Panel that takes longer to defrost - Anybody come across this?

[ktech]

I have had a customer suspect that something is wrong with one of the LG panels we installed a couple of years a go. They have sent the photo (attached) as I have not visited site yet. System performance is reported as ok although the customer has not check any data on the inverter, it's a SMA 4000TL so will be able to get some good information from it and can do the usual checks on the panel.Issue is on a frosty morning this one pane takes longer than others to defrost.

 

[Marvo]

Wild uneducated guess at panel orientation maybe.

 

[telectrix]

condensation got in?

 

[ktech]

Some type of delamination was one of my thoughts but surely the whole string would go down.

The customer was originally informed by a 'helpful' passer by that knocked on her door to inform one of the panels was not working, he could tell all this from looking at the panel. I've spoken to him on the phone and he thinks it's because 'the panel is not heating up properly to defrost it' - they have panel installed at his council works office so he has a wealth of experience. The customer has noted that this person calls then modules and not panels so must know what he I taking about. He then followed up his drive by with a letter to the customer with a copy of the photo. He now seems to be her new best friend and they have a telephone call each evening to discuss the issue!

 

[SolarCity]

I'd say condensation would be a good bet. Not a good advert for LG if this is the case.

Could it be that the panel isn't plugged in with the other panels?

 

[whinmoor]

Probably condensation. That panel must have a slightly different efficiency to the others.

Either that or something in the loft? Maybe a wodge of insulation directly under that panel?

I would have suggested faulty optimiser or micro-inverter but you've already ruled out that possibility.

There was a vaguely similar discussion on this forum a couple of years ago. You might get some ideas from this:

http://www.electriciansforums.co.uk...ergy-forum/51112-interesting-observation.html

 

[Gavin A]

the most efficient panel will defrost last, the least efficient first, as the excess power from the most efficient panels gets fed into the least efficient as resistive heat.

ie if one panel is 260Wp, the lowest is 250WP, the average is 255Wp, with 10 panels, in full sun 50W (ish) would be being fed into the lower rated panels (cells actually) as heat, no heat would ever be fed into the highest rated panel - well maybe once all others have defrosted they may end up at higher outputs due to the glass being clear.

something like that anyway.

The more variable the efficiency of the panels, the more noticeable it will be.

 

[Marvo]

I like that explanation Gavin, it's both technically sound and very elegant from where I am. I guess the panel would be at it's most efficient just at the point where it's defrosted and therefore transparent but before it's temperature has increased. Therefore the last panel to defrost would be the most efficient for a while until it warms up but that warm-up time would be somewhat arrested by the fact it's the most efficient panel because it's still relatively the coolest.

I'm getting the hang of this solar stuff, I should be an expert by the time it arrives en mass in my country

 

[Gavin A]

I've got an array on our side porch roof with our landing window directly above it, so I've spent a bit of time watching how it works.

This process can even melt snow, or more precisely once the top of the panels are clear it will feed small amounts of current into the covered cells to melt the base layers of snow, then the snow starts sliding off the panels one by one.

 

[ktech]

Great explanation Gavin, thanks.

I just hope my customer doesn't ask me to replace all the remaining panels to suit!

 

[Gavin A]

just tell her that's how you tell which panel is best.

though you could find someone asking you to restring it after monitoring which were the last 50% of panels to lose their ice.

 

[Worcester]

Can't quite get my head around it at this time of the morning, though Gavin's explanation sounds so technically good it will even shut here 'knowledgeable' neighbour up