Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations KootK on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Rooftop solar system fires 1

Status
Not open for further replies.

Just found this for solar fires in the UK.

Not many of them in the grand scale of things.

Interesting that isolators are the main culprit.

The Germans prefer to have a DC disconnect on the actual inverter. Hence I don't have them. There is no fusing in the UK to my knowledge on series strings. But I presume an inline locally install fuse box has the same risk as an isolator. I don't have either.

Connectors next on the list must admit I used sunclix because that's the make on kostal inverter. And the panels came with two mc4 connectors for every 10 panels one male and one female which where also sunclix (panels assembly done in Estonia by NAPs)

My inverter is mount on a concrete cellar wall.And there is a linked fire alarm down there. Many installations you see in solar groups they are in people's attics mounted on fibre board and battery sitting under them on wooden flooring.

But me doing that is more luck than judgement.

Wonder if some of these roof fires are linked to the old school firemen's isolators which are just a remote isolator.
 
This has been an interesting discussion. Two of my three kids have solar on their houses and we're thinking about it. However, when I saw the heading "Solar system fires", my first thought was, "Which planet is it this time?" [smile]

============
"Is it the only lesson of history that mankind is unteachable?"
--Winston S. Churchill
 
Yes, a very confusing title. Probably too late for the OP to put something useful there instead.

Keith Cress
kcress -
 
Yes "solar system" definitely brings to mind the sun with planets around it.

Add fires and I'm wondering if it's a solar storm or coronal mass ejection event. Some transformers got close to catching on fire during previous events in 1989. We are currently at a peak in the 11 year cycle of solar activity. The results of a solar storm just hit earth yesterday
On the original subject, my gut feel is workmanship / connections. I have been involved in infrared monitoring of electrical equipment for the past 20 years (at least on and off). The vast majority of electrical events (infrared findings, occasional smoke, and rare fires) reported at our power plant were related to connections. Combine that with the current solar environment where we have had high rate of installation over the past decade with demand for lots of new workers and the result is potential for a lot of installation workmanship events. Just my SWAG fwiw...



=====================================
(2B)+(2B)' ?
 
I see the issue, although most read my first post and understood. At any rate, I have asked management to change the title to "Rooftop solar system fires". Watch this space.
 
hokie66 (Structural) said:
I don't think Australia is on top of this issue, either. It is reported that about 10% of house fires in Australia are started by rooftop solar. But that could be because domestic installations are done by cowboys.

Most of the installations are done by cowboys unfortunately.
There were also different 'best practice' aspects that have since been reversed due to issues, such as use of polarised breakers and their subsequent withdrawal.

Having had to do utility mandated inspections on a few installations, there's plenty of opportunity for cowboy workmanship to creep in, and that's without the often seen incorrectly selected equipment that isn't suitable for duty, but still used as the installers don't know any better.

The interesting thing with solar panels is when a short occurs there's enough current flowing to get things really hot but not enough to melt conductors, blow clear or trip breaker magnetic elements. This results in the short sitting there for ages until the material around it gets hot enough to burn.

EDMS Australia
 
I must admit as a mechie my gut feeling is to reduce the number of connections in a systems because its doesn't matter if we are talking electrical, hydraulic, pneumatic they are in the main the failure points. Especially if they involved onsite fitting.

Its one of the reasons why I don't like and would avoid any of these modules that sit under the panels be it optimisers or rapid shut down isolators. You double up on the amount of connectors and they will be dissimilar makes between module and panel.

The cowboy stuff goes beyond electrical.

A workmate had a system fitted 2 months ago and ask me to go look at it. he has panel corners sticking high of the roof ridge line, And the wires under the panels are spaghetti with some of them dangling and touching the roof. Others are not touching but the connectors are the lowest point of the loop. Which is just asking for roof problems and water ingress into the connectors. The corners proud of the ridge line he can do nothing about and apparently is local permitted. but the big boss came to see the wires dangling and was extremely upset. The workers turned up on a Sunday unpaid to fix them.

Solar groups in UK are showing pictures of 100mm racking bracket screws which should be through into the trusses just sticking through the moisture board it was lucky the owner spotted the bumps in the insulation. court case now in progress for a new roof. MSC which is meant to be the trade oversight of standards and quality have washed their hands of it and say its a civil matter. Quite what you get for paying 15% more for a certified MSC installer I can't really work out.



 
From Tugboat's link to the IEEE article:

PV_Installation_Fires_xtu3k5.png


Are "lack of drainage" and "DC arc in combiner box" not installation errors?
The other factors could also be exacerbated by poor workmanship during installation.
 
Combiner boxes are notoriously found to be totally crappy quality in many cases. You can end up with a lot of amps running thru them at the hottest part of the day. I can see a lot of them stressing out and smoking.

Keith Cress
kcress -
 
It would be good to show which particular country you are describing.

That which is done in one country may not be done in another.

For example: whatever a "combiner box" is, in the US they should not be a problem due to "a lot of amps running thru them".

If y'all want to talk about crap installations in far away places (which you may well inhabit), please note where those far away places ARE.


spalso
 
We don't have combiner boxes here just straight strings to inverter no fuses and no rapid shutdown.

They are the place where multiple strings come through, that's where the fuses/breakers are located, Surge protection for the strings, string monitoring if it's not done and the rapid shut down isolators if required if it's in the attic and also the earthing bus bar and grounding protection stuff. Some also do the paralleling of strings up as well. If there is DC charging of batteries involved there is a whole load of other stuff in there as well.

They are used a lot in the USA due to a code regulation you have. Before about 10 years ago USA mainly used inverters with transformers and required a separate ground spike to the solar which was bonded across to the main house spike. That then went to the combiner box to be used for the surge protection, array grounding and to the inverter DC side.

The transformer less inverters get the ground from the AC side and as such don't require a dedicated earth spike of their own for the DC side. But the fusing, surge protection, Dc breakers, firemans isolator etc are still being put into combiner boxes. They developed over the years into quite an industry and heavily sold as major safety required items. Each box has colossal amounts of connection points between various components handling some very high currents and voltages. And they like to get them as small as possible which is one of the selling points.

The old transformer inverter setups with battery with all the trunking etc you could easily fill up 2m2 of wall with all the bits and bobs that are now included in the purpose made combiner boxes which come in at under 0.5m2.

German codes require all that external stuff to be included inside the transformer less inverter. Europe has never really used transformer isolating inverters. Plus, it's a different setup because we have 230V/400V 1/3 phases and USA has your split phase setup.

The above is what I know on the subject of combiner boxes and I will admit it may be incomplete or wrong in some parts especially surrounding the earth required onto isolating transformer setups.

Some solar setups are works of art with the wiring and various bits and pieces, You look at mine with 5 cables coming down a flexitube from the ceiling into the inverter and it feels very like you have turned up to a gun fight with a potato gun when everyone else has military assault rifles. The combiners can be 500$ plus. Some say but you need to be able to isolate the strings. Only time I have ever disconnected mine after installation in 4 years, was to boot up my second inverter and register it with the grid for the new barn. And due to that being delayed building it and paperwork deadlines, I did a Boeing and borrowed a string of the first inverter to power it up. Connected the AC to the house and internet registered it and then left it for 3 days until I got the confirmation from the grid that paperwork was now complete and then put it back the way it was and the new inverter under my bed.

 
Am I right about what combiner boxes are itsmoked?

I am talking about an ex soviet state in the EU, with knowledge of how they do it in Germany, UK and a little bit about France mainly due to mates that have systems installed in those countries.

In fact, I should be getting sales bonuses of Kostal and BYD for the number of systems installed I have been involved with. And on a hit list for Enphase and Solar Edge.
 
"spsalso; Not a problem!

Chinese crap combiner boxes installed in the USA."

Thanks. And what is the name on those "combiner boxes"? And the model number?

While the UL is not high on my list of integrity-first organizations, I do imagine that every "combiner box" would have to have UL approval to be installed in a solar system.

I can imagine that if an installer installed a non-UL approved item, and it started a fire, the insurance company involved would suddenly divorce themselves from the need to pay out to their insured.


spsalso
 
We get lots of issues with Hoffman brand enclosures on our boats. They have a stainless steel unit that uses square section o-rings integrated into the heads of the screws that hold the cover on. They use buna-n rubber for the o-rings. After about 5 years the o-rings have multiple cracks and the enclosure starts admitting water from spray. I have had to go through my fleet and replace all o-rings with silicone and have also had to replace the gasket with EPTFE tape. Hoffman is a premier brand, we don't have any better choices.
 
And as someone has said before most of the solar stuff is in plastic enclosures these days.

And there is a huge difference in temperature range specification. The 5$ each MC4 connectors are -40 to +85 C the 1.50$ are 0 to 60 degs C.

Not that most consumers would ever look at a tech sheets.
 
Maybe a photovoltaic system installed in our solar system.

An insurance company that had to pay because of a fire from a non-rated box, would likely go after the inspected, installer, and maybe even the finance company. Substandard crap is not allowed, according to code.
The homeowner would not know this, and a licenced installer/electrician/inspector would.

Hoffman boxes are a premium brand, and are not the cheapest, but it might be salt water that you are fighting. The steel boxes work well in the mostly dry country.
 
Sparweb mentioned drainage. How do these combiner boxes drain?
 
They don't from what I have seen they are basically fancy consumer units with various ratings which as tug says are not lifetime guaranteed. They seem to like bling fancy photo's of complicated fancy electrical gear with vast marketing verbose nonsense about implied safety benefits with very little actual meat of case requirements for having the features. So its lots of DIN modules with heavy duty wires going between various points. They look impressive I will admit. But the engineering case for having a lot of the stuff is nice in theory but just the addition of 30-40 connections running at high voltage and in the US case high currents seems to create more incidents than they protect against.
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor