real poer question 1

[jcraft]

This is puzzling me so hope you can help. We have injection molding machines in our plant. We have a power analyzer hooked up to one to monitor the power consumption and stuff. Looking at the load profile for about 2 hours on this machine, it draws anywhere from 2.5 kw to 20kw depending on how the heater bands and pump motor are cycling. We have occasional peaks up to 25 kw for 1-2 sec. intervals. the data was taken at 1 second samples. The strange part is that during this 2 hour period about 4 - 6 times for one second the real power is actually negative. We looked at the actual data table and for one second it goes negative then it is back up to the standard kw. This is what we read from the data table for this one second duration. Apparent= 16.6kva, real power = -1.44 kw, Ractive = 16.54 Kvar and PF = -.09 ( Pf is usually .4 to .87). This is puzzling why the Kw would be negative and the PF goes into the negative for such a short duration. Any light shed on this would be wonderful. Thanks

 

[buzzp]

Sounds like the heater is supplying power back to the grid based on your readings. The inductive properties of the heater make this possible. Thats the best I can come up with, assuming the data is actual (no errors in hook up , etc)

 

[jcraft]

we have done another test with just the heaters and these heaters are almost totaly resistive in nature. The Pf is 99% when doing just the heater loads and they are about 2kw woth of heat.

 

[bacon4life]

What kind of heaters are used for injection molding? I thought heaters were usually resistive.

 

[jraef]

Any chance that you have power factor correction capacitors on the motor, and when the motor turns ff the caps are left on?

Eng-Tips: Help for your job, not for your homework Read faq731-376

 

[jghrist]

You may be getting some angular error in the CTs. At very low power factor, a small angular error could cause an indication of negative real power.

 

[waross]

Have you rechecked the connections on the analyser?
Do you have severe voltage dips in the plant?
I would guess the pump to be 20 Hp to 25 HP. Is that close?
One explanation may be that some external mechanical force is overspeeding the motor for a moment. Is this physically possible with your pump? Do you have a valving sequence that can do strange things to the flow through the pump?
If you had a mechanical load, and showed me those readings, I would say the load was overhauling the motor, but with a pump it is puzzling.

 

[davidbeach]

I think that jghrist is on to something. Voltage combined with noise for current can produce some very interesting, but totally meaningless results.

 

[jcraft]

That is some interesting things to think about. It's hard to comprehend though how real power can be negative. If the machine is operating normally and you are consuming real power to do work, how can it ever go negative.

 

[rcw retired EE]

When an induction motor stops, it can act like an induction generator as it spins down until the energy stored in the magnetic windings is dissipated, usually just a few cycles. If there are caps across the motor it might become self excited for a short period of time, maybe enough to fool the data logger.

 

[Senselessticker]

I agree with rcwilson. The simplest answer is there are some motor loads (most likely while spinning down) are putting "Watts" back into the system (just like a generator). Having a the pf show up negative during those times supports the idea. Depending on how the equipment "reads" pf...a neg. pf would just indicate a shift from lagging to leading (or motoring to generating, inductance to capacitance, etc...)

I don't have much real world experience with such a matter. But I know theoritically there is no reason why a motor can't act as a generator from time to time depending on what the load is doing.

 

[waross]

An induction motor ussually acts as an induction generator in two situations.
1 An over-hauling load drives it faster than syncronous speed.
2 There is a fault on the system. Hence my query as to severe voltage dips in the plant.
What-ever the cause, I'm going to sit back and listen for a while. This may be a good learning opertunity for some of us.
yours

 

[stevenal]

I've seen this occur when the monitor current probes chosen for the job are for high currents and the actual load is much smaller. Is your load well within the probes' range? And some require batteries. Batteries good? Battery contacts?

 

[Marke]

If the motor is spinning down, it wil act as a generator, but that is because it is disconnected from the supply. If it was connected to the supply, it would not be spinning down. How can a disconnected motor put power back into the supply?

Best regards,

Mark Empson

http://www.lmphotonics.com

 

[jghrist]

Good question, Marke. I was just about to be convinced that my metering error answer was wrong, but now I think I'm right, at least about the negative power part. The reason I question myself is that I can't figure out where 16.54 kvar of reactive load would come from with negligible real load. Even if the motors were all running with no load, I can't imagine them having that much reactive load considering that the peak real load is only 25 kW including heaters. I think maybe a wiring problem in the measuring circuit is more likely.

 

[rcw retired EE]

Mark - You are correct about the energy in the motor having no place to go as it spins down - unless it is still connected to another load or part of the system. Without knowing the circuit arrangement and the location of the data logger in the circuit it is difficult to say what is happening. Was some switching occurring upstream?

I have taken oscillograms and recordings in industrial power systems that show induction motors supplying power to other loads. Usually it is when an upstream breaker or transfer switch opens. It never lasts long. (5- 10 cycles except for large motors > 2000 kW.)

I have seen where starting a large motor seems to suck power from the operating motors on an isolated system powered by diesel generators. Frequency dropped during the motor starting so the on-line motors would be running faster than synchronous speed momentarily and act like induction generators. It would be very unusual to see that occur on a utility supply, unless something in the process caused the pumps to accelerate.

I think you are right -Meter error due to low currents is the most likely explanation.

 

[waross]

rcwilson
I agree with you.
The text books used to say 1 or two cycles.
In support of your observation of 5 to 10 cycles duration;
I think that the time is set by the motor coasting down. Remember that when it is acting as an induction generator, the energy that is regenerated is aupplied by the rotational inertia of the motor. The inertia doesn't last long. Large motors > 2000 kW. have a lot of inertia.
Many years ago, I took a class out to the shop to demonstrate regeneration, or induction generation if you prefer.
We coupled a DC motor to an AC motor. Our various machines were in the 3 Hp. to 5 Hp. range. We connected a load bank in parallel with the motor. We connected an analyser in the circuit. We started the motor. We then energised the DC motor.
As we increased the torque and the speed we had a good example of the motor current reducing and then actually transfering power to the resistance bank. Then the power to the motor and resistance bank was interrupted, whether intentionally or accidentally I don't remember. The voltages and currents all changed, but stabilized at new values and continued stable. The motor was still delivering power to the load bank.
Years later, different institution different class, different mix of equipment in the shop. I tried to repeat the demonstration. The system would regenerate while the motor was being over driven, but when the main power was interrupted the voltages and currents decayed to zero almost instantly.
I have seen a couple or other examples over the years.
I am trying to visualize how valve operation could push a pump over speed. Any scenario I can imagine is not very probable or believeable.
I'm with you on meter error on this one. Possibly a sampling error.
yours