Multilin Standard Thermal Curve 1

[gokulkrish2]

Hi all.

I have a Question regarding the multilin 469 relays which we use for motor protection which protects a 1000HP motor which runs a Crusher.

We do the installation of the multilin protection scheme through the standard way of putting in required input parameters it asks about the motor.

But At this standard thermal curve(which it assume with the motor parameters) we are able to start the motor only 3 times per hour. If we start more than 3 times, the thermal limit exceeds and the multilin dont let us start. But the RTD attached to the motor never heat up. So i want to find out if the standard thermal curve assumed by the multilin is good or whether it needs to be customized according to the application.

If it needs to be customized what we should do.

Hopefully i could get some inputs from you guys.

gokul

 

[dpc]

The RTDs give indication of the temperature of the stator. It's likely that the motor is rotor-limited during starting. Excessive starts can cause the rotor to overheat and the stator RTDs will not detect this. The RTDs provide indication of motor overheating during normal running operation. They are not generally too helpful for protection during starting.

Three times per hour for 1000 hp motor seems excessive.

I'd check with the motor manufacturer regarding the allowable starts and go with their recommendations.



"The more the universe seems comprehensible, the more it also seems pointless." -- Steven Weinberg

 

[ScottyUK]

Maybe not that excessive - the 2.8MW two pole HP boiler feedwater pumps at the plant I recently escaped from managed 3 evenly spaced starts per hour.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[gokulkrish2]

s, scotty is right and we do space the starts. Actually the first time we start the motor, it reaches its 90% of the thermal capacity. Then we give it a shot after 5-10 min but the third try will only be given after the thermal limit comes to less than 10%

Coming to my question, is it by any chance that the thermal limits of the motor can vary according to the application of to what pupose its used? I personally feel no. but my manager wants me to research about it. So i want to double check and make sure.

gokul

 

[dpc]

The motor doesn't "know" what its application is. If it gets too hot too frequently, it will fail. The relay thermal replica is supposed to model the temperatures inside the motor based on the motor current.

If you can get data from the motor manufacturer related to cooling times, allowable starts, etc, the model the relay uses can be improved a great deal.

Schweitzer has a fairly new motor relay out that claims advanced thermal modeling capabilities. I'm including a link to one of their technical papers on motor modeling. You may need to register with SEL to get access to it.

http://www.selinc.com/sel-710.htm

http://www.selinc.com/techpprs/6338_AdvancedThermalMotor_EL-MZ_20081006.pdf

"The more the universe seems comprehensible, the more it also seems pointless." -- Steven Weinberg

 

[gokulkrish2]

Thanks for the links.
I will go through them and get me informed.

will get back to you guys in case i dont understand something

thank you once again.

gokul

 

[lansford]

You should get the thermal damge curve from the mfgr and see how it matches the standard curves in the relay.
JIM

 

[dpc]

Just bear in mind that thermal protection during starting requires different considerations than thermal protection for normal running overloads. The relay has to be able to combine the two conditions and to detect when the motor is in a start condition.

"The more the universe seems comprehensible, the more it also seems pointless." -- Steven Weinberg

 

[roydm]

I think it also depends on the conditions the motor is starting under, obviously starting a crusher that has tripped on overload and is bogged is a lot worse than one that has been shut down empty. I have seen the rotor bars melt on bogged crushers after multiple start attempts. If the crusher has a clutch the strain on the motor will be much less.
i agree the limit of 3 x per hour is very annoying at times.
Roy

 

[jraef]

Just to address your earlier question regarding the RTD readings being low but not allowing the motor to start. RTD inputs can be used to bias the thermal model, but only in ADDITION to the thermal damage curve, not in lieu of it. In other words, a high RTD temperature reading might cause a quicker trip even though the thermal model would not otherwise indicate it, but if the thermal model is already indicating excessive temperature, a low RTD reading will be ignored and will not override the thermal model. The reasons are exactly as dpc stated in his first response, you are not able to directly see the rotor damage curve with the RTDs.

I have seen a system for large motors that has RTDs for the rotor an a wireless communication system back to a protective relay, but the motor must be big enough to house the components in the rotor assembly. Even then, I don't know that I would completely trust direct measurement like that over a very complex tried-and-true thermal modeling systems as is used in the Multilin and SEL relays.


"If I had eight hours to chop down a tree, I'd spend six sharpening my axe." -- Abraham Lincoln
For the best use of Eng-Tips, please click here -> faq731-376

 

[jraef]

This was it.

http://www.accumetrix.com/demo/mgp.html

"If I had eight hours to chop down a tree, I'd spend six sharpening my axe." -- Abraham Lincoln
For the best use of Eng-Tips, please click here -> faq731-376

 

[rockman7892]

I have seen the NEMA standard for most larger motor that states the motor can have (2) cold starts and (1) hot start.

I've been trying to figure out a way to use the thermal model and thermal capacity to control this (2) cold (1) hot standard. This would mean that for the two cold starts each start should only use up 50% of the motors thermal capacity. I have heard in some cases that in order to do this you need to set the tric curve somewhere inbetween the hot and cold thermal damage curves for the motor.

Does anyone have any experiene with setting these models to use the thermal capacity to drive the Nema recommended number of starts?

 

[gokulkrish2]

can you guys please tell me what is the thermal curves which the motor manufacturer should provide to us.

When i requested for the curves the motor manufacturer has provided me with to curves, one for running and one for starting, and both of them are %full load current VS time.

Now i am assuming that all i got to do now is to see that the curve assumed by the multilin is below that so that it gives less space and time or above it right.

Or are there any other curve which actually gives the thermal capacity of the motor.

I am attaching the curves for perusal.

gokul

 

[jraef]

Yes, that looks to be the Thermal Damage Curve, although it is not labeled as such. So you are correct, you want the trip curve and stall point that you select to be under that damage curve. You don't want the motor to be exposed to currents in excess of the time and currents shown.


"If I had eight hours to chop down a tree, I'd spend six sharpening my axe." -- Abraham Lincoln
For the best use of Eng-Tips, please click here -> faq731-376

 

[rockman7892]

For instance if we set the multilin curve just below the "hot" motor thermal damage curve but are still using up about 80% thermal capacity on a cold start, then how can we ensure a second consecutive start (2 cold) if the thermal capacity is already at 80%. How would we bring this thermal capacity used on startup down to somewhere around 50%.

Can we move the multilin curve above the "hot" thermal damage curve as long as we stay belwo the cold?

 

[gokulkrish2]

Also, i want to throw some light on the motor inrush.
Shouldnt we have to make sure that we have our curve above the motor inrush to avoid false trips.

Where should i look for the motor inrush curve. When asked the motor manufacturer he says he has never heard of that term before.

gokul

 

[dpc]

Take 1.6 times the locked rotor amps for about 0.1 seconds. That should be conservative.

You should buy a copy of the IEEE Red Book. This is all covered in there.



"The more the universe seems comprehensible, the more it also seems pointless." -- Steven Weinberg