Motor starting current and genset size 1

[Munequip]

We have had a problem at a sanitary lift station and would appreciate any comments or suggestions.

The station is equipped with two (2) 28.2 HP (21.0 kw)submersible pumps and a 80 kw diesel generator with an 200A automatic transfer switch. The generator is sized ot operate two motors on a staggered start sequence.

The pumps operate fine when connected to utility power. When the pumps are operated via the stand by generator, the across line motor starter will "chatter" briefly (less than 1 second).

The submersible pump motor has a full load amp rating of 32.3 amps at 480 volts, and operates at 3450 RPM. The published locked rotor current for the motor is 245 amps. The NEMA code rating for the motor is "H".

During operation from the utility power, the in rush current has been measured between 380 amps and 420 amps (depending on leg and test). This was measured with a Fluke model 453(?).

When the generator attempts to start one of the pump motors, the control voltage will drop from 120 volts to ~ 75 volts. The control transformer is 5 kva. Two different transformers have been installed, with the second unit incorporating taps to try to provide some adjustment. The only items requiring control power at this point are the coils of the motor starter. There have been two different manufacturers of motor starters installed in the control panel

The submersible pump manufacturer states that the inrush current is with in an acceptable limit of 10 to 15 times FLA. They also state that since the units operate on utility power the issue is with the gen set.

The generator has had the voltage regulator changed, and the voltage has been adjusted to provide between 460 volts to 502 volts. The same results have occured. During the last round of testing , one of the units stopped "chattering", but the other unit has not (for no apparent reason or adjustment).

We would appreciate any type of response as we are currently baffled.

 

[GGOSS]

Howdy DoggyDog,

I don't dispute what you say about auto-trannys at all, in fact I have often offered them in lieu of soft starters. However the advantages they present (in terms of further reducing in line current draw) are limited to what I refer to as being 'light load' applications.

Take for example that we have an auto-tranformer starter with standard tappings of 50, 65 & 80%, a motor with an LRC of 6 x FLC and a load that requires 25% torque (very rare)to achieve rated speed. The line current draw will be; 3 x FLC with a soft starter, you could use the 50% tapping on the transofrmer and this would give a line current draw of 1.5 x FLC not taking into consideration transformer losses of-course

If we now consider at a load that requires say 49% start torque, the line current draw will be; 4.2 x FLC with a soft starter, the 65% tapping on the transformer could be selected however as this will not allow the motor to achieve full speed before the transition to full voltage, when the trnasition occurs you can expect the a line current draw somewhere between 2.5 to 6 x FLC. Alternativelt you could select the 80% tapping which would give a line draw of 3.84 x FLC excluding transformer losses.

From the above you will note that the difference in line current draw via a soft starter or auto-transformer starter are similar for heavier applications. The advantage that the soft starter provides is that you can program things like current ramp starts (low initial current setting ramping to maximum current setting) which basically gives the governor in a gen-set more time to respond to the changing load condition thereby providing better overall performance.

The other significant point to note is that the transformer will provide a reduced start torque right throughout the acceleration period. With the right soft starter (closed loop controlled current), the available motor torque increases as the motor approaches rated speed making it a far better option for starting difficult loads.

So in short, an auto-transformer starter can reduce the starting current to levels below that of a soft starter in some applications. There are however several limitations as noted above.

Regards,
GGOSS

 

[jraef]

Well stated GGOSS.

Yasas!

"Our virtues and our failings are inseparable, like force and matter. When they separate, man is no more." Nikola Tesla
Read the Eng-Tips Site Policies at faq731-376
Member, [blue]P[sup]3[/sup][/blue]

 

[LionelHutz]

I have no idea why they all insist on providing frequency diviation detection functions with softstart.

Lets put it in the simplest terms. You need to time between zero crossings to calculate the line period. Then, you need to figure out where in the cycle you want to fire the SCR and time from the zero crossing to the firing point. Using analog circuits or a processor you'll always end up with some limits on the range of timing you can use. Since you have limits on the timing range you also have limits on the operating frequency range.

The other significant point to note is that the transformer will provide a reduced start torque right throughout the acceleration period. With the right soft starter (closed loop controlled current), the available motor torque increases as the motor approaches rated speed making it a far better option for starting difficult loads.

This is where the big advantage goes to a soft-starter. A pump follows a squared load function, ie the load increases by the square of the speed increase. So, as you accelerate the motor you need more and more torque. Using a soft-starter you can ramp up the torque output to better match the pump load requirements.

I've been involved in applying soft-starters on many generator/pump applications and you've got lots of generator for those two pumps. We've even developed software to control the power instead of the current since a generator is more affected by power drawn then kVA drawn. Using tricks like this has allowed us to start motors on generators down to a 1:1 kW rating ratio. We like to see the generator sized >=1.5x the motor though.

 

[abcd3286]

I have some thoughts based on 4 years operating electrical generation and distribution on submarine (turbo DC and AC motor and diesel driven alternators / generators)

1. When the transfer switch switches how is load picked up engine NOTE I did not say alternator - on purpose
2. How large is that load?
Each of these 20 KW motors starting load is 120 KW. You said the genset was 80 (??). I think you would grunt to at 50% overload.
3. You make no mention of what frequency (engine speed) does on starting these loads.
4. The engine just may not be able to handle the load you are imposing on it. IT CANNOT START EVERYTHING AT ONCE
5. What is switch over time on transfer switch? How long does it take to get engine up to speed.
6. You have to be smart and selective when loading down any generator.
7. All you have been doing is what is called "Easter Egging" in the Navy -- you are guessing and throwing parts at the problem.

Put some CTs in there and a frequency detector and use a high speed data logger to record at intervals of 1 to 5 msec(current voltage frquency) what is happening.

If these two pumps are the largest load I would start them first then start everything else. This is called selective loading.

Dan Bentler

 

[abcd3286]

One other thing I forgot to mention.
If engine speed drops too low then voltage WILL drop no matter how good the voltage regulator is.

We started 75 KW feed pump on 250 KW -- 250 VDC diesel gen. The first time we tried it the engine just shut down. We first thought we had stalled it. What we found out there was a solenoid in the governer fed from 250 VDC. Voltage dropped when starting the feed pump of course,
the solenoid dropped out, the governor "dropped out" and the engine stopped. We used the manual override on the governor when starting feed pumps on the diesel - worked fine.

I believe your genset is just fine, the motors are just fine, you just have a problem in how you are loading the engine.

Change your loading sequence.
OR
use soft starts
OR
use VFDs which may also give you a dollar returne by saving energy when you don't have to run both pumps at full speed.

Dan Bentler

 

[aolalde]

Munequip:

I identify two problems in your application:

1- Your submersible pump motors have very high inrush current, 13 times the full load current. For 420 amperes measured while starting on the utility line, the power demand at 480 volts is: kVA i = 1.732*480*420/1000 = 349.2
The first motor load, kVAm = 1.732*480*32.3/1000 =26.85, adds to the inrush load.
When the second motor starts the demand is 376 kVA with low power factor.

2- The generator is to small and possible “Stand- By type”
On average a maximum current of 3 times the generator full load current is allowed for 10 seconds before thermal damage is developed to the generator windings. Assuming the generator has 0.8 power factor your max current is 376/100 = 3.76 times the full load current. That is above the safety limits and the expected voltage drop is higher than normal.

Probably that generator is good to start two 28.2 HP motors but “Nema design B type” with 6 to 6.5 times inrush current, but not the extremely high inrush submersible pump motors.
Since 13 times Ifl is twice the 6.5 normal inrush current, I guess you need around a 200 kVA Generator to minimize the voltage drop and prevent damage to the generator.

I wonder if the pumps will start with reduced torque, so the recommended inrush reductions of different types could work, other way the generator size has to be increased.

 

[jraef]

My interpretation of his "10 to 15 times FLA" statement about inrush as being his meter picking up the true inrush, or magnetizing current, of the motor as opposed to his Locked Rotor Amps, and that only lasts a couple of cycles. He likely had his Fluke set for Peak reading, and the reading rate was set so high that it picked up and recorded magnetizing current.

Whenever I see odd HP sizes like that in the US on submersible lift station pumps, 9 times out of 10 it is a Flygt pump. They use their own custom motor, but it has a Design B torque/speed curve for the most part. LRA is 6x FLA, nothing special.

"Our virtues and our failings are inseparable, like force and matter. When they separate, man is no more." Nikola Tesla
Read the Eng-Tips Site Policies at faq731-376
Member, [blue]P[sup]3[/sup][/blue]

 

[digitrex]

My suggestion is to make simple modification initially which is easier to do instead major change in the starter. Just service the existing contactor. I give my experiences on contactors:

1. I encountered some large contactors chattering upon closing for a few seconds. We opened up the contactors and cleaned the internals. The chattering noise disappeared.

2. We are using UK voltage (415V) and contactor coil is supposed to be 240V. But in our market here, 220V coil is very popular and by strictly complying to nominal voltage of 240V it will means special order and long delivery time. However, use of 220V coil makes us happier since the drop-off voltage is generally lower, around 45% - 65%. For more than 10years of use of 220V coils(we have a few hundred LV motors) on 240V source, we haven't encounter any failure.

Try the above 2 simple approach first before you attempt on more complicated solutions.

 

[DoggyDog]

jraef,
Being less expensive and actually working are two different things! This is eng-tips not sales-tips! By drawing less current, there will be less loading on the alternator and prime mover = closer to rated voltage and frequency. BTW, Autotransformer starters are available at this power.

GGOSS,
I agree with your statement completely. Your calculations also prove my point (more succinctly than I could ever do!). Would you also agree that a submersible pump is considered a 'light load' application?

 

[digitrex]

Perhaps you may want to ask how do I know our contactor coils' drop-off voltage is around 45..65%.

Voltage sag/dips is something that we cannot avoid. We have transient disturbance recorder installed at our infeed station, therefore we know how bad are the voltage dips.

When there is a voltage dip, and some motors drop off line, we will investigate the contactors' drop-off voltage by use of a single-phase variac. We will improve it by cleaning the contactor internal components or just replace the whole contactor.

 

[Laplacian]

I disagree that the UPS is the best solution to the problem based off maintenance requirements of the UPS itself.

A constant voltage transformer is more suited for this application and can hold the voltage to 90% rated output with 70% nominal input voltage continuously at full load. They can ride through 3ms at zero input voltage. They work even better at lesser loads or if oversized.

The CVT's also require no periodic maintenance and are suitable for the same hot environment outdoor motor starters are usually in.

For a starting point, check out the Sola-Hevi duty MCR and CVS series. I think they patented or invented the CVT?

 

[jraef]

DoggyDog,
So you are saying that RVSS starters don't work? You are entitled to your opinion, but I beg to differ. Also, I found only one price shown on-line for RVAT starters that small, and it was almost 4 times the average RVSS price of equal rating, US$2,500 vs US$650. Even if that price was off by a factor of 2, twice the price is still a LOT more expensive, not a little. So I guess if you HAD to waste money based on fear of electronics, at least do it grandly, or "2 grand"-ly as this case may be.

"Our virtues and our failings are inseparable, like force and matter. When they separate, man is no more." Nikola Tesla
Read the Eng-Tips Site Policies at faq731-376
Member, [blue]P[sup]3[/sup][/blue]

 

[Marke]

Hello Doggydog

Having been actively involved in the industry for many years, I can confirm that while auto transformer certainly can be used as starters of induction motor, and that the line current is reduced by the transformer action such that the line current drops by the square of the voltage relative to the full voltage starting current, there are many circumstances where the benefit of this action becomes masked by the constant voltage torque curves.

There are a number of different types of starter available and all yeild different charascteristics.
The star delta and auto transformer starters are constant voltage. The reduced voltage start torque curve is equal to the full voltage start torque curve multiplied by the square of the voltage reduction ratio over the entire speed range. The load curve often follows a square law curve and is low torque at low speed but rises rapidly as the motor approaches full speed.
If you take a set of motor speed torque curves and then plot the 50% tap curve against the start torque curve for a pump, you will usually find that the motor does not have sufficient torque to accelerate the motor to full speed.

I have posted some example curves to

http://www.LMPhotonics.com/start_curves/start_curves.htm

These clearly show that the autotransformer in these cases is not superior to the soft starter in reducing the line current. There are certainly situations where the auto transformer does provide a lower line current, but there are many that do not. The best applications for the autotransformer to provide a superior line current during start, are primarily inertial loads where there is little work current requirement.
One of the problems with the AT starter on gensets is that it increases the KW loading on the engine more than a soft starter does due to the power dissipated in the transformer.

Best regards,

Mark Empson

http://www.lmphotonics.com

 

[ScottyUK]

Marke,

Good post as normal, but the last line

One of the problems with the AT starter on gensets is that it increases the KW loading on the engine more than a soft starter does due to the power dissipated in the transformer.

puzzles me. Are you talking about the small losses due to I[sup]2[/sup]R in the windings and the magnetising losses in the transformer? I'm surprised that they are larger than the V[sub]F[/sub].I and switching losses in the solid-state switches of the electronic unit. The benefits of the solid-state starter are many, but this seems a very marginal one!


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

One day my ship will come in.
But with my luck, I'll be at the airport!

 

[Marke]

Hi ScottyUK

There are reasonably high losses in the autotransformer during start. They are intermittently rated with a limited number of starts per hour. The losses of the soft starter are definitely lower at around 4 watts per amp.
The primary resistance starter is probably the worst starter to use on a gen set due to the very high KW losses in the resistors which really load up the engine.

Best regards,

Mark Empson

http://www.lmphotonics.com

 

[easpe]

Aolalde is correct starting KVA with those motors is to much for a 80 KVA genset. Volt drop is greater then normal design causing contactor coil field to colapse most likely. Good design would suggest different way of starting, differnt pump or bigger generator. However if motors start with ups, CVT or other method suppling contactor coil see how long everything last you might luck out depending on outages.