Locked rotor definition 2

[RaulChavez]

Hi,

I´m working with a motor that blew during commissioning.
this is a 1000cv@4000Vac motor with a locked rotor time = 3.5sec.
According the vendor recommendations, we can do 3 starts per hour: one start each 20 minutes.
According the relay records, this motor was started many times in a short period, less than 5min between starts.
It´s common that during commissioning operation guys do many starts and nothing happens.
My question is, is possible that in motors with a short rotor locked time, consecutive starts can blow the motor?

 

[edison123]

Yes, both the stator and rotor are thermally and mechanically stressed severely during starts. So, multiple starts within a short span will damage these windings.

"It´s common that during commissioning operation guys do many starts and nothing happens."

Well, now it has and hopefully your "commissioning guys" have burnt their fingers !!!

*Nothing exists except atoms and empty space; everything else is just an opinion*

 

[RaulChavez]

Thanks, I´m trying to do an inquiry about this damage, I´m quite sure that the problem was due the consecutive starts, but I want to be sure that the ANSI protection N 66 (consecutive starts) becomes a critical issue when we work with motors with a very short locked rotor time.

 

[LionelHutz]

Your question depends on many factors. For example, If you're using a soft-starter and the motor is not coupled it could be possible to start the motor a number of times without causing damage. Each start could be much less than locked rotor current and take less than 3.5 seconds so it would take a number of starts to exceed that 3.5 second locked rotor rating.

However, from your question it sounds like the vendor said 20 min between starts with the attached load and this was exceeded. You now know why you were told not to do that.

Basically, the whole motor protection relay scheme is important. A good thermal model, time between starts protection, unbalance protection etc are all important. Skip one and you could get a damaged motor like you're seeing.

 

[jraef]

It should also be noted that generally the smaller the motor, the more starts-per-hour are acceptable. So when your commissioning guys made that statement, if their previous experience was with small motors it could be true from their perspective. I have run into this quite a bit when doing startups of MV motors. I have even had technicians claim it was no problem because they put their hand on the motor frame and declared it to be "cool"! What they don't understand is that the first point of damage is often the rotor because it can only dissipate it's heat across the air gap into the stator or out the shafts, which takes a lot longer. The starts-per-hour rating is more often related to the rotor damage curve because once the rotor goes, the stator follows quickly.

But that does not excuse their actions in my opinion. Most technicians and electricians with any amount of real-world experience will know that all motors have a starting condition specification and it only takes a moment out of their day to ask a question when they are not sure. What they did was a classic assumption, which gets it's name from making an ASS out of U and ME!

 

[RaulChavez]

According the motor specification. This motor was projected for a direct start. The start torque is 260% and the start time is 0.5sec!!!
So that the combination of high start torque in a very short period (0.5sec)will explains the short locked rotor time.
Maybe for this motor the minimum time between consecutive starts might be more than 20 minutes.

 

[JERRYROBINSON]

NEMA MG1 1998.20.12.1, 20.12.2 treat this for standard. Any deviation from this should have a nametag defining the starting limitations.

The lock rotor time and acceleration time ( 2 different values) are based upon Nema WK^2 limitations or the actual load, and are basically advising the user the point where the motor can / will fail from overheated components; rotor or stator, typically rotor.

I agree with the other responders that your vendor was not educated in the proper start-up proceedures. Was this a manufacturers start-up engineer, or a vendors representative? In most cases on a machine above 1,000 hp it is adviseable to request a start-up proceedure from the manufacturers engineering department.

 

[electricpete]

I agree with the other comments. You should understand the limitations for starting a motor which are derived from the manufacturer or the specs used to purchase the motor, and also have to consider the load inertia and torque during start.

For smaller motors driving standard inertia's, limits are given by NEMA here:

http://www.joliet-equipment.com/allowable_starts.htm

the situation is less standardized for larger motors. I think NEMA only requires one start hot or two starts cold.... not much definition of how much cooling off time is required beyond that.

Off the top of my head, if the start time is truly 0.5 sec and the safe locked rotor time is truly 3.5 seconds, then I think you should be able to do 6 starts in a row without violating any thermal limits (assuming your actual voltage is equal or lower to the voltage assumed by the 3.5 second limit). Does anyone agree/disagree with that?


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

 

[RaulChavez]

The great concern is: how prevent the motor blow again, and also define who will pay the motor repair. Tomorrow we will start this repaired motor uncoupled, I will take the current vs time graphic and probably next Wednesday we will start this motor with load.
I have read something about the material used to build the rotor bars, when its needed a high torque and a high acceleration those might be made in aluminum cast

 

[itsmoked]

I would add I have not seen a single motor (>5HP) that could sit LR for 3.5 seconds without shocking amounts of smoke pouring out of them! So your 3.5s number is even highly generous. I wouldn't dream of trying starts spaced closer than 20 minutes on your motor. EVER!

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[jraef]

The important point is, you cannot really make generalizations about motors above the NEMA frame sizes (low voltage, 250HP). For medium voltage motors like yours, you absolutely MUST use the manufacturer's data, which should look something like this;

SKM white paper link

If the data is no longer available, ask another MV motor manufacturer of a similar motor for theirs. It will probably be close.

As to a solution, you should have a Motor Protection Relay on that motor. They are expensive (at least the good ones are), but you already know the cost of NOT having one!

JRaef.com
"Engineers like to solve problems. If there are no problems handily available, they will create their own problems." Scott Adams
For the best use of Eng-Tips, please click here -> faq731-376

 

[GGOSS]

This may appear a little harsh but it is the esponsibility of the commissioning engineer to ensure the motor protection relay is programmed correctly. If the thermal withstand curves for the motor are not available or if the motor protection relay cannot be programmed to protect the motor adequately, should they not tread carefully?

Regards,
GGOSS

 

[nsv]

"It´s common that during commissioning operation guys do many starts and nothing happens.", but actually things do happen, you just don't notice. The high temperatures reached at consequtive start-up's eat away the lifetime of the insulation and the rate of deterioration goes exponentially with the temperature.

NSV

 

[waross]

In the Cowern papers on the Baldor Motor web site, there is a formula for calculating RMS horsepower for a motor under various loadings during the duty cycle. The formula recognises that a rotor self cools much better when it is turning.
The formula uses a factor of 2 or 3 to weigh the cooling time of a stopped rotor.
If the cycle is a running motor that is unloaded and is stopped and restarted, it will withstand more starts per hour than a motor that is started and then stopped, restarted and stopped again.
I have seen statements that rotor heat must be conducted away through the shaft, and that rotor heat must be disipated across the air gap to the stator. I suggest that rotor cooling in a running motor is primarily by air flow past the shorting rings on the ends of the rotor bars.
This is particularly true in the smaller motors where the squirrel cage and the cooling fan blades are cast in one piece of aluminum.
respectfully

 

[dpc]

Might be worth a look at least for background on motor thermal models.

http://www.selinc.com/techpprs/6164.pdf

http://www.selinc.com/techpprs/6024.pdf

http://www.selinc.com/techpprs/6023.pdf

 

[jraef]

waross,
Re: dissipation through the shaft and air gap etc.
I was referring to small NEMA frame motors and didn't make the distinction that it's only TENV motors. Those details were not germane to my point, that being that feeling the outer frame with your hand is not a good indicator that a motor is "cool", not to the subject of this thread, MV motors, so I didn't take the time to go into that detail. Maybe I should have..

 

[waross]

Hi jraef;
I wasn't challenging you. I have seen such comments in other threads also.
The point that I was trying to make is that if one motor is running steadily and stopped and immediately restarted every 10 minutes, the motor may be expected to be cooler and less likely to fail than a similar motor that is started every 10 minutes and then immediately stopped.
I have re-read my reference and Mr. Cowern refers to the motors inability to cool itself when it is stopped, without particular reference to the rotor.
I apologise if I mis-spoke.

Here is a link to the Cowern Papers by Edward H. Cowern P.E., of Baldor Motors.

http://www.baldor.com/pdf/literature/PR2525.pdf

Starting on page 43 he discusses RMS loading of motors.

A point for discussion is the action of thermal modeling relays.
Do thermal models differentiate between the cooling of an unloaded running motor and the slower cooling of a stopped motor. If not, must extra care be taken when commisioning to avoid multiple short starts that may "fool" the thermal model and overheat the motor without the condition being reflected in the thermal model?
comments?
respectfully

 

[electricpete]

"The point that I was trying to make is that if one motor is running steadily and stopped and immediately restarted every 10 minutes, the motor may be expected to be cooler and less likely to fail than a similar motor that is started every 10 minutes and then immediately stopped"

Yes, that is absolutely true and not so obvious to non-motor folks. If you think only about the heat input-side and not the heat removal mechanisms, you would come to the opposite (wrong) conclusion. I have explained it to some folks that do machine balancing and they were very surprised.

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

 

[jraef]

I was involved in the design of a yet-to-be-released motor protection relay, and yes we differentiated our cooling time constant calculations based upon the run time of the motor. As far as I know however, that feature was only available in very sophisticated relays such as the Multilin 369/469, Schweitzer 701 and maybe a few others in that range of products. It took a fair bit of processing power to accomplish. To my knowledge, the (relatively) inexpensive Solid State Overload Relays / MPRs currently available have yet to incorporate that ability. All of the competing products I investigated were using a fixed cooling time model, which definitely presents that as a problem.

But I will say that most erred on the side of being conservative, i.e. basing their cooling time constant on a stopped motor, not one that had been running longer. The problem with that then would only be that the motor may be off-line longer than necessary if it tripped on OL or that it may trip on OL sooner than necessary with repeated starts. There were a few however that had what I considered to be very short cooling time constants for the motors they were intended to protect, shorter than many large motors have published on their data sheets. It's something that a prudent engineer would want to investigate if truly concerned about proper motor protection.

 

[waross]

Thank you electricpete and jraef.
Respectfully