Power Factor Correction on Transformers?? Will there be a KW reduction?

[prs1jazz]

Good morning. I hoped you all could help me out with a few questions about PFC on transformers. I am the COO of a couple plants in the north east and south west and have been on a mission to reduce our energy savings (beyond lighting, VFD's, controls etc... focusing in on inductive loads). Recently we have been doing a study at our Investment Casting facility to try to save KW on transformer based injection molding equipment using a technology that has extremely fast switching on each leg without capacitance to eliminate KVAR. (trust me when I say that I have ran into several black box technologies out there and vetted the living heck out of them.... this is not one of them.) This technology actually operates with the parasitic load of a lightbulb and without capacitance without disrupting the frequency. Understanding that KVAR doesn't lead to KW decrease, we thought on transformer based applications where core loss occurs that the streamlining of KVAR would help eliminate the burn of the step down. It was very interesting that at night when the machines were off but the transformer was running that we still saw a 9-12% decrease in KW. I tried PFC directly on a small current wax press but actually saw an increase but it could have been due to the extremely low current (under 6 amps) and that the switch was not engaging properly.

My question is, am I dreaming of KW reduction with PFC on transformer based applications? If I am not, should we be as close to the transformer as possible or the switch gear?

You all are much more knowledgeable than our group, so I want to thank you all in advance for the help for my amateurish question.

Have a great weekend!

 

[kwired]

Are you trying to correct the PF of the transformer itself?

The PF of that transformer will change as the load on it changes. Of course exactly how it changes will depend on the PF of the load as well.

A transformer that has primary energized but no load should have a low power factor in comparison to same transformer with rated full load @ unity PF.

 

[prs1jazz]

I should mention that the wax injection press only has a small 120v step down transformer from 480v.

One me other question would be "what happens when you PFC a transformer?"

 

[ron]

PF correction is usually only worth it if you are getting PF penalties from the utility.

http://static.schneider-electric.us...keycode=n846v&promocode=16319&promo_key=16319

 

[prs1jazz]

Thanks Kwired for the reply. Yes, we are trying to correct the transformer so that during the entire low current to high current cycles it would perform at unity.

 

[prs1jazz]

Hi Ron. We do understand that PF doesn't matter on billing unless there is a charge or demand KVA purposes. What I am trying to better understand is how transformers react under correction.

 

[GoldDigger]

Are you trying to correct the PF of the transformer itself?

The PF of that transformer will change as the load on it changes. Of course exactly how it changes will depend on the PF of the load as well.

A transformer that has primary energized but no load should have a low power factor in comparison to same transformer with rated full load @ unity PF.

The PF of the transformer will indeed change with load, but if the load is always PF of 1, then all the capacitors need to do is balance the magnetizing (zero load) current of the core and it will automatically correct all of the partial to full load PF values.

 

[prs1jazz]

We are correcting on the primary side of the transformer. Thanks again!

 

[Besoeker]

Good morning. I hoped you all could help me out with a few questions about PFC on transformers. I am the COO of a couple plants in the north east and south west and have been on a mission to reduce our energy savings (beyond lighting, VFD's, controls etc... focusing in on inductive loads). Recently we have been doing a study at our Investment Casting facility to try to save KW on transformer based injection molding equipment using a technology that has extremely fast switching on each leg without capacitance to eliminate KVAR. (trust me when I say that I have ran into several black box technologies out there and vetted the living heck out of them.... this is not one of them.) This technology actually operates with the parasitic load of a lightbulb and without capacitance without disrupting the frequency. Understanding that KVAR doesn't lead to KW decrease, we thought on transformer based applications where core loss occurs that the streamlining of KVAR would help eliminate the burn of the step down. It was very interesting that at night when the machines were off but the transformer was running that we still saw a 9-12% decrease in KW. I tried PFC directly on a small current wax press but actually saw an increase but it could have been due to the extremely low current (under 6 amps) and that the switch was not engaging properly.

My question is, am I dreaming of KW reduction with PFC on transformer based applications? If I am not, should we be as close to the transformer as possible or the switch gear?

You all are much more knowledgeable than our group, so I want to thank you all in advance for the help for my amateurish question.

Have a great weekend!

No.

 

[kwired]

The PF of the transformer will indeed change with load, but if the load is always PF of 1, then all the capacitors need to do is balance the magnetizing (zero load) current of the core and it will automatically correct all of the partial to full load PF values.

I kind of suspected that. I guess one could install a capacitor to correct the transformer at no load, and then install additional correction at low PF loads supplied by the transformer. Otherwise you need some system at/near the service that can be adjusted to the right correction level depending on current net PF for the service.

 

[prs1jazz]

Sorry for the ignorance but if the Pfc installed on the primary side of the transformer can fluctuate with the load on the secondary side of the transformer, would you still benefit from eliminating core loss hence a decrease is KW? Thanks again guys... Much appreciated.

 

[Besoeker]

Sorry for the ignorance but if the Pfc can fluctuate with the load on the secondary side of the transformer, would you still benefit from eliminating core loss hence a decrease is KW? Thanks again guys... Much appreciated.

You can't eliminate core loss.
Unless you reduce applied voltage but then you wouldn't get rated voltage out.

 

[prs1jazz]

Thank you. Would there be any benefit to PFC on transformers then?

 

[gar]

140824-1556 EDT

prs1jazz:

Core loss is a resistive load. Power factor correction has nothing to do with changing core loss (core loss is a resistive load, heat) except in very minor residual ways that are insignificant.

A change in the input current from the reactive (inductive) load resulting from the magnetizing current to the transformer is the only thing correctable with power factor correction, and since this reactive current is produced by a non-linear load the correction is not very good with capacitors. This magnetizing current is a small factor when compared with the full load current of the transformer. Power factor correction does nothing significant to currents within the transformer. It only modifies the current on the feeder lines to the transformer-capacitor combination.

To reduce transformer losses --- (1) turn power off to the transformer when it is not in use, (2) buy a lower loss transformer, (3) figure out how to use less energy in the end process and then buy a smaller high efficiency transformer.

I do not believe you are a trained electrical engineer, and therefore these concepts may be difficult to understand. You have to be very careful when someone tries to sell you on some energy saving technique.

One illustration is the application of a power factor correction capacitor to your home. Great claims are made for large power savings. These are false claims. Under very controlled test conditions (hard to perform) you probably won't see +/- 1% change and if any change occurs it is the result of peripherial factors. You might actually see an increase in energy use rather than a reducetion.

The typical demonstration is an unloaded fractional Hp induction motor, very poor power factor. When loaded the motor power factor is much better. An ammeter is used to measure motor current in this demonstration. Without power factor correction the current is high, 4.71 A and 552 VA, and with the correction capacitor the current is much lower, 1.33 A and 158 VA. Note: the experiment is done with an ammeter and not a wattmeter. As a residential customer you do not buy amperes or volt-amperes, but rather watt-hours (energy). Thus, the experiment is not of a parameter that you buy from the power company, and the demonstration is a fraud. The power input in my experiment with no capacitor is 140 W and with the capacitor 142 W. This is from an experiment I ran to duplicate what the "snake oil" salesmen demonstrate, and the data is in my book on "Electrical Energy Measurement, Conservation, and Methods to Reduce Your Electric Bill". See http://beta-a2.com/energy.html for a little more information.

.

 

[prs1jazz]

Good evening and thank you for the response. I am very aware of black box aka snake oil devices and haven approached several times. Most of them average the phases as well like capacitor banks. This technology is different though working on each phase under maximum power transfer e developer has a doctorate in quantum physics owning a large electrical engineering group and spent 8 years putting this technology together in europe. As I mentioned, it doesn't use capacitance to make the correction on each phase.

You are absolutely correct in saying that I am not an electrical engineer. I have been around enough to understand the basics but as they say the devil is in the details. I should have said copper loss not core loss. That was my mistake. Could you help me with the question of elimination of Kvar before the transformer? What would transpire when this occurs? Thanks again for all the help gentleman. I am doing my best to be the greenest foundry in the states with the most advanced technology.

 

[gar]

140824-1950 EDT

prs1jazz:

I believe your last post has now completely changed the question. Several aspects are not clear relative to the new question.

You are a foundry. That may mean large electric arc furnaces or other electrically heated furnaces, and in turn very large electrical loads. Possibly megawatts of required electrical power. I do not know what the current waveform of an arc furnace looks like, but it should be largely resistive.

Assuming these are your loads and that they are on the output of the transformer, then the first big questions are:
(1) What is the magnitude of the load? KVA rating of the transformer?
(2) What is your input power factor?
(3) Is this a distortion, or displacement, or combination power factor?
(4) Where would the power factor correction equipment be located?
(5) Do you presently pay a low power factor penalty?
(6) Do you presently have a poor power factor?

If there is no power factor penalty, then the only saving possible would be a reduction in the losses in all of your equipmrnt on the load side of the power company meter. The closer the power factor correction equipment is to the revenue meter the lower is the saving until it is zero with the PFC at the meter. You want the PFC as close to the load with bad power factor as possible.

In our part of the country I would think that preheating a melt with gas would be a lower total cost than doing a total electric furnace heat.

You need to provide more information to us for comment, and you really need to hire an unbiased consultant to study how much saving you would likely get with the proposed PFC equipment. That equipment would be described as an active power factor correction device.

An active power factor correfction device will have its own loss. This loss must be subtracted from the savings to be achieved with the device to determine any net improvement.

.

 

[prs1jazz]

Thanks again for the quick response. We do lost wax investment castings. This consists of wax and plastic injection presses in which patterns are shelled, melted out and induction furnaces are used to pour alloy. The study started with wax injection presses which operate at about a .55-.6pf. Understanding you can't save kw correcting a load, we did see beneficial changes when a transformer was introduced. The current wax presses use between 15-30 amps each (we have several) each equipped with a 480v-120v transformer for the pump. We would like to locate the pfc right on the equipment or breaker box. We have very pour power factor at both facilities. :thumbsdown:

 

[GoldDigger]

What PF penalty are you paying right now and how does that penalty compare to the rest of your bill?

 

[prs1jazz]

It depends on which facility utility we are concentrating on. We have several in all different billing methods. Everywhere from KVA demand, KW-KWH, and down south a PF penalty. That's why I am concentrating in KW.

 

[GoldDigger]

It depends on which facility utility we are concentrating on. We have several in all different billing methods. Everywhere from KVA demand, KW-KWH, and down south a PF penalty. That's why I am concentrating in KW.

If you are concentrating on the KW, and do the correct math you should see that any change in metered KW as a result of PF correction will be lower than putting the same effort into reducing the VD in the wiring. And even that too may not be worth doing.