Power factor, voltage drop, and SKM PowerTools

[jmbelectrical]

Using SKM PowerTools, I was performing a load flow analysis on a project to ensure that voltage drop levels were being kept to a maximum of 3% at panelboards and switchboards. (My state's energy code requires that it be kept to a maximum of 5% at the end load, and our goal is to distribute it such that it is kept at 3% or less to the furthest downstream panelboard,with the remaining 2% occurring along the branch circuit.)

Before going any further, here are a few items that I feel are worth mentioning:

1. The system type is 480/277V, three-phase, four-wire, 60Hz.
2. This is an educational facility, and I was assuming a power factor of 0.9 (lagging) for all loads. Loads generally consist of LED lighting, receptacles, electric space heating, VFD-driven exhaust fans, VFD-driven air handling units, VFD-driven pumps, and air-cooled chillers. Loads were modeled as one "block" connected to each panelboard or switchboard.
3. Feeders with an ampacity of 100A or less are copper with THHN or THWN insulation. Some are in PVC conduit and others are in EMT.
4. Feeders with an ampacity greater than 100A are aluminum with THHN or THWN insulation. Some are in PVC conduit and others are in EMT.
5. I do not have much experience with SKM PowerTools. I typically use a simple, Excel-based, voltage drop calculator.

For most of this exercise, it seemed that the results I was getting were unremarkable. As expected, several, rather long (Anywhere from 200 to 700 feet) feeders with light to moderate loading required increases in conductor sizes in response to excessive voltage drop.

And then I ran into a different situation. A 4,000A switchboard with a demand load of about 2,900A, located about 150 feet from the utility transformer, had a voltage drop of around 3.5%. Its service feeders were comprised of 12 sets of 600kcmil aluminum conductors with THWN insulation. Increasing the conductors' size to 750kcmil helped, but the voltage drop still exceeded 3%. Changing the conductor type to copper brought it down to almost exactly 3%, but this was problematic since the voltage drop exceeded 3% at the downstream panelboards. The utility transformer can only accept a maximum of 12 sets of conductors, with a maximum conductor size of 750kcmil. I had hit a roadblock.

Upon bringing this to my boss's attention, he recommended adjusting the power factor of the load to unity, citing similar experiences on other projects. Doing so brought the voltage drop down to under 1.5%. Using the Excel-based voltage drop calculator referenced above yielded similar results. It seems that it assumes a similar power factor.

This leads me to the following questions:

1. I've never taken power factor into account when performing voltage drop calculations in the past. I know that the effective impedance of a conductor changes depending upon the power factor of the load, and I know that the voltage drop due to a conductor's reactance becomes more of an issue when dealing with larger conductors. In your opinion, when should power factor be taken into account?
2. Was my original, assumed power factor of 0.9 reasonable or was it too conservative for a facility of this type?
3. Is assuming a power factor of unity the right thing to do in this scenario?
4. Does SKM PowerTools have any "quirks" that may have skewed the results? Is there a possibility that I modeled something incorrectly?

Thanks to all for any input provided.

 

[waross]

Our code is similar.
On a simple example:
A 10 volt drop at 100% PF on a 120 Volt circuit will result in a voltage of 110 Volts
A 10 volt drop at 0% PF on a 120 Volt circuit will result in a voltage of 119.6 Volts
The voltage drop in the conductors alone should get less as the PF drops, not greater.
It looks as if your software is including the transformer regulation in the voltage drop calculations.
Our code, by inference is concerned with the voltage drop in the conductors and does not consider the transformer regulation.
Two questions:
a) Does your code include transformer regulation in voltage drop calculations?
b) Does your software include transformer regulation in voltage drop calculations?

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

 

[jmbelectrical]

waross,

Our energy code does not require us to include transformer regulation in voltage drop calculations.

As for the software, SKM PowerTools does not appear to include transformer regulation in its voltage drop calculations. I could be mistaken, though. I'm still getting familiar with it.

 

[waross]

The voltage drop getting greater instead of less at a lower power factor is a strong indication that the software is including transformer regulation in the calculations.
The code regulates what you have direct control over, that is the voltage drop in the conductors.
However, many times we must calculate the voltage sag when a large motor starts. The voltage sag will include the regulation of the transformer along with the voltage drop in the conductors.
I strongly suspect that your software is calculating the total voltage sag.
I would do a quick sanity check.
Double the length of one of your feeders and run the software again. If the software is calculating only the drop in the cables the result will be double.
If the software is including the transformer regulation then the change will not be linear with cable length.

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

 

[bacon4life]

1)In small conductors where the voltage drop is primarily due to resistance, the voltage drop is only slight larger as the power factor drops from unity to 90%. For very large conductors were the reactance dominates, the voltage drop increases significantly as the power factor drops from unity to 90%.

3)I have never seen a commercial building at unity power factor, so I disagree with your boss's recommendation of unity power factor. Something higher than 90% is likely appropriate if all of the drives and the lighting is inherently power factor corrected.

Waross,
Are you comparing 2.4 MW + 0 MVAR (2900@100%pf) to 2.16 MW + 1.05 MVAR (2900A@0.90 pf)? From a utility perspective, I assumed the real power would be constant so we should instead compare 2.4 MW + 0 MVAR (2900@100%pf) to 2.4 MW + 1.16 MVAR. My reading of your statement seems to conflict with the conventional guidance that installing capacitor banks to improve voltage drop.

 

[dpc]

The power flow simulation is heavily dependent on what you assume for the loading. In a power flow calculation, you are trying simulate actual loading, actual voltages and actual currents. This is much different than a NEC load calculation. The NEC rules for calculating loads that are used for sizing feeder breakers and feeder conductors are extremely conservative and actual peak demand will almost certainly be less than this.

Power factor 0.9 to 0.95 is a reasonable assumption.

 

[RodneyDixon]

To exclude the voltage drop across the source transformer go to Balanced Studies > Load Flow Setup -- you will see the first option is a checkbox, untick it.

 

[waross]

I have assumed that you are concerned with code compliance rather than the actual voltage at the load.
Please read carefully the NEC section concerning allowable voltage drop.
I am familiar with the Canadian code. There is much harmony between the CEC and the NEC but in specific cases you must refer to the actual wording of the applicable code.
The voltage drop limits in the CEC are similar to the limits in the NEC.
What you need to check is the method of calculating the voltage drop.
The CEC refers to table D3 which shows the allowable conductor distance for a 1% voltage drop at 120 Volts for a range of currents.
Instructions are given for extrapolating for other % voltage drops, other voltages and other currents.
The important aspect in regards to code compliance is that neither the power factor nor the inductance reactance of the circuits are considered.
The CEC also allows the use of a correction factor when the current is less than the allowable current of the conductor.
eg: For a conductor rated at 90 degrees C, loaded to 40% of the allowable current the maximum allowable circuit distance may be increased by a factor of 1.06.
And there are exceptions for dedicated circuits feeding equipment that is rated to operate at a voltage lower than 95% of the nominal voltage and some other circuits.
Code compliance rather than actual voltage at the load.
What says the NEC?

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