Transformer sizing for step up and step down transformers

[jaggedben]

...
Call the transformer company, any one of them, and tell them you have 80A/66kVA of 480/277V PV inverter power that you want to feed into the grid at X thousand volts delta. ...
l

This is the kind of thing that gets the moderators get mad at you, and bloats the thread. He doesn't have a delta service (we can tell that from the one clear photo he posted), and getting a new service may not be legally or financially feasible for getting credited for the PV output, depending on the state's net metering regime.

Try to temper your suggestions with the caveats that would be necessary to ensure they would be valid suggestions. Or just ask questions instead.

 

[PVfarmer]

This is the kind of thing that gets the moderators get mad at you, and bloats the thread. He doesn't have a delta service (we can tell that from the one clear photo he posted), and getting a new service may not be legally or financially feasible for getting credited for the PV output, depending on the state's net metering regime.

Try to temper your suggestions with the caveats that would be necessary to ensure they would be valid suggestions. Or just ask questions instead.

I hear ya-
I wasn't saying he had delta service at all- I was attempting to reference the SolarPro article about how utilities "prefer" delta primaries *when* the service is 3ph.
I'm taking into account that my (similar) farm project will be going from the 7200V to 120/240V single/split phase service xfmr to a 12.47kV (grid delta voltage) to 480/277V 3 phase xfmr, because of the PV inverters.
It's worth it.

I did tell him to make sure he can even get 3 phase service (hear that zman?) where he is before anything else, if he want to use 3ph inverters.

I'm trying to tell zman that if he shows his POCO the plans for the mystical 3ph to 1ph PV setup, they'll say...um, nope.

It could be the exact opposite where he is, but where I am, getting a new 3ph service for over 50kVA of PV output (and not going 1ph) is definitely encouraged, if not 100% required.
I'll spare you the copy + paste, but I've read the pdf from my POCO a few times.

 

[ggunn]

I'm not here to argue with people I just feel I am right in saying this is possible...

But that's just it. You are not right, and when you keep insisting that you "feel" that you are, in spite of folks who DO understand this stuff telling you what is wrong with your idea, you are obviously here to argue with people. In the immortal words of Scotty, "I cannae rewrite the laws of physics, Cap'n!", and neither can you. I know it sounds romantic for the guy who doesn't understand how things work to come in with an "out of the box" concept and prove all the experts wrong. That happens all the time in the movies but virtually never in real life.

I really don't mean to be hard on you, but this belief you express is based upon your ignorance of the fundamentals.

 

[electrofelon]

PVfarmer,

You are really overthinking and overcomlicating things with all your discussion about utility primary voltage and utility transformers. You have nothing to do with the utility transformer. The utility will tell you how much DG you can have. They will tell you how much it will cost if you want more. If you want to upgrade (or add) a 480 volt service, they will tell you how much it will cost.

Those MV transformers you linked to: In the vast majority of cases, you WILL NOT have a primary service. Of course there are customers that do have primary utility service, but they are going to be very large customers. Probably none of us will do a PV system where the utility gives us a primary service. Remember that although a, say, 1 meg PV system sounds really big to most of us that is only 1200 amps at 480V. That is still in panelboard territory. Most supermarkets (where I have looked) have a 750K-1 meg transformer. It seems you dont have a lot of experience with utilities and services.

 

[iwire]

Try to temper your suggestions with the caveats that would be necessary to ensure they would be valid suggestions. Or just ask questions instead.

:thumbsup:

 

[PVfarmer]

zman- Hang in there, buddy. I asked for a Pre-Application Report from the POCO here- that's how I found out that 3 phase was available at the farm here.

PVfarmer,

1 You are really overthinking and overcomlicating things

2 The utility will tell you how much DG you can have. They will tell you how much it will cost if you want more. If you want to upgrade (or add) a 480 volt service, they will tell you how much it will cost.

3 Those MV transformers you linked to: In the vast majority of cases, you WILL NOT have a primary service.

Thanks.

1 Yes, I've been known to do that. I'm not here to argue about my own MV credentials. I do know for a fact that zman's inverter installation manual says "MV transformer required".
And it seems to be a fact that his service xfmr at the farm is 50kVA or less, and 1ph.
These are both huge problems with 66kVA of 3ph PV power!
The reason I mentioned grid voltage is that with the "MV transformer required" thing, zman should realize that he *has to* have those 480/277V inverters connected to a xfmr secondary *and* whatever voltage his POCO gives him on the primary side- to make the inverter work.
If he can't get 3ph, he'll still need to upgrade to at least a single 75kVA service xfmr if he wants to feed in 66kVA of single phase PV.

2 I know those things, as far as my situation is concerned.
The point I'm trying to make in general is that zman hasn't seemingly talked to his own POCO about this, and it would be crazy to order expensive equipment before talking to them, that's all.
As far as my farm project, hopefully it won't cost anything for the xfmr, which is 12.47kV delta to 480/277V wye.
If the farm here did have 4160V grid, buying that 4160V delta primary xfmr might have been an option- but the grid is 12.47kV- so the POCO engineer designs everything up to the pad mounted xfmr.

I don't know why the POCO would say "consult us" about 4160V below (from ESB 750, I see RI and MA in your sig)- it would be helpful if they were a little more specific. "Appropriate" delivery voltage is kinda vague.

3 Yes, most people don't get MV service- but if zman has a 1000 foot driveway at his farm project, the wires would stay at whatever his grid voltage is from the grid to the service xfmr at his barn there.
There's only one person who can tell him what his voltage and KVA are- the connection department guy from his own POCO.
I thought the xfmr for my farm project was 50kVA- it's really 25kVA, which I found out by talking to the connection dept. guy on the phone. Now it's going to be 150kVA, and 3 phase, which is nice.


Non-standard services include, but are not limited to: 25 Hertz, 2 phase systems, 2 wire 120 volts,
240 volts delta, 460 volts wye, 480 volts delta, 600 volts delta, 2400 volts, 4160 volts or 4800 volts
services.

While 2400 volts, 4160 volts, or 4800 volts are no longer standard, they may still be available
at certain locations; consult the Company.

Customers now receiving non-standard service shall not expand the use of such service, except in
very limited circumstances at the sole discretion of the Company.
Customers with an existing non-standard service requesting a service change shall consult with the
Company to obtain a standard single or three phase 60 Hertz service at an appropriate delivery voltage.

 

[electrofelon]

zman- Hang in there, buddy. I asked for a Pre-Application Report from the POCO here- that's how I found out that 3 phase was available at the farm here.



Thanks.

1 Yes, I've been known to do that. I'm not here to argue about my own MV credentials. I do know for a fact that zman's inverter installation manual says "MV transformer required".
And it seems to be a fact that his service xfmr at the farm is 50kVA or less, and 1ph.
These are both huge problems with 66kVA of 3ph PV power!

maybe. That is up the the utility. they typically have no problem loading their transformers to 120%. Also depends on if that 66KVA is DC or AC (I dont feel like reading back through the thread to refresh myself).

The reason I mentioned grid voltage is that with the "MV transformer required" thing, zman should realize that he *has to* have those 480/277V inverters connected to a xfmr secondary.....

No. the inverters need to be connected to an electrical service or SDS having a configuration compatible with the inverters. I dont mean to nitpick - yes the service will be the secondary of a utility transformer - It just seems more confusing when you put it that way.

2 I know those things, as far as my situation is concerned.
The point I'm trying to make in general is that zman hasn't seemingly talked to his own POCO about this, and it would be crazy to order expensive equipment before talking to them, that's all.

I concur. Having an accepted interconnect agreement before ordering stuff is smart.

As far as my farm project, hopefully it won't cost anything for the xfmr, which is 12.47kV delta to 480/277V wye.
If the farm here did have 4160V grid, buying that 4160V delta primary xfmr might have been an option- but the grid is 12.47kV- so the POCO engineer designs everything up to the pad mounted xfmr.

If the grid was 4160, wouldnt that have been the only option??

I don't know why the POCO would say "consult us" about 4160V below (from ESB 750, I see RI and MA in your sig)- it would be helpful if they were a little more specific. "Appropriate" delivery voltage is kinda vague.

Im sure there are some utilities out there that are real flexible and will give you whatever voltage they make a transformer for, but I think you will find most wont do anything "abnormal". I tried to get a single phase 480 service from national grid and they wouldnt do it.

3 Yes, most people don't get MV service- but if zman has a 1000 foot driveway at his farm project, the wires would stay at whatever his grid voltage is from the grid to the service xfmr at his barn there.

Its still utility MV conductors and a LV service. He probably needs to give the utility an easement but there is still a service point, utility lines, and customer lines.

 

[ggunn]

Isn't all this totally beside the point? His assertion is that he can land a three phase inverter on a single phase service. All this talk about MV interconnections is completely irrelevant; no one is going to bother with interconnecting a measly 66kW at MV.

 

[PVfarmer]

zman -???

Your POCO has some sort of distributed generation dept- that's where you start, with the Pre-App Report.
Do it, man!

Isn't all this totally beside the point? His assertion is that he can land a three phase inverter on a single phase service. All this talk about MV interconnections is completely irrelevant; no one is going to bother with interconnecting a measly 66kW at MV.

1 maybe. That is up the the utility. they typically have no problem loading their transformers to 120%. Also depends on if that 66KVA is DC or AC (I dont feel like reading back through the thread to refresh myself).

2 No. the inverters need to be connected to an electrical service or SDS having a configuration compatible with the inverters. I dont mean to nitpick - yes the service will be the secondary of a utility transformer - It just seems more confusing when you put it that way.

3 I concur. Having an accepted interconnect agreement before ordering stuff is smart.

4 If the grid was 4160, wouldnt that have been the only option??

5 Im sure there are some utilities out there that are real flexible and will give you whatever voltage they make a transformer for, but I think you will find most wont do anything "abnormal". I tried to get a single phase 480 service from national grid and they wouldnt do it.

6 Its still utility MV conductors and a LV service. He probably needs to give the utility an easement but there is still a service point, utility lines, and customer lines.

ggunn-
Yes, you're probably right there. It would just be nice if the POCO would say "4160V is available for PV connections OVER 250kW"...or something!
It could possibly be relevant- there is in fact a section of the application- "Is the customer providing an interconnection transformer?"
I would be surprised if anybody ever answered "yes", sure.
But it is mentioned. If you had 250kW (or any amount really) of 480Y PV output and a 4160V grid, the PV would have to go through a 480y --> 4160V xfmr- this is just a question of who owns it, customer or POCO.
It's stuff like this link that gets me, um, distracted.
So by the below (loosely), you'd possibly be saving $8800 going MV with 250kVA of PV, but only $880 with 25kVA of PV.
Saving $8800 seems worth considering.

The following example compares the cost of running 2,500 kVA for 100 ft at 480 V with 600 V cable and running the same amount of power for the same distance at 13.2 kV with 15 kV EPR cable:

• Running 480 V, 3,000 A for 100 ft requires eight sets of 3.5-in. conduit, each with three 500 kcmil phase conductors and one 500 kcmil ground conductor, and costs around $100,000
• Running 13.2 kV, 110 A for 100 ft requires one 4-in. conduit with three 2/0 phase conductors and one 2/0 ground conductor, and costs around $12,000.

http://www.csemag.com/single-articl...ibution/9ea49999f50f77b358677b5c3c283fff.html

electrofelon-

1 Right, they load them to 120% to step down to supply customer loads. The farm here draws 120A from a 104A/25kVA xfmr. But can you just do that backwards for stepping up?
Even if you could do it, it seems like it would affect the xfmr efficiency and therefore PV output kind of a lot.
The POCO doesn't want to be paying people for power going out thru the meter, and then lose a lot of it because the xfmr is undersized.

2 I don't get how any grid tied PV inverter can be connected to *only* a SDS.
Here's my simplified interpretation- I keep seeing the term "infinite impedance" applied to the grid.
GTIs *must* have that infinite output. Period.
If you had a 25kW GTI connected to a SDS only, wouldn't the loads from the SDS have to be exactly = to the PV output at all times?
If the loads at high noon are 24.6kW which = the inverter output, maybe ok. But what about at 3PM when the loads are still 24.6KW and the inverter is only putting out 15kW?
Or if the loads are 2kW at noon because everybody is at lunch, where's all the PV going to go?
Inverters can adjust the DC side somewhat by MPPT, but 100% of the DC is converted to AC and has to go somewhere.

3 That would really be the only way to go. It's the POCO's grid- they have the final say.

4 Well, yes. I'm saying the option there would be who actually owns the xfmr. The POCO needs that easement because they own the wires going over or under the customer's property and usually the xfmr.
But the customer owning the xfmr is a possibility- check out that link about MV connections- they are becoming more common as of 3 years ago..."On average, primary metering can save up to 5% on kWh charges. "
5% is nothing to sneeze at!

5 That's exactly what I mean by NG being more specific! I was thinking about 277/480V 1ph. They list 277/480V and 120/208V 1ph as "available" on the chart on pg 71. (ESB 750)
BUT! If you look at the chart on pg 37 *and then* note 3 on pg 38...277/480V 1ph is only available off of a 480/277V 3ph service!

Note 3: (277/480V 1ph is available...)
3. Where the present service is three phase, 4 wire, 480 wye/277 volts. Three-phase, 4 wire, 480 wye/277 volts is Commercial and Industrial use only.

6 Back to #4...isn't the definition of "service point" the service drop? As in the POCO usually owns the wires (aboveground) up to where they attach to the building, and the customer owns from there thru the meter and to the main switch and on?

But regardless, the important thing here is that zman talks to his POCO to find out if 3ph is available, before even designing a 3ph system, much less buying anything!

edit:
Also depends on if that 66KVA is DC or AC (I dont feel like reading back through the thread to refresh myself).

He wants to use two of the 33kW SE 3ph 480 *wye* inverters. So his proposed 480V delta xfmr isn't the right one.

 

[PVfarmer]

I tried to get a single phase 480 service from national grid and they wouldnt do it.

Hey, about that, and the note #3 thing...

I discussed this with my engineer. With my farm project, there's a "rule" that the house and "agricultural" part have to be separately metered. Ok. That would happen with a 3ph "main" service and additional 1ph 277/480V service, with the barn area being 3ph and the house being 1ph. (or it could be the barns at 208Y/120V 3ph and the house at 120/208 1ph)

"Where the present service is three phase, 4 wire, 480 wye/277 volts. "
...you can get 1ph 277/480V.
So the "present" (according to NG rule) 4 wire service being wye, as in A, B, C and N, and you get the 480V from A-B, B-C, or A-C.
You get the 277/480V 1 ph additional service from say A, B and N however.

When you add PV into the system, the inverters would be supplying the barns' varying loads on A B C, as well as the house's varying load on A and B- which would cause A and B PV output to differ from C output at the service xfmr. Not balanced.

If the PV inverter supplies both the service step up xfmr AND a single "loads" step down xfmr at the inverter nominal voltage (3ph wye 480/277V *with* neutral for both LV side of service xfmr and HV side of loads xfmr), it is more balanced.

But then in my case, with a single loads xfmr, there's only one meter for barns and house- can't do that.
But, the POCO is ok with new 480/277V for the barns, and leaving the existing 120/240V service where it is, for the house only. (Two accounts- residential and commercial)
Which is pretty much ideal.

 

[GoldDigger]

PVfarmer,
In your response to # 2 you seem to be assuming that all SDSs are independent power sources such as generators. The GTI would be connected "only" to an SDS even if that SDS is just a customer owned transformer off the POCO service. The poster did not want to limit the comment to direct service interconnections. You are correct that small independent power sources could be a problem, but not, as you state, even in the case where local loads are greater than the PV output!
Also, FWIW, the grid impedance is not infinite (or a good approximation), it is zero. The conductance is infinite.

 

[PVfarmer]

1000V DC and accessibility

1000V DC and accessibility

I truly feel getting over 1000volt dc system is the biggest hurdle.

I asked a while back in this thread, but more specifically...

zman said his system was ground mounted and over 600V DC. What are the "complications" with over 600V DC, as long as it's outside and not attached to a building?
Obviously, everything underground is inaccessible.

This part after the italics from the 1000V DC link is saying (outdoor) DC combiners over 600V have to be kept locked?

And could someone confirm this? ->The over 600V DC ends at the inverter(s) DC input(s). Which code section is re: how far the 600v DC can go to an inverter which is inside an electrical shed? Is it 10 feet? Or 5 feet? And it must be metal conduit for that short indoor DC run?
Am I in the right neighborhood with the 10 (or 5) foot rule there?

Thanks.

SolarPro magazine-Apr/May 2013
Section 110.31, “Enclosure for Electrical Installations.” In
indoor installations, electrical equipment in systems above
600 V—inverters, combiners, switches and so on—must be
metal enclosed and marked with appropriate caution signs
if they are accessible to unqualified persons. Alternately,
the equipment can be rendered accessible only to qualified
persons by installing a lock or locating the equipment in
an access-controlled vault, room, closet or other walled or
fenced-in area. The upshot here is that inverters, combiners
or switches with nonmetallic enclosures may be subject to
different requirements than those with metallic enclosures.
While it stands to reason that components properly evaluated
and certified to UL 1741, whether in a metallic enclosure
or not, should have the same installation requirements,
an AHJ may interpret this differently.

In outdoor installations, electrical equipment behind
a locked fence, wall or screen is considered inaccessible to
unqualified persons. Per Table 110.31, a fence must have
10-foot clearance to live parts. Equipment that is accessible
to unqualified personnel, whether nonmetallic or metal enclosed,
must be designed “such that exposed nuts or bolts cannot be
readily removed.” If the enclosure has a door or hinged cover and is
mounted less than 8 feet above the ground level, it must be kept locked. Note that this section also
references Article 225, which in turn references Section 300.37
(both of which are discussed later in this article).
http://www.sma-america.com/fileadmi...gn/SolarProMagazine_1000VDC_AprilMay_2013.pdf

 

[zman990]

I asked a while back in this thread, but more specifically...

zman said his system was ground mounted and over 600V DC. What are the "complications" with over 600V DC, as long as it's outside and not attached to a building?
Obviously, everything underground is inaccessible.

This part after the italics from the 1000V DC link is saying (outdoor) DC combiners over 600V have to be kept locked?

And could someone confirm this? ->The over 600V DC ends at the inverter(s) DC input(s). Which code section is re: how far the 600v DC can go to an inverter which is inside an electrical shed? Is it 10 feet? Or 5 feet? And it must be metal conduit for that short indoor DC run?
Am I in the right neighborhood with the 10 (or 5) foot rule there?

Thanks.

SolarPro magazine-Apr/May 2013
Section 110.31, “Enclosure for Electrical Installations.” In
indoor installations, electrical equipment in systems above
600 V—inverters, combiners, switches and so on—must be
metal enclosed and marked with appropriate caution signs
if they are accessible to unqualified persons. Alternately,
the equipment can be rendered accessible only to qualified
persons by installing a lock or locating the equipment in
an access-controlled vault, room, closet or other walled or
fenced-in area. The upshot here is that inverters, combiners
or switches with nonmetallic enclosures may be subject to
different requirements than those with metallic enclosures.
While it stands to reason that components properly evaluated
and certified to UL 1741, whether in a metallic enclosure
or not, should have the same installation requirements,
an AHJ may interpret this differently.

In outdoor installations, electrical equipment behind
a locked fence, wall or screen is considered inaccessible to
unqualified persons. Per Table 110.31, a fence must have
10-foot clearance to live parts. Equipment that is accessible
to unqualified personnel, whether nonmetallic or metal enclosed,
must be designed “such that exposed nuts or bolts cannot be
readily removed.” If the enclosure has a door or hinged cover and is
mounted less than 8 feet above the ground level, it must be kept locked. Note that this section also
references Article 225, which in turn references Section 300.37
(both of which are discussed later in this article).
http://www.sma-america.com/fileadmi...gn/SolarProMagazine_1000VDC_AprilMay_2013.pdf

That was badass post pv farmer I said same thing ahj is in 2011 goes right to 690.7.

Sent from my HTC One_M8 using Tapatalk

 

[GoldDigger]

The two problems I see with PV over 600V are that it is not allowed at all at a residential site and that if not under 2014 code cycle it would require use of MV wiring methods.

 

[PVfarmer]

PVfarmer,
In your response to # 2 you seem to be assuming that all SDSs are independent power sources such as generators. The GTI would be connected "only" to an SDS even if that SDS is just a customer owned transformer off the POCO service. The poster did not want to limit the comment to direct service interconnections. You are correct that small independent power sources could be a problem, but not, as you state, even in the case where local loads are greater than the PV output!
Also, FWIW, the grid impedance is not infinite (or a good approximation), it is zero. The conductance is infinite.

Yes, yes- I get it now- totally my fault.
You mean Separately Derived System, correct?
For some reason I saw SDS and thought "Secondary Distribution".

When he said-
"the inverters need to be connected to an electrical service or SDS"...

um...what does he mean exactly then? I thought any GTI has to be connected to POCO service *and* any SDS.
As in, GTIs can output the PV power to both grid *and* any SDS, or 100% to grid, but never 100% to just a customer xfmr and loads.

Just a really simple hypothetical example- if you had a 4000w SMA Sunny Boy 240V inverter (GTI) and two Sunny Island 4048s (hybrid) with plenty of batteries, the 4000w GTI would only run until the batteries were full if it was off-grid. On-grid, it wouldn't have to shut down, as the SIs would send it's extra output to grid when the batteries were full and the load was < PV output.
Correct?

 

[PVfarmer]

The two problems I see with PV over 600V are that it is not allowed at all at a residential site and that if not under 2014 code cycle it would require use of MV wiring methods.

Thanks a lot- it's 2014 here.
Now, as far as residential- do you mean on residential buildings, or as in as per local property zoning, or only on a commercial electrical account, or... all of those? Something else too?

Aren't all 1000V DC rated inverters also 3 phase? And 480V to boot? The chart I have for 2014 lists them all as such.

 

[PVfarmer]

That was badass post pv farmer I said same thing ahj is in 2011 goes right to 690.7.

Thanks, dude.
Now, if I were you I'd be asking GD what he meant by "if not under 2014 code cycle it would require use of MV wiring methods."

Because you aren't and that sounds like it could be pricey, but I'm not exactly sure what the details are there.

Not sure what you mean by "goes right to 690.7". There's a bit more to it!
That tells you if you're over 600V, sure. But nothing about connecting it if you are.

the rated open-circuit voltage (Voc) of the series-connected PV modules,
as corrected for the lowest expected ambient temperature.

 

[GoldDigger]

Yes, yes- I get it now- totally my fault.
You mean Separately Derived System, correct?
For some reason I saw SDS and thought "Secondary Distribution".

When he said-
"the inverters need to be connected to an electrical service or SDS"...

um...what does he mean exactly then? I thought any GTI has to be connected to POCO service *and* any SDS.
As in, GTIs can output the PV power to both grid *and* any SDS, or 100% to grid, but never 100% to just a customer xfmr and loads.

Just a really simple hypothetical example- if you had a 4000w SMA Sunny Boy 240V inverter (GTI) and two Sunny Island 4048s (hybrid) with plenty of batteries, the 4000w GTI would only run until the batteries were full if it was off-grid. On-grid, it wouldn't have to shut down, as the SIs would send it's extra output to grid when the batteries were full and the load was < PV output.
Correct?

Correct.

 

[PVfarmer]

Also, FWIW, the grid impedance is not infinite (or a good approximation), it is zero. The conductance is infinite.

Correct.

Maybe they (the POCO) mean infinitely small? As in so close to zero it isn't a factor re: rating purposes?

And thanks again!

9.3 AVAILABLE FAULT CURRENT
For equipment rating purposes, the following tables list the maximum fault currents available at the Company’s
transformer secondary terminals. These fault currents are based on the lowest impedance of transformers
the Company procures and on infinite supply impedance on the primary side.

Also- this is interesting, from the article SolarPro posted earlier, I'm glad I'm on 20114 NEC!:

I use the term MV here to describe electrical system components rated between 5 kV and 38 kV. This definition corresponds to common US utility-distribution voltages. In practice, voltage class definitions are somewhat tricky to pin down. They vary from industry to industry and from one set of codes or product standards to another. On one hand, The Authoritative Dictionary of IEEE Standards Terms (IEEE 100) defines medium voltage as “a class of nominal system voltages greater than 1,000 V but less than 100,000 V.” On the other, Article 490 of the National Electrical Code defines high voltage as “more than 600 volts, nominal.” While the NEC does not include a stand-alone definition of medium voltage, Article 328 details the Code requirements related to medium-voltage or Type MV cable. As described in NEC Sections 328.2 and 328.10, MV cable is rated at “2001 volts or higher” and is “permitted for use on power systems rated up to and including 35,000 volts, nominal.”
http://solarprofessional.com/articles/design-installation/basics-of-medium-voltage-wiring

 

[zman990]

I ws just saying it is possible and it is

Sent from my HTC One_M8 using Tapatalk