×
INTELLIGENT WORK FORUMS
FOR ENGINEERING PROFESSIONALS

Log In

Come Join Us!

Are you an
Engineering professional?
Join Eng-Tips Forums!
  • Talk With Other Members
  • Be Notified Of Responses
    To Your Posts
  • Keyword Search
  • One-Click Access To Your
    Favorite Forums
  • Automated Signatures
    On Your Posts
  • Best Of All, It's Free!
  • Students Click Here

*Eng-Tips's functionality depends on members receiving e-mail. By joining you are opting in to receive e-mail.

Posting Guidelines

Promoting, selling, recruiting, coursework and thesis posting is forbidden.

Students Click Here

Jobs

Switchgear fault after start of motor
2

Switchgear fault after start of motor

Switchgear fault after start of motor

(OP)
This event comes from a nameless anonymous generating facility.

We have studied this event and drawn some preliminary conclusions altthough can't quite explain everything. I offer some of the facts for your consideration. Let's see if you come to same conclusions as us.   

System Description (see slide 1)

Motor data: 7000 hp, 13,200v horizontal sleeve bearing motor driving centfigual pump. Locked rotor current is 1560A at 100% voltage.   DOL start.

Electrical system is shown on slide 1.  The motor in question is fed from 13.8kv standby bus 2C, which is fed via tie breaker from aux bus 2C, which is fed from X winding of 25/13.8kv transformer.

Upstream Transformer data:  LTC on hi-side. 20% impedance at 84mva.  Delta/grounded wye.  secondary neutral point is grounded through 20 ohm resistor.  That neutral resistor current is sensed by bus time-overcurrent relay which never tripped during this event.

Motor breaker data: GE Magne-blast circuit breaker

Cable between bus and motor: 800 feet of 750 mcm single-conductor cable.

We have no current traces available for this system but we do sense/record voltage at 1-second sample intervals.  This voltage is sensed from open-delta / grounded 2/3 wye PT's connected to the winding that corresponds to primary phases B to C.

Description of "normal start" (slide 2)
Prior to start the voltage is at it's normal value of 14,100.
The motor draws the high starting current for approximately 4 seconds before it gets up to speed.
During the period when motor is starting, the 13.8kv buses drop from intiial 14,100 volts down to 12,900 volts.

Analysis:  this voltage drop from 14,100 to 12,900 is about exactly what we expect based on transformer impedance, assuming the LTC does not respond in this short time interval.  I calculated cable voltage drop and it is small comparison... I think voltage at motor terminals during normal start is 12,700 – 12,800 volts.

Raw data for fault start (slide 3)
The raw data is there for your review.  I think it cannot be explained unless we consider there is a 2 second error/difference in time-stamps for different channels recording things that occurred at the same time. Normally I would think that 1-sec samples can be at most 1 second off relative to each other, but these voltages are recorded by different CPU channels and our computer engineer says that events that occur virually simultaneously on different CPU's may be time-stamped up to 2 seconds apart

Fault start with time-scale adjusted for simplicity(slide 4)
Timeline:
Notation: "timeA/timeB" to denote event that occurs during 1-second interval between timeA and timeB
Up until 10:21:53, we see normal pre-start boltages
At 10:21:53/10:21/54, the pump breaker closes and all voltages change in a manner to a similar start.  During the next 4 seconds, the pump discharge pressure increases in a manner identical to a normal start.  OPerators report hearing the motor come up to speed.

At 10:21:57/10:21/57, the pump breaker opens (indicated by aux contact).
Following breaker open, the bus voltages do not recover to their initial values as they do during slide 2 normal start, but they gradually recover toward normal over the next 8 seconds.

At 10:21:06/10:21:07, all buses drop to a very low value, and the tie breaker opens (tripped on time-overcurrent... the only protective device that actuates this breaker)

At 10:21:07 / 10:21:08, all buses except Sby 13.8 and 4kv 2C return to normal. This indicates that tie breaker opening removed the fault for all buses except those two buses which are now isolated.

At 10:21:14 – 4kv 2C bus voltage returns as expected when energized by alternate source not shown on this drawing.

Post event-inspections
[*]The motor breaker cubicle door is blown open and cubicle is bulged outwards. The breaker is slightly cocked
[*]All motor relays indicate red flag, however these are electromechanical relays prone to flagging upon vibration.
[*]The breaker cubicle includes 3 compartments: breaker cubicle, line side compartment, load side compartment... all 3 compartments have heavy soot although heaviest in the breaker compartment.
[*]Line-side to load-side checks of breaker confirm it is open
[*]good megger checks cannot be obtained due to heavy soot.
[*]The only evidence of actual fault to ground or phase-to-phase is in the breaker cubicle.
[*]Motor breaker inspection: A and C phase arcing contacts have moderate wear although not significant enough loss of material to affect their function. In contrast B phase arcing contact surface looks pristine.   A and C phase arc chutes show lots of molten material, in contrast B phase looks pristine.  All 3 phase main contacts show not much damage except possible evidence of overheating (dull finish).   There is evidence of heavy arcing to ground at all three phases directly behind the arc-chutes (the arc chutes open toward the back of the breaker).  There is also possible slight arcing to ground at the front below bottom of arc chuts although no damage to integrity of arc chutes.
[*]Tie breaker inspection (the one that tripped second).  B and C phase main contacts show pitting. All arcing contacts and arc chutes in good conditions.
[*]Relay testing – all motor cubicle relays were tested and responded in specification.
[*]CT testing – all CT's meggered sat and passed excitation test
[*]Motor testing – All testing passed. Insulation resistance 10+gigaohms with PI > 2. DC step voltage test to 24kvdc with phases separated. Performed surge test to 15kv (test set limitation) - sat.  Winding insulation resistances differ by less than 1% - sat.   Motor terminal box inspection shows nothing unusual.  Endwinding inspection shows nothing unusual (no evidence of internal fault or movement from motor feeding a fault outside of the motor)
[*]Cable testing – 10+megaohms. Visual inspection shows nothing unsual.

Discussion
I do have more data avaible if desired – breaker photos, complete info on motor protection, almost-complete information on system protection, voltage PT diagrams, etc etc. Didn't want to go into info overload... I provided what I thought was relevant although of course I may be missing significance of some things.

One thing that we struggle with is what caused the breaker to go open to begin with: everything in the protection zone downstream of the protective CT's tests in the load side compartment of breaker cubicle tests sat.

Interested to hear your thoughts on partial initiators (what caused the motor breaker to go open) as well as what happened after the motor breaker went open.

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

RE: Switchgear fault after start of motor

(OP)
Also I should mention the as-found condition of breaker primary disconnects stabs was firmly seated. The force of event seem to have moved the breaker off of it's side-rails but it was held in place by the primary disconnects. Inspection of these disconnects showed nothing unusual

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

RE: Switchgear fault after start of motor

(OP)
The breaker positions (motor and tie) are shown on the right scale (0 or 1)

For motor breaker: 0 means open, 1 means closed (like a breaker A contact).  The 2-second interval when this item is 1 indicates breaker closed for 4 seconds based on aux contact position.

For tie breaker: 1 means open,  means closed (like a breaker A contact). The breaker is closed at beginning of the event and transitions to tripped at 10:21:05.

My apologies for not adjusting these to provide consistent notation for breaker position.  

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

RE: Switchgear fault after start of motor

(OP)
Also the arc chute return straps (steel) that carry the arcing contact current were melted in all 3 phases.  A and C melted at both ends and fell out... B phase melted near middle and remained in place.

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

RE: Switchgear fault after start of motor

Could the close latch be slightly worn allowing the contacts to pop open during starting inrush?

RE: Switchgear fault after start of motor

(OP)
Sort of like a contact bounce?  I'll have to think about that.

The basic scenario we are thinking about is:

[*]10:21:53/10:21/54 Motor starts

[*]10:21:57/10:21/57  (just before reaching normal speed) – motor receives trip signal for some unknown reason (failure #1) and motor breaker opens.

[*]Immediately after contact opening, phases A and C restrike for some unknown reason (failure #2) and continue to carry current through plasma, single-phasing the motor.

[*]Voltage rises slowly either due to motor slowly accelerating or due to tapchanger response.

[*]The arcing within the breaker increases and eventually goes phase-to-phase to ground, resulting in tie breaker trip at 10:22:06/10:22:07

Failure #1 is initial trip of motor breaker – have no idea why since everything tested good downstream of breaker.

Failure #2 is motor breaker phases A and C failure to interrupt once it interrupted, as evidenced by lots of molten material in A and C phase arc chutes.    Why?  I have no idea. One thing is that we don't' normally open the motor under locked-rotor-current.   The nameplate says "rated short circuit amps" are 28KA, which I assume means it is rated to interrupt 28KA.  Locked rotor current is much less at 1500A, but it is also inductive.
 

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

RE: Switchgear fault after start of motor

Have you checked all of the adjustments on the breaker? There are 9 sets of adjsutments on a Magneblast and pretty much all of them could cause this sort of failure either directly or indirectly.  

RE: Switchgear fault after start of motor

Electripete...

Per your original post, you are reading an aux contact as an indication of the breaker opening; this reflects mechanical position of the breaker, but does not imply (at least in my mind) that a trip signal was recieved.

The 28 kA is interrupting rating; this is different than the close and latch rating, but may not be relevant anyway.  These ratings are based on a breaker in perfect operating condition.  A bit of wear in the wrong place can drastically afffect in-service capabilities, as opposed to ratings.

As I understand it, closing contacts attempt to separate due to the forces of the arc; a certain amount of mechanical force is needed at the contacts to maintain metal-to-metal contact at a certain close and latch Amperage rating.  If wear or misajustment allows the close latch to not quite lock completely, the escalating arc at one or more contacts will force the breaker back open.

By now, there is too much vaporized metal in the arc to snufffed by the arc chutes...  Or so I imagine the sequence to have been.

RE: Switchgear fault after start of motor

(OP)

Quote:

Per your original post, you are reading an aux contact as an indication of the breaker opening; this reflects mechanical position of the breaker, but does not imply (at least in my mind) that a trip signal was recieved.
I left out one piece of info.  In addition to all the electromechanical relay flags which we observed and attributed to direct vibration affects, 86 relay actuated.  This 86 relay is a stout rotary relay which cannot be directly actuated vibration. The only way it could have actuated in absence of trip is by closing of one of the protective relay contacts that feed it.  It is possible but not probable in our opinion that the vibration caused closure of one of the protective relay contacts.  In any case, you're right... could have been a trip or could have been something else involving breaker, switch, etc.

Quote:

Have you checked all of the adjustments on the breaker? There are 9 sets of adjsutments on a Magneblast and pretty much all of them could cause this sort of failure either directly or indirectly.   
The breaker is in bad condition from the fault. We are going to attempt to do some checks including time-travel checks, but don't have a lot of hope that the as-found condition of the breaker represents the condition at time of fault since the breaker has been through a lot of heat and mechanical agitation.

 
 

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

RE: Switchgear fault after start of motor

I would check to see when the last time the adjustments were checked. Some testing companies ignore the adjustments, which are vital to proper operation.

RE: Switchgear fault after start of motor

(OP)
I will check on that.  The categories of malfunctions which could cause restrike would be limited to:
1 – arcing contacts open to slowly
2 – After opening, main or arcing contacts bounce back too far off of their open causing clearance momentarily too low
3 – arc chute fails to interrupt due to contamination, moisture, coil problems etc.
Is that correct or have I left some out?
 

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

RE: Switchgear fault after start of motor

(OP)
One thing that was mentioned about puffer style breakers such as these is that you only get one puff which occurs during opening. So if there is a restrike for some reason, it will likely continue and the breaker has no more capability to extinguish it without that puff

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

RE: Switchgear fault after start of motor

(OP)
If anyone is interested, the principle of interruption for these breakers is discussed on pages 7 and 8 of GEI
88764 (I will attach them if I can figure out a way to excerpt those pages... the entire document is too big to attach).   The combination of air and electromagnetic forces push the arc toward the back of the arc runners where the two arc runners (upper and lower) are further apart. It is similar to the principle demo'd by a "Jacob's ladder" which the arc walks along the two electrodes in the direction of increasing spacing. The coils somehow enhance this effect.  

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

RE: Switchgear fault after start of motor

Hence the name mangeblast. Have you check the operation of the coils and puffers?

What type of operating mech do you have?

Red Flag This Post

Please let us know here why this post is inappropriate. Reasons such as off-topic, duplicates, flames, illegal, vulgar, or students posting their homework.

Red Flag Submitted

Thank you for helping keep Eng-Tips Forums free from inappropriate posts.
The Eng-Tips staff will check this out and take appropriate action.

Reply To This Thread

Posting in the Eng-Tips forums is a member-only feature.

Click Here to join Eng-Tips and talk with other members!


Resources