force (negative) spring constant, Kf, from measurements of transient UMP at known eccentricities and shaft elastic restoring stiffnesses, Ke, (with and without stator
deflection softening effects) and determines a stability constant Ks=Ke/Kf. The condition to prevent rotor touching the stator is (Ks+1/Ks-1)*sigma < g where sigma is initial rotor displacement from the center of the stator and g is the mean length of the airgap. For his 10 kW wound rotor test motor with 15.7 in. stator OD, 9.0 in. rotor OD, 4.5 in. shaft OD, 6.0 in. core length and 0.030 in. air gap length, he finds the maximum permissible startup rotor eccentricity to be 62% with stator deflection softening (about 28% with an infinitely stiff stator) and the vibration frequencies to be 123 Hz and 138 Hz with and without stator deflection softening. Since the measured transient UMP for the cage rotor motor was (at 10% eccentricity) 269 lbs. vs 284 lbs. for the wound rotor
motor, the cage rotor machine would not vary much in permissible eccentricity. Frequency differences may depend on rotor core stiffenening effects on shaft deflection which Bradford does not discuss. All of Bradford's numbers are in British Imperial units, not US units.

 

[jbartos]

Suggestion to the previous posting: The posted link appears to be forbidden.

 

[vanstoja]

It was not forbidden via the search category &quot;Unbalanced magnetic pull&quot; Try that and look for Tenhunen's name possibly on the 2nd or 3rd page of the listings. The isbn number appearing in the link indicates that the two Helsinki UTech papers are from a book of unknown title. This is not the first time for me that an exactly reproduced link from an aol/google search could not be accessed from eng-tips. I don't know why that happens. Has anyone got an explanation?

 

[electricpete]

I think that you are accessing some material through ProQuest which verifies your subscription by examining your ISP.

We can get access to the root page:

http://lib.hut.fi/index-en.html

There are many electronic journals indexed but they require password to access.

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

 

[jbartos]

Comment: HUT library’s Ewelib e-books (insists on authentication).

 

[gordonl]

I haven't reviewed all posts in the thread, but on long steam machines (from pedestle to pedestle) the rotor settles into a bowed position and must be brought up to speed slowly to allow warm up and the bow to fall out or else rub occurs.

The temperatures in steam rotors wouldn't compare to that of motor rotors, but perhaps it occurs to a smaller extent.

Food for thought.

 

[jbartos]

Suggestion: With certain number of poles and of stator and rotor slots in cage motors, peculiar and deleterious behavior may be observed when the machine is started. For example, with the number S1 of stator slots equal to the number S2 of rotor slots, the machine may refuse to start at all, a phenomenon called cogging. In some cases, excessive vibration may be set up at subsynchronous speeds, due to pulsation of mechanical force in the teeth, generating noise, e.g. medium pitched howl. The excessive vibration may cause the rotor rub on starting.