Solar panel fires!

[J.P.M.]

Well, apparently that is penny wise and pound foolish. If 90% of electrical failures are caused by improper connections, those industries that do not provide the proper tools are just foolish and some would say dangerous.

The NEC and manufacturers provide the torque specs and it would be wise to comply.

Depending on frequency and gravity of the failures, if I was looking at a spot where a large %age of those failures seemed to be concentrated, I'd take a long, global look at it and see if I could figure out why.

[sensij]

Depending on frequency and gravity of the failures, if I was looking at a spot where a large %age of those failures seemed to be concentrated, I'd take a long, global look at it and see if I could figure out why.

From a "global" perspective, this is why I design my machines using components with spring terminals as much as possible, and require ferrules on stranded conductors. It isn't perfect, but at least you don't have to trust someone to use a torque wrench correctly.

[J.P.M.]

From a "global" perspective, this is why I design my machines using components with spring terminals as much as possible, and require ferrules on stranded conductors. It isn't perfect, but at least you don't have to trust someone to use a torque wrench correctly.

Understood. I was thinking more of any system where the failures were concentrated in a very lopsided and possibly identifiable way. Sort of like forensic system engineering.

[DanS26]

Most all Electricians are self employed contractors. Your typical Solar Contractor is not even an Electrical Contractor or knows much about electrical. It is a General Contractor, Home Improvement, Roofing Contractor, and some EC's milking the market with the free money as a side job until the bubble bust next year. They in turn hire roofers and an EC to do the work on Sub Contract basis. Solar City is a Prime Example.

High voltage connections on the roof and improperly torqued connections. Sounds like a fire to me!

So what does the NEC do....require arc fault detection. OK.

But the real problem is untrained installers not using the correct techniques.

[DanS26]

From a "global" perspective, this is why I design my machines using components with spring terminals as much as possible, and require ferrules on stranded conductors. It isn't perfect, but at least you don't have to trust someone to use a torque wrench correctly.

sensij.....yes it's in the design....if electrical manufacturers built in the proper torque it could be significant.

If the connections "clicked" like your locking gas cap at the right torque, it would be a vast improvement.

[Sunking]

But the real problem is untrained installers not using the correct techniques.

Well if you have been around sola rcrews are like landscapers, No Comprendo.

[inetdog]

Well if you have been around sola rcrews are like landscapers, No Comprendo.

"Green side up!"

[SunEagle]

Well, apparently that is penny wise and pound foolish. If 90% of electrical failures are caused by improper connections, those industries that do not provide the proper tools are just foolish and some would say dangerous.

The NEC and manufacturers provide the torque specs and it would be wise to comply.

I agree that some companies do not think about getting the correct tool for it's maintenance people or provide them enough funding to get their own "specialty" tools.

That was one of the reasons I pushed for a Thermal IR camera and developed the entire PM program around using it to find potential issues including poor terminations. We spent about $10,000 on equipment and training to set up the program we probably saved 10 times that in the first year by finding issues before they caused a fire and downtime.

[Spiffy Solar]

Squirrel problem is very real

What kind of maintenance is there to do with the wires under the panels and in conduit all the way to the inverters? I thought solar panels and inverters were pretty much maintenance free.

I've got hundreds of installs under my belt as well and laughed at the salesman trying to sell us squirrel guard in the warehouse one morning. That's just how it was reported, personally I think it was a failed connection in a bp panel which has happened a lot lately.

You may have "hundreds of installs under your belt" and just be working in areas without a squirrel (or pigeon) problems. I talked to one installer whose first two installs, ever, got hit by squirrels. That was in Minnesota.

It is a relatively rare problem, but to a customer with squirrel damage, it's not a problem to be scoffed at. And that customer is going to want that system protected.

Arching from squirrel damage has definitely started fires, too. I heard directly of one in Maryland, from the installer. Fortunately, the homeowner came home, saw smoke and was able to take care of it quickly.

The real question is, how often does it happen? National solar leasing companies report that it's about 1% in some areas (Long Island, for instance). And that's just in one year. Other areas may never experience the problem at all. But undoubtedly, the more solar that gets installed, the more problems we'll see with critters.

My own neighborhood has a huge pigeon problem. That's how my business got started.

Here's the back of a panel, scorched by arching caused by a squirrel chewing on the wires. Fortunately, they burned themselves out (lost contact) before more damage was done.
BlackenedModuleBacking-tiny.png
This particular home had extensive damage. The squirrel nest itself was left for a long time and the moisture it retained caused the roof to rot. There was a hole in the roof, beneath the nest.

[donald]

Is high voltage DC from the roof in conduit in all states?

[inetdog]

Is high voltage DC from the roof in conduit in all states?

On the roof it can be PV wire or other UV safe and mechanically protected cable types. Once it goes inside, especially in an attic, it has to be a metallic sheathed cable or in a conduit, at least until it gets to the first fused disconnect.

The details are in the National Electrical Code, which at least in this subject area is usually adopted without amendment by states or local jurisdictions.
What will vary will be the revision cycle (year) of the code that is adopted.
PV wire was not introduced until the 2011 cycle, for example. (See section 690.31 for a start.)

[SunEagle]

Is high voltage DC from the roof in conduit in all states?

Pretty much that is the standard way of running the DC wiring once it is past the panel array and racking.

The wires can be attached using approved "tie wraps" or "clips" to the racking under the panels but you can't run loose wires along the roof so they need to be in some type of "raceway" or conduit.

[posplayr]

I was glad to hear that this thread finally converged to electrical connections being the primary problem and away from other esoteric things like static electricity.

Any connection which is based on an interference fit or smashing of a ductile material like copper is going to be susceptible to oxidation at the point where the current is supposed to go from one contact to the other. The more corrosion the more resistance. The heat generated will be based on P=I^2*R and as Temp goes up so does R increasing T so the whole thing is unstable. You have to keep the R below certain levels.

To a first order, higher voltages are going to be safer as the I^2 effect is reduced by a factor of 1/4 for a doubling in the voltage (AC or DC). But it seems like in a string the DC voltages are even higher than your typical AC.

My question is what is standard procedure in solar installs for electrical contacts (other than set screw torque specs)? Are there sprays (e.g. DeOXit) or greases (Dielectric) used on contacts?

I know house wiring typically has bare wire contacts unless (some water proof silicon filled screw nuts are used) and there are still high currents 10-20 amp. While I'm not an expert, but I'll guess anyway; the AC probably avoids as much corrosion build up at the contact as a DC current might.

[inetdog]

I was glad to hear that this thread finally converged to electrical connections being the primary problem and away from other esoteric things like static electricity.

Any connection which is based on an interference fit or smashing of a ductile material like copper is going to be susceptible to oxidation at the point where the current is supposed to go from one contact to the other. The more corrosion the more resistance. The heat generated will be based on P=I^2*R and as Temp goes up so does R increasing T so the whole thing is unstable. You have to keep the R below certain levels.

To a first order, higher voltages are going to be safer as the I^2 effect is reduced by a factor of 1/4 for a doubling in the voltage (AC or DC). But it seems like in a string the DC voltages are even higher than your typical AC.

My question is what is standard procedure in solar installs for electrical contacts (other than set screw torque specs)? Are there sprays (e.g. DeOXit) or greases (Dielectric) used on contacts?

I know house wiring typically has bare wire contacts unless (some water proof silicon filled screw nuts are used) and there are still high currents 10-20 amp. While I'm not an expert, but I'll guess anyway; the AC probably avoids as much corrosion build up at the contact as a DC current might.

The compression connectors are designed to produce what is called a gas tight connection, where there is no opportunity for moisture, oxygen, or anything else to infiltrate between the compressed and form fitting metal surfaces.
For switch and breaker contacts, there is more of a dependence on contact wiping action during the make to break through a thin oxide layer. Copper oxide also breaks down at a very low voltage (hence its early use in rectifier stacks). If the switch is not cycled occasionally the contacts may lose conductivity or weld into position.

[firerescue712]

Really?

That is why electrical codes are getting tougher on Solar. Lots of fires.

First thing Firement do with a house fire with solar is get on the roof with a fire ax and smash all the panels with a fire ax. It is the only way to turn them off. That is why CA a electrical building codes requires setbacks on solar panels so fire fighters have enough room to walk around the roof and smash panels.

When Diets & Watson warehouse burned down in NJ, all the fire department could do is watch because the place was covered in solar panels. Not a complete loss, the Fire Fighters had free nice hot meals while they watched. Lots of BBQ on the house of course. .

As a professional firefighter, I never cause or allow any unnecessary damage caused by uneducated or inexperienced firefighters. I have been a firefighter since 1978. Although solar is rare in Arkansas, measures are always taken to investigate and gather information on new products and ideas. Then we are prepared to accurately address the situation. One thing many forget, solar panels add a weight load to roofs. I am worried about a roof collapse more than an electrical issue from a properly installed solar array. More firefighters are killed/injured from structural collapse than from electrocution.

Many fires occur to improperly installed systems. Once a solar panel starts to burn, it is quickly consumed by flame. Wires overheat due to inadequate sizing, improper type for conditions, loose connections, broken insulation, nail/screw causing a short. Life is dangerous. Danger cannot be completely prevented.

Education will help those making decisions and discussions about solar installations, electrical safety, fire prevention, or anything else that they do not understand. After all, we are all born knowing very little. People not involved with firefighting or procedures used to fight fires do not understand why fires are not always extinguished. Safety of people is always number one. Material things can be replaced, but not the life. If anyone thinks they can do it better, try joining a fire department and try it out in person. I guarantee they will get a new perspective on fires.

Sorry for the rant. Now,,,back to our normally scheduled forum.