low output until all panels in sun

[davey_b]

I have an electricity monitor on our 4kw pv system so I can see what the output is and I noticed today that about noon when all but the last 2 panels were in full sun it was outputting 0.3kw. As the sun moved and that last 2 panels came into full sun it went up to full output of over 3kw in about 10 minutes.

We have a solarmax 4200s inverter but I forget what the panels are. My question is if most of the panels were in full sun why wasn't the output high? Why would the 2 end panels need to get sun for it to give a decent output?

Any thoughts much appreciated.

Dave

 

[FB.]

My question is if most of the panels were in full sun why wasn't the output high?

Because a solar array is only as good as the weakest link.

Why would the 2 end panels need to get sun for it to give a decent output?

The shaded cells are not producing as much electricity, so they work at a slower rate than the cells in sun, creating a bottleneck which forces all the other panels to work at a lower rate. Usually bypass diodes in each panel would help to allow power to shunt itself around the bottleneck to some extent.
There may also be problems with the inverter tracking an incorrect optimum voltage, since this optimum will keep changing as panels come into and out of the sun.

 

[FB.]

There may also be problems with the inverter tracking an incorrect optimum voltage, since this optimum will keep changing as panels come into and out of the sun.

To clarify:

When daybreak arrives, no panels have sun but they have enough daylight to generate an output to wake the inverter. 16 panels running at their optimum 30V each for a total of 480V.

Then the sun comes up and twelve of the panels are in full sun while the other four activate bypass diodes because the shaded four are creating a bottleneck. Twelve panels running at optimum 30V each means the array would be best at 360V, but the inverter only scans a small voltage difference either side of what was the optimum at startup, so it doesn't see the new (and very different) optimum voltage so continues to run the array at 480V and therefore way off the optimum voltage for maximum power. As panels come on-stream or off-stream during the day, the optimum voltage will be jumping about all over the place and power production will be well below expectations.

Some inverters will periodically check a wide voltage range for a change in the maximum power voltage. This feature was introduced about a year ago when the tracking of sub-optimal voltage peaks was found to be a problem.

The bottom line is that solar arrays were never intended to be installed where shading occurs. Certain inverter and panel configurations can get mostly around the problem.

 

[SolarCity]

Do you have a picture of the array that you can show us?

It sounds like you're suffering from heavy shading which could seriously affect your potential yield.

What inverter do you have and how long have you had the system installed?

 

[Earthstore]

In my opinion your inverter is a little too large, and your system has not been designed with the shading problem that you have, there are ways to negate the effects of shading but sadly some companies that were installing PV did not have a clue.

From the information that you have given so far, you could have increased your total
generation by having a slightly smaller system and not mounting panels in the shaded area, which would have cost a little less to install but your peak output would be slightly lower.

From what you say, maybe I would have installed a 3.5KWP system and maybe a 3-3.5KW inverter.

As said before some pics would be helpful, panels shaded and unshaded, and maybe inverter too.

I hope this makes sense to you

 

[davey_b]

We had the system installed last September. There was issues with low output and they came back in April and looked at it and output was better. Then I noticed that when it was at peak power it would suddenly drop to 0 and then climb back to nearly 4kw. It was saying "Vac 10 min too high" and "Frequency too high" (see video of this error here: Solarmax error log - YouTube and status log: Solarmax status log - YouTube)

The inverter was replaced but now this inverter is saying "Failure fan2" so I'm going to contact solarmax about this tomorrow.

It's a Solormax 4200s inverter and we have 16 Canadian Solar CS6P panels.

Here is a picture of our array.

As you can see we're getting full sun, it's pretty much straight on the panels and there is a wee bit of shading on the end panels and we're only getting 0.37kw!

 

[The Solar King]

What is the orientation of your house, i.e. is it south facing or near to it?

How many panels could be moves to the left of the solar thermal array?

 

[davey_b]

House is south facing, slightly south west.

Could probably move 2 panels but seems crazy that it can't be set up to cope in this situation and generate a decent amount when it has a bit of shading.

 

[jason121]

Do you have a log burner ?:gettree:

 

[jerriais]

I'd definitely go for Tigo optimisers in this sort of shading case

Products | Tigo Energy

They'll make sure each panel is always producing at it's maximum and you can also monitor the whole system online without changing your inverter.

Ask you local installer if they can do it, or if you're in SE England we'll be happy to quote. I'd imagine it wouldn't take long to recoup the cost considering how much power you're losing now through the shading and poor string design.

Matt T
SunSmart Energy Ltd

 

[davey_b]

Right so am I correct in saying that with this inverter and panels there is no way to get this working properly without either replacing it or retro fitting an optimiser? If that's the case I'm minded to tell the installer to remove it as it's not fit for purpose.

 

[whinmoor]

You've had the system for nearly a year now. Did your quotation say how much output to expect per year? Presumably you're not on target to hit the prediction? If you're miles off, it might be worth having the installer around for a "chat" about why it's under-performing (assuming they are still trading).

Would it be possible to relocate some of the panels to the left or under or above the thermal? Even then, optimisers would probably be a good idea with all those trees.

 

[davey_b]

Right, thinking of going back to installer and suggesting they either install optimisers on each panel or remove the system. Will optimisers like these http://www.buypvdirect.co.uk/Solar_Edge_Power_Box resolve this problem?

On a separate issue solarmax say I need a 2nd replacement inverter as the fan has failed in this one!

 

[davey_b]

Just thinking, if they're having to replace the inverter anyway would it be as easy to install an inverter that has this technology built in? If so what products do that?

 

[The Solar King]

This is one of those where hindsight is a wonderful thing. Was the Solar Thermal installed first? Ideally I would have swapped the position of the two as the ST is less affected by shading than PV and it is easy to compensate by increasing the size of panel used. As it is you appear to have 6SqM of ST which is pretty hefty unless you are using a large amount of hot water.

The PV should have been modelled in something like PVSol to assess the effects of shading and optimise panel placement.

Before going any further, get the issues of the inverter sorted, to ensure this is not the root cause of the problem. If it is shutting down when hot due to lack of fan cooling, the system is hardly going to perform as intended. Where is the inverter located? A roasting hot attic won't help.

With regard to optimisers, they will make some difference, but they are unlikely to come for free. The enhancement in performance needs to be assessed against the incremental cost. Some optimisers require a matching inverter. Solar Edge are about top launch a inverter independent unit, and as yet I have not seen any pricing.

As already suggested, check what performance prediction you were given when you agreed the purchase. Do talk to your installer. It is in everyone's interest to resolve this positively and amicably.

 

[davey_b]

We were predicted 3180kwh per year and since the end of September we've done 682khw so far. The most we've done in a day is 16kwh, this is on a 4kw array, and we're not really generating more than about 0.5kw anytime before 2pm!

 

[The Solar King]

There is something not right here as this performance prediction for a south facing 4kW system suggests quite a bit of shading has been taken in to account. Even in sunny Edinburgh an ideally placed unshaded 4kW system will produce 3800kWh/a.

The system does not appear to be functioning correctly. Shading may be only part of the problem. Get your installers back and get them to sort the problem.

 

[davey_b]

Yup! The installations manager is off this week but I'm going to email him and away his response. I'm planning to give them 3 alternatives.

1. Fit optamisers
2. Fit micro-inverters on each panel
3. Remove and refund along with lost income.

We'll see what they say.

 

[solarfred]

You don't need PV Sol to tell you this installation won't work. Clearly the trees are blocking the southern (SE?) aspect and in winter will probably block 90% of any available sun even though no leaves. You could take the end two panels and fit them landscape under the thermal but frankly this system was never going to produce anything like maximum output. What time of day was this pic taken??

 

[davey_b]

I understand and accept that it would not get full output because of shading but when the panels are only getting slightly shaded and we get 0.37kw I don't think that's acceptable. This photo was taken at 1.30pm yesterday.

 

[solarfred]

You should certainly have one string all the way to the left so only one gets the worst of the shading but the output number does suggest something is terribly wrong. It needs to be checked to ensure a connector has not popped and one string is down thus leaving a shaded string? Hard to say. You could turn off one string and see if the output changes?

 

[The Solar King]

It may all be on one string.

 

[solarfred]

Yes and if it get's to 3kw it can't be disconnected either! It has to be a shading issue taking out potentially two middle panels of the string if it is one.

 

[solarfred]

If it is one string then it is right on the voltage limit of the inverter and if the temperature is low it will likely exceed the limit as these panels are 37voc at STC. Maybe something to do with it. So how many inputs are there from the panels? 2 cables or 4 cables?

 

[SolarCity]

The output figure of 0.37kW does seem low to be honest.

I would be amazed if the panels at the right end of the array were actually providing any benefit whatsoever. A proper shading analysis would have pointed out which panels would receive significant shading - something, incidentally, that the new MCS method won't do.

I've not got time to check the system parameters but I will when I get back home.

 

[davey_b]

If it is one string then it is right on the voltage limit of the inverter and if the temperature is low it will likely exceed the limit as these panels are 37voc at STC. Maybe something to do with it. So how many inputs are there from the panels? 2 cables or 4 cables?

I'll need to check by looking at the inverter but I'm pretty sure there are 4 inputs from the panels to it.

 

[whinmoor]

A photo of the inverter would be interesting whilst you're there...

 

[davey_b]

It's this one here: SolarMax S series | SolarMax

If you want a photo of it in situ I can do that though.

 

[Gavin A]

you've been sold the wrong inverter. For a string inverter in shading like this you need either Aurora Power-One, or SMA TL or HF inverters, or similar which have a full voltage range scan function that stops it sitting on a false peak like yours is, and enables the bypass diodes to function correctly to bypass the shaded sections and allow the rest of the array to operate at near enough full capacity.

they have also probably undersized the cable, causing the inverter to raise the local grid voltage beyond the protection limits causing it to cut out - either that or you have very high grid impedance on your incoming supply / high grid voltage anyway, or maybe a lot of other solar installs on the same line.

I'll post up a graph of a system we've been monitoring that has hard shading coming across it from 5.30pm, without this function enabled it was doing as yours is, with it enabled it's reducing output by around 10-20% as the shading runs across the array.

 

[whinmoor]

I think we'd all be interested in seeing how it's been wired up...

No experience with Solarmax but we'd have used a Power One in this scenario. Or perhaps Solar Edge. Probably would have positioned the panels differently as well.

 

[davey_b]

I think we'd all be interested in seeing how it's been wired up...

No experience with Solarmax but we'd have used a Power One in this scenario. Or perhaps Solar Edge. Probably would have positioned the panels differently as well.

I'll get a photo for you tonight. What in particular are you looking for a photo of?

 

[whinmoor]

I'll get a photo for you tonight. What in particular are you looking for a photo of?

Just the whole thing. Is it in the loft directly below the panels or somewhere else? Where the isolators are, how they connect to the inverter, what cable has been used...

It's probably been done correctly and only the shading is impacting on output but if there's something wrong, you've got a few dozen experienced installers who might spot it.

 

[davey_b]

OK, that all makes sense.

I spoke to a guy called Iain Garner from Tigo this afternoon about their optimisers. That was really helpful and it seems that the kit that they have could help us overcome this issue. Also solarmax have now said that we'll need a new inverter since there is the issue with a fan failure! Hopefully we'll get there in the end.

 

[davey_b]

Ok, here's some photos. It's in the loft just below the panels.

 

[oliver s]

Looks to me that the pos and neg are the wrong way round in the inverter if the tape on the cables are correct? only looks like 1.5 flex to the AC isolator 2.5 at best, should be 6mm, 4mm at worst. Are there any other isolators for the DC supply from the array. In all doesn't look too good would definitely get them back

 

[whinmoor]

Are there any DC isolators or are they somehow built into the inverter? Doesn't look like it to me.

There should really be a label on the inverter but that's small fry compared to other issues. The four cables flying thru the loftspace aren't best practice either. They should really be clipped or run in trunking and clearly labelled.

No fireproof board behind the inverter either...

 

[davey_b]

There's an isolator between the meter and the consumer unit but that's it as far as I can see.

Do you mean the cable that runs from the inverter to the ac isolator in the loft is too thin? The grey twin and earth? What effect would that have? Could it cause the inverter to restart as shown in the videos I posted earlier?

Thanks guys, this is a big help and feel I'm getting to grips with it.

 

[oliver s]

The main concern is the white cable from the bottom of the isolator which should go to your consumer unit via another isolator and the generation meter this would cause conciderable volt drop and loss in generation but the main issue is the positive and negative the wrong way round from the panels this would cause the problems you have, don't be tempted to swop them round as there are no isolators and would be very dangerous, it definitely needs looking at by someone who knows what they are doing

 

[jason121]

Best not call the original installer then

 

[oliver s]

I would agree with that

 

[davey_b]

What do you mean about the white cable? At the moment that runs to the generation meter, then to an isolator and then to the consumer unit.

 

[solarfred]

The three connectors on the far left are positive DC inputs and the three on the right are negative DC inputs. There is a DC isolator inbuilt on the unit. No way of knowing if the cables are wrongly marked. It's just a nasty lazy installation.
How did they test each string without DC isolators - well of course they didn't. Normally if you reverse the input polarity the unit will not work but this puts out 3kw so who knows what is going on.
Is there a test report sheet in your customer pack?

 

[whinmoor]

What do you mean about the white cable? At the moment that runs to the generation meter, then to an isolator and then to the consumer unit.

Does it connect to the main house consumer unit or did they install a dedicated "mini-consumer unit"?

 

[davey_b]

No test report sheet that I can see.

Connects into main consumer unit.

 

[oliver s]

The white cable is way too small it should be 6mm the same as you would use for a shower/ cooker supply.

 

[davey_b]

The white cable is way too small it should be 6mm the same as you would use for a shower/ cooker supply.

Would that cause the inverter to give the errors shown in the videos?

 

[The Solar King]

DC isolator should be on left side of unit, slightly recessed, not in view on photos.

 

[1Justin]

We had the system installed last September. There was issues with low output and they came back in April and looked at it and output was better. Then I noticed that when it was at peak power it would suddenly drop to 0 and then climb back to nearly 4kw. It was saying "Vac 10 min too high" and "Frequency too high"

The inverter was replaced but now this inverter is saying "Failure fan2" so I'm going to contact solarmax about this tomorrow.

It's a Solormax 4200s inverter and we have 16 Canadian Solar CS6P panels.


As you can see we're getting full sun, it's pretty much straight on the panels and there is a wee bit of shading on the end panels and we're only getting 0.37kw!

Davey_B. I can vouch for that, and I have the same. It's pretty much expected as has been mentioned already. I have trees to the East.

I don't think many people realize the extent of a very small bit of shade till they see what you have there.

I did the same set of photos with mine one morning last summer, and I have a series of photographs every ~ 15 seconds as the panels come out of morning shade of a tree. I am on 2.3KWP, and exactly like your system, the reduction is massive, probably 90% loss, even with just a minute piece of shade left on one of the panels, - maybe half an A4 sheet. Loose that last leaf shadow, and Pow.. right up there!

I am on single string, some unspectacular Yingli's, and fairly conventional 4 year old Fronius, nothing extraordinary.

 

[davey_b]

DC isolator should be on left side of unit, slightly recessed, not in view on photos.

Do you mean the switch on the side of the inverter?

 

[solarfred]

S series | SolarMax