Do solar panels ever fail because of shading?

[SenileOldGit]

I have been reading up about bypass diodes, and I noticed that the guarantee for my Canadian Solar panels states that they mustn't be in shade for more than 100 hours, or something like that, and I have a large tree that is currently shading my panels for several hours of the day, which I am going to have severely cut back - but has anybody here ever had a panel fail because of shading? I presume that first the bypass diode(s) have to fail, which would then force the electricity from the unshaded panels to go through the solar cells in the shaded panel itself, thus causing excess heat in the solar cells?
I presume that as soon as a bypass diode fails, you would see a marked drop in the output of your string, because that panel really will 'bring the other panels down' to its level - I wonder how long a panel can last when a bypass diode (or all three bypass diodes) have failed, and it's in shade? Have any tests been done of this? It should be pretty easy, buy a £200 panel, or a secondhand one that isn't too old, and remove the bypass diodes, put it into a string, and see what happens to it.

 

[MaximusAnonus]

Care to share the link where it states the shade and the 100 hour mark? I've never read anything like that.

 

[SenileOldGit]

https://www.bimblesolar.com/docs/CS_Manual_Standard_Modules_en_en-v2.9.pdf

Page 8:

"Modules shall not be permanently shaded (including
partial shading, spot shading, even shading or uneven
shading) under any circumstance. Permanent shading
includes shading of the same cell, cell row, or module
portion for extended and repeated periods of time (e.g.
more than 200 daylight hours over the warrantied
service lifetime). Power dissipated in fully or partially
shaded cells will result in power loss, reduced yield and
can cause localized overheating, which in turn may
negatively impact the module service lifetime.
Permanent shading may cause accelerated ageing of
the encapsulation material and place thermal stress on
the bypass diodes. This would void the module’s
warranty unless properly mitigated through the use of
Module Level Power Electronic (MLPE) devices."


200 daylight hours over the warrantied service lifetime is not very long at all, so I thought it was very worrying when I read this about my Canadian Solar panels.

 

[MaximusAnonus]

huh seems like DAH has something about that too
well this is new...

I've never heard about such a failure tho...

 

[RCinFLA]

Many cheap panels have insufficient sized/heatsinking on bypass diodes. I would predict many installations that get significant daily partial array series string shading during strong sun illumination time of day have some bad bypass diodes.

When in bypass due to partial panel shading, a bypass diode can get subjected to about 5 watts of heating due to current through diode and diode voltage drop. Partial shading in early morning or evening is not so bad because illumination level is low and panel current is low.

A pellet diode in a small, sealed junction box with no air circulation will get fried.

Some cheap panels put two pellet diodes in parallel in a poor design attempt to get their current capability greater. In parallel, current on diodes are not shared equally.

Best panels have TO220, or similar packaged diodes with some metal mount heatsinking.

 

[SenileOldGit]

Would it be possible to add your own bypass diodes in a separate box on each of your panels? (Mine are ground mounted, so easy to get to.) That way I could implement whatever sort of diode I wanted, with whatever heatsink I wanted.

 

[sunsurfer]

The other issue is due to partial heating/expansion of the panel and affects on the sealing edges. Lots of panel failures are do to weather intrusion between the layers. Uneven heating over time could cause a breach to the sealing areas due to different rates of heat/cooling expansion to different parts of the module.

 

[RCinFLA]

Would it be possible to add your own bypass diodes in a separate box on each of your panels? (Mine are ground mounted, so easy to get to.) That way I could implement whatever sort of diode I wanted, with whatever heatsink I wanted.

DC and moisture create electrolysis. It is important to keep junction box sealed of water intrusion.

If junction box has a gasketed, screwed on lid you can open then you can improve bypass diodes. Do it within existing junction box.

I use TO220 packaged diodes with some added heat sinking. Remember TO220 heat sink tabs are usually electrically connected to a connection of internal diode so you have to keep heat sinking tabs electrically isolated between each bypass diodes. There is usually one bypass diode for each pair of long cell rows, ie, six cell rows panel has three bypass diodes.

 

[wattmatters]

Solar panel warranties are a jumbled mess when it comes to shading. It's a nonsense frankly given shading is unavoidable for most domestic use scenarios. Many such warranty terms would be in breach of Australian consumer law. Don't know about other countries.

Most Solar Panel Warranties Voided By Shading

You may be surprised to learn what most solar panel manufacturers' warranties indicate about shaded solar panels. Find out here.

www.solarquotes.com.au

 

[rcrracer]

Solar panel warranties are a jumbled mess when it comes to shading. It's a nonsense frankly given shading is unavoidable for most domestic use scenarios. Many such warranty terms would be in breach of Australian consumer law. Don't know about other countries.

Most Solar Panel Warranties Voided By Shading

You may be surprised to learn what most solar panel manufacturers' warranties indicate about shaded solar panels. Find out here.

www.solarquotes.com.au

Maybe Andy at the down under solar garage will read your post. He's getting ready to replace his garage panels, some of which are failing. He is all the time talking about the clouds and trees shading. Seems like he has some Canadian Solar panels.

 

[SenileOldGit]

So nobody here has had a panel fail because of shading, and nobody knows of this happening ever? I think that might be possible, because modern solar panels are so well made (I believe) compared to earlier panels from ten plus years ago, and I can't find any reports of panels failing anywhere, when I search on Google, but perhaps I'm doing it wrong.

 

[RCinFLA]

So nobody here has had a panel fail because of shading, and nobody knows of this happening ever? I think that might be possible, because modern solar panels are so well made (I believe) compared to earlier panels from ten plus years ago, and I can't find any reports of panels failing anywhere, when I search on Google, but perhaps I'm doing it wrong.

I doubt many here have checked to see if they have open bypass diodes. If you don't have partial shading, the bypass diodes are not called upon and you will have normal PV output.

All PV cells have some shunt defects. If the defect shunts are bad enough or get subjected to high enough reverse bias from partial shading, and do not have a bypass diode limiting the maximum reverse bias, the defect spots can get very hot. Typically, bypass diodes are placed every 20 to 30 series connected cells to limit maximum reverse bias to 10-15v.

If you are lucky, it just melts a hole in the panel plastic backside seal, or it can destroy the cell.



At the worse you can have a fire.

 

[wattmatters]

Maybe Andy at the down under solar garage will read your post. He's getting ready to replace his garage panels, some of which are failing. He is all the time talking about the clouds and trees shading. Seems like he has some Canadian Solar panels.

Andy bought all his current panels super cheap second hand from sellers getting rid of their old PV systems. He has/had zero idea on their providence.

 

[Substrate]

Wow - the whole *point* of subdividing panels with their own bypass diodes, was so that when a cell or cell's DID get shadowed by this, the amount of current going through these now shadowed "reverse bias" diode cell(s), (think of each cell as loosely akin to a diode itself) was limited to near half the panel's total current output and should survive!

Yeah, back in the day when panels weren't subdivided, cells got destroyed by super-long consistent shadows or debris in summer where shadows never seemed to move fast near the noon hours and took the brunt of the whole panel. Hence the subdividing bypass diodes of today.

Where this IS, or shall I say WAS a major concern was with THIN FILM panels that by design, did not incorporate any bypass diode subdivision. Even though a long consistent shadow across them at the same place had plenty of surface area to dissapate this reverse-bias heating condition, it was never a good idea. Mostly during long noon sun hour ranges where the shadows move slowly.

We're talking extremes here. Yeah, good idea to clean your panels, and not let a permanent bird dropping sit over a cell for 20 years.

I won't argue with their warranty department, but it *looks* like they are carrying over a thin-film concern, or perhaps their own cells weren't designed to handle a 50% reduction in reverse-bias (shadowed condition) current purposely designed for that reduction by the use of the bypass diodes in the first place.

They may have inadvertently exposed a weakness in their cell selection, and a further bypass diode circuitry to divide the panels into 3rd's or 4th's might be advised. But I don't think so - maybe a mixup between the concerns of thin-film vs crystalline has been made by the warranty guy. Maybe worth calling them to ask for an explanation.

 

[SenileOldGit]

I doubt many here have checked to see if they have open bypass diodes. If you don't have partial shading, the bypass diodes are not called upon and you will have normal PV output.

All PV cells have some shunt defects. If the defect shunts are bad enough or get subjected to high enough reverse bias from partial shading, and do not have a bypass diode limiting the maximum reverse bias, the defect spots can get very hot. Typically, bypass diodes are placed every 20 to 30 series connected cells to limit maximum reverse bias to 10-15v.

If you are lucky, it just melts a hole in the panel plastic backside seal, or it can destroy the cell.

View attachment 116905View attachment 116907View attachment 116906

At the worse you can have a fire.
View attachment 116904

I think "and do not have a bypass diode" is the important thing here - as far as I am aware, all solar panels made today have bypass diodes. So I presume shading isn't anything to worry about, in terms of panel damage - otherwise I think there would be a lot of evidence of panels being damaged by it. (I mean panels made in the last five or ten years - which have bypass diodes.)

 

[Substrate]

IANAL, but this seems like legalese marketing with a get out of jail free card.

For most people, some sort of shading is going to occur, which when you read the material basically voids 99% of the owners out there.

However, the last line of that material spells it out to me as leverage:

"This would void the module’s
warranty unless properly mitigated through the use of
Module Level Power Electronic (MLPE) devices."

AH, there's the upsell. Unfortunately, stuff like this has a tendency of becoming urban-legend which some might think apply to panels that are designed well in the first place.

So to maintain your warranty, looks like you need to have their MLPE devices.

 

[efficientPV]

This is some vindication of what I have said for years, going into full bypass or near bypass is bad for cells. Charge controllers should have a setting for minimum MPP voltage for a string. Making long strings and thinking the controller will sort things out is nuts. Looks like manufacturers are beginning to see the failure statistics for this poor practice.

 

[RCinFLA]

The most common panel configuration is six long rows of cells with each down and back row pair path from junction box having a bypass diode. This configuration allows all the bypass diodes to be located in junction box on one end of panel.

For this common panel configuration, the bypassing is one third of the panel. A bypassed row pair will have about a negative half volt, from conducting bypass diode, that subtracts from the illuminated two row pairs still remaining.


Each 'red' bypass diode in forward conduction bypass is carrying 8.5 amps of the string current with about half a volt diode forward voltage drop resulting in about 4.2 watts of diode heating. This is what cooks the bypass diode unless they have sufficient heat dissipation capability.

Realize in morning or late afternoon, when partial shading is almost inevitable, there will not be full illumination on panel so the panel current will be much less, and less stressful on bypass diodes heating.

Easy way to check bypass diodes is to check for their forward conduction with a CV/CC power supply applying a reverse voltage to a shaded panel. If any bypass diode in panel is open, then the reverse applied power supply will not be clamped in voltage.

 

[Maitake]

This is some vindication of what I have said for years, going into full bypass or near bypass is bad for cells. Charge controllers should have a setting for minimum MPP voltage for a string. Making long strings and thinking the controller will sort things out is nuts. Looks like manufacturers are beginning to see the failure statistics for this poor practice.

Ahhh so that's what the min start voltage on the MPPT is!
What is a good setting as absolute or percentage of VOC?

 

[SenileOldGit]

This is some vindication of what I have said for years, going into full bypass or near bypass is bad for cells. Charge controllers should have a setting for minimum MPP voltage for a string. Making long strings and thinking the controller will sort things out is nuts. Looks like manufacturers are beginning to see the failure statistics for this poor practice.

What are the failure statistics? My original post asked if anybody here had had a solar panel fail because of shading, nobody has said yes so far. What does "This" mean, in your sentence "This is some vindication of what I have said for years"?
Why is full or near bypass bad for cells, when, as long as the bypass diodes are working, the cells aren't taking the current?