Starting a 4500 hp motor while it was coasting down 2

I work in a refinery. We recently inadvertently tripped a 12 kV circuit. The 12kV circuit was reclosed within 30 seconds. One of the loads on that circuit is a 4500 hp motor that is a driver for an air compressor. The motor has a breaker for a starter. The motor's breaker never opened during the outage. Its trip/close power is from the substation. Thus, one of the loads that was picked up as soon as the breaker was reclosed was the 4500 hp motor. When the tripped circuit was re-energized the motor was still coasting down. The motor's normallly runs at 1800 rpm. At the time the breaker was closed it had coasted down to 450 rpm. My questions are, should I change the motor control circuit so that upon loss of main power to the motor to trip the motor? If so/or not so why?

 

[jbartos]

Suggestions:
1. Verify whether or not the motor was designed for a full voltage across its terminal start. If you do not have any soft or reduced voltage-starting arrangement, then it was.
No action needs to be taken.
2. If you happen to have a reduced voltage starting arrangement, then the situation needs to be rectified. The motor swithgear must open when there is a power outage or a deep voltage dip. This is done by the undervoltage protection/relay.

 

[JBUDA54]

On a recent project have I gained knowledge in this subject, but I believe in any case reduced voltage start or not, there should be a Undervoltage sensing relay. This prevents the motor pumping excessive energy back into the system in instances such as the one described. The undervoltage relay typically has some sort of time delay setting that will ride through momentary voltage dips such as other large motors starting. I believe all motors above 2300V should have this feature.
It has been brought to my attention by a collegue of mine that it is possible in the case described that the motor (in generator mode) will have it's own induced voltage while coasting down due to it's own inertia. If the induced voltage from the motor is out of phase with the bus voltage (upon the upstream breaker being reclosed) the inrush current to the motor will be drastically higher than what the motor was designed for causing the motor stator to endure excessive energy than it's original design. This can ultimately damage the motor.

 

[peterb]

Assuming that this is an induction motor, there is no sustained generator action (unless there are capacitors installed) beyond a few cycles after supply removal, as there is no means of maintaining the motor field. Once the motor terminal voltage has collapsed, it is safe to re-energize it without regard to incoming supply voltage phasing. This is the basis used for delayed bus transfer schemes, where closing to the alternate supply is delayed until the bus voltage has decayed to a safe value (typically <30%).

I would agree that an undervoltage relay may be appropriate in order to ensure that the breaker doesn't remain closed indefinitely - you would want to have a controlled re-start when supply is restored. Also, any reclosing on the 12 kV circuit must either be delayed or check that the feeder voltage is zero, so that there is no possibility of out of phase closing while the motor terminal voltage is still decaying and above acceptable limits.

 

[electricpete]

I agree with peterb on the subject of reclosing out of phase. I remember an EPRI study which indicated the &quot;danger zone&quot; in terms of period of time power was removed before reenergization of a motor was typically 3-10 cycles. If less than 3 cycles, the motor voltage hasn't yet had time to drift out of phase with the source. If more than 10 cycles the motor voltage has dropped in magnitude to the point where out-of-phase reclosing is not serious.

There are lots of reasons to trip the motor on undervoltage:
#1 - Prevent huge starting surge due to simultaneous restart of multiple motors when the supply is reenergized.
#2 - Prevent possible personnel or equipment hazard due to motor starting unexpectedly when power is returned.
#3 - Prevent possible motor damage due to prolonged operation at reduced voltage (although thermal overloads should handle this pretty well anyway... undervoltage trip provides some redundancy).
#4 - Immediate restart of a large motor after tripping can be fairly severe duty since motor is already at operating temperature and has to withstand heating from a starting surge on top of that. I think NEMA requires motors to be designed for this but there is no need to subject the motor to this unless it is required.
There are probably a few more reasons.
Also might consider something of a delay in the response to prevent unwanted trips on momentary power interruptions.

 

[electricpete]

Some more on item #4 above - even though it didn't happen this time, it might be possible that power can start and trip several times... potentially subjecting the motor to several attempted starts during a short period. This can do damage to rotors of large motors, especially when driving high inertia loads.

(NEMA standard only requires one start when initially at temp or two starts in succession).

 

[electricpete]

I mentioned the EPRI study regarding acceptable power interruption times (<3 or >10 cycles between loss of power and restoration of power for &quot;typical&quot; motor).

I found much more relevant info on that subject.

NEMA MG-1-1998R1, for large motors - section 20.34.1

&quot;Slow Transfer Reclosing&quot; - power interrupted for >1.5 times the &quot;open circuit ac time constant&quot;. This is the recommended approach since it allows residual flux/voltage to decay to acceptable levels.

&quot;Fast Transfer Reclosing&quot; - power interrupted for <0.5 times the open circuit ac time constant. Voltage should not have drifted that far apart by this time.

Section 1.60.1 - &quot;Open Circuit AC Time constant&quot;
= (Xm+X2)/(2pi*f*r2) where Xm, X2, r2 are equivalent circuit parameters (per-phase). f power frequency.

So 3-10 cycles was the EPRI typical number. NEMA gives info on calcuating limits for a specific motor.

 

[electricpete]

I'm not 100% sure I have properly represented the Fast Transfer/Reclosing. Here's what they say:

&quot;A fast transfer or reclosing is one which occurs within a time period shorter than 1.5 open circuit ac time constants. In such cases transfer or reclosure should be timed to occur when the difference in motor residual voltage and frequency and the incoming system voltage and frequency will not result in damaging transients...... it is recommended that the electromechanical interactions of the motor, driven equipment, and power system be studied for any system where fast transfer or reclosure is used&quot;

Sorry I have left out about a paragraph that seemed irrelevant to me... getting a little tired of typing.

The bottom line is it seems clear they're recommending a study to see if the fast transfer/reclosure option will work. I'm not exactly clear on the word &quot;should&quot; in the 2nd sentence of quoted section. Are they saying that the user should apply phase-checking relays? Or are they saying that limiting the duration to <0.5*Tau &quot;should&quot; [is expected to] result in a situation where the voltages are still approximately in phase. My reading of EPRI was the latter interpretation.... but I suspect that's not what NEMA is saying.

 

[electricpete]

All of the quoted stuff in my last 2 messages comes from NEMA MG-1-1998 (not EPRI).

 

[electricpete]

One more correction: 1st message of 8/22 was correct that >1.5 Tau was for slow xfer/reclose and <0.5 Tau was for fast xfer/reclose. 2nd message on 8/22 incorrectly said <1.5 Tau for fast xfer reclose.


By the way - I haven't figured out how those stars appear immediately after I post.

The reason that I'm making so many posts is that I don't proofread my posts until after they're posted, not that I'm trying to rack up stars !

 

[CSCEE]

You lost some motor life when you reclosed in on a coasting motor. Severe mechanical shock can occur to the motor and shaft when the restart happens. Use an undervoltage relay with a timer to block timing until at least 10 seconds after the trip. It is best to wait until the motor is fully stopped before restarting to avoid motor damage. The number of starts per hour should be limited as well.

 

[electricpete]

CSCEE - You mention a block for 10 seconds... but the original messages states breaker was reclosed &quot;within 30 seconds&quot; which leads me to believe probably greater than 10 seconds anyway.

From what I have read, I believe there is no danger as long as the duration of power interruption is greater than 1.5 open circuit ac time constants of the motor which is on the order of 10 cycles (regardless of the fact that the motor probably has not stopped spinning during that short time).
The nature of the transient toruqes have to do with the energy left in the electric/magnetic fields, not the energy left in the rotating inertia... and decay of the electric/magnetic energy is primarily influeced by electrical quantities (it would not be terrbily incorrect to assume that mechanical energy does not decay signficantly in the time frame in which the electrical/magnetic field energy decays substantially to zero). So again, from my perspective, the fact that the pump was still spinning is not really of any significance... only the duration of power interruption. I think this is born out by NEMA MG-1's recommended practice.

That's the way I see it from here. I'm interested to know if you disagree with any of my statements. No doubt I've been wrong quite a few times before.

 

[electricpete]

To add just a little to my recent message:

The NEMA requirements for slow transfer/reclose are based soley on the open circuit ac time constant which is computed from electrical motor parameters. Load inertia/load torque which affect the mechanical coastdown are conspicuosly absent. This helped me reach the conclusion that mechanical coast-down is irrelevant in determining the minimum alloweable duration of power interrutpion.

Also, the apparent contradiction between EPRI and NEMA and the ambiguities in NEMA which I noted in my 8/22/01 posts all center around fast-transfer reclosing. Today's discussion of minimum duration of power interrutpion is related to slow-transfer closing.... for which case the NEMA and EPRI guidance appear consistent and unambiguous.

 

[gwc43]

Some interesting discussion here! Let me add a couple of things. I'm going to assume from the description of &quot;coasting&quot; that this was not a reciprocating compressor but a turbine type.
1. If this is a power to close and power to trip type breaker, consider a control power source from a UPS system in order for the under-voltage device to work. I need more info on equipment.

2. Consider installing a &quot;lock-out&quot; relay device.

3. Consider installing modern motor protection (SR-469 Multilin or like) for proper motor managing.

4. You didn't mention any effects this may have had on the compressor. Normally these motors start with the compressor un-loaded. Does your control scheme have an instant dump on power loss?

5. Other equipment like an oil skid that may provide lubrication for the motor and the compressor. Some motors in this size range will need a DC powered back-up pump just for lubrication during &quot;coast-down&quot;.

 

[jbartos]

Small encore: Please notice that the auxiliary system at power plants does have many large motors designed with Full Voltage across the Line starts. If there happens to be some short interruption, then many of these motors (some of them still coasting down) will continue to operate as usual. It appears that the squirrel cage induction motor is rugged enough to withstand this duty, possibly on an occasional basis only.

 

[peterb]

Just to confirm what jbartos said above, I have worked on the design of several thermal power stations and can state that it is quite normal for station auxiliaries to be transferred from the unit bus to the startup bus in the event of a bus supply failure. The transfer is usually delayed until the running bus voltage has decayed below about 30% (an undervoltage relay is a permissive for the transfer), in line with electricpete's posts above. The starters are either circuit breakers or DC held contactors, fed from the station battery, so they will not droput due to the loss of AC.
As you say, jbartos, these motors are designed for across the line staring and this condition does not impose any higher duty. I don't believe that there will be any loss of motor life under these conditions.

 

[jbartos]

Suggestion to the original posting marked by ///\\\:
My questions are, should I change the motor control circuit so that upon loss of main power to the motor to trip the motor?
///Based on the above postings, standards, and studies it may be prudent to add recommended protection to the motor by inhibiting its start until its terminal voltage is in compliance with the manufacturer and standards recommendations.\\ If so/or not so why?
///It is recommended by the standards, results of studies, and possibly by the manufacturer. However, the industry practice appears to be not so meticulous, when it comes to the squirrel cage induction motors.\\\

 

[electricpete]

I would summarize my input by saying that the 4 reasons I gave in my first 8/8/01 message remain true. The type of motor damage discussed in item #4 of my 8/8/01 message was overheating damage (primarily in rotor) due to multiple restarts with motor initially at operating temperature. Time required between restarts would be on the order of 15 minutes for a large motor. (NEMA standard only allows one immediate restart if the motor was initially at operating temperature... any subsequent restart without cooling should only be undertaken with advice from motor manufacturer)

Later on we talked about a different type of possible motor damage from high torques due to out-of-phase reclosing if the duration of the power interruption was too short (<1.5x open circuit ac time constant.) That is a different type of damage.

The two types of damage mentioned (#1 - thermal and #2 - high torque due to out of phase reclose) result from separate types of conditions and must be given separate consideration, although the protection might overlap (for example undervoltage trip protects against both scenario's).

 

[horsedragon]

To All,

Protecting the motor is one thing to think about. You have to consider the inertia load on the motor and protect the equipent the motor is connected to. I have seen 5000lb rotating assemblies bent due to motor reenergization. I have also seen coupling bolts and keys sheared off. I work for a large equipment manufacturer and If we find out you have been doing rolling restarts your waranty is null and void. Typically if I see bolts sheared off on a coupling design with a service factor of 2.5X (most couplings are 1.5x) I start looking at the power source.

Chuck

 

[Skogsgurra]

I was really surprised to see JB back again - until I happened to look at the date: August 2001.

That apart, I think that you make things a little bit too easy for you Chuck. There needs to be a more detailed investigation - restarting an induction motor that has no soft-start equipment should be fully acceptable when power has been away for 30 seconds - even ten seconds will be OK since the rotor time constant seldom is more than a few seconds.

An undervoltage lock-out is always good and necessary in many cases. The guarantee question, however, should not be solved in such a hasty way.

Gunnar Englund

http://www.gke.org