General Purpose Transformer Theoretical Setup

Status
Not open for further replies.
Twoskinsoneman

Twoskinsoneman

Senior Member
Location
West Virginia, USA NEC: 2020
Occupation
Facility Senior Electrician
My mind was wandering today. I've not have much experience with transformers. Anyway this is what I was thinking.

What if I wanted to run a single 15a branch circuit a very long distance away. Let's say 600ft away. It could just be a single receptacle or maybe lights or whatever. Could I run NM from a 15amp breaker in my panel to a 120vac-600vac general purpose transformer, then run the 600ft at 600vac to another 120vac-600vac transformer?

What are the rules governing this.
Things that come to mind are.
Where do I need OCPDs?
What needs to be bonded?
What table can be used to gauge the wires?
What rules apply to 600vac that wouldn't apply to a standard 120vac circuit?

Purely theoritical. Thanks
 
In theory this would work. As you know this is the exact technique used for long distance power transmission. From previous web discussions, I understand that some cell sites use this technique for their _service_, not just for a single circuit.

I believe that the biggest thing neglected in this sort of installation is the impedance of the transformers. The higher voltage will help with the voltage drop of the conductors, but the transformers themselves have their own built in voltage drop, which is likely to be several percent between no load and full load.

I would probably suggest a 'buck-boost' autotransformer arrangement to compensate for voltage drop, rather than a step-up - step-down arrangement. Even better if you could get some sort of automatic tap changer to regulated the voltage at the load.

I'll let others comment on the code issues with the step-up - step-down arrangement.

-Jon
 
Twoskinsoneman said:
Could I run NM from a 15amp breaker in my panel to a 120vac-600vac general purpose transformer, then run the 600ft at 600vac to another 120vac-600vac transformer?

What are the rules governing this.
Things that come to mind are.
Where do I need OCPDs?
What needs to be bonded?
What table can be used to gauge the wires?
What rules apply to 600vac that wouldn't apply to a standard 120vac circuit?
Click to expand...

I have designed circuits like this. In general, ignore the fact that one of the voltages is 600V and just design the system as you would any other 2-wire circuit fed by a transformer.

OCPD are required on the pirmary side of transfromers. OCPD are optional on the secondary side (of course the conducotrs need properprotection per article 240).

Bonding and grounding does not change, 150V to ground must be bonded/grounded, and if the circuit is not bonded then detectors are required.

Use the standard tables for ampacities which are dependent upon the conductor insulation and installation.

No special rules, all equipment must be rated for the applied nominal voltage. If you are backfeeding a transfromer then you need to contact the manufacturer
 
Jon,
I would probably suggest a 'buck-boost' autotransformer arrangement to compensate for voltage drop, ...
Click to expand...
That only works well for a constant load. If the load changes a step up transfromer at the supply end and a step down at the load end is a better choice.
Don
 
don_resqcapt19 said:
Jon,

That only works well for a constant load. If the load changes a step up transfromer at the supply end and a step down at the load end is a better choice.
Don
Click to expand...

I was thinking this because at 600v the highest current I could pull from a 15a circuit is 3amps. If math is close I could run the 600 ft distance in 16AWG CU. That's a pretty good savings.

Does that sound right...
 
Don, I recognize that the buck/boost would not compensate for a variable voltage drop. That is why my next sentence mentioned the use of a tap changer to compensate for the variable voltage drop.

I understand how the higher voltage and lower current for the same power will mean lower percentage voltage drop in the _conductors_.

But unless I misunderstand the concept, the impedance of the transformer(s) will also influence the voltage drop of the system.

As I understand the 'impedance' rating of a transformer, it is the percentage voltage drop from no load to full load. For small transformers such as this, the impedance rating would probably be between 3% and 6%.

Say select a pair of 2KVA transformers for the present application, and we get boxes with 5% impedance. We adjust things so that at no load the voltage delivered to the distant receptacle is 120V. With a 15A load at the receptacle, the voltage will have dropped some 9% (approximating here) because of the impedance of the transformers alone, in addition to any conductor voltage drop.

A solution, of course, is to use larger transformers, so that their % loading is lower, and thus the impedance drop is lower. If we select 20KVA 5% transformers, then the voltage drop due to impedance is less than 1% from the transformers, in addition to the voltage drop from the conductors.

So rather than having oversized conductors to mitigate voltage drop, you need oversized transformers.

Do I understand this correctly?

-Jon
 
winnie said:
As I understand the 'impedance' rating of a transformer, it is the percentage voltage drop from no load to full load. For small transformers such as this, the impedance rating would probably be between 3% and 6%.

Do I understand this correctly?
Click to expand...

In a transformer the term % impedance represents the % of rated voltage required to provide rated current into a short circuit. Even though we call it IMPEDANCE and use the short hand %Z it is actually % voltage (which is %IZ).
 
Yup, I did misunderstand the impedance rating in its specifics, though I believe that I have the general concept correct.

As Jim points out, by definition the 'impedance' of a transformer is measured into a dead short circuit with reduced supply voltage. Clearly this is a different condition than nominal supply voltage with a normal load.

My corrected, though still limited understanding:

The 'impedance' of a transformer will be highly inductive, quite unlike normal load impedance. The transformer impedance will result in voltage drop as I suggested above, but the voltage drop will not be in phase with the voltage drop across the load. Vector math will be needed to determine the voltage drop across the load, and the voltage drop will almost certainly be less than the percentages given above, less then half for normal load power factors, perhaps much less for pure resistive loads.

So still an issue, but significantly less of an issue than I imagined.

-Jon
 
iwire said:
16 AWG?

No that would be a code violation 310.5, maybe 14 AWG.
Click to expand...

I see that section states "except as permitted elsewhere in this Code".
Is there not somewhere that would allow smaller than 14 for this situation. We're talking about 3amps. Since it would have it'a own OCPD on the secondary of the first transformer, is there an exception some where?
 
Twoskinsoneman said:
Is there not somewhere that would allow smaller than 14 for this situation. We're talking about 3amps.
Click to expand...

No I don't believe there is a section of code that would allow 16 AWG for 600 Volts in this application.

What your talking about is run of the mill feeder or branch circuit wiring, nothing special or unusual that would have exceptions to 310.5.

Would you really want to run 600 volts on 16 AWG?

I would not, I would probably use a minimum of 12 AWG.

It's not the load of 3 amps that is the worry, the worry in my mind is what happens during a ground fault, does the OCPD open before the 16 AWG melts back from the fault.
 
peter d said:
Where exactly does the transformer impedance value come into play? For determining fault current?
Click to expand...

Fault current is one of the situations.

In the case of a bolted fault at the transformer, the only thing restricting current flow is the impedance of the transformer and the characteristics of the primary supply. A common approximation is to treat the primary supply as 'infinite', and simply calculate the available current flow as limited by the transformer itself.

The transformer impedance is an impedance that is in series with the load, just like the resistance of the wires feeding the load is in series with the load.

In the same way that the resistance of the wires causes a voltage drop, the impedance of the transformer causes a voltage drop. The _difference_ between the two is that the resistive voltage drop of the wire is always in phase with the load current, whereas the voltage drop of the transformer has a strong inductive component. This inductive voltage drop is out of phase with the load current and has much less effect on it.

I tried to find information on the relative contribution of resistance and reactance to transformer impedance. A google search for "transformer impedance X/R ratio" provides quite a number of interesting hits, including a link to a GE white paper http://www.geindustrial.com/publibrary/checkout/White Papers|GET-3550F|generic

Page 42 and 43 have tables of transformer impedance, however one of the other search results notes that the tables are dated but still useful. http://www.appanet.org/newsletters/ppmagazinedetail.cfm?ItemNumber=17451&sn.ItemNumber=0

A caution that brings us back to the original question in this thread: from what I've gathered, smaller transformers tend to have a greater proportion of resistive to reactive impedance, and thus a greater proportion of the transformer impedance will be seen by the load as voltage drop.

-Jon
 
Yes, Fault current and when you parallel transformers.
The only other point where impedance comes into play that I can think of is transformer inrush when a transformer is first energized. Small KVA transformers as well as those that are incapsulated and low Z. It is not uncommon to see 1-2%Z. Low Z = high inrush to magnetize the core = a possible problem with tripping the pri. OCPD when the transformer in first energized.
 
Status
Not open for further replies.
Top