New solar install, wrong microinverters used

jmdenver said:

Is 4:1 a "proper" Hoymiles design? I mean I can live with it I'm just very concerned about future headaches with warranty/insurance/POCO and aggravated that they willy nilly changed things up without saying "boo" about it and then just make up excuses (maybe they are legit, aside from the glaring fact it saved them money) and take no ownership/responsibility for it.

Click to expand...

They sell 4:1, it has UL1741 certifications, it was also properly submitted to all the extra relevant organizations in my highly regulated state, including the 1741SA exemption and all that.

When I say proper I mean in the sense of, 1:1 with Hoymiles is pretty damn silly, most of these non-Enphase companies get their fundamental strengths from simplifying the hardware with N:1 inverters. Why would I pay the same with Hoymiles and take a higher, i dunno, geopolitical risk from a future trade war cutting off warranty replacements? It'll take bigger problems in the world for Enphase to stop supporting their inverters.

That's independent of how the installer conducted business. I can only speak to how I analyzed 4:1 vs 1:1 installations when I picked my hardware.

You probably have bigger warranty problems like your installer disappearing, which is a high probability event anyway (I assume cut-rate installers in California have a limited half-life and can't be trusted to provide support for 25 years. OTOH, do you really want to pay +50-75% for the ones that have been around 20-30 years, vs keeping the cash in hand and investing it). And at that point you hope Hoymiles has better domestic support network. This is the devil's bargain with buying less common brands (which I was OK with personally and explicitly documented in my own project planning records, you can search my post history here for it about a year ago. There are worse lesser brands than Hoymiles, they are better capitalized than NEP AFAICT, but both have high geopolitical exposure).

jmdenver said:

I definitely want to avoid any more screwing around on my roof. I got a new roof solely for the purpose of getting solar (the old one probably had 5-10 years left on it) and certainly don't want to have more holes drilled / filled / then re-drilled.

Click to expand...

Yeah that part of the situation sucks too. You are right to feel upset/not sure what to do here...

So what is the delta to the IQ8A bids?

If you can find other of their customers that have the same issue, you may have a better case. Not sure how you would do that.

my concern is with RSD. Are these hoymiles units providing Rapid Shut Down? Does the 4x1 configuration meet the NEC code year requirements that the system was installed under?

skyking1 said:

my concern is with RSD. Are these hoymiles units providing Rapid Shut Down? Does the 4x1 configuration meet the NEC code year requirements that the system was installed under?

Click to expand...

Yes, why wouldn’t they?

I installed last year

Of course they won’t comply if you do something weird like I was considering, which was pulling long DC cable through the attic to panels not within the sub array.

zanydroid said:

Yes, why wouldn’t they?

I installed last year

Of course they won’t comply if you do something weird like I was considering, which was pulling long DC cable through the attic to panels not within the sub array.

Click to expand...

4:1 under a continuous array is by my reading of NEC identical to 1:1 wrt the controlled conductor voltages and physical protection rules (IE, since conductors are within the array you get to have 60V after RSD. The input voltage limit is 60v. Going by memory with the numbers)

Just asking the most important question for me. That's significant when the installer makes a change that could put you crossways with code.
The 1 to 1 micros, there is no question. The power does not get past the edge of each module.

skyking1 said:

Just asking the most important question for me. That's significant when the installer makes a change that could put you crossways with code.
The 1 to 1 micros, there is no question. The power does not get past the edge of each module.

Click to expand...

The 2:1 and 4:1 modules do not do any series connections, so the wire voltage is the same as the panel voltage. The 4:1 are tricky in that they parallel together two inputs, and I’m not sure if they disconnect these on shutdown. But it should be fine for voltage to carry across the 2P provided the voltage of one panel is under 60V

There are some other brands that do series wiring, and in that case your question is more applicable. I’m not sure how they handle series wiring that go over 60v, easiest would be to direct connect the panels to the module and then break the series connection on shutdown rather than have the panels be connected in series directly.

HM-1500 has 2 mppt. If you got individual modules, then you would have 4 mppt for 4 panels. Don't know if that would make a production difference if one panel is partially shaded.

zanydroid said:

When I say proper I mean in the sense of, 1:1 with Hoymiles is pretty damn silly, most of these non-Enphase companies get their fundamental strengths from simplifying the hardware with N:1 inverters. Why would I pay the same with Hoymiles and take a higher, i dunno, geopolitical risk from a future trade war cutting off warranty replacements? It'll take bigger problems in the world for Enphase to stop supporting their inverters.

That's independent of how the installer conducted business. I can only speak to how I analyzed 4:1 vs 1:1 installations when I picked my hardware.

Click to expand...

With the exception of a basic understanding of risk dealing with a Chinese company going forward most of the rest of that is Greek to me. 4:1 may have been just as good/better in some respects - had they quoted them and the permit/POCO approvaI been based on them I wouldn't have known any different/cared.

zanydroid said:

You probably have bigger warranty problems like your installer disappearing, which is a high probability event anyway (I assume cut-rate installers in California have a limited half-life and can't be trusted to provide support for 25 years. OTOH, do you really want to pay +50-75% for the ones that have been around 20-30 years, vs keeping the cash in hand and investing it). And at that point you hope Hoymiles has better domestic support network. This is the devil's bargain with buying less common brands (which I was OK with personally and explicitly documented in my own project planning records, you can search my post history here for it about a year ago. There are worse lesser brands than Hoymiles, they are better capitalized than NEP AFAICT, but both have high geopolitical exposure).

Click to expand...

That was my main concern with warranty... having no paperwork reflecting what they installed I imagine Hoymiles would be a nightmare dealing with if I had to go to a different installer. (not to say they won't be a nightmare regardless) This company has been around for 7 years (not 30 but not 6 months lol) at least for their roofing. Not quite as clear on the solar division.

zanydroid said:

Yeah that part of the situation sucks too. You are right to feel upset/not sure what to do here...

So what is the delta to the IQ8A bids?

Click to expand...

Not sure if talking about specific prices is ok here so I'll steer clear. Enphase (they never quoted that option but based on their reply Thursday it would have added maybe 15%. (to the TOTAL price, not just inverters) I probably would have gone with that had they offered it but... As is, with the tax credit this really isn't a terribly substantial investment- all relative of course, but fairly low 5 figure.

Sorry for burying the lede, but... I went on a bit of a rampage yesterday as it seemed they had no intention of doing anything. If all it would take was a bad Google review to get a response within an hour promising to install 16 more 350s (already had one, they would just take out the 1500s and replace with the 350s) I would have done that a few days ago and foregone the BBB, FTC, Attorney General's office and a half dozen trade organizations they are part of. Sorry, but not really, since had they said "sorry, we messed up, we had good intentions but understand your concerns and will make it right" to start with it would have been a non-issue. I was able to call off the dogs in most cases but I guess a BBB complaint can't be removed / but pretty simple to have them say it is resolved and I concur.

Anyway, they were out today and replaced it all. They gave me a slightly different excuse this time, blaming their software, but I'm happy. I assume there will be no issue with the final county inspection and POCO approval.

All the new 350s are showing up in the app, one of the guys out to install said it might be tomorrow for their office to "send in the serial numbers" but apparently not, and then I might have to flip the switch at the AC disco to register them. They ARE showing a different "alarm" message than the original 350 and 1500s did (after they were powered on for a bit last Wednesday) : Alarm Status : Grid undervoltage, Grid underfrequency, no voltage on grid point". (the original 350 simply said, and still does, "grid disconnected") So I have an email out to them to see if flipping the AC disco on, let it sit for a bit, then back off will resolve that. I understand "grid disconnected" until the POCO allows it to go live / aside from a brief test, but the other warnings are a little freaky.

So, hopefully this is all resolved... bit more time should tell.

Thank EVERYONE for their insights and thoughts!

Excellemt outcome!! As the customer, you were right to insist on getting what was designed and permitted. Thanks for reporting back.

Let me pick your brains about one more thing before I get back with the installer letting them know everything now reports online. (they told me to flip on the AC disco for 1-15 minutes, took a while but they all eventually did, then turned it back off)

This is the map of energy produced, as you see there is significant variance between panels but three on the South facing roof are ~50% lower than the other ten and two on the East side are about 25% below the other two.

I don't want to obsess about that based on a few minutes of data (as I recall it looked kind of similar when they tested that original inverter configuration back on the 10th, but the only panel I remember specifically was very low was the bottom left one)

Anything obvious that pops out to explain? Misalignment, something with the wiring, etc) There is no shade (I regret that in the summer lol) with the possible exception of a bit cast by the chimney at the top left of the Southern facing roof late in the day. Amazing how much dust/pollen/bird crap I got in 5 days! Some scuff mark looking things on the two bottom right panels - may well just be areas where the dust was wiped off when they removed/reinstalled them. Supposed to rain pretty heavy later so I'll take a look again tomorrow. Might have to fire up the drone also, haha. Thanks again all!

There are 3 panels the chimney could have shaded. To shade further panels, it would have had to shade a closer one.

Do trees ever cast a shadow? Has to affect upper row before lower.

There seem to be two panels with absolutely no excuse for being at 50% or less. And one other around 70%.

The 4 panels on right of diagram - how about a picture of where they are on the roof? I'm guessing back side but can't be sure.

new install with all new equipment. It is not a good thing, but you can get bad gear out of the box. Do not obsess about it. Gather data for a while and then inform the installer of the discrepancies.

Hedges said:

There are 3 panels the chimney could have shaded. To shade further panels, it would have had to shade a closer one.

Do trees ever cast a shadow? Has to affect upper row before lower.

Click to expand...

The closest trees on the front/south side of the are across the street .. I've never noticed any shading beyond perhaps the first 5 or 6 feet of my driveway. (note the highlighted tree looking thing at the front/right corner of the house is just a flowerbed - well, weed bed lol) so it produces no shading on the roof.

This is a better birds-eye.. as you see on the right/east side (where the four panels are) there is noting close enough / or positioned in a way to ever result in shade. The only shade that might impact me would be on the left/west side which may to some extent impact the front/south panels late in the day (same as the chimney) but that's not a big concern.

Hedges said:

There seem to be two panels with absolutely no excuse for being at 50% or less. And one other around 70%.

The 4 panels on right of diagram - how about a picture of where they are on the roof? I'm guessing back side but can't be sure.

Click to expand...

I don't have a birds-eye photo but the schematic above shows the front/south and right/east configuration.

I doubt this is apropos of anything related to the generation variations but it seems they are set up in two "arrays" and two "strings"- just thought I'd toss that out there.

I didnt see it mentioned here, but my math gets you are also losing 400W of production as well...

Supervstech said:

I didnt see it mentioned here, but my math gets you are also losing 400W of production as well...

Click to expand...

That's based on the inverter swap between the 1500s and 350s, yes? That's fine by me just to avoid future hassles having a configuration that doesn't match the permit/POCO approval, etc. And that is DC not AC - based on my admittedly rudimentary understanding of the difference that would only be a factor on the (relatively few? over the course of a full year) days where we have ideal weather conditions and the panels are producing 375w as opposed to 350? (they technically are 410w panels - assuming they installed the right ones of those, lol)

"String" has little meaning for microinverters. They are just in parallel, can't exceed wattage allowed for the wire and breaker. (Unlike for series string inverters, where all panels need same orientation for same current.)

The four panels on different roof face are similar to each other, lower than the many on main face. That is fine.

Four panels on the main face are lower than they should be. The lower left one is OK if chimney cast a shadow on it, otherwise not OK.

Vendor needs to address this. Either faulty connection (which could be overheated already) or faulty equipment.
Assuming the panels do get good sun, as they appear to.

Confirm with multiple readings. Don't know if they occasionally do an MPPT sweep. Graphs over time for some brands show brief dips.

Hedges said:

"String" as little meaning for microinverters. They are just in parallel, can't exceed wattage allowed for the wire and breaker. (Unlike for series string inverters, where all panels need same orientation for same current.)

The four panels on different roof face are similar to each other, lower than the many on main face. That is fine.

Four panels on the main face are lower than they should be. The lower left one is OK if chimney cast a shadow on it, otherwise not OK.

Vendor needs to address this. Either faulty connection (which could be overheated already) or faulty equipment.
Assuming the panels do get good sun, as they appear to.

Confirm with multiple readings. Don't know if they occasionally do an MPPT sweep. Graphs over time for some brands show brief dips.

Click to expand...

I pointed those out to the installer, he agrees they look off. (well he specifically mentioned the three most egregious) I told him I wasn't basing anything on a 15 minute test (or two) but wanted to be clear I'd be paying close attention once the system goes live.

He said they would keep an eye on them as well (I believe they can remotely monitor them the same way I can) and will swap them if they "are not producing as they should out of the gate".

Presumably those would be a warranty claim on their behalf... assuming the panels are defective and they didn't mess up the wiring/connection. 4 out of 17 seems to be a high % of failures right out of the gate though.

Time being it's just a waiting game for POCO to switch them on. Since they got right on the inverter swap (admittedly only after bad reviews) I'll give them the benefit of doubt for the moment.