Retrofitting soft starters to existing motors with high inertia loads - engineering process

[123MB]

Hi All

So I have a client with several three phase induction motors ranging from 220kW to 335kW. The motors are 2 pole star connected and were built in around 1985. They are started DOL at the moment. The client's supply authority has read him the riot act and he has asked us to design a replacement MCC for him that will include soft starters for the existing motors.

The motors are driving blowers which I consider to be 'high inertia' loads (I am not sure if this classification is relatively correct, but anyway). A DOL start on these things ranges from 8 to 15 seconds for the 220kW to 335kW respectively.

Now, the issue I have is determining the best process to specify the new soft starters. My technical design requirement for the reduced voltage start was going to be a current limit start at %300-%400 FLA but I haven't based this on any quantifiable process (i.e. relating the desired voltage dip to the motor starting current).

The client has zero information on his plant and there is no documentation for the existing motors (speed-torque curves, etc.). All I have is the existing plant in front of me and whatever observations I can make from that.

I have gathered some very rough 'rules of thumb' from technical references around the place which, when summarized, basically say that, for a constant load inertia, 50% starting voltage gives rise to %400 start time. So, considering 50% of the DOL start voltage (i.e. %400 current limit, assuming %800 DOL start current) will lead to worst case 60 second start time (400% of the worst case DOL start time of 15 seconds). I have two problems with this:

1) The customer has specified AB SMC flex starters which have a maximum start time of 30s (plus a 10s stall time) before tripping out
2) The process/reasoning I have applied, summarized above, I dont believe would stand up to any engineering scrutiny.
3) This doesn't give me any idea of wether or not the motor will be able to apply the required initial torque to start the load at the reduced voltage

So, can anyone give me any quantifiable tips, processes, calculation references etc. that I can use to extrapolate data from the plant in front of me in order to better specify the soft starter and guarantee it's performance at the required reduced voltage. Keeping in mind I have no technical data on the motors, loads, etc.

Maybe an approximation/rule of thumb based on measured DOL start currents and durations? I am pretty desperate as you maybe can tell.

In an ideal world, what information would you ask for to deliver this solution, assuming any information was available?

 

[Marke]

Hello 123MB

You are correct, the fans are typically a high inertia load, but in addition to the inertial load during start, there is a "work" load which is roughly proportional to the square of the speed.
If you reduce the start current too much, you will have insufficient torque to accelerate the fan to full speed because of the work torque speed curve oof the fan.
If the fans are started with closed shutters, then the work torque will be reduced. If the fans are started free ventillating, then you may require 450% start current to accelerate the fans freely to full speed.

Plot the speed torque curves of the motors at various start currents and over this, plot the speed torque curve of the fan (ignoring the inertial torque) and you can determine the minimum start current to accelerate the fan to full speed.

Best regards,
Mark

Mark Empson
Advanced Motor Control Ltd

 

[123MB]

Thanks for your reply, mate

 

[Steven1984]

Hello 123MB,

Mark is correct in what he is stating and there are companies around who will be able to help you with this application.

With high inertia fans of this size using a soft starter with the option for a current limit start will be a better solution than a standard timed voltage ramp.

With a current limit start you may be able to reduce the starting current to 350% the FLC but with more information on the load we should be able to make more recommendations.

Steve

 

[Steven1984]

123MB - Please feel free to contact me at the website in my signature.

Steven Atkey

http://www.fairford.com

 

[LionelHutz]

It's not possible to predict the required starting current or time without the motor and load speed vs torque curves and the motor speed vs current curves. Don't believe anyone that suggests it is.

You 1/2 current = 4 x accel time rule of thumb is very rough. 1/2 current would mean 1/4 the torque and the motor may stall or it may be very close to stalling which would make that rule very wrong. The problem, as already stated, is that the load keeps increasing as the speed increases and more motor torque goes to powering the load (moving air in this case) and less remains to keep accelerating the load. The acceleration time and motor heating will increase rapidly as the motor approaches stall conditions (little or no torque left to accelerate). There's a often cited NEMA formula for calculating the acceleration time of a motor which is about as equally useless as this rule of thumb.

Closed dampers on the fan would help your cause immensely. This can lower the load so it only reaches 40% of rated torque as opposed to reaching 100% of rated torque as the motor accelerates to full speed. This leaves more torque available for acceleration.

A soft-starter that only allows 40 seconds of acceleration would not be correct for this application. Also be careful of the soft-starter ratings. A 335kW starter which is rated for 350% current for 30 seconds certainly won't work on the 335kW motor.

 

[123MB]

Thanks guys -

Steven1984, unfortunately, I don't have any more information - I can gather information from site if I am convinced it will help. If I am convinced it won't (which I am leaning towards) then I will confess the inherent uncertainties in soft starter selection to the client. Ultimately he doesn't have the information so it's his doing in part.

LionelHutz - the SMC flex technically allows 30s at %600 FLA on a current limit start, so the starter is rated for atleast that... but I do not know the absolute rating of the starter - AB seem to be very guarded on information like this. Thanks for the information regarding the curves I shall have to have a look and see what similar information I can find for my application.

 

[jraef]

Also just so you know, you can override the accel time limit on the SMC Flex, but it may require you to up-size the starter. If you do that though, you might run smack into the thermal damage curve of the motor. If properly programmed, the motor thermal overload protection will of course prevent damage, but it still results in a failed application.

Without the torque speed curves of the fan, it's going to be a crap shoot (gamble). The only sure bet would be to use a VFD as the soft starter, but what are they willing to spend for a guaranteed solution? If the variable speed however can provide any additional benefit, that might be more palatable.

"Dear future generations: Please accept our apologies. We were rolling drunk on petroleum."
— Kilgore Trout (via Kurt Vonnegut)
For the best use of Eng-Tips, please click here -> faq731-376

 

[Steven1984]

Thanks 123MB

Soft starting is an excepted form of motor starting and we have done many installations on very large fans (larger than 335kW) with good results reducing the starting current dramatically compared to DOL and Star/Delta.

LionelHutz is right it is not possible to know exactly what the starting current will be without motor and load speed torque curves. We can however, make an educated estimation of the potential reduction compared to DOL and thus the current limit level to set to start your fan successfully.

I hope this helps.

Steve


Steven Atkey

http://www.fairford.com

 

[Marke]

The starting curves for the fan will be fairly generic, so we can make some assumptions there based on whether the fans are shuttered during start or not, we can always assume worst case.
The motors are a different story as there are considerable differences between motors and the motor characteristics have a major impact on what happens.
If the motor curves are available, then that is a tremendous advantage. If the motor curves are not available, we can still make some valid assumptions based on the LRC (LRI) and the LRT of the motor. This is commonly available information.
If this is not available, then the LRC can be measured as the DOL Start current and that will give a clue!

Best regards,

Mark Empson
Advanced Motor Control Ltd

 

[Drivesrock]

What is the system voltage, will there be a bypass contactor and how many starts per hour or consecutively could there be per blower?

How about a larger than 335kW soft starter with plenty of overload capacity and use it to start a motor and then the bypass contactor switches that directly to the mains and then use the starter for the next motor? Perhaps one large starter will be more economical than individual ones if bypass contactors are to be used anyway? And now you will have more flexibility with yr current limits and times. Some time delay may have to be allowed inbetween starts of the softstarter to allow its thyristor heatsink to cool down - is this acceptable to your process?