Backfeeding transformers with grid interactive micro-inverters

Status
Not open for further replies.
This is my first post and I am hoping to get some insights on using a 150KVA - 480V(Delta) / 208V/120V (Wye) dry type transformer in a PV system using grid interactive inverters (specifically Enphase M250 micro-inverters) -

The situation is this:
Enphase M250 microinverters generate a 3-phase 208V output and connected to the low voltage (208V) side of a 150kva dry type transformer. These micro-inverters won’t power up until they sense the presence of the utility voltage which will need to come from the 480V side (primary) to the 208V (secondary) applying the 208V to the micro-inverters. In this specific application the 480V side will be the first side of the transformer to which a potential is applied which will in turn provide any inrush currents magnetically charging the transformer. Once the micro-inverters see the 208v potential on the secondary, they will then start producing an output which will be shared with the utility – through the transformer. So while power transfer will go back and forth between the 208V and 480V sides depending on the load sharing capability of the micro-inverters, the transformer will remain supplied from the 480V side, even when the micro inverters stop producing output after the sun sets. The primary (480V side) will remain the dominant source of power in the system.

Problem:
The transformer manufacturer specifies that when their transformer (Dry type 150KVA) is wired to be back fed or in a step up configuration the X0 connection should not be connected and not establish a neutral on the LV side. The reason given is:(my commentary are provided in red)

1) the wye connection (120V) on the LV side is normally not needed when stepping up voltage (In my application the LV (208V) side needs to supply the 120V needed to power a Envoy communication gateway- which isn't possible without establishing a neutral on the LV side)
2) to avoid creating a direct path to ground and possibly generating parasitic currents on the case which may be harmful (not sure why potentials would be developed on the case of the transformer if it is properly grounded)
3) in the case of a short circuit or atmospheric discharge on the LV side affecting the HV side because the two sides not being isolated (again, not sure why there would be an issue with interaction between LV and HV sides if case is properly grounded - no neutral will be created or used on the 480V side)

I am hoping to get any insights on the hazards with connecting the LV side in a wye configuration with a neutral, that can help me to further understand the manufacturers concerns with connecting the X0 terminal of the transformer when stepping up voltage, or in my case, back feeding the transformer?

Thanks,

Badger
 

jaggedben

Senior Member
Location
Northern California
Occupation
Solar and Energy Storage Installer
I would contact Enphase and get their ideas. Hopefully you haven't already bought the transformer; they may recommend another manufacturer.
Or maybe they can talk to the manufacturer and get them to agree this isn't a typical step up situation.

BTW, in addition to powering the Envoy, you also need the neutral to get the inverters to turn on.

Also, are you aware that Enphase has a 480V micro-inverter now? I believe it also requires a neutral, however.

As far as theory goes, all I can say is that I don't believe the Enphase inverters put any current on the neutral. The only current on the neutral would be 5w for the Envoy which is not a step up application.
 
Location
California
I would second checking with Enphase. I do believe the neutral is required for voltage reference (minimal to no current). The 480V version outputs between phase to neutral so it definitely needs a neutral connection. Their spec sheet is listing a strange voltage range: https://enphase.com/sites/default/files/C250_DS_EN_US_web.pdf (I think there may be a typo).

In your application, the neutral is definitely required. The only reason I can think about with the manufacturer's concern is that it could be used as a grounding transformer. On some other inverter applications (generally central), you would not need/want a neutral connection. Some inverters require an ungrouned wye but Enphase is not one of those inverters. You might see what the transformer manufacturer says if you tell them the neutral is required.
 
Thanks for the responses:

As recommended I had checked with Enphase and for the M250 - Enphase doesn't consider the application to be a step-up voltage application which would then use a distribution transformer to be connected in its standard configuration (Delta primary - Wye secondary). The neutral on the secondary is necessary for the micro-inverters to operate and is primarily a communication pathway instead of current carrying (NEC 2008 Article 690.62 does allow sizing where the conductor is "used solely for instrumentation, voltage detection, or phase detection, and connected to a single phase or 3-phase utility-interactive inverter") which is the case with Enphase micro-inverters. The neutral is required and needs to be connected for the Enphase micro-inverter to work.

Jaggedben -you stated that because the only current on the neutral is ~ 5W (for the envoy) this is not a step up application. Can you provide more information on what can be used to determine when an application is "step-up"?

As far as Enphase recommending specific transformers in this instance with the M250 they don't. This is unfortunate. However, the C250 micro-inverter mentioned earlier does provide an output that is more easily transformed to 480V and uses an auto transformer (which Enphase does provide recommendations). Those auto-transformers convert the C250's 422Y/244V output to a 480Y/277V output - and some autotransformers are recommended in Enphase's C250 Commercial Design Guide (https://enphase.com/en-us/support/c...commercial-pv-using-c250-microinverter-system)

I have also been in contact with Eaton whose response would seem to indicate they have never dealt with this before... and they would see this as a field installation question, not a design application question...???...

PV Farmer - the service is 480V/277V.
 

GoldDigger

Moderator
Staff member
Location
Placerville, CA, USA
Occupation
Retired PV System Designer
Keep in mind that although the power flow is from GTI to service the magnetizing current in the transformer will still come from the service side. That removes some of the potential concerns about using a transformer "backwards" (if that is part of this problem.)
It does mean that treating the GTI side of the transformer as the "input" is not appropriate.
 

jaggedben

Senior Member
Location
Northern California
Occupation
Solar and Energy Storage Installer
Jaggedben -you stated that because the only current on the neutral is ~ 5W (for the envoy) this is not a step up application. Can you provide more information on what can be used to determine when an application is "step-up"?

I don't think I stated what I meant and I might need to take it back. What I was surmising was that whereas a wye primary powered by the grid might be able to put a lot of current on the neutral, the Enphase inverters at 208V aren't capable of that. Thus concerns about the neutral being connected might be quite different.
I'm not an engineer though, so take it from me with a grain of salt.

BTW, this is true for the M250 at 208V, but might not be true for the C250 which connects L-N. See below.

Sorry if this is OT, but why would Enphase design their C250 to deliver 422Y/244V output instead of 480Y/277V?
Cheers, Wayne

At first I had the same reaction; that's totally weird and shocking. Then I realized that the C250 connects line-to-neutral, and suddenly it all made sense. The C250 and M250 are the same internally. The only difference is the firmware and the fact that the C250 connects line to neutral. The L-N voltage of the C250 is basically the same as the L-L voltage of the M250. Rather than engineer an new inverter capable of outputting 277V, they just took their existing product and tweaked a connection, essentially needing zero physical product development. The thing may or may not take off as product, but they essentially sunk no money into trying, so why not try?

I have no inside knowledge, mind you, but Enphase has done this sort of thing before. The old M190 came in a single phase and three-phase version; the only difference was the wiring connection. Then they externalized the wiring differences to the trunk cable when they developed the M215 and M250. Now they've re-internalized one small wiring difference to create a commercial product that is really the same as the residential product. I don't even know why they have different product numbers for the M250 and C250 trunk cable, but maybe the voltage rating is actually different.
 

Smart $

Esteemed Member
Location
Ohio
Thank you and I agree. The transformer will be connected, protected and grounded as if the 480V (HV) side is the input.
FWIW, don't forget these are designed as voltage transformers. Inverters are not a voltage source, though they do raise the voltage a bit to push out current.

Another consideration is knowing the stiffest voltage source remains on the high side. Even if you push the full rated 150kVA in reverse, the high side voltage won't sag like it would in a step up scenario with a 150kVA 480V load. In fact, the high side voltage will be a tad higher than at low or zero power transfer.
 
PV Farmer - the service is 480V/277V.

You might want to browse page 12 here, and ctrl-f the whole booklet for the word delta.
This part below is concerning RI, and the requirements are different in MA, NY and CT (all in this booklet).
The way I'm reading this, by primary of the step-up transformer, they mean the customer side.
So if you were over 500kW of DG, you could NOT use a xfmr with 480V delta in RI.

The way I'm reading it, when they're talking about step-up xfmrs, the primary is the customer side?

5.3.4 Neutral Stabilization and Grounding
Where the Customer is served from a Company four-wire multi-grounded neutral
distribution circuit, adequate grounding must be provided to ensure neutral stability
during accidental isolation of the line from the main system. This may require an
additional ground source. Adequate grounding can be provided either by the use of
wye-delta main power transformer (see Section 5.5) or by installing an appropriate
grounding transformer. The effects of such grounding on the Company’s ground-relay
sensitivity shall be limited.
As a minimum, the Customer’s generation equipment shall be grounded in accordance
with the latest requirements of the National Electrical Code (NEC). For specific
installations, refer to the applicable sections of this document and the Company’s ESB
750. The Company requires ground protection on any system that can be a generation
source and to protect transformers that can be paralleled and supplied from two sources.
When generator tripping is needed to sense ground faults on the Company’s Distribution
EPS:
• The winding arrangement of the Customer’s DG facility transformer shall be such
that the Company’s system remains effectively grounded (see Section 5.5.3).
• The Company may require that the grounding impedance be limited to the highest
value suitable for neutral stabilization (see Section 5.7.2), or to limit generator ground
fault contributions (see Section 5.7.5).
• A “zero-sequence” voltage or “3V 0 ” scheme will be required on the primary side of a
delta primary wound transformer supplying the DG facility; see Figure 4

5.5.3 Primary and Higher Voltage Served DG Customers
The Company reserves the right to specify the winding connections for the Customer’s
“step-up” transformer between the Company’s delivery point and the Customer’s DG
facility’s output voltage as well as whether it is to be grounded or ungrounded. The
Customer shall provide their interface “step-up” transformer’s specifications and ratings
for the Company’s review in the impact and detailed studies.

5.5.3.1 Effectively Grounded, Four-wire Multi-grounded 3-phase Wye EPS:
The Company requires that all interface transformers be configured to have a
wye connected primary winding with a fully insulated neutral, and the secondary
winding to have a delta connection.
• If infeasible, an alternative that is subject to Company acceptance review is a
primary wye grounded - secondary wye grounded transformer requiring a
grounded source permitted under specified conditions.
• Any DG or aggregate DG below 500kW in a Customer’s facility may be
permitted to utilize a primary delta - secondary wye grounded transformer.
• Where any DG or aggregate DG is 500kW or greater in a Customer’s facility,
ungrounded transformers (i.e. primary delta or Yg-Yg with ungrounded
source) are not permitted on any Company-owned effectively grounded, four-
wire multi-grounded 3-phase wye circuits.

5.5.3.2 Not Effectively Grounded, Three-wire 3-phase EPS:
The Company only allows the connection of delta primary interface transformers
on three-wire not effectively grounded 3-phase primary and high voltage circuits.
Refer to Section 5.3.4 for grounding requirements such that the generator does
not contribute to line-ground faults on the Company’s EPS.
http://www.nationalgridus.com/non_html/shared_constr_esb756.pdf
 
quote_icon.png
Originally Posted by GREG-BADGER

PV Farmer - the service is 480V/277V.


This seems like the Sunny Tripower deal, where they say 480V but the inverters are really putting out 3 legs of 277V L-N.
So are the Enphases really only putting out 3 legs of 120V L-N? The 480V or 208V listed in specs are "nominal", meaning named because they exist, but they aren't really used?

If so, this may be an idea- while I am NOT an engineer, if you do a bank of 3 single phase xmfrs, wired 208/120 wye, so getting one leg of 120V L-N each from Enphases, and putting out one leg of 277 L-N each towards the grid...
you wouldn't have to use delta at all?

Three of these, set up to run comfortably bi-directionally? It does say reverse connectable...


http://www.temcoindustrialpower.com/products/Transformers_-_General/HT0075.html
 
No, the C250s are putting out 244V (nominal) L-N.

Sorry, I was talking about the M250s putting out 120V L-N and 208 L-L.

You have to use either an isolation or auto xfmr to go to 480/277V with the C250s according to page 12.
http://enphase.com/sites/default/files/C250_manual_EN_NA.pdf

It would be awesome if someone could clear this up for me.
Transformers A and B posted below.
Xfmr A has an "empty dot", otherwise known as a circle, on the left side of the H2-H5 side and the same dot on the RIGHT side of the X1 X2 X3 X4 side.
Xfmr B has a SOLID dot in the same places-
is this dot telling us that there is a removable strap between the electrical neutral of the xfmr and the ground connection of the enclosure?
If so, the solid dot means there IS a removable strap, and the open dot/circle means there is NOT a strap?
Or do I have that backwards and/or completely wrong?


A-
http://www.temcoindustrialpower.com/products/Transformers_-_General/HT0025.html
wiring for A-
http://attachments.temcoindustrialpower.com/Wiring_diagram/transformer_wiring_HT_diagram31.pdf

B-
http://www.temcoindustrialpower.com/products/Transformers_-_General/HT4085.html
wiring for B-
http://attachments.temcoindustrialpower.com/Wiring_diagram/transformer_wiring_HT_diagram2.pdf
 

GoldDigger

Moderator
Staff member
Location
Placerville, CA, USA
Occupation
Retired PV System Designer
Sorry, I was talking about the M250s putting out 120V L-N and 208 L-L.

You have to use either an isolation or auto xfmr to go to 480/277V with the C250s according to page 12.
http://enphase.com/sites/default/files/C250_manual_EN_NA.pdf

It would be awesome if someone could clear this up for me.
Transformers A and B posted below.
Xfmr A has an "empty dot", otherwise known as a circle, on the left side of the H2-H5 side and the same dot on the RIGHT side of the X1 X2 X3 X4 side.
Xfmr B has a SOLID dot in the same places-
is this dot telling us that there is a removable strap between the electrical neutral of the xfmr and the ground connection of the enclosure?
If so, the solid dot means there IS a removable strap, and the open dot/circle means there is NOT a strap?
Or do I have that backwards and/or completely wrong?


A-
http://www.temcoindustrialpower.com/products/Transformers_-_General/HT0025.html
wiring for A-
http://attachments.temcoindustrialpower.com/Wiring_diagram/transformer_wiring_HT_diagram31.pdf

B-
http://www.temcoindustrialpower.com/products/Transformers_-_General/HT4085.html
wiring for B-
http://attachments.temcoindustrialpower.com/Wiring_diagram/transformer_wiring_HT_diagram2.pdf

1.
... completely wrong?
The dot is a polarity marking. It tells you which direction the current will flow in the secondary when DC flows into the winding with a dot on the primary side. Not particularly important unless you are paralleling transformers or using three phase.

2. The M250 will either provide 240V nominal line to line or 208V nominal line to line. It does not source current through the neutral at all in either configuration. The N terminal is just there to allow it to monitor the balance of the grid 120/240 so that it can shut down if there appears to be a grid problem (like lost neutral to the house.) I believe that this check is UL required now for inverters with L-L output into a 120/240V three wire single phase service.
 
1. The dot is a polarity marking. It tells you which direction the current will flow in the secondary when DC flows into the winding with a dot on the primary side. Not particularly important unless you are paralleling transformers or using three phase.

Ok, good to know, thanks. But why the solid dot as opposed to a circle-what's the difference there?
Nothing at all?
 

ggunn

PE (Electrical), NABCEP certified
Location
Austin, TX, USA
Occupation
Electrical Engineer - Photovoltaic Systems
Don't recall the tree. Can you sum up the fallacy you claim in a few words... so as to avoid the tree...?
Current sources don't need to raise voltage to push current. There is voltage drop between an inverter and the service, but the inverter doesn't need it to function; it's the effect of current flow, not the cause. An inverter would be perfectly happy feeding superconductors with zero resistance and therefore zero voltage rise.
 
Status
Not open for further replies.
Top