Diodes, matrixing and shading?

Sun plumber

New member
Joined
Jul 20, 2017
Messages
21
I saw an interesting video of a guy who rewired some broken solar panels to kind of bypass the bad spots:

It seems to me the same rewiring might reduce the effects of partial shading on an array.

I'm thinking of using two parallel panels with a cheapo MPPT charge controller (looking at mpt-7210a and CKT-EV-600).

It seems like the panel diodes get removed in the video. Is that a good idea?

I would love to see a schematic for a 72 cell panel and an explanation of what the diodes do when they are working and when they are just along for the ride.

I want to try to get the maximum performance out of a small array so if I can reduce the effect of shading, that is great but I don't want to create a dangerous electrical situation.

I'll use the best wire and connectors I can afford to maintain it's weatherproofing and minimize IR drop (no ugly hacks).
 
Parallel panels are not affected in same way as serialized panels are. If 1 panel in a series have shading it affects the whole string. But if 1 panel in a parallel of 2 have issue it does not affect the other panel directly. 2*250w panels in series could potentially go to 0 if 1 panel fails or are shaded. If 2*260 are in parallel instead and you have 1 that fails you may have 260w total instead. (Numbers are a bit exaggerated)

You have the diodes there to protect from backfiring into the other panel. If you have 2 panels in parallel and 1 of them are shaded the diode blocks current from rushing backwards into that panel.
The work he did was for "fixing"the issue with the broken serialized cells in the panels.

How much shading do you got? The best way to optimize shading is with optimizers or microinverters on each panel.

If you wanna go the cheaper way look at his other video:
Where he shows the matrix principle. It can be done with normal Y connectors.

Doing this internally on the same panel i dont see any reason. Unless the panel is brooken.
 
On an RV shading could be different every day. I suppose you have more control over it than a house but some times compromise will happen and solar will lose.

I just want to have the best chance of charging my batteries if I get stuck with partial shading.
 
If you're doing the setup on an RV, then the Y-connectors in the video daromer showed would be perfect. Your runs are very short
 
For a big series and parallel array that is the way to go and easy.

With two panels, connected in parallel, they are already Y'd.

The idea is to do the same thing with the individual strings of cells in the panels.

On an tangentially related note, I saw another video, this time using super capacitors:

It's a little simplistic but if I stuck three small 16v super capacitor banks between the terminals in the junction box and connected all four points for that matrix connection, the caps could provide the current when one string gets shaded.

The main issue I can think of is that when the sun goes down, the caps will try to keep running the charge controller until they are flat, then it will take a while to charge in the morning.
 
Sunplumber said:
On an tangentially related note, I saw another video, this time using super capacitors:

Have just watched that video .... it's a schoolboy who doesn't understand what he's doing ... at 6:30 he shows it doesn't (effectively ) work ...

Capacitors are a non starter for many reasons ..

For an RV , because your wire distance is so short the optimum is to have many small"12V"panels in parallel, designed to charge a 12V battery these have about 20V open circuit voltage , 17 V operating voltage
 
Yeah, I can tell there are many things wrong with what he did but if I'm just trying to minimize a dip in voltage/current when a bird or plane fly's over so the mppt controller doesn't get fouled up, I think there is some promise there.

I don't know if I would want to add enough cap to handle shading from a cloud passing over but a little bit of cap should be cheap and might give me maybe 20-30% more watts if there is a lot of intermittent shading.
 
I got some help with capacitor math and at a 6 amp rate, a 16 farad super capacitor module will sag one volt in 2 2/3 second. That is for one string of cells getting shaded, if both parallel strings get shaded the time would halve.

There are 63 farad modules available also so those times could be quadrupled for about twice the price.

Those times are ideal, reality is probably worse.

A bird or a plane probably takes less than 2 2/3 seconds to fly over but I don't know if 1 volt of sag will cause the mppt controller to foul up or if slowing down the sag is enough to keep it happy. I imagine it will keep the controller happier than cutting the current in half instantaneously.
 
Where do you live where birds/planes could interfere with your panels for longer than a second? The only thing you really need to worry about shading are trees. Well, and clouds, kinda. Even in cloudy days, the needed solar energy is able to penetrate the clouds (not 100%, obviously, but a good amount)
 
This is a corner case but let's say a small flock of birds fly's past. None of them shade a cell for a second but the cumulative shading might drop the cap voltage a few volts.

More to think about is capacitor leakage and it seems like the charging rate is low if the caps pull the panel down to a low voltage.

I'll try to figure out how big those issues are and if they can be overcome.
 
Hrmm. I dunno. I dont think a few seconds would really make a huge difference. The bounce recovery would be really quick as the shadows from the birds would be moving quickly from panel to panel. So there might be a slight sag, but nothing to shut it down.
Unless you're referring to a huge flock of birds that block out the sun for almost a minute :p Now that's a sight I'd love to see.
 
Watching videos of the cheap mppt charge controllers, it seems like a momentary shading can foul up their output for 20-30 seconds some time (mpt-7210a).

String together enough moments and you could lose minutes.

Over a day it might be a few hundred watt hours (around 60 mppt foul ups=200 watt hours).
 
Hmmm, interesting. That is something to think about. I didn't know those cheaper MPPT controllers would go nuts like that. I thought they all had some kind of regulator to minimize rippling.

But you have a good point there. If shadows happened 10 times a day, and the controller is basically offline for 20 seconds, 3.5 minutes is lost. Over the course of a month 1.75 hours is gone. So, if you would of been using a kW during those times, that's a 1.75kWhrs lost. That's a lot. So yeah, I see your point there.
 

That video shows the problem.

The panel gets shaded at about 0:07, output goes near zero, then it starts to recover around 0:27 with a very slow ramp up (full power is around 400 watts).

It would be nice to see how much shading the panel actually got but if you look at the panel voltage display you can see it sag a bit and then drop to 10v right as current goes to zero, then the voltage snaps right back but it still takes 20 seconds before it starts to recover.

I don't know if that is just a charge controller design flaw or if all mppt controllers will behave similarly but it seems bad enough to try to fix.
 
Wow, that is a lot of sag. Hrmm

I'm not sure, but you could check out Julians Iletts MUPPET design. He has software posted and the whole workings. It's his MPPT controller and it works pretty good. Altho, the part I'm unsure of, is how well it rebounds from shadowing. If you have all the parts available, you could build a quick mock up with a test panel and see how it performs. Or ask Julian if he's done that test or not.

But yeah, I see what ya mean. That is a long time. Way to long.
 
Sunplumber said:
Yeah, I can tell there are many things wrong with what he did but if I'm just trying to minimize a dip in voltage/current when a bird or plane fly's over so the mppt controller doesn't get fouled up, I think there is some promise there.

Why would you want to do that ???? theshading problem has nothing to do with birds or planes ...They fly high and wouldn't have any effect .....

The problem is, if you have a leaf or dirty patch on the panel ...if one of those 100x100 mm cells which make up a panel, is 30% covered, the whole panel , and all those in series with it are delivering only 30%

Shadows from distant objects buildings trees ,are less of a problem since they are moving constantly .
 
Birds nor planes does affect it over long term. yes cheaper MPPT and fake mppt like the 7210 will take time to recover. Its always about what you want to pay. Either you buy the cheap MPPT charger and loose some power and perhaps get more panels or buy a proper one directly.

If you get a proper charge controller from start then you dont have to think much about the shading. And if you have a tree in the way either optimize for it or cut it down :D
 
Being close to the flight path of three airports, I can tell you that a jet in the glide slope can create a big shadow.

The birds tend to fly at about tree height too.

Perhaps my concern about airplanes is overblown because once I'm on the road, I can move to avoid the pattern, I'm just lazy so I would rather have a system that maximizes the output.

Right now I'm looking at 1,000 watts and I'm thinking about perhaps 1,500 or even 2,000. That would need 2,3 or 4 7210 charge controllers for roughly $80, $120 or $160 plus another $40 to have a spare.

Is there a real MPPT solar charge controller without shading issues in that price range?

Super capacitors and parallel wiring would cost about $80 per 500 watts so the total cost of charge controller for 1 kw is $240. The panels would be about $800 per kilowatt so all up a little over $1/watt, then add batteries and an inverter.

That seems pretty cheap to me. The nicer charge controllers seem pretty expensive, possibly more than I want to spend on my whole solar array.

I haven't seen anything in the middle yet. Maybe I'm not looking in the right place.
 
I would say no. There are no proper MPPT controllers in that range that does what you want. If you do think planes are an issue and still want to use 7210 charge controllers i would say the 7210 cause 10x more issues in terms of efficiency than the planes ever will be able to cause. Unless you have a helicopter hovering over your panels :) Just because a plane create a big shadow it only lasts for a very very short time in general. Its then better to just make sure that the MPPT charger recovers fast.

800USD per 1kw panels sounds fair. perhaps a bit high. I pay 600USD here :) And proper charge controller cost from 200USD if you ask me. For instance a simple one as the PCM60x will do what you want and be pretty efficient compare to the cheaper ones out there. Its around 240USD i think.
 
I'm looking at more expensive panels because an RV has limited roof area and high efficiency cells let me save space.

I will look in to that controller, thanks.
 
Back
Top