CaySol
Solar apprentice
Apples to oranges then. Hard to suggest 60v arrays for the rest of us.
Very long PV wire
- Thread starter Rocketman
- Start date
They say it can happen pretty quickly, not years. Up to 70% loss. Also, the bifacials may be more susceptible.
I wonder though, cause how often have you heard anyone with a problem with PID? I post a thread asking cause I wonder if its really an issue for our small setups vs solar farms. It might be a big nothing burger.
very true, multiple small arrays wont' work for everyone.
robbob2112
Doing more research, mosty harmless
So far in reading on this forum I have only ever seen you talk about PID damage, unless it is a response to you.
Gotta dig into what caused the fires. Modules for sure are more likely to catch fire at higher voltages due to insulation failures, for obvious reasons. Ive seen that many times.
I have been researching it allot recently and unfortunately, no one really knows 100% what is going on with it. There was a thread a while back by a gent from Italy that thought he had it but I don't know if he ever tested. Panels have to be marked in order and tested separately.
I am suspicious allot of the cheap panels by the pallet from solar manufactures are from PID damaged farms. Decent ones are resold used.
I deep dived on it a long while back and its a fascinating subject. Here is a brand new panel (mine) that is already eager to die an early death. Check out that bad boy cell. Don't worry, I'm not pushing much power through that cell, its a small array.
If you ref the blue in the target box @28c then your looking at about 37c Delta. It was a cool day. It felt noticeably warmer. Its brand new and with low power passing through it will be fine. I would not put it in a high power array.
Yeah that's one thing about flir cameras. They tend to make things look worse then they really are.
I wouldn't worry about a 7 or 8 deg delta T. In the UL test the module is first heated to NOCT and then test cells are subjected to both light at 1 sun plus the heating from being reverse biased. The hot cells typically get to around 100C or so, or maybe around 50C above the surrounding cells.
That kind of temp will eventually discolor the EVA and wrinkle the backsheet but it's not gonna catch fire. You can get the same effect on a perfectly healthy cell by adding the right amount of shading from a bird turd or leaf so if cells were going to catch fire from that it would happen all the time. Thank the bypass diodes for that not happening.
Does that module make full power?
I wouldn't worry about a 7 or 8 deg delta T. In the UL test the module is first heated to NOCT and then test cells are subjected to both light at 1 sun plus the heating from being reverse biased. The hot cells typically get to around 100C or so, or maybe around 50C above the surrounding cells.
That kind of temp will eventually discolor the EVA and wrinkle the backsheet but it's not gonna catch fire. You can get the same effect on a perfectly healthy cell by adding the right amount of shading from a bird turd or leaf so if cells were going to catch fire from that it would happen all the time. Thank the bypass diodes for that not happening.
Does that module make full power?
PhysicsbyTom
New Member
10 AWG solar wire would handle it but to be on the conservative side, I would go with #8 AWG. I am assuming you are also using Victron Inverters.
Ok sorry I misread the scale. Yeah at 38 Delta T that bypass diode should probably be on and the module power down by a third if there are 3 diodes. You can tell because the diode will be hot too. Not gonna catch fire though.
I've averaged 163 kwh/d so far in Jan with 3 days over 200. My peak was 244. The two EVs is what makes it jump up. It's a 60 mile round trip to anywhere.
The forward bias on the diode likely not there yet. Thats the problem with bypass diodes, its only a partial fix.
Delta T is more than double and you can see how its warming that half of the panel. I will have to check it out this summer when its 100F+. If Victron wants to send me 450/100 I will test it under some real loads vs the wimpy string with the 100/30 @12v
Diodes fully fix what they're meant to fix which is keeping weak or partially shaded cells from destroying the module or starting fires, ie, pass the UL test. Module performance would be better with more diodes of course, but UL doesnt care about performance, just safety. No manufacturer is going to put more diodes in there than they need to to pass the test, costs $.
If you want to test how hot that cell can get you can just take that module out of the circuit and short the leads. Then put it in full sun. It will try to push Isc through the bad cell which will turn on the diode.
While you're at it you can pick a cell in a different part of the module, make up some little pieces of cardboard of various sizes with tape on them and stick them on the cell. You're simulating a bird turd. That will drive the cell in reverse. By adjusting the size of the cardboard piece you can max out the power dissipation in the cell. Start around 10% of the cell area, that should be pretty close to worst case. It will probably end up hotter than your bad cell.
You dont need to wait for a super hot day btw, just add the ambient temp difference, you'll be close.
Edit: I don't think the bad cell is heating up that whole half of the module, glass is a poor heat conductor so the hot spot should look more like a bulls eye, like it does just to the left of the hot spot. I think what you're seeing there is either a little breeze cooling the left side of the module or maybe a radiation effect from whatever is behind there.
If you want to test how hot that cell can get you can just take that module out of the circuit and short the leads. Then put it in full sun. It will try to push Isc through the bad cell which will turn on the diode.
While you're at it you can pick a cell in a different part of the module, make up some little pieces of cardboard of various sizes with tape on them and stick them on the cell. You're simulating a bird turd. That will drive the cell in reverse. By adjusting the size of the cardboard piece you can max out the power dissipation in the cell. Start around 10% of the cell area, that should be pretty close to worst case. It will probably end up hotter than your bad cell.
You dont need to wait for a super hot day btw, just add the ambient temp difference, you'll be close.
Edit: I don't think the bad cell is heating up that whole half of the module, glass is a poor heat conductor so the hot spot should look more like a bulls eye, like it does just to the left of the hot spot. I think what you're seeing there is either a little breeze cooling the left side of the module or maybe a radiation effect from whatever is behind there.
Last edited:
This is the OP. Thank you for making a long and varied thread (and it kinda kept on topic
).
For those that wondered why I was looking at Aluminum 4awg wire - it’s all about the cost/foot. (With really long lengths the cost per foot really matters).
These are the costs from a “random” website - wireandcableyourway.com
10/2 copper UF $1.19/foot
8-8 aluminum URD $.97/foot
4-4-4-4 aluminum URD $1.55 (but you get 2 arrays)
Now I didn’t know I needed a ground wire - but a single 6 aluminum URD is $.43/foot.
So even though #4 aluminum is way bigger than I need it comes out to $.78/foot - plus boxes on each end to swap it back something connectable.
And I also found out that network cable is a no-go in this trench (way too long).
Now I just need to decide if we are giving up nomadic lifestyle (we full-time RV) and if we want the issues with this property.
).For those that wondered why I was looking at Aluminum 4awg wire - it’s all about the cost/foot. (With really long lengths the cost per foot really matters).
These are the costs from a “random” website - wireandcableyourway.com
10/2 copper UF $1.19/foot
8-8 aluminum URD $.97/foot
4-4-4-4 aluminum URD $1.55 (but you get 2 arrays)
Now I didn’t know I needed a ground wire - but a single 6 aluminum URD is $.43/foot.
So even though #4 aluminum is way bigger than I need it comes out to $.78/foot - plus boxes on each end to swap it back something connectable.
And I also found out that network cable is a no-go in this trench (way too long).
Now I just need to decide if we are giving up nomadic lifestyle (we full-time RV) and if we want the issues with this property.
I'm with you on using direct bury Al over that distance. Put some antioxidant paste on the terminations and retorque them a couple times.
If you go with 6/2 Al you might be able to use a couple IMO isolators at your array and make the transition from your module interconnects to your Al conductors in those. You should have disconnects at the array anyway. Pretty sure the IMOs can take 6 awg, not certain the terminals are spec'd for Al though.
If you go with 6/2 Al you might be able to use a couple IMO isolators at your array and make the transition from your module interconnects to your Al conductors in those. You should have disconnects at the array anyway. Pretty sure the IMOs can take 6 awg, not certain the terminals are spec'd for Al though.
mrnudie
New Member
depending on what state you reside code varies enormously on earthing, regardless imho, a local earth rod to the pv array frame is essential and it should also have a surge diversion earth point, that is bonded back to another earth rod at the mppt controller. bonding is massively important to reduce differential voltages induced from lightning during a storm. Do not have separate eath rods that are not tied together. Voltage differentials can and will kill lots of electronics.
I notice the suggestion of micro inverters, i would not go that path either. Just more points of failure and if they are part of the panel structure even worse…
all “power” runs must have an earth, better yet for a long cable run is armoured cable that has a shielded or sheaf that can be earthed.
mrnudie
New Member
a
an earth “circuit” is not for fault current ( that is what a earth neutral bond is for) it is to provide protection to people and hopefully from stray voltages aka storms and so on. regardless an array should have its own earth rod that is bonded back to another at the mppt or pv controller. there are no guarantees a local strike can generate enormous (stupid!) voltages across even bonded earth points…
This is what I do but IIRC, it is no longer recommended to retorque the connections.
@Rocketman When running a low pv voltage Growatt, I had over 200' of 2/2/4 al urd in the ground. Like you mentioned, it was overkill but cheaper than the alternatives. FYI, it is a pita pulling that stuff through conduit.
Yeah there is a debate about that. The argument is that you get cold flow in the conductor so when you retorque it, it will be compressed further, then cold flow some more, etc. But OTOH if it's actually gotten loose you have to retorque it or it will overheat. Probably the best thing would be to do an IR scan under load and if it's abnormally hot retorque it, if not leave it alone.
It sure is. Also the harder the pull the easier it is to cause insulation damage. It's standard practice to megger all buried conductors on commercial and utility scale PV installs and insulation failures on pulled conductors are quite common. Another reason I favor direct burial over conduit, as long as you don't have a colony of those subterranean murder squirrels....
I don't have one myself anymore but megger meters aren't very expensive these days, it might be worth buying one to check the conductors on those long runs. Way better to find out before backfilling the trench.
They should help with PID because they keep the PV modules at a low voltage relative to ground.
Not sure what you mean by help with grounding. You still and always have to earth ground your support structure and module frames. If you mean do they help with damage from lightning induced transients the answer is no, you've now placed a bunch of electronic devices out in your array so you've made that worse.
Personally I wouldn't consider using microinverters on a ground mount or any other situation where RSD wasn't a requirement, unless your shading is really egregious. But thats what chainsaws are for.
Not sure what you mean by help with grounding. You still and always have to earth ground your support structure and module frames. If you mean do they help with damage from lightning induced transients the answer is no, you've now placed a bunch of electronic devices out in your array so you've made that worse.
Personally I wouldn't consider using microinverters on a ground mount or any other situation where RSD wasn't a requirement, unless your shading is really egregious. But thats what chainsaws are for.
Similar threads
