Motor cooling down logic ?

Hello

To prevent over-heating of motor supplyer rate them for 3-4 start per hours of full load and direct online starting...

To respect that, a very basique plc logic can freeze any restart for 15-20minutes (3x20minutes=3start max per hour )

This is a pain when the motor ran for a long time already cooled down from it's last start while we still need to wait the maximum time...

I have seen relay/controler with an evolutive logic taking count of how often the motor get started in the same hours etc permitting to restart it just after few minutes for the first attempt and increasing the delay for the second etc...

Do someones know how this logic is based ?

I'm looking to a logic decreasing the 1200 secondes locking time as long as the motor run
20minutes:
300sec + (900sec-1sec each 1or2 sec of continius runtime)
Ranging from 5 to 20 minutes...

Is there any easier way to do that or other parameter to take into account?


By the way i have seen programmer that never goes inside an eletrical room to see blown device, starting and stoping good horsepower motor many times under a very short period doing test like in the lab!

You could perhaps increase an estimated temp with x degrees at every start attempt, and decrease it by multiplying with ...hmm... 0.999 every x seconds. You will have to take the environment temp in account. If the next start attempt will exceed the max limit, it should be rejected.

And I think I've seen a thread about this issue before a couple of years ago. I'll try to find it.

Kalle


EDIT: Eureka! http://www.plctalk.net/qanda/showthread.php?t=58504&highlight=motor

EDIT2: What I tried to suggest in my first sentences is better explained at the very end of the thread above.

The allowed motor start/stops per hour specification is a motor selection criteria, not a motor control mandate. If you anticipate that the motor is going to be frequently started and stopped as a part of normal operation then select a motor that meets the required specification instead of preventing operation and loss of production. If the specification might be exceeded only rarely, and not as a part of normal operation, then I would not worry about it too much.

If you do have a genuine need to limit start/stops, I don't recommend over complicating it. A simple timer/counter should be sufficient for most general cases. The one or two times when its not the perfect solution its still good enough - four starts in 2 minutes followed by a fifth ten minutes later is probably OK. Most motors are really quite robust and can take a little bit of abuse.

A way that I've seen it done was on a 360A 3ph 415v motor on a YD started refrigeration compressor...

'Start' a 15 minute "delay-off" timer every-time the motor started which would inhibit a restart whilst timing out. If the motor stopped after 7 minutes then 7-8 minutes later the motor was available to restart.

Perhaps I've missed the point somewhere here... it seems simple to me.

(Perhaps the application is more critical than a fridge plant.)

.

Jeff23spl said:

Do someones know how this logic is based ?

Click to expand...

It is based on the estimated motor temperature.

The clue is of course not to burn the stators winding isolation and not to fry the rotor.

You could use a simple startup prevention timer as suggested by Alaric and Silva and top it with putting a thermistor inbetween the stator poles.

Kalle

This is always a tricky situation - consider a motor that starts, and stops within one second because of a process problem/alarm. Why should it be prevented from starting again after the n'th attempt. It is a heat problem, and there wouldn't be much heat build up in short runs like that.

A slightly more complex approach, but which mirrors the reason why motors have a "starts per hour" restriction, is this one I have used several times....

1. Each "Start" adds a constant to a register...

2. The register decrements continuously to zero at two timebases, depending on whether the motor is running or not (heat dissipation is different).

3. Prohibit another start if the adding of the constant in step 1 exceeds a limit value.

It's a simple one to code, but can be difficult to determine the constant (step 1), the decrement rates (step 2), and the limit (step 3), but can be calculated quite readily.

I don't have access to sample code (might be able to get it), but that's not important as each situation is different.

You can apply the same algorithm to your DOL contactors if they have a restriction too.

thanks guy, you're approching of what i am looking for. I like the simulation of the temperature rise at each start and cooling down...but how to figure out the parameter is the clue.

Alaric give good points too but in my case, It's a repeated design where the motor doesn't need to be restarted often in normal operation but it is manually operated in manny case(from a scada system)....So i have to protect it from stressed operator. HP range is usually 100-800hp so in some case, just 1 more time can be enough to fry somethnig expensive!


I actually use 1 "off delay" to lock it out on the run feedback to be safe with the worst case.
I will always need a minimum delay of 5 min for the mechanic but i would like to play with the other 15 remaining minutes...

The repeated start happend mostly at commissionning or when something goes wrong with the plant and peoples goes crazy to put it back online after manny try...
In all case, their is no major issue to wait 20 minutes and sometimes it give them the time to look at what make it going wrong....but when the guy is waiting at the countdown, it looks like to be the end of the world
It is even worst when the MCC protection relay tells it's ok to restard after 5-10 minutes while my plc still lock it out for another 10minutes !

i found something interresting about minimum motor nema design

have a look on the page 22 of this link:
http://www.conae.gob.mx/work/sites/CONAE/resources/LocalContent/2391/1/images/MG10.pdf

Their is a start number per hours with also a minimum cool down delay even for the first attemp to restart.

This mean we can't restart a motor 2-3 time in a row even if it is rated for 3 start per hours....We have to wait a little bit

It's also stated to be carrefull with thermistance or RTD as it's just represent the stator winding temperature with a time response for the heat transfer. Also sometimes the rotor bars may have got more degrees than winding and takes more times for this heat to reach the motor case than the winding...

I'll look at these needs and figure out a simple logic to be safe and a little bit more user friendly than just a minimum off delay

I got something, give me your feeling:


I keep my lockout time-out but play with the preset setpoint:
A common "timer off" drived by a "run feedback" falling edge disable the motor start while timing...

The timer preset is forced to a 300sec minimum value alltime
Preset value is incremented by 600s each time the motor start (run feedback rising edge) (Add 600)
Preset is decremented by 1 each 3 sec of the motor running time. (preset=preset-1 each 3 secondes of running time)
Preset is decremented by 1 each 2sec or resting time when lockout time is finished.
　

If we start once and stop right away, a 10minutes delay would be issued and then 20,30,40 and so on making it safer than 3 start per hour
If we start and run for more than 30 minutes, the preset will be decresed to 0 so the 5minute minimum will be used next for resting time So 3starts per hours protection still respected
If we stop at first for 20minutes, (resting10+10more) the preset is decreased to 5min and then incresed by 10 to get 15minute upon next start and in this case we would still be ok

This thread has gotten fairly complex already so I'm sorry to add another element to the equation. When an induction motor starts, there is a large initial inrush current that exists until the motor and load are able to spin up to about 85% of full speed. At that point, the rotor starts synchronizing with the stator field and the current fails rapidly to normal running levels.

It is the heating effects of this inrush current (often 4 to 16 times nameplate FLA) that causes the start delay requirement on motors. I'm told that most of this heat develops in the rotor so rtd's and thermistors in the stator windings aren't of much help in measuring it.

Starting motors with reduced voltage starters will help some but not as much as would seem likely. This is due to the benefits of reducing inrush magnitude being cancelled, to some degree, by the increased time spent in inrush.

Operation on a VFD however, eliminates virtually all inrush considerations since the VFD starts the motor nearly synchronously at 0hz and brings the frequency up as the motor accellerates. There certainly are other heat issues in motors on drives but start cycles is not one of them.

I would take the start cycle spec and do your programming to enforce it as written. It isn't a pleasant thing to sit and wait but, that's the choice that was made when the motor was specified. You can't program around a problem that was built into the machine. If you just can't wait, then it is time to shop the market for a motor with a higher starts per hour rating. Brace yourself! It won't be cheap!

This situation kind of reminds me of a Simpson's episode where Homer get up, sticks his chest way out, puts on his most manly voice, and announces with great pomp and authority "in this house, we will follow the laws of physics!" Well, yes, I guess so! Hurray for you, Homer, for taking such a compelling stand!

In my plant I have an array (DINT 10 elements) which stores the last 10 times the motor was started (successful or unsuccessful). To populate this DINT, I have a routine that runs every 500ms which takes the clock time and compresses it into a single DINT e.g. 20100317170325 = 2010, 3rd month, 17th day, 17:03:25.

On my motor page I have a login only part which allows you to change 2 values for each motor. One being number of starts per hour and the other for time between starts. e.g. 600hp motor = 3 starts per hour and 12 min delay between starts.

I wrote an add-on which grabs these values and sees if the motor is allows to start when the start button is pressed and lets them know if not i.e. too many starts per hour wait 44 mins or 3 mins until restart.

As normal, the entire facility doesn't like that the place is shutdown to wait for a motor to cool, so people can login (logged into alarms) and change the number of starts or start delay, but management ok is required for this.

I still think the best bet is monitoring winding temperature. This can be quite economical with Klixon type devices
http://klixon.com/klixon/thermal-protector-7am.htm
although I prefer RTDs because you can adjust the trip point.

It is true that rotor temps are hard to measure, but in my expereince if the stator is too hot so is the rotor.

good points dickdv but are you sure a soft start will reduce the temperature rise inside the motor and would permit more start per hours ?
Yes it reduce the inrush maximum level but i believe it extend the time to get the motor at full speed...Less instant heat for a longer delay give the same energy? And sometimes a softstart can keep a motor in the starting process for too long when it doen't get configured right...
Sometimes i heard peoples askind a softstart to save energy !!!Even the peak demand read by the electric counter will integrate a short high peak or a long small peak as the same increase...

Jeff, as you can see from my earlier post, a softstarter, properly configured may and often does not make much change in the total motor heating. You are correct to figure (only a little conservatively) that a softstarter will be of no help.

Tom, I used to figure that way about Klixons and motor temp but, after some very enlightening discussions with a fellow NTT instructor who has been in induction motor design his whole life, I've had to adjust my thinking a bit.

It's safe to say that if the stator is hot, the rotor is also hot but the rub comes when you consider the inverse condition. When the rotor is hot, the stator may NOT be.

This condition can exist with rapid DOL starts or overloads that pull the motor below the breakdown torque point. It even can happen when large amounts of heat are transmitted into the motor directly thru the shaft.

These conditions are much less possible when on an inverter.

DickDV said:

When the rotor is hot, the stator may NOT be.

Click to expand...

Point taken - thanks, Dick.