480V, 1PH, 3W (Center Grounded) Feeder

[hillbilly1]

As some know ,this is my favorite method. Have to run the numbers for your case to see if this is advantageous or just more wire is the way to go.

“Neutral floated” on the load side disconnect is kind of a misnomer, It’s not really “Floating”, it’s just not bonded at that point. 480 volt transformers seem to be easier to get off the shelf here in the states than a 600 volt.

 

[electrofelon]

Transformer losses are practically negligible. As far as voltage regulation goes, it’s actually improved vs running a long feeder at the utilization voltage. This is why the utility transmission/distribution systems are not done at the utilization voltage. They take advantage of the larger impedances found with higher voltage systems and the voltage division principle to minimize line losses.

Biggest downside as you’ve mentioned is the cost of two transformers and of course the installation requirements that come with this type of system.


Sent from my iPhone using Tapatalk

$100 per year no load losses for typical resi setup. Personally I don't consider that negligible (and that is just no load losses). Voltage regulation stinks after three transformers (2+utility). I have this setup at my house by the way so I have first hand experience.

 

[xptpcrewx]

$100 per year no load losses for typical resi setup. Personally I don't consider that negligible (and that is just no load losses). Voltage regulation stinks after three transformers (2+utility). I have this setup at my house by the way so I have first hand experience.

I don’t know your specific installation but it is common in industrial installations (as well as some commercial installations) to have over 2 transformation levels. For VR, when the system is designed, adjusted and operated properly, its not an issue. Obviously if VR is an issue at the service to begin with it will propagate and get worse the further you go downstream.

Post #14 said it best. Need to run the numbers to see if it’s advantageous. There will be some length at which using transformers will or will not make sense.


Sent from my iPhone using Tapatalk

 

[electrofelon]

Post #14 said it best. Need to run the numbers to see if it’s advantageous. There will be some length at which using transformers will or will not make sense.


Sent from my iPhone using Tapatalk

Exactly. My point is just that step up step down is not as much of the fairy tale everyone thinks it is, and typically people jump to this too soon. Also, I have a hard time seeing how going to 480 or 600 would be worth it. If you're going to jump in, go to MV and stop messing around.

 

[mbrooke]

I agree it’s may not be the best solution in every application; however, let’s not forget that transformers have the benefit of tap adjustments to overcome voltage drop. Also, if the transformers are supplying 2w systems, the primary OCPD can be used to protect the secondary conductors.


Sent from my iPhone using Tapatalk

Right, taps help raise voltage. However I feel that under low load you are going to be dealing with high voltage.

 

[xptpcrewx]

Right, taps help raise voltage. However I feel that under low load you are going to be dealing with high voltage.

As long as the utilization voltage doesn’t exceed 105%, it’s acceptable. Utilization voltage is also acceptable at 90% per NEMA nameplate standards. So if voltage regulation goes from 105% to 90% of nominal voltage, then something is very very wrong.


Sent from my iPhone using Tapatalk

 

[mbrooke]

As long as the utilization voltage doesn’t exceed 105%, it’s acceptable. Utilization voltage is also acceptable at 90% per NEMA nameplate standards. So if voltage regulation goes from 105% to 90% of nominal voltage, then something is very very wrong.


Sent from my iPhone using Tapatalk

Agree. However, remember that POCO voltage can legally vary + or - 5%.

If boosting +3% and the POCO has a spicy day at +5% you could in theory go over +5% at the cabin.


Me personally I would go the direct wire route. Sizing the wire such that it will pass enough current to open the source OCPD in no more than 5 seconds. (Don't forget about leaving enough fault current for the branch circuit too) EGC sized no less than half the phase wires.

 

[xptpcrewx]

Agree. However, remember that POCO voltage can legally vary + or - 5%.

If boosting +3% and the POCO has a spicy day at +5% you could in theory go over +5% at the cabin.


Me personally I would go the direct wire route. Sizing the wire such that it will pass enough current to open the source OCPD in no more than 5 seconds. (Don't forget about leaving enough fault current for the branch circuit too) EGC sized no less than half the phase wires.

True. This is where the 5% voltage drop rule of thumb comes from (service to the branch circuit). Just need to confirm VR doesn’t exceed 5% and you would be fine for any allowable POCO variation.


Sent from my iPhone using Tapatalk

 

[mbrooke]

True. This is where the 5% voltage drop rule of thumb comes from (service to the branch circuit). Just need to confirm VR doesn’t exceed 5% and you would be fine for any allowable POCO variation.


Sent from my iPhone using Tapatalk

I trust you.

Though I'm still personally confused. Under light or no load, a POCO voltage of 126 volt with a +2.5% customer trafo tap would output around 130 volts. I mean it might be ok, but kind of steep.

 

[xptpcrewx]

I trust you.

Though I'm still personally confused. Under light or no load, a POCO voltage of 126 volt with a +2.5% customer trafo tap would output around 130 volts. I mean it might be ok, but kind of steep.

Thanks. I am not advocating for tap adjustments where the POCO voltage is already at the upper limit or is historically known to reach those levels. Whether an installation has light or no load really depends on the application. For example, applications where some amount of base load is always present would allow for a constant +2.5% tap adjustments without going above utilization limits. Also, the utility voltage profile at a particular location in the POCO network may never allow the voltage to ever get to the upper limit of 105% - like at the end of radial distribution. Another reason the voltage may not ever get that high is if the POCO or facility is using an OLTC. The best way to analyze this type of problem is by performing a load flow analysis with the historical voltage profile over the course of a year.

Not that I've ever seen it, but how about an automatic remote motor/pump application that involves switching-in the motor feeder and transformer (with a fixed tap set above nominal) only when in operation?

 

[mbrooke]

Thanks. I am not advocating for tap adjustments where the POCO voltage is already at the upper limit or is historically known to reach those levels. Whether an installation has light or no load really depends on the application. For example, applications where some amount of base load is always present would allow for a constant +2.5% tap adjustments without going above utilization limits. Also, the utility voltage profile at a particular location in the POCO network may never allow the voltage to ever get to the upper limit of 105% - like at the end of radial distribution. Another reason the voltage may not ever get that high is if the POCO or facility is using an OLTC. The best way to analyze this type of problem is by performing a load flow analysis with the historical voltage profile over the course of a year.

Not that I've ever seen it, but how about an automatic remote motor/pump application that involves switching-in the motor feeder and transformer (with a fixed tap set above nominal) only when in operation?

Agree with everything however IMO I would never assume that a POCO won't go the upper limit. The supply substation could be set to 105% if feeder loading gets bad enough or extra circuits need to be pickup up during an emergency. Granted the practice is typically to boost to around 123 volts max and use voltage regs and caps banks to prevent from going to 105%, but you never know especially with all the solar and cogen.

But, in the end, you know your POCO better than I do.

 

[xptpcrewx]

Agree with everything however IMO I would never assume that a POCO won't go the upper limit. The supply substation could be set to 105% if feeder loading gets bad enough or extra circuits need to be pickup up during an emergency. Granted the practice is typically to boost to around 123 volts max and use voltage regs and caps banks to prevent from going to 105%, but you never know especially with all the solar and cogen.

But, in the end, you know your POCO better than I do.

I don't recommend assuming these things either. I try to use standards and empirical data to inform my decisions. Keep in mind, even if you have line compensation at the sending end of the line, the loads closest to the sending end are the limiting factor (because the voltage here cannot be exceeded), so even with the OLTC or voltage regulators boosted at their upper limits (due to loading conditions), customers at the far receiving end of the line would naturally see less voltage. Also think about if the customer is the one who owns/controls the OLTC and transformer.

One thing I should add about voltage regulation and tap adjustments. In reality, the IZ drop has a magnitude and angle and is NOT the same thing as the allowable voltage drop (which is a projection difference of sending and receiving end voltage phasors). Depending on the power factor of the circuit, the IZ drop may appear to exceed limits but will actually have a sending to receiving end voltage drop of 0 volts!

 

[mbrooke]

I don't recommend assuming these things either. I try to use standards and empirical data to inform my decisions. Keep in mind, even if you have line compensation at the sending end of the line, the loads closest to the sending end are the limiting factor (because the voltage here cannot be exceeded), so even with the OLTC or voltage regulators boosted at their upper limits (due to loading conditions), customers at the far receiving end of the line would naturally see less voltage. Also think about if the customer is the one who owns/controls the OLTC and transformer.

One thing I should add about voltage regulation and tap adjustments. In reality, the IZ drop has a magnitude and angle and is NOT the same thing as the allowable voltage drop (which is a projection difference of sending and receiving end voltage phasors). Depending on the power factor of the circuit, the IZ drop may appear to exceed limits but will actually have a sending to receiving end voltage drop of 0 volts!

Right- though not all customers will be at the far end. But again you know your system better than I do.


0 volts as in no voltage drop? Like a low level ferranti rise?

 

[xptpcrewx]

Right- though not all customers will be at the far end. But again you know your system better than I do.


0 volts as in no voltage drop? Like a low level ferranti rise?

Zero volts difference in projection magnitudes - but non-zero IZ drop (usually seen with leading power factor loads).

Not talking about the Ferrari effect, since that is more specific to long transmission line applications.

If I get some time, I can put a derivation and example together. They key thing here is capacitive load or not, the IZ drop is NOT the same thing as the voltage drop (projection) when determining the difference between sending and receiving end voltage. Most people calculate IZ, IR or IX drop without knowing this result is almost always much larger than than its projection (the actual voltage drop). IZ, IR or IX is a simplistic overly conservative approach.

To think about it correctly, the cable and load power factors should be considered. Refer to NEC Chapter 9, Table 9, Ze formula in the last sentence of Note 2 (for power factors other than 0.85).


Sent from my iPhone using Tapatalk

 

[mbrooke]

Zero volts difference in projection magnitudes - but non-zero IZ drop (usually seen with leading power factor loads).

Not talking about the Ferrari effect, since that is more specific to long transmission line applications.

If I get some time, I can put a derivation and example together. They key thing here is capacitive load or not, the IZ drop is NOT the same thing as the voltage drop (projection) when determining the difference between sending and receiving end voltage. Most people calculate IZ, IR or IX drop without knowing this result is almost always much larger than than its projection (the actual voltage drop). IZ, IR or IX is a simplistic overly conservative approach.

To think about it correctly, the cable and load power factors should be considered. Refer to NEC Chapter 9, Table 9, Ze formula in the last sentence of Note 2 (for power factors other than 0.85).


Sent from my iPhone using Tapatalk

Example will be epic Love learning this stuff.