Power Factor Correction, Leading Power Factor, and a few other concern 1

[rhatcher]

I want to install power factor correction at my facility. I have a single 13.8kV feeder with utility meter feeding 7 substations with various low voltage secondaries. I would like to install one capacitor bank on the secondary of one the the substations to compensate for the entire facility. I have examined several months worth of power data (15 minute interval) and determined that my kvar consumption varies between 150 and 300, with 200 being the nominal value for most of the time, 300 being rare, and 150 occurring about 25% of the time. There are no drives or sensitive electronics (other than PC's) on the system. I want to install a 200kVAR capacitor bank on the secondary of one of the transformers (3000kVA) to achieve near unity power factor during normal operations (about 75% of the time). This means:
- 25% of the time I will have a leading power factor for the entire facility (meaning at the 13.8kV feeder).
- 100% of the time the secondary bus of the 3000kVA transformer where the capacitor is installed will have a leading power factor.
(If you are noting a discrepancy then you are right. I have about 8MVA of capacity but my average consumption is only 250-300kVA. Why that it true is a long story that is off-topic)

My questions are:
- Will the leading power factor cause overvoltages on the substation secondary where the cap is installed or on the 13.8kV feeder? My feelings are that the 200kVAR cap is orders of magnitude smaller than the capacity of the transformer or the feeder so there should be no effect. Am I right?
- What are the general pitfalls (hazards) of running a leading power factor? (There is no penalty from the utility).

 

[busbar]

Aside from bus-voltage issues, study of the serving utility’s tariff/rate schedule may be in order, particularly if leading power factor at the metering node produces any penalties. Basic non-vendor-specific considerations at …Application of Shunt Power Capacitors

http://www.dca.ufrn.br/~ricardo/files/shuntcap.pdf

http://www.dca.ufrn.br/~ricardo/files/capshunt92.pdf

 

[jghrist]

The voltage rise caused by the capacitors will be:

% volt rise = (kvar·XL)/(10·kV²)

Where: XL = the inductive reactance of the line and transformer (ohms)
kvar = three-phase capacitor size
kV = line-to-line voltage

This voltage rise is above the voltage as reduced by load voltage drop. That is, if there is 5% voltage drop with no capacitors, and 6% voltage rise caused by the capacitors, then the net voltage rise will be 1%.

Another pitfall of leading pf is increased losses. If the pf in any section of the circuit is different from unity, the current (and I²R losses) will be higher than if the pf were unity.

It will probably be less expensive to install the capacitors on the 13.8 kV than on the secondary side of one of the transformers.

 

[asymptote]

As jghist points out any leading or lagging Vars will involve losses due to I*IxR which equal revenue lost. One very important consideration must be the system resonance with attendant overvoltage issues as the PF swings from lagging to leading and back. A more correct approach as taken in industry would be to have various stages of correction switched in to maintain a lagging PF in the region of .95 but not unity.

 

[rhatcher]

busbar, thanks for the great link, a star for you. Thanks also to jghrist and asymptote for their responses. I am going to educate myself on the info busbar provided before continuing this thread.

 

[reactive]

If you send me the following I will tell you whether there are any potential problems.
- 13,8kV 3ph fault level.
- kVA, LV voltage and %Z of the transformer where you would like to connect the PFC.
- How are you metered, two quadrant (no leading VArs) or four quadrant? I assume you want to install PFC because you are charged for kVA maximum demand.

Regards
Bryan

 

[stevenal]

rhatcher,

Think of capacitors as reactive power generators. If I understand correctly, you wish to install this generator on the secondary of one substation. Part of the reactive power will flow normally to the load, but the majority will be backfeeding this one substation to feed the others. Does this one substation have sufficient capacity for this extra kVA? You will be increasing the real power loss at this one substation as jghrist said.

From the loss perspective, the ideal location for capacitors is at the individual reactive loads. Consolidation has its advantages too. If you wish to consolidate, consider the 13.8 side.

 

[rhatcher]

reactive, here is the info you asked for (to the best of my ability...)

- The available 13.2kV fault current is unknown, but I do know that I am located in an area with very high power density. By this I mean that there is probably 30-50MVA of other (end user) transformers within a 1/4 mile radius of me (in addition to mine). The entire complex (surrounding 500 acres or so) is fed from 2 separate utilities using some sort of tie arrangement. This being the case, for the purpose of fault calculations on my system I would consider the 13.2kV as an "infinite bus". However, I have not previously considered this as significant to this project since I plan to install the caps at the 480V level using a current limiting breaker already present in the switchboard as a spare.
- The transformer where the caps will be installed is a 3000kVA, 13.2kV/480V delta-delta with 5.38% impedance.
- The utility metering reads kW and kVA with kVAR and pf as calculated values. The utility has already stated that there is no penalty for leading pf. In fact, I don't think they can tell one way or the other anyway since they only see kW and kVA.
- Billing is based on kWH and on a demand (penalty) charge for peak kVA during any 15 minute interval. The demand charge is the majority of the bill, kWHs are cheap.

In addition:
- The 3000kVA substation's connected load is intermittent use with a maximum of 500kVA consumption but normally is anywhere between 0-200kVA.
- My total power consumption varies somewhat but nominal values look like this (based on 15 minute interval report):

Loaded: 200kW 230kVAR 307kVA 0.66 pf
Idle: 144kW 167kVAR 221kVA 0.65 pf

My system is loaded a minimum of 16 hours a day for a minimum of 5 days a week. The remainder is idle time. The load cycle is that in the morning everything is turned on and at the end of the second shift (16 or more hours later) everything is turned off until the next morning. There is minor load variation throughout the day but there is no routine cycling of the load (ie. fluctuating from loaded to idle)other than that. I want to install 200kVARs of capacitors to correct my system to this:

Loaded: 200kW 30kVAR 202kVA 0.99 pf (lagging)
Idle: 144kW -33kVAR 147kVA 0.97 pf (leading)

This would represent a savings of $13,000 per year (minimum) in kVA demand charges. However, I would run a leading power factor when idle.

What do you think?

 

[busbar]

Understanding that kVA is a scalar [directionless] quantity, don’t forget to account for leading and lagging conditions at the metering node will increment equally. Id est—a ‘demand’ of kVAR delivered bills the same as a ‘demand’ of kVAR received. The accepted definition of ‘kVA delivered’ versus ‘kVA received’ is based simply [and potentially erroneously] on the associated real-power direction at that moment, during which KVAR-flow direction is not differentiated with kVA and kVAh metering.

Thank considered, one issue that may surface based on a quick read of your numbers is that it may be possible during low-load periods that ‘kVA demand’ during leading PF may possibly exceed that during lagging PF. That would be a hard chunk of A utility energy bill to have to pay for.

 

[reactive]

rhatcher

I have assumed a fault level of 100 MVA. Based on this you will have no series resonance problems.

Although it is not the neatest solution, your idea will work just fine. The transformer you are connecting to will operate leading I would imagine for most of the time (depending on it's own load) resulting in some increase in I2r losses, however this is negligible.

As long as your utility are happy with leading, no problem. Installing 200kVAr at 13,2kV is not cost effective.

Just one thing, is 307kVA your monthly maximum demand (MD)(the highest 15-min period in a month). If not you should be correcting for the MD in order to get the largest financial benefit. Take your MD and correct to 0,995 lag and then do your sums again. What size of PFC do you need now? What is your MD charge?

You may find that the savings can justify a larger bank.

Sorry to complicate the issue.

 

[jbartos]

Suggestion: Reference:
Donald Beeman, "Industrial Power Systems Handbook," First Edition, McGraw-Hill Book Company, Inc., 1955,
Chapter 8 Power-Factor Improvement by W.C. Bloomquist

 

[jbartos]

Suggestion: Visit

http://energy.navy.mil/publications/pnl/pwrfc.doc

Power Factor
If you are concerned about power factor charges on your plant’s electricity bill. Call your utility. They can tell you how much you are being charged for your power factor, how much you stand to save by correcting it, and give you an estimate of how much it will cost you to do so

Electrical utilities often have a charge for low power factor on their monthly bills. This charge can be confusing unless you understand what power factor is. Power factor is the ratio of useful power to total power in an alternating current (AC) circuit. In alternating current produced by electric generators, voltage and current in the circuit rise and fall like waves, with the current wave typically lagging the voltage wave. As current flows in the wire, a magnetic field forms around the wire in proportion to the magnitude of the current and in a direction that depends on the direction of the current flow. As the current rises and falls, the magnetic field changes, inducing a counter-voltage in the circuit. This counter-voltage lowers the instantaneous voltage in the circuit and makes it appear that the voltage wave is leading the current wave.

The magnitude of the magnetic field surrounding a wire is generally small, however, if the wire is wrapped around an iron core such as in a motor or transformer, the magnetic field may be many times larger, resulting in greater changes in the magnetic field over the cycle. Items in a circuit that increase the magnetic field generated also increase the inductance of the circuit, causing the current to lag farther behind the voltage in the AC cycle.

The addition of capacitance in an AC circuit has the opposite effect of inductance, causing the current wave to approach, and in extreme cases even lead, the voltage wave.

At any point in the AC cycle, the instantaneous power in the circuit is the product of the voltage and the current. The resultant power is measured in watts. By measuring the instantaneous power at even time increments over the cycle and averaging the result, you could obtain the true power in the circuit. It is difficult to measure the instantaneous power, however, because the voltage and current are changing from positive to negative and back to positive 60 times a second. Instead, true power is calculated using the root-mean-square (RMS) values for current and voltage. The RMS voltage or current value is what you measure with a standard AC voltage or current meter.

Reactive Power in a circuit represents work done at the generator to compensate for power losses resulting form the inductive loads on the circuit. Reactive power does not represent useful work done by the circuit.

Apparent power is calculated as the product of the RMS current and voltage in a circuit and is often expressed as EI. Apparent power is the power the utility supplies at the electrical meter.

Power factor is the ratio of true power to apparent power in a circuit. In AC circuits, the true power is less than or equal to the apparent power, and so the power factor is always less than or equal to one.

Engineers often use the power triangle to describe the relationship between these quantities. In this triangle, the length of the hypotenuse of a right triangle represents the apparent power, one of the acute angles is used to represent the phase difference, the adjacent side to represents the true power, and the opposite side to represents the reactive power.

If E is in volts and I is in Amperes, power is calculated in Watts and reactive power is calculated in Volt-Amperes-Reactive or VARs.

Because utilities lose some of their generating capacity to the reactive power load, utilities try to keep the reactive power low (and the power factor high) in the loads they service.

Low power factor can also be caused by electronic equipment that generate harmonics of the power supply frequency and affect the current or voltage profile in the electrical circuit. In particular, electronic dimmers and electronic power supplies that rely on solid state diodes or silicon control rectifiers (SCRs) can develop these harmonics. Low power factor caused by these loads is generally not a problem unless one-half or more of the electrical load can be attributed to electronic loads.

Utilities charge for low power factor using several different methods. Often, utilities measure reactive power in commercial or industrial loads and have a charge based on the average amount of reactive power (in kilo Volt-Amperes-Reactive or kVAR) measured during a billing period. At other times, they may charge an additional billing cost if the average power factor falls below a certain value (generally between 0.8 and 0.9) in the billing period. The utility may simply charge on the basis of apparent power (in kilovolt-Amperes or kVA) instead of kilowatts (kW) used in a plant. At other times there may not be power factor charges at all. If there is a charge, it should be apparent on the bill. If the bill is unclear, call your utility and they should explain the power factor costs to you.

Repair of low power factor involves the addition of power factor correction capacitors at your site. The total amount of capacitance required must be sized specifically for your electrical load. In general, you would like to install enough capacitance to raise your power factor to 95%. The best source of information on how much capacitance you need and how much it can save on your electric bill is the local electrical utility.

Remember, if you are concerned about power factor costs on your electric bill, contact your local utility. They too have an interest in solving power factor problems at your site.

For further information
contact
Your local electrical utility,
Your Engineering Field Division, or
Energy & Utilities Department
Code 22
Naval Facilities Engineering Service Center
Port Hueneme, California 93043-4328
Telephone 805-982-3486
Autovon 551-3486
FAX 805-982-5388