KVA and voltage drop

[rlcguy]

Hello,

As an electrician I have always been taught that voltage drop was compensated for by increasing conductor size, shortening the length of the circuit or using a smaller load. But the other day our staking engineers quoted a customer a 25kva transfomer vs a 15kva unit to compensate for the voltage drop.

Have any of you used this methed? They have a piece of software that you can change kva, wire size or circuit length to get desired voltage drop. I am curious to hear peoples thoughts on this. Personally I can't get past the idea of the wire being the bottle neck due to resistance.

As a side note the bigger transformer was only $100 more than the smaller one. Bigger wire would have been a couple hundred more so they saved the customer money if their assumption/software is correct.


Thanks for your thoughts and comments,

RLCGuy

 

[charlie b]

I don’t believe this is a valid approach to resolving VD issues. The KVA rating of a transformer is based on the ability of the transformer to reject heat to its surroundings. That has nothing to do with what happens to voltage further downstream. A higher rated transformer might (I don’t know for certain) have a lower internal impedance. If that is true, then for the same load current the voltage at the transformer secondary would be higher (i.e., less voltage drop internal to the transformer windings). But if there is a concern over VD in a circuit, this difference would be insignificantly small. That is, going with a higher rated transformer would not significantly improve the voltage at the load, and you would still need to use larger secondary conductors.

 

[jrohe]

Hello,

As an electrician I have always been taught that voltage drop was compensated for by increasing conductor size, shortening the length of the circuit or using a smaller load. But the other day our staking engineers quoted a customer a 25kva transfomer vs a 15kva unit to compensate for the voltage drop.

Have any of you used this methed? They have a piece of software that you can change kva, wire size or circuit length to get desired voltage drop. I am curious to hear peoples thoughts on this. Personally I can't get past the idea of the wire being the bottle neck due to resistance.

As a side note the bigger transformer was only $100 more than the smaller one. Bigger wire would have been a couple hundred more so they saved the customer money if their assumption/software is correct.


Thanks for your thoughts and comments,

RLCGuy

I do not see how a larger transformer can offset the voltage drop. If I have a 150-foot, 120 volt circuit loaded to 15 amps and the circuit is served by a panelboard that is served by a 30 kVA transformer, I'm going to call for #8 wire to keep voltage drop below 3 percent. If I change that transformer to a 45 kVA transformer, it does not change my voltage drop calcs; I'm still going to require a #8 wire.

I suppose it is possible that the 15 kVA transformer does not have any tap adjustments but the 25 kVA transformer might, in which case they could offset the effects of voltage drop by using a higher tap setting.

They could also be recommending the larger transformer to mitigate the voltage dip experienced when larger motors start.

 

[jumper]

I am gonna go with the guys above. Never heard of a just using a larger tranny for VD issues. The availability of taps mentioned above might be a viable reason.

 

[Carultch]

Hello,

As an electrician I have always been taught that voltage drop was compensated for by increasing conductor size, shortening the length of the circuit or using a smaller load. But the other day our staking engineers quoted a customer a 25kva transfomer vs a 15kva unit to compensate for the voltage drop.

Have any of you used this methed? They have a piece of software that you can change kva, wire size or circuit length to get desired voltage drop. I am curious to hear peoples thoughts on this. Personally I can't get past the idea of the wire being the bottle neck due to resistance.

As a side note the bigger transformer was only $100 more than the smaller one. Bigger wire would have been a couple hundred more so they saved the customer money if their assumption/software is correct.


Thanks for your thoughts and comments,

RLCGuy

The KVA of the transformer can only affect the starting voltage before the voltage drop through the secondary conductors and feeders comes in to play. So instead of starting with nominal 240V, you might end up starting with 245V at the transformer secondary, due to a transformer that operates at a lower temperature and lower internal resistances relative to the load. You would still end up having roughly the same voltage drop through the circuit conductors, but it is a voltage drop from what starts slgithly higher.

Transformers are analogous to gear systems, in that they trade the two quantities that carry power. Gear systems rigidly constrain the ratio of speeds, and if 100% efficient, the ratio of the torques would be the inverse. The output torque is slightly lower than theoretical as a result of inefficiencies, but the speed ratio still is theoretical no matter what the efficiency.

Unlike gear systems, transformers do not rigidly constrain the ratio of either voltage or current. Both ratios will differ slightly from theoretical due to inefficiencies. Transformers are more efficient at lower than nameplate loads, which means the secondary voltage will be slightly higher, within a tolerance limit, than what it would be at full load.

 

[Jraef]

If a transformer is UNDER sized for a given load, saturation under load can cause voltage drop, so increasing the size of the transformer would indeed reduce voltage drop CAUSED by the transformer. That happens a lot in motor starting applications where the transformer is driven into saturation by an extreme current draw from motor starting. But if the transformer is not CAUSING the voltage drop, it cannot solve it.

 

[rlcguy]

I appreciate all of the comments and am in agreement with all that has been shared. I have been unable to find anyone who can back up or vouch for the theory put forth by the software our staking engineers are using.

 

[GoldDigger]

I appreciate all of the comments and am in agreement with all that has been shared. I have been unable to find anyone who can back up or vouch for the theory put forth by the software our staking engineers are using.

If the software the engineers are using calculates a voltage drop within the transformer itself (not unlikely, but often small compared to other voltage drop sources), then that software can also calculate what the equivalent voltage drop would be for a larger (or more efficient) transformer.

The difference between the two, independent of whatever other voltage drops may be present in the circuit, will be what you can theoretically save by substituting the "better" transformer.

If the voltage drop they calculate is in fact significant, it may well be that they are underestimating the actual performance of the smaller transformer that they are comparing.


The internal resistance of the transformer can be calculated knowing the rated size of the transformer and the %Z of the transformer. For the same %Z, the larger transformer will have lower internal resistance.

Now if POCO will automatically pull larger service conductors to go with the larger transformer, there may be something more to the argument.

 

[Smart $]

I appreciate all of the comments and am in agreement with all that has been shared. I have been unable to find anyone who can back up or vouch for the theory put forth by the software our staking engineers are using.

I'll back up what Jraef said...

 

[topgone]

If a transformer is UNDER sized for a given load, saturation under load can cause voltage drop, so increasing the size of the transformer would indeed reduce voltage drop CAUSED by the transformer. That happens a lot in motor starting applications where the transformer is driven into saturation by an extreme current draw from motor starting. But if the transformer is not CAUSING the voltage drop, it cannot solve it.

Agree.
In my greenhorn days, I was tasked with attaching recording voltmeters/ammeters to distribution transformers where the community reported issues with VD. The whole idea is that if the terminal voltage of the transformers get really low and amp recordings are up, the problem is the installed transformer capacity.

What the poster above pointed out, if the transformer has built-in taps, it would be more desirable compared with no taps at all.

 

[mivey]

Hello,

As an electrician I have always been taught that voltage drop was compensated for by increasing conductor size, shortening the length of the circuit or using a smaller load. But the other day our staking engineers quoted a customer a 25kva transfomer vs a 15kva unit to compensate for the voltage drop.

Have any of you used this methed? They have a piece of software that you can change kva, wire size or circuit length to get desired voltage drop. I am curious to hear peoples thoughts on this. Personally I can't get past the idea of the wire being the bottle neck due to resistance.

As a side note the bigger transformer was only $100 more than the smaller one. Bigger wire would have been a couple hundred more so they saved the customer money if their assumption/software is correct.


Thanks for your thoughts and comments,

RLCGuy

Perfectly valid. With conductor only, you are only looking at part of the overall voltage drop. There is also drop in the transformer (see Carultch's post).

As an example with a 10 kVA load, suppose you compare a 15 kVA unit with 1.8% Z with a 25 kVA unit with 2.2% Z. For the 15 kVA, the transformer drop can be roughed as 10/15*1.8% or 1.2%. For the 25 kVA you can rough it to be 10/25*2.2% or 0.88%. Upsizing can help and may be cost effective in some scenarios.

There are more accurate formulas you would use with more detailed load info but you get the idea.

 

[Carultch]

Perfectly valid. With conductor only, you are only looking at part of the overall voltage drop. There is also drop in the transformer (see Carultch's post).

As an example with a 10 kVA load, suppose you compare a 15 kVA unit with 1.8% Z with a 25 kVA unit with 2.2% Z. For the 15 kVA, the transformer drop can be roughed as 10/15*1.8% or 1.2%. For the 25 kVA you can rough it to be 10/25*2.2% or 0.88%. Upsizing can help and may be cost effective in some scenarios.

There are more accurate formulas you would use with more detailed load info but you get the idea.

How does impedance relate to full load efficiency?

 

[kwired]

You may find cases where the lights dim more on the smaller transformer when the AC or other larger motor starts up then with the larger transformer - less impedance in the transformer results in less voltage drop before even considering drop in the conductors.

But you can't increase transformer size to compensate for a long conductor run the drop across the conductor will not change.

 

[rlcguy]

There aren't variable taps available on these transformers. They are just your run of the mill distribution padmount transformers.

 

[ggunn]

But you can't increase transformer size to compensate for a long conductor run the drop across the conductor will not change.

Likewise, if the voltage drops across the terminals of the transformer under load, then nothing you can do with the conductors will compensate for it.

 

[mivey]

How does impedance relate to full load efficiency?

part of it is resistance.

 

[Carultch]

part of it is resistance.

If the impedance of a transformer is 5%, does that mean that the efficiency is 95%?
Is it more likely to expect higher or lower efficiency, than (100%-impedance)?

Are there any formulas to determine theoretical full load efficiency, knowing the impedance?

 

[wirenut1980]

I agree with Mivey, just as conductor size contributes to voltage drop between the source and the load, so do transformer size. Upsizing the transformer with the same nameplate %Z will result in less voltage drop across the transformer given the load is the same.

If the transformer nameplate %Z is 5%, that means you will have 5% voltage drop through the transformer when the transformer is fully loaded.

 

[winnie]

If the transformer nameplate %Z is 5%, that means you will have 5% voltage drop through the transformer when the transformer is fully loaded.

This is not quite true, but is a good approximation to start with.

5% impedance means that with the secondary shorted, full current will flow with 5% voltage applied to the primary.

So in a very real sense, essentially 5% of the EMF has to be dropped across the transformer at full load.

But that is not a pure resistive drop; it has a large inductive component.

This means that the voltage seen by the load won't drop by the the full amount caused by the transformer impedance; some of it shows up as a change in phase angle. The voltage drop seen by the load also depends on the power factor of the load, and if the load is capacitive could also be seen as a voltage rise.

-Jon

 

[GoldDigger]

So in a very real sense, essentially 5% of the EMF has to be dropped across the transformer at full load.

But that is not a pure resistive drop; it has a large inductive component.

Since the magnetizing inductance of a transformer is in parallel rather than in series with the reflected downstream impedance it seems to me that, as is the case with a motor, the idling current of a transformer has a large inductive component while the full load or fault load current of the same transformer has a relatively small inductive component.
That same factor applies to the voltage drop caused by the resistance of the secondary and primary windings, which together contribute to the %Z figure.