[RE-wrenches] Inverter Failure Due to Generator Backfeed (?)

[Jason Szumlanski]

Mac,

That's what I was thinking, too. The inverters should be resilient to AC
input variations.

I learned something else... the inverters had lightning arrestors installed
and connected to DC inputs that still had dummy fuses in them. Later when
they were discovered by someone else totally fried (the dummy fuses), they
were swapped for proper fuses and then the inverters worked for another
year. I suspect that ALL of the DC and AC dummy fuses were probably left in
by the original installer, then swapped out when the lightning arrestor
damage was discovered. The plot thickens.

To simplify I said Eaton, but it is actually an Eaton that was privately
labeled as an ASP (now defunct). I tried to get help from Eaton in this
scenario before and was unsuccessful. They didn't have parts or spares, and
didn't seem inclined to honor the warranty for the ASP branded product.

​It's a long story - part of the fraudulently UL labeled module fiasco from
a couple of years ago...

Anyway, the customer agreed to buy some SMA TL inverters to put the saga
behind him. Good move...

Jason​


On Sat, Jan 30, 2016 at 8:35 AM, Mac Lewis <maclewis1 at gmail.com> wrote:

> Hi Jason,
>
> I really don't think the generator is at fault.  I think of these grid
> tied inverters as current sources, not voltage sources.  If they don't
> "like" the voltage at their AC input terminals, its their job to disconnect
> from that voltage source and reconnect when the voltage is within their
> specs.  It's certainly not impossible and I would recommend having a
> contactor remove the PV input when the transfer switch is in generator
> mode.  I'm sure if you tell Eaton that there was a generator connection,
> they would be much less inclined to honor any warranty that there may be.
>
> One caveat to this is if there are lots of non-linear loads and the
> generator wasn't sized to account for this you might see voltage distortion
> that may cause problems.  I'm not sure how an inverter decides if a voltage
> source is worthy of connecting to or not, so wave-form distortion may not
> be picked up, and I'm not sure if there is any input filtering.  Here is a
> very good and succinct article on the effect of non-linear loads on
> generators:
> https://www.idc-online.com/technical_references/pdfs/electrical_engineering/Generators_and_nonlinear_loads.pdf
>
> On another note, if there are a lot of inductive loads, activation of a
> transfer switch could cause a voltage spike in the system.
>
> Good Luck
>
> On Fri, Jan 29, 2016 at 9:37 PM, Jay <jay.peltz at gmail.com> wrote:
>
>> Hi jerry
>>
>> Can you give any insight as to the failure mechanism?
>>
>> I guess I don't understand why bad AC would damage the inverter, I mean
>> besides the obvious bad equipment design.
>>
>> Jay
>> Peltz power.
>>
>> On Jan 29, 2016, at 3:18 PM, Jerry Shafer <jerrysgarage01 at gmail.com>
>> wrote:
>>
>> We have seen something very close, it appears that the gwnny will always
>> win this war.
>> Jerry
>> On Jan 29, 2016 1:14 PM, "Jason Szumlanski" <
>> jason at floridasolardesigngroup.com> wrote:
>>
>>> I'm looking at two Eaton PV250 inverters that failed simultaneously as
>>> far as we know (not my install). Everything checked out normal, but no
>>> display and no power output. I suspect both boards are bad. Upon digging
>>> into the scenario I discovered that the owner had a whole-house generator
>>> transfer switch installed a while back, after the PV was installed. The PV
>>> is a load side interconnection, so calamity should have been expected.
>>>
>>> What are the odds that the generator caused the inverter boards to fail?
>>> I was always under the impression that it was more likely for a generator
>>> component to be damaged than an inverter component in this scenario. It
>>> could be coincidence and the inverters may have just failed around the same.
>>>
>>> Note:  ~10kW PV, 25kW Generator
>>>
>>> Side note: The PV was interconnected at a 50A breaker on a 200A bus bar
>>> in the main distribution panel and somehow passed inspection by an AHJ that
>>> shall not be named... Over 120%, and also not the correct OCPD rating
>>> anyway for these two inverters. Passed plan review and inspection...
>>> Contractor out of business... surprise!
>>>
>>>
>>> Jason Szumlanski
>>> Florida Solar Design Group
>>>
>>
>
> --
>
>
>
> Mac Lewis
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.re-wrenches.org/pipermail/re-wrenches-re-wrenches.org/attachments/20160130/d529206e/attachment-0003.html>