Critique of my friends 18KW solar install

micahv

Howdy,

So I have dabbled in Solar installs off and on over the years and am by no means an expert. Anyhow, when I showed up at my friends farm, he had recently gotten a 18KW ground mount install. When I looked at his inverter display I was curious as to why it had such low power output for full sun in a Texas summer. I grabbed my clamp meter and found a dead string, my friend walked by and tapped a wire which sparked and faulted the inverter. So a 500 volt exposed conductor blowing in the wind...  This was an exposed wire in which the MC4 connector had melted. I assumed this may have been due to poor crimping causing excessive heat. He said this happened once before, he managed to find one of the installed that no longer worked for this company who came out and fixed it.

I also noticed the array had no grounding that I could find, and cable management was quite poor. Many of the cables appear to be stressed with sharp bends and potential to rub against surfaces. This is in rural Texas, so I figure code compliance is more on the honor system. He did say the local electric cooperative came up before it was connected to the grid, but I suspect the mostly just looked at the grid tie by the meters.  

Another thing that stood out for me with the AC run back to the meter was a utility pole was installed and the wire was run such that is cast a shadow on the array mid afternoon. Also they didn't use weather heads just some sealant at the top of the pipe where it went up and then back down to the meter. Just a lot of red flags in my unprofessional opinion.

This system is:

4 strings of 15 Trina 310DD05H.05(II) - Pmax: 310, Vmp32.9, Imp 9.43, Voc 40.2, Isc 9.83

2 strings going going into each of the 2 Fronius Quotro disconnects which tie into a Fronius Primo 15.0-1

My thought was I should redo all the field crimped MC4s, with a proper crimper,  clean up the wiring and add array grounding. 

What would folks recommendation be to ground this system? I assume I am not missing anything with why this array has no equipment ground? I believe this is an iron ridge rack? It may be that the mounting hardware is bonding the modules to the rack? If that is the case I assume the rack should still be bonded together and a ground rod installed?

Cable management? a variety of cable clips and smooth the bends of the cables. Should I worry about the lack of weather heads in the AC home run? 

I also wondered about if this array is a little large for the inverter?

I'm not currently at my friends farm, but will be in about a week so would like to get a jump on sourcing some hardware before I get there.

Thanks for any suggestions.

 

Anyhow, the gist of the system configuration is

BB.

Hmm... If he paid for this to be proffessionaly installed (in the last year or so), he should be calling up the installer. There have been lots of fires around the country started by "arc faults" (poor crimps/connections--That start arcing as the connection fails).

Would be sending them articles about Walmarts/Amazon/etc. buildings that have caught fire (in you friend's case--Grass land fire in dry weather?):

https://www.instituteforenergyresearch.org/renewable/solar/walmart-and-amazon-sue-tesla-solar-panels-blamed-for-fires/
https://www.businessinsider.com/tesla-solar-panel-fires-walmart-amazon-safety-regulator-investigation-foia-2021-3?op=1
https://duckduckgo.com/?q=solar+array+fires+at+walmart+amazon&ia=web

Also--Need to check that the "exposed to sun" wiring is rated for UV exposure...

All wiring that leaves a "box" or enters a "box/home/etc." need to have proper sealing and drip loops. On two of my homes over the years, I have had poor or no drip loop to the weather head to these homes (not related to solar power)--And over the decades a few drips of water manage to follow the cable into the drop (past the weather head) and down into the meter socket in one home, and into the AC panel in another) and the corrosion was just enough to cause a mess (corrosion, heat from connector resistance, "blinking lights in home").

I understand wanting to help your friend--And long term, that may be the only workable answer--But the original installer left a dangerous mess and you and your friend don't want to get hands on and "dilute/spread out liability" at this point.

Example of an installer (good faith?) that bought non--Listed (non-UL/NRTL) panels that caught fire on a roof (fortunately, fire department was able to stop the fire before roof fire became a house fire):

https://forum.solar-electric.com/discussion/3375/panel-fire-question/p1

Improper solar array wiring is a huge nightmare out there... And Arc Faults are difficult to detect/prevent. Normal circuit breakers don't work. There is for Arc Fault detection/shutdown hardware available--Don't know what your friend has.

Regarding grounding... I guess metal pipe in concrete filled post holes? Don't know what is in the ground (earth bonded posts through to concrete/earth ground--UFER):

https://www.ecmag.com/section/codes-standards/what-ufer-ground-concrete-encased-grounding-electrodes
https://en.wikipedia.org/wiki/Ufer_ground

If there is lightning in the region... More grounding does not hurt. May also want to look at surge protectors:

https://www.solar-electric.com/residential/circuit-protection/surge-protection.html

Would need to read the Fronius inverter installation manual to figure out what other grounding would be needed. Can't always assume that grounding the Array Negative lead to a ground rod is the correct thing to do... With modern "transformerless" GT inverters, both legs of + and - are "hot" with respect to ground (as I understand).

-Bill 

Dave Angelini

Get us another couple pictures ! 

 Off the top, in NA, each panel needs a #10 minimum ground wire, no matter what the topology of the inverter is. If the rails were designed to provide a ground, there would need to be a ground wire on each rail. Rail to pipe is not code for grounding.

The array looks armature and non code compliant.  As Bill said, try and get to the installer. 

Even if the pipe is done right underground, it can't be the only source of grounding for code in NA.

This array would not be standing near Fort Myers Florida today. If it is Iron Ridge, try and get a model number to look it up.

micahv

Thanks for the info Y'all.  He has tried to get to the installer but they have not been very responsive and a quick look at the Better Business Bureau shows this is not surprising.  I'm not sure if it is worth the effort to get the installer who clearly doesn't know the intricacies of doing a solar install. I suspect they will fade into non-existence if it hasn't already happened.

So upon return to the farm and further analysis I have discovered these installers basically fumbled about everything possible.

So I will just start with the rack for this post.

The most glaring and difficult to resolve issue is the use on 2" conduit instead of schedule 40 pipe or mechanical tubing. I question also if the piers were dug deep enough as well, as one of them is heaving up pushing the bottom of one row about an inch above the other row.

The array is about at 20 deg, which assuming they used schedule 40 pipe would require 18 piers according the the design tool which is the number installed.

I did confirm with Iron ridge we do indeed need to add a 10# equipment ground. I believe I just need to bond the rack at one point as all the modules and rack interconnect bond everything together.

Anyhow, I did clean up the wiring the best I could, add drip loops to where the home runs enter the disconnects at the inverter. In hind-site I probably would have gotten the cable ties from Iron Ridge instead of zip ties to supplement the heyco cable clips but it is much improved from where it was. I know ideally I would have used a cable tray or something for the home runs but I just used a lot of zip ties for now. This has been a learning experience for me as far as the art of cable management.

I also don't think anything was torqued to spec, and I suspect there isn't a spec since conduit is used.  I did torque the UFO brackets when I removed a few rows of panels to rotate 180deg to get better cable runs.

I don't think he would be willing to pursue any action against the installers it is probably not worth the effort nor likely to end favorably.   I don't think we want to go through the effort of rebuilding the rack which I estimate to cost $2000 for all the schedule 40 pipe and cement as we would need to dig new holes. This was just a quick estimate I did and was kind of surprised how much that pipe costs. 

Some of the module brackets are a little close to the end of the rails, which would require removing all the panels and moving the modules up a little.  I know ideally this entire rack needs to be taken apart and replaced. There are no animals and very few people are out here otherwise fencing it off would also be required or concealing the cables better? 

Anyhow, any other suggestions on the racking sh*tshow?

Thanks,

I'll follow up more with the electrical issues I found concerning.

Dave Angelini

I think you have the right attitude. This is a mostly Home Depot array without the engineering.

As long as it does not start a fire or electrocute a bus full of school kids,  let it be.

You might want to add some lateral braces or X the back poles. Below is an array in the Caribbean 10 years back.
It could have the back X braces but since it is all Sched 40 galvanized, and not as big as your friends, it has passed a few big tests with hurricanes.

Also notice the pipe is continuous thru the tee's.

Graham Parkinson

Yikers - at a minimum, Good that Dave noticed that the conduit doesn't run through the tees.  A classic design error, similar to what caused the tragic failure of that hotel lobby floor years ago.

I'd add some tie plates to hold those conduit sections together at at the tee's.  You might split some appropriate sized pipe that will fit over the conduit with a zip disk and clamp both halves over the joints above the tees and then drill and through bolt to connect the conduit through the gap over the tees.  Otherwise, any kind of uplift from wind on the high side will pull those joints apart.   

I'd also get busy and add more hold down piping on the high side too, or add hold down wires, turnbuckles connected to deadman anchors in the ground.

At least that "Solar Contractor" installed some useful "temporary scaffolding" that will hold the panels until your friend gets them secured.