Consider that the soft starter is in circuit only for a short period, so anything it saves can only be saved during this period. Your payback period will be very long. It is possible with certain loads to use a device similar to a soft starter but with different control electronics to reduce the voltage to the motor and thus reduce the current supplied to it. Marke's FAQ covers this very well: FAQ237-971: Energy Savers for Induction Motors.
 

----------------------------------
  Sometimes I only open my mouth to swap feet...

In general, soft starters are not used to save energy, but often save money in the form of reduced maintenance costs. Y-Delta starters are very hard on the mechanical systems of your motor and driven load because when they transition from Start to Run, there is a spike of torque that occurs, sometimes even higher than if you had started across the line (DOL)! Soft starters will reduce or eliminate that damage from torque spikes.

But as to energy savings, not in 99.99999% of applications. If the soft starter has the "Nola Energy Saver" circuit in it as discussed in the FAQ, there is a very slight possibility of a minor amount of savings IF the motor is unloaded for long periods of time, but not if the motor is being used within its normal operating range. As Marke says in the FAQ, you can only save energy that is being wasted. If your motor is already running efficiently,  you are not going to be able to save much more than that.

If the motor is unloaded for long periods of time however, there is no better energy saver than the Off switch. A side benefit of a soft starter then is that it becomes less of a problem to turn motors off when not needed because there will be less electrical and mechanical shock when re-starting them.

JRaef.com
"Engineers like to solve problems.  If there are no problems handily available, they will create their own problems."   Scott Adams  
For the best use of Eng-Tips, please click here -> FAQ731-376: Eng-Tips.com Forum Policies

Hello mobex

When it comes to saving energy, you can only save a portion of the energy that you are wasting. Induction motors are inherently relatively efficient and are not the highest loss component in most systems. There are some out there, who are misinformed and claim that the induction motor draws close to it's rated power under all load conditions, this is not true. The induction motor is efficient down to below 50% of it's shaft load. Energy can be saved on overfluxed motors operating at near zero shaft load, but relative to the motor rating, this can be quite small.
There are also some who claim that by reducing the start current, you will save energy. This is also not true.
You will use the same amount of energy to start the load, just over a longer period.
I have also seen claims that the soft starter will reduce maximum demand penalties by reducing the start current. This is also not true in most cases, as the maximum demand metering is typically base on a half hour integral and so a few seconds has little effect. Infact, the area under the time current curve is greater with the soft starter so if anything, it will increase the average slightly.

What the soft starter will do, is reduce the torque and current transients. This will reduce disturbances on the supply and mechanical damage. This can result in the machine being turned off more frequently and energy can be saved by not operating the machine under idle conditions.

Best regards,

Mark Empson
http://www.lmphotonics.com

Could a SS allow a lower peak charge and so, indirectly, save some scatch?

Keith Cress
Flamin Systems, Inc.- http://www.flaminsystems.com

Most utilities have a "Demand Charge".  It's doubtful that one motor with a soft strart would make a difference but you won't know until you check.  
See what kind of demand charge your paying now.
The damand charge is what they can make you pay for the system they have to build to serve your load.
If you have a bunch of high horsepower motors and turn them all on at 7:30 every morning a few soft starts may save you some money.

Hi Keith

It depends on how the peak charge is metered. Over here, the peak charge is a maximum demand charge and is the maximum half our integral. In that case no.
If you have a peak charge tariff that is base on a small time interval, less than thestarting time of the machine, then yes you could save money. Most maximum demand (peak charge) tariffs that I am familiar with are either half hour or quarter hour integrals.

Best regards,

Mark Empson
http://www.lmphotonics.com

My experience with demand charges is the same as Marke's.
The demand ios over either a 15 minute or 30 minute interval.
Motor starting is almost completely transparent to the demand meter.
respectfully

Interesting - I always labored under the fear of a big motor start. Thanks for the clarification.

Keith Cress
Flamin Systems, Inc.- http://www.flaminsystems.com

Keith,
There are a few utilities here in California who use a ratcheting instantaneous demand meter, so a reduced voltage start of any kind will make a difference if the starting power of a large motor exceeds the demand limit they impose. Both of the utilities I am aware of are small 2nd tier suppliers who buy their bulk power from the bigger guys and resell it. None of the big utilities here such as PG&E, SMUD or SCE use that system however, they are all on sliding 15 minute demand windows.

jraef,

Can you provide any more information on the ratcheting instantaneous demand meter? Manufacturer for example? Even this type of meter must have a response time unless the goal is to charge or credit the customer for every sub-cycle transient that they didn't necessarily cause.

The purpose of demand charges is to recover the cost of building a system that can meet that demand. The system components used have thermal characteristics that will permit the short starting currents with negligible additional heating. Of course flicker is another issue, and soft starters might be required for this reason.

How much will a demand meter respond to a short peak?
An estimate based on time constants and rise times:
We used to calculate the rise time of a circuit. 5 rise times us accepted as close enough to 100%. The quantity measured will increase to 63.7% in one rise time.
If the meter uses a 15 minute demand period the rise time will be 15/5 and the meter will respond to 63.7% of a demand that persists for 3 minutes. That is it will indicate 63.7KW for a 100 KW load.
We can reiterate and use 3/5. In .6 minutes (36 seconds) the meter indication will be 63.7% of 63.7% or 41% of the total demand.
Again, 36/5 or 7.2 seconds, .63.7^3 = 26%
And again, 7.2/5 or 1.4 seconds, 67^4 = 16% of the total demand.
Now the energy demand of a starting motor is concentrated in the first part of the starting cycle and is highly reactive, so the demand meter responce to a fast starting, motor (One or two seconds start time) may be in the range of 10% or less of the maximum demand of the motor. Bear in mind also that we are concerned with energy demand, not current demand. Although the starting current may be 5 or 6 times the running current, this current is highly reactive and the energy or KW demand will be much less than the apparent energy or KVA demand.
I must retract my earlier statment that motor starting is transparent to a demand meter. A fast starting motor will have a small effect on a demand meter, a slow starting application will have a greater effect.

Bottom line, if you need to know;
You can break the starting period into small time blocks and calculate the progressive responce of the demand meter to each small segment of the starting curve (after correcting for power factor which will also be a changing value). this is a lot easier with spread sheets than the old days with slide rules. Your result will be based on a number of assumptions. (Assumed starting curve versus actual starting curve, assumed instantaneous power factor versus actual power factor).
For a more accurate indication of the actual demand meter responce to motor starting, go and look at the demand meter while someone starts the motor.
respectfully

Over here, the maximum demand metering is current based, not KW. The tariff is "maximum demand KVA"

The effect on the maximum demand meter is dependent on the area under the current curve.
If you have a high start current for a short time, it has a lower area than a reduced current for a longer time, so I believe that the reduced voltage starter can actually increase the Max demand reading rather than reduce it provided that the integral time is much larger than the starting time.
For example, if we look at a DOL start of an inertial load, it may take say 10 seconds to start DOL at say 600% current.
To start the same load at 300% current, is going to take 40 seconds. (the current is half, so the torque is a quarter and I assume zero work torque). The energy consumed is equal, but the average current is double.
I believe that these people advertising and promoting soft starters as a means to reduce demand charges are totally incorrect except in the instances referred to by Jraef which I have not actually found myself.

Best regards,

Mark Empson
http://www.lmphotonics.com