Inrush, transformers, and potential energy in induction motor.

[fastline12]

Have a few odd question and figured I would roll them all into one.

1. I believe a running but unloaded 3P motor on a electric service has the effect of damping inrush seen by the power company when another motor starts up? IE, the already running motor helps start the next motor. Is there a way to calculate or quantify the potential energy held within a running induction motor?

2. Would it be accurate to say an isolation transformer behaves in a similar way to hold potential energy and would work to snub inrush currents seen on the primary side? Nevermind the obvious difference in current due to any step up/down in voltage. Just focusing on inrush that might be seen on the secondary, but not so much on the primary.



We have a limited electric service that cannot have nasty inrush spikes so as long as we keep our current draw as consistent as possible, our power company will be happy. In thinking about snubbing other motor starts, we considered just getting a rather large 3P motor to run as an idler to help snub inrush currents. What I would like to do is start the big motor on a small VFD and ramp it up VERY slow so as to not trip the VFD on over current. Once the motor is to speed though, I need to switch the motor right to grid power and shut down the VFD. just using the VFD as a soft start. I know they make soft starters but but sure how soft they really are and we would be starting a good size motor and would not be able to support the inrush requirement.

 

[itsmoked]

Yes, other motors do provide some buffering. They do this by briefly letting their loads drive them. That means it all depends on the load they're running. Some loads might not give back much or any, others might work quite well. The best would be a completely unloaded motor driving a heavy sheave or flywheel. However an idling motor has hideous power factor and is usually undesirable.

Yes, a transformer shields the primary side by attenuating what would normally be sent back as the transformer has impedance that reduces how much impulse makes it back. However. Energizing transformers can cause very large hits all by themselves.


It's not easy and I have no idea how to switch a motor from VFD to DOL while it's running. VFDs generally do not like having their outputs messed with while they're running. Typically it destroys their output stages. You're not supposed to put contactors between them and the motors.

If you really aren't going to use them other than for a starter you should just use a soft-starter as they are designed for that or you should use a variable voltage auto-transformer type starter that ends up being DOL once fully started.



Keith Cress
kcress -

http://www.flaminsystems.com

 

[fastline12]

Smoked, I had the very idea about the variac for soft starts but thought I might try to do something more automated. What would you think about some wye/delta switching along with a series choke that would be switched out once up to speed? I cannot remember the calculation of how much current or power the motor would have if I employ starting with low voltage in delta configuration.

 

[itsmoked]

If I recall the power jumps from about 1/3 to full on a WYE-DELTA starter.

But, if you're trying to avoid jolting the supply then Y-Δ is an unmitigated disaster. There are several threads in here that illuminate the issue with Y-Δ starting. The short story is that Y-Δ starters are "open transition" starters unlike the auto transformer type I suggested. That open transition means the motor is disconnected from the network briefly during the transition from Y to Δ. That means the motor will be generating on it's residual magnetic field and completely out of sync with the power grid when it is re-connected in Δ. That makes it a crap-shoot as to what happens. Statistically you will re-connect them 180 degree out at some point. That will cause whopping big shocks to the network that are far greater than just a standard DOL start.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[waross]

A synchronous motor may provide some voltage support. An induction motor will not provide support to the grid unless it is operating as an induction generator. To operate as an induction generator the motor must be running over speed.
A motor will back feed energy if the frequency drops enough. A motor WILL NOT back feed energy if only the voltage drops.
If there is a fault on the supply to a motor, the grid frequency as seen by the motor may drop to zero. Now the motor will effectively be running over speed with respect to the grid frequency and will back feed into the fault. This won't help you to lessen motor starting surges.
Transformers do not store potential energy.
A load on a transformer will cause a voltage drop on the output of the transformer. When the voltage to a motor is lower, the current is lower. Bearing in mind that the starting surge of a motor may be 6 times the FLC, a small transformer may reduce the surge to only 5.5 times the FLC. Not much help there.
The action of a transformer in reducing surges is the same as the action of a series reactor.
The effective series impedance of a transformer is influenced by the power factor of the load.
You may use the regulation of the transformer to calculate the series impedance at a specified power factor.
You may use the impedance voltage of the transformer to calculate the series impedance under short circuit conditions.
The effective impedance of the transformer during motor starting will be between those figures.

A large motor and a soft starter will be expensive and won't really help.
Spend the money on soft starters or VFDs for problem motors. That will help start the motor but it won't help start the other motors.

By the way, you may be better to use a small VFD and a small motor to spin a large, unloaded motor up to speed to avoid the inrush.
That will avoid the pitfalls that itsmoked mentions.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[fastline12]

Waross, it was my understanding that you cannot sidestep the inrush of an induction motor by spinning it to full speed because the coils are not yet energized and still present a low impedance condition when connected? This makes me want to test it though just out of curiosity!

 

[waross]

The majority of the inrush is caused by the high slip frequency and resulting low impedance of a stationary motor.
The speed is critical. If you spin a 1760 RPM motor to only 1720 RPM you may still see a short surge of about 200%.
You want to be between rated speed and synchronous speed. 1780 RPM would be good for a 1760 RPM rated motor. 1800 would be better.
As well as the motor starting surge there is also a motor starting transient, similar to a transformer surge but very short.
The transient is so short that it is ignored or not detected by many protection devices.
It may be interesting to check for residual voltage at the motor terminals of a spinning motor. I expect the residual voltage to be negligible. If the residual is high enough to cause issues this may be mitigated with a sync check relay.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[itsmoked]

I'd just live with the VFD running the motor as designed. Skip the any DOL transition. The VFD will provide the lowest possible footprint back to the grid.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[TugboatEng]

What is the horsepower of the motor?

 

[fastline12]

I will try to explain more. We have enough 'amps' in the electric service, BUT this is a rural area and the PC has demanded that we not have anything more than 10HP motors on the line. I find it ridiculous, but I will act like I will comply. The way I see it, as long as we are not hitting the line with nasty inrush spikes, we might be running a bunch of resistive ovens for all they know.

My thought process was start, run, and idle a big motor when there is lots of activity on the line to help buffer the inrush seen at the meter by the PC.

Now regarding loads, most machines run on VFDs so starting current becomes very easy to predict. The nastiest load we have is only a 30HP accelerating to over speed in 3sec. so it will pull about 180A or so during accel but do this many times per hour! This is a machine tool. There is also general support equipment such as 10HP compressor that will start across the line unless we add a VFD, and other smaller loads, but there is no way to ensure two motors don't start at the same time. Pretty much bank on that happening.

My thoughts were to deploy or have in hand several tactics to help here. Idling motor, possibly slow the accel rate on the motors, possibly double the capacitance for the DC tank on each big VFD with delay on make switching so those caps won't energize at the same time as the factory caps. I am """hoping""" the caps can provide an easy way to store up some good power and on our line regen drives, these caps will help how much power is sent back to the line.



All of this is in effort to simply "smooth" the curve regarding amperage demand seen by the PC. If we are pulling a near constant current, I understand that they have intelligent regulators on the line that will calibrate and be happy. It is when the voltage keeps changing from all the different load changes that starts to cause issue.

 

[fastline12]

OK, I did some single phase testing at home because I have several phase converters around and already know what they pull and they are easy to test with.



Inrush is all on the single phase side here. All 240V
The inrush seen directly on the dedicated service wire to a 10HP converter = 155A
Inrush to that same converter seen at the meter, to include lights, air conditioner and such running = 130A
Inrush to that same converter after starting two other 10HP converters (total of 20hp idler on the line) = 90A at the meter.


that seems to be a pretty decent difference! I wonder what would happen if I had 50-100HP of idler on the line??

 

[waross]

Motors may not give the same results as converters.
Why?
There are a lot of capacitors in converters.
Capacitors draw leading reactive current.
Motors draw a lot of lagging reactive current when they are starting.
Leading reactive currents will offset lagging reactive currents.
Converters are a basically a special motor and some capacitors.
The converter motors are not aiding the starting motor.
The capacitors are supplying a lot of the reactive current drawn by the starting motor.
Why not just use capacitors instead of a motor. Yes that will work somewhat, BUT.
Many years ago (40 or 50 or more years ago, it was an old text book when I was young.)
The use of capacitors was considered to reduce the starting surge of motors.
There were a number of reasons why the technique did not become common.
Capacitors, too much capacity will drive the voltage up and shorten the life of incandescent lamps. This was more important way back when.
A lot of capacity was needed for a very short time.
As I remember the scheme there were three banks of capacitors that were disconnected in steps AS THE MOTOR ACCELERATED.
You needed at least two more contactors.
Cutting the capacitors out of the circuit at the proper time was problematic with the equipment available at the time.
The number of critical adjustments and failure modes made the system less reliable than other systems.
If the system failed you may not realize the failure had happened until you got a large demand charge on your power bill.
And the big one;
Other reduced current starting methods gave more "Bang for the buck."
It works on paper, but I've never seen the scheme in the field.
I'll be interested to hear if any other forum members have seen such a starting method used.

Listen to itsmoked; Spend the money on VFDs.

A typical 10 HP three phase motor will draw about 25 Amps at 230 Volts. This may draw 150 Amps when starting. With an "Out of the box" VFD you can probably start the motor without exceeding 30 Amps.
No need to mess with the capacitors, and they will not supply reactive current back through the rectifier to reduce the starting current of other motors.
Note on air compressors:
If you ramp the speed up slowly on a compressor, the unloader may load the compressor before you are up to speed and pssibly stall the motor. You need the flywheel effect at full speed.
The solution is not to mess with the starting method but to change or modify the unloader so it does not load up too soon.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[TugboatEng]

$35,000 ish buys a top of the line 99kW diesel generator set that will probably run your loads. You'll still want a soft start on that 10hp and 30hp motor.

 

[fastline12]

LOL, we might get to the point of using a genny if we piss off the power company too bad!

On an interesting note, I have a 3 phase, 10HP motor that I tested for fun. Ran it across the line to see inrush, then disconnected it for about 2sec, then reconnected while still very close to sync speed. At about 2sec disconnected, when I reconnect the motor pulls almost the same as unloaded amps.

Was just an interesting test that Waross was right on about. If the motor is spinning at sync speed, the inrush is minimal.

 

[waross]

Be care full with momentary disconnections. That has been known to break motor shafts or spin the rotor on the shaft.
It is more of a concern when PF correction caps are directly connected to the motor, time is less than 2 seconds and the load has high inertia.
When the motor is disconnected, the caps tend to hold up the residual voltage. If the motor is reconnected when the back EMF is out of phase with the grid there may be a violent current and torque transient.
This is by no means a common occurrence but it can happen.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[LionelHutz]

You're doing this on a 1-phase supply, right?

Have you considered a MG set, possibly with a big flywheel as well? Then, you just have to get it going and once it's running, any transients in your shop will be reasonably well buffered from power grid. For 1-phase, the prime mover can be built similar to a rotary phase converter. The big issue with this is that it would not be very energy efficient. The last time I metered an arrangement like this, the MG set was drawing 6kW continually and the operation would use 20-30kW as the various woodworking equipment was operated. So, the place got to 80% efficiency best case and used quite a big of wasted energy due to no equipment operating with the MG set left on.

Another thing that can help is using a single active front end with all the VFD's running from it. Basically, the grid to capacitor bank is a single converter and then the motors each have an inverter that runs off the common capacitor bank. This allows regen energy from a stopping a VFD to be directly used as power by the other VFD's instead of being converted back to the grid and then drawn from the grid by another VFD. Cutting out the grid really helps if the grid is single phase.