Grounded WYE from Inverters?

Status
Not open for further replies.
R

RichardM

Member
Hello, I'm working on a PV project and had a few questions that I can't seem to find any good information on. Maybe someone on here has the know how to help ;)

Do inverters (in terms of large PV), when utilizing a WYE out with no sub-component transformer, generally provide a grounded WYE or ungrounded WYE?

Secondly, (why) is it necessary to combine the ground from such a grounded WYE with the ground on the secondary side of a step-up transformer? Couldn't they simply be grounded individually in such a setup?

This site is great, I'm learning so much just by reading interesting queries and questions (I'm a student), hopefully someone can help me understand this.

Thanks already!
 
Last edited:
RichardM said:
Hello, I'm working on a PV project and had a few questions that I can't seem to find any good information on. Maybe someone on here has the know how to help ;)

Do inverters (in terms of large PV), ~~
~~(I'm a student), hopefully someone can help me understand this.
Click to expand...
How large is LARGE? Normal plain vanilla would be grounded in a small cup/cone - in the 5 gallon bucket - not sure.... But wye would suggest grounded. As delta would suggest not... Any other info to share? Spec's? Diagrams?
 
The 250kw inverters I installed had separate installation and maintenance manuals that explicitly warned not to bond XO on the isolation xfmr side.

Each inverter MFG is different, and the OEM manuals must be used to make these connections. In my case the inverter OEM voided all warranty and warned of catastrophic failure or explosion if these connections were wrong.

There were several other criteria that would void the Inverter warranty if not followed explicitly, including using qualified & trained installers.

One criteria that gave the engineers the most trouble was "No direct sunlight." Some of the inverters were exposed long enough in the sun to shut down, and covers had to be added.
 
e57 said:
How large is LARGE? Normal plain vanilla would be grounded in a small cup/cone - in the 5 gallon bucket - not sure.... But wye would suggest grounded. As delta would suggest not... Any other info to share? Spec's? Diagrams?
Click to expand...


The inverters are xantrex gt500's (so 500kW). The 500kW transformers are obviously grounded Y, maybe something got lost in translation when the question was asked..

So there is no "standard" way that these large inverters connect usually? I would have expected systems to be relatively similar for comparable inverters.
 
RichardM said:
So there is no "standard" way that these large inverters connect usually? I would have expected systems to be relatively similar for comparable inverters.
Click to expand...

No Way. You usually need detailed prints just for the pad mount, right down to the concrete compression strenth, much less torque values for lugs, listing for connector type, AL/CU, etc, etc..

The 250kw Xantrex was the most back assward thing I've ever seen. If you can't match model with downloaded Installation & Maintenance manuals from the Xantrex website, then get on the phone.
 
RichardM said:
The inverters are xantrex gt500's (so 500kW). The 500kW transformers are obviously grounded Y, maybe something got lost in translation when the question was asked..

So there is no "standard" way that these large inverters connect usually? I would have expected systems to be relatively similar for comparable inverters.
Click to expand...

Having seen a few alternative energy projects there is no such thing as a standard design. In fact there are no standard designs for any energy project. I see no reason the transformer could not be Delta. It depends on the design. There may be a "standard way" to connect a 500 KVA transformer but I doubt it. It will depend on your available space, grid integration scheme, Relaying, Equipment types you choose, and may other factors. My advice to you is to find an experienced Professional Engineer and an Experienced Electrical contractor who will help you out. As a consultant you may be either but if you don't know the answers call one of your buddies and take them to lunch it might be the best 40 bucks you ever spent.
 
ramsy said:
No Way. You usually need detailed prints just for the pad mount, right down to the concrete compression strenth, much less torque values for lugs, listing for connector type, AL/CU, etc, etc..

The 250kw Xantrex was the most back assward thing I've ever seen. If you can't match model with downloaded Installation & Maintenance manuals from the Xantrex website, then get on the phone.
Click to expand...

I think you are misunderstanding me, I am not talking about a COMPLETE standard setup.. just for a WYE connection.

Given that we want a WYE, and that we know the inverter will be providing WYE, is there any reason why it would be ungrounded WYE, and not grounded WYE? And Why ;)



MJJBEE said:
Having seen a few alternative energy projects there is no such thing as a standard design. In fact there are no standard designs for any energy project. I see no reason the transformer could not be Delta. It depends on the design. There may be a "standard way" to connect a 500 KVA transformer but I doubt it. It will depend on your available space, grid integration scheme, Relaying, Equipment types you choose, and may other factors. My advice to you is to find an experienced Professional Engineer and an Experienced Electrical contractor who will help you out. As a consultant you may be either but if you don't know the answers call one of your buddies and take them to lunch it might be the best 40 bucks you ever spent.
Click to expand...

That's not really part of my question, I am only referring to a WYE connection.. I am merely trying to understand the reasoning behind having a grounded WYE, or ungrounded (if that even exists from inverter output).

I have plenty of people that know what they are doing that will be checking my work to make sure it works out right in the end. But as of right now, I'm trying not to utilize their help and understand this stuff myself.

So again, any reason why an inverter output would be ungrounded WYE for such a large system? Wye, or wye not? :grin:
 
Alright, here's what I've learned now.

The confusion basically came from the customer as they didn't give a clear spec of the inverters. They were going to utilize new transformers from Xantrex that just came out that eliminated the isolation transformers.

So what happens then is that the step-up transformer will act as the isolation transformer, which means you will have to have a floating three phase between the inverter and transformer. The previous project I worked on was using two inverters per 1MVA transformer, and had isolation transformers on the inverters + switchgear.

So there we have it. If you have an inverter with a built in isolation transformer, then you have a grounded WYE/Delta, otherwise it has to be floating. This is obviously pretty cool, because you get that extra 1% of efficiency from removing the iso transformer, which is quite a big deal when we are talking about big power.
 
The output of the inverter is "premises wiring" and if the voltage of the inverter output falls within the rules in 250.20, then the output must be grounded. The manufacturers instructions cannot supersede a code rule.
 
Or 250.30.

The inverters I worked on did use a grounded conductor on the AC output, the manuals described it as a reference, which if required could automatically match a miss wired phase rotation with utility interactive services.
 
ramsy said:
No Way. You usually need detailed prints just for the pad mount, right down to the concrete compression strenth, much less torque values for lugs, listing for connector type, AL/CU, etc, etc..

~~~ If you can't match model with downloaded Installation & Maintenance manuals from the Xantrex website, then get on the phone.
Click to expand...

I'd say the same.... And that an ungrounded wye is a fickle animal and would not be a choice I would like to make short of advace by someone with a stamp on the job with a clear paper trail to CYA.
 
Is this inverter interconnected with the utility service? In other words, does the inverter supplement the utility power and both the service and the inverter can feed the load at the same time?

If this is the case, then there are some NEC issues at hand from Article 250.... parallel grounded conductor path and all that stuff. Is the inverter actually a seperately derived system? No common neutral connection to the service neutral?
 
crossman said:
Is this inverter interconnected with the utility service? In other words, does the inverter supplement the utility power and both the service and the inverter can feed the load at the same time?
Click to expand...

Yup, it is a net-metering setup.


If this is the case, then there are some NEC issues at hand from Article 250.... parallel grounded conductor path and all that stuff. Is the inverter actually a seperately derived system? No common neutral connection to the service neutral?
Click to expand...

Yes, separately derived. It HAS to be ungrounded WYE, it's tied in with the DC side ground on one end, and has the floating WYE to the stepup on the other. As long as the transformer is grounded with the necessary provisions (less than 25Ohm impedance etc.) and that is tied in with the utility, then there are no problems apparently.
 
RichardM said:
Yes, separately derived. It HAS to be ungrounded WYE, it's tied in with the DC side ground on one end, and has the floating WYE to the stepup on the other. As long as the transformer is grounded with the necessary provisions (less than 25Ohm impedance etc.) and that is tied in with the utility, then there are no problems apparently.
Click to expand...
What is the voltage at the "ungrounded wye"?
 
Don said:
What is the voltage at the "ungrounded wye"?
Click to expand...

This sounds similar to the inverters I saw, which use one iso xfmr between a DC input bus and AC output bus.

This isolation xfmr was factory connected w/floating neutral, not serviceable by non-factory personnel. Metering inverter L-N at AC output would show SDS voltage, dictated by the panel array design. Metering the floating factory L-N, at the isolation point, might show an unrefined wave form.
 
ramsy said:
don_resqcapt19 said:
RichardM said:
Yes, separately derived. It HAS to be ungrounded WYE, it's tied in with the DC side ground on one end, and has the floating WYE to the stepup on the other. As long as the transformer is grounded with the necessary provisions (less than 25Ohm impedance etc.) and that is tied in with the utility, then there are no problems apparently.
Click to expand...
What is the voltage at the "ungrounded wye"?
Click to expand...
This sounds similar to the inverters I saw, which use one iso xfmr between a DC input bus and AC output bus.

This isolation xfmr was factory connected w/floating neutral, not serviceable by non-factory personnel. Metering inverter L-N at AC output would show SDS voltage, dictated by the panel array design. Metering the floating factory L-N, at the isolation point, might show an unrefined wave form.
Click to expand...
Does anyone have an actual "process" and connection diagram?

From what I am gathering from the manufacturer's online literature, the 3-phase inverter is delta connected to a wye system... in which case I can see why RichardM is calling it an ungrounded wye. The literature available online is very vague.
 
A relevant note of caution to the OP, regarding damage to inverters.

Unsolved inverter or panel damage can halt further contracts from panel MFG's, who aren't clear about design vs installation errors. Equipment selection errors and change orders on either side can ultimately result in the MFG suspending new projects until the problems are resolved.

For example, after several successful installs my GC inherited an ongoing PV system, ground fault from a NECA contractor. The same fault value jumped between each 250kw inverter, but never occurred on both at the same time. After taking their electrical in house, me & the GC repaired smoked PV panels, smoked combiners, ground faults, and megger failures but this did not fix the persistent ground fault.

With further projects suspended, my contractor relied on their talent's credit for remaining material on existing projects, until much of that talent quit. Lots of NECA side workers from prior contractor wanted to help, but when I suggested they approve it with their local they never showed up. The problem was ultimately fixed by Xantrex factory technicians who repaired the inverters.

The PV system panel frames were bolted to the roof, perhaps to building steal, but inverter electronics are shielded from these transients including lightning. So, if the electronics are isolated from higher voltages & system transients, damaging inverter electronics probably requires shorting something across that factory isolation point.

So, shorting the floating factory L-N at the isolation xfmr point, even with a meter, may not be advisable if it can damage inverter electronics.
 
Last edited:
Smart $ said:
Does anyone have an actual "process" and connection diagram?

From what I am gathering from the manufacturer's online literature, the 3-phase inverter is delta connected to a wye system... in which case I can see why RichardM is calling it an ungrounded wye. The literature available online is very vague.
Click to expand...

MFG may not disclose internal design, beyond what is needed for proper installation, and i/o connections.

IMO disclosure of trade designs by publication or patent guarantees a complete lack of protection. Whether or not you believe China's economy rode in on pirated Microsoft software, as early as the 1980's the patent & copyright process was a smorgasbord for Asian entrepreneurs.

The fines were either completely avoidable or dwarfed by returns in local markets. Think of global entrepreneurs as corporate members of the World Trade Organization, with diplomatic immunity to common law in foreign nations.
 
ramsy said:
MFG may not disclose internal design, beyond what is needed for proper installation, and i/o connections.

...
Click to expand...
I wasn't asking for disclosure of proprietary information. A block diagram depicting the situation under discussion and system interconnection would suffice.
 
Status
Not open for further replies.
Top