Rejigging my PV

[wildebus]

Thinking of doing something with the Solar as not really touched it since fitted the panels around Oct '17.
Been musing on adding a tilt, or having a sliding double-decker setup (and have actually got an linear actuator 'in stock' in order to save having to get up on the roof).
But .... I don't think I actually need that much more solar to make it worthwhile to be bothered to go to that much hassle, especially with a decent sized battery bank.
So .... maybe just add a couple more panels to the existing array instead, rejig how they are all cabled up and just leave it as a "fit & forget" setup?

Step 1: Whip off the panels so I can give both them and the Roof a damn good clean!

You would think the roof under the panels would be fairly clean but not really and the small exposed area between each panel was really mucky!

Bit more of a close up here

It looks just about as dirty as when I first got the van and spent a day cleaning the roof!

Luckily it only looked as dirty and this time only took a couple of hours

(still need to remove the rear rack to get to the roof under that to clean it)

Now a message to Robert @RAW ... What I have done here reinforces why I am like a stuck record at you whenever you wonder about how to mount panels on your new Sprinter ... Because I started off with initially fitting a rail system onto the roof, it took me less then 30 minutes to unbolt, uncable and remove all 4 panels off the roof using a single 17mm spanner. If those panels were secured directly and individually into the roof, it would be a whole new ball game! If I were changing the panels for different ones of a different size, all I need to do to mount the new ones would be to slide the captive bolts along the tracking to suit the new size, rather than having to fill in existing holes and drill new ones (or spend an age scraping off old adhesive and squirting out new glue).


Next step is working out what there is room for on the roof to add extra panels and decide how to cable them all up ... Stay tuned

 

[Millie Master]

As for cleaning the roof, indeed anything you would like to mention, I honestly do find that the Bilt Hamber produced (UK company UK produced) Surfex HD is utterly brilliant.

I bought a 5 ltr. bottle of it and now I have used it on numerous different jobs including the cleaning of Millie's filthy roof at a significantly reduced strength (following the instructions), I now have a squirty bottle filled with a Surfex HD/water mix at about a 10 water to 1 Surfex HD and after giving it a shake before using it each time, I use if from cleaning the alloy wheels, squirting a goodly amount onto our induction hob before agitating and finally for squibbing into tea/coffee and calc stained mugs to bring them up like new.

Totally off topic, but I thought you might like to know

Phil

 

[wildebus]

As for cleaning the roof, indeed anything you would like to mention, I honestly do find that the Bilt Hamber produced (UK company UK produced) Surfex HD is utterly brilliant.

I bought a 5 ltr. bottle of it and now I have used it on numerous different jobs including the cleaning of Millie's filthy roof at a significantly reduced strength (following the instructions), I now have a squirty bottle filled with a Surfex HD/water mix at about a 10 water to 1 Surfex HD and after giving it a shake before using it each time, I use if from cleaning the alloy wheels, squirting a goodly amount onto our induction hob before agitating and finally for squibbing into tea/coffee and calc stained mugs to bring them up like new.

Totally off topic, but I thought you might like to know

Phil

Sounds good .
I should have mentioned that my "go to" cleaning product for this kind of ingrained muck is AutoSmart G101. Superb product. I made a fairly stong mix of 1:10 of G101 with hot water in a spray bottle (hot water makes it much more effective - might be the same for Surfex? maybe try a hot mix to see?).
Also, like Surfex (I am guessing, seeing as you use it for cups), G101 is food-safe and is great on kitchen counters and especially stainless steel.

What I haven't cleaned yet are the big lumpy spots of underseal on the roof (yes, underseal on the roof. It got there when I was spraying the wheel arches and the underseal can exploded. So to get underseal on the roof when spraying under the van, you can imagine how much mess it made of the side of the van whilst traveling in all directions )

 

[Millie Master]

Personally Dave, I would use a rag soaked in petrol to dissolve the underseal on the roof, and then when it is all good and clean I would give the entire roof a darned good canubra based polish.

Phil

 

[wildebus]

Personally Dave, I would use a rag soaked in petrol to dissolve the underseal on the roof, and then when it is all good and clean I would give the entire roof a darned good canubra based polish.

Phil

I'll give that a whirl tomorrow I used thinners on the side of the van after the disaster happened, but that was when it was still all pretty wet. I thought I cleaned all the underseal but some must have gone under the panels there.
Got some good wax (can't recall the name but I think it is a canuba wax) that I use on wheels and was planning to apply to the roof to try and keep it from getting so bad.
EDIT: This is the wax I use ... https://amzn.to/2xCHwAG - Collinite 476S.

 

[wildebus]

Put a 400W radiator on for a while in the van and then turned on the 2kW Water Heater on around 10PM - that ran for around 10 minutes and will kick on again before sunrise tomorrow. Plan is to take some power out the batteries so the solar will be doing more than just ticking over so I can evaluate the setup I am trying right now.

Test 1 is one half of the possible new array: Got a 100W panel + 100W panel + 30W panel. Will be able to check the total power and very easy take out any panel from the array and see what effect it has on the total.
What that should do is tell me the difference between the EcoWorthy 100W panel and the 30W Victron Panel in terms of harvesting.
What I will also do if I remember is swap one of the 100W EcoWorthy panels I fitted in Oct 2017 with a new 100W EcoWorthy panel and see if there is any difference between the two (any degredation in the 2.5 years? or improvement in cell manufacturing maybe?)

I will also out of curiosity setup the panels in both Parallel and Series and compare the results. Right now ready for the morning the three panels are setup in Parallel Mode as that should be the best way for these specific panels.
Basic rule is that you ignore the wattages, but look at the Voltage & Current specs and combine based on the closest matches. My 100W panels are 21V and 5Aish; the 30W Panel is 22V and 1.3Aish - so the voltages are very close, which means you connect in Parallel and the final voltage drops to the lowest in the array. If the currents were closest, you would connect in Series and the total current would drop to the lowest in the array. So when I try it as a Series configuration, I should, if the mathematics is right, reduce the maximum potential power pretty dramatically! (not a desirable result!).

 

[RAW]

I will also out of curiosity setup the panels in both Parallel and Series and compare the results. Right now ready for the morning the three panels are setup in Parallel Mode as that should be the best way for these specific panels.

I am testing out those cheap flexi panels still, connected in series at the moment as the outputs are pretty much matched.
Also if you wanted a big panel, not sure it would fit though (2015 x 996), like standard commercial panel size, Ian at Xerogrid is selling these for around £130 a pop

Datasheet here

 

[wildebus]

Looks good value.
Not keen TBH on fitting a big single panel. Not convinced it would be robust enough for a vehicle and all the shaking and vibration it would suffer.

 

[wildebus]

Today looks like a good day to do some comparisons - currently virtually cloudless so whilst ultimate performance may not be measurable due to panel orientation etc, it should be fairly easy to identify the effect of making changes by comparisons.

First test today .... Does adding a 30W panel to a 100W+100W array make a difference?
The day started with the 3 panels all connected and the graph below shows the progress as the day started.

Where I have marked the timeline with *** is the point where I disconnected the 30W panel from the array and there is a definate drop in power harvested. The trend line continues to rise but at a lower wattage point.
(PS. the regular spiked dips should be ignored - that is the Venus GX sampling the data just at the same instance the MPPT controller momentarily turns off to record the voltages)

I will shortly put the 30W panel back and remove a 100W one ...

 

[wildebus]

Personally Dave, I would use a rag soaked in petrol to dissolve the underseal on the roof, and then when it is all good and clean I would give the entire roof a darned good canubra based polish.

Phil

The petrol worked well, although I had to make sure I cleaned the area immediately of the residue after as within a few seconds there would be a brown smear left that was really hard to shift!
Sorted now though will be taking the roof basket off tomorrow and then washing the roof properly with some shampoo and then give it a good wax ready for the panels to be re-fitted (altering the fixings on the panels as well at the same time in order to close up the gaps by pairing the panels).

 

[wildebus]

Did a few tests in the morning. I did a change in the afternoon but it was hard to really evaluate as it coincided with the sun being blocked by a large pine tree (Happens each day and costs me a potential 50Ah of charge each time).

The day started off pretty cloudless so could get a very consistant initial reading. The back end of the day was similar with limited cloud and that allowed me to superimpose the kind of harvesting that could be expected if there were no shading from anything and no changes made to the setup.
This also meant it was quite easy to see the differences when making a change. The Graph is below with the annotations on it

The Purple lines show when harvesting is impacted by shade -
1) the house roof in the morning (I park north of the house and towards the west side, so don't get any of the early sun)
2) there is a large pine tree (taller than the house) WSW of the van and that really hurts the harvesting in the early to mid afternoon
But with the curve imposed on the graph, the impact of taking panels away can be clearly seen at the non-shade times.
The effect of changing a mis-matched panel array from an all-parallel setup to an all-series setup co-incided with the pine tree effect so it was very hard to judge. I'll have to repeat that test another day between 9AM and Noon to really judge that difference.

I am wondering when I was changing all the connections around if I connected the 30W panel back properly at 12:00 as the harvesting from 12:00 to 12:30 is a direct pattern continuation from the earlier line when I first removed the 30W panel.

 

[MarkJ]

Your pine tree reminds me of a local issue we have round here. A broadband company is putting up masts around the village (which they are entitled to do under recent planning law changes, but that's another story). However one of them has been put under an oak tree so its top is lost in pleasant, soon-to-be-leafy branches. I guess the plan said put it here, so that's where they put it.

The solution put forward? Cut down the oak tree....

 

[wildebus]

Nota good solution there is it
I would not mind cutting down the pine (branches break off in high winds and nearly land on the van as well!).

 

[wildebus]

Took the roof basket off, and then after cleaning the underseal with petrol, I then gave the roof a proper wash with car shampoo and then applied some Collinite 478S Wax (I use this on the Alloy Wheels, including the inner wells, to make them much easier to wash) on the entire roof.

Clean Roof! by David, on Flickr
Much better!
You can see from the shadow cast by the house roof at 10:30 that where the van is parked, the window of opportunity for solar harvesting is much shorter than the actual daylight duration

Next Step - redo the cables on the smaller Victron panels so they reach the main group and the block connectors

Victron PV Panel by David, on Flickr

Using 4-way combiners to make parallel groups of 3 panels

Cabling Up! by David, on Flickr
Also using some dielectric grease to keep the connections free from oxidisation.
All these connectors sit under one of the panels for ease of maintenance rather than being dotted around the roof.

Then it was time to drop on the last panel over the connectors and enable the charger again ....

 

[wildebus]

As my memory is pretty unreliable when it comes to remembering what I have done, made some notes in case I need to check anything later on ...
So now I have got 6 panels, and added a tag on the photo so I know which is which:

Clarence - PV Panels by David, on Flickr
Waiting on a set of brackets for Panel "P2 -A" to secure (I took them off this panel so I could use them for the two new panels I installed)

And wiring as follows:

2020 Clarence Solar by David, on Flickr
Setup so I have a pair of 100W Panels and a 30W panel all connected in Parallel (Parallel 1 = P1); Same again for Parallel 2 (P2). These would each appear to now be a 12V (22Voc) 230W Panel. Then connected the two Parallel arrays in series, which doubles the voltage, so now the Victron MPPT Controller sees a 24V (44Voc) 460W Panel.
The Victron MPPT 100/30 is rated at 440W, so 460W is marginally higher then the rating, but this will not matter as both the Voltage and the Input Current of this "Panel" are within its parameters. If I get full harvesting power (which will happen in the summer at high noon a few times), all that will happen is the output may be capped at 440W and I waste 20W. I would expect in reality the Controller will be speced to deliver slightly above the rating, so I may still get that extra 5% above the quoted maximum anyway.

 

[wildebus]

So ... Why not just do a tilt setup? And any shading issues?

Shading first ...
This is the layout on the roof now:

Clarence - All back on by David, on Flickr
At the front there is a Light Bar just ahead of the first panel. I positioned the panel so there will only be a shadow from that light bar when the sun is very low, by which time the harvesting will be pretty gone anyway.
At the rear there is a roof basket and a fan with a rising cover left of the two small panels. Similar to the light bar, the roof basket will only cast a shadow when the sun is very low so no real impact (if there is tall stuff on the basket it could be a different matter, generally no effect).
The fan cover when closed is not a problem, but if the cover is raised, then that will cast a shadow (if the sun is on the left of the van).
However, because of the way the panels are connected, I will only lose the output of the 30W panels so no worse off than before.

Tilting...
I have considered fitting a tilting setup for a long time, but each time came to mind generally decided against for a couple of reasons.
1) I just can't be arsed to get on the roof and raise and drop panels. Even if the lifting was electric, I would still need (or want anyway) to go up and secure from wind gusts
2) With a tilting setup you have to park in a orientation that suits the tilt unless you have a setup that will tilt in muliple directions. I really didn't want that constraint. I might want to park so I can have the side door opening to a particular view (or making sure I avoid a particular view for that matter!). Sure you could just not do the tilt in that case but you may end up with a setup only optimised 25% of the time.
I know quite a few people have tiliting setups and for them it is the dogs dodahs, but just not for me .

I'll shift the van in a day or so to try and position away from the House and Pine Tree shades and see how she goes in a full day

 

[wildebus]

Waiting on a set of brackets to secure one of the brackets (more of that later). When they arrived today, fitted and then shifted the van about 8 foot eastwards and 4 foot north. Much difference? Well, I would say so!
Moved it about quarter to four ....

Also checked the situation with potential shadow when the fan cover is up ... Raising the fan drops the harvesting by 12W which is pretty acceptable.
And of course that will only be the case when the sun is low AND to the left of van.
So rejig complete and I *think* has been worth it, but tomorrow, with the van parked out of the house and (mostly) tree shade (mostly as the tree is VERY tall so hard to be clear from it) I will be able to evaluate a full days of harvesting

 

[SquirrellCook]

I see you using terminal grease Dave. Would that be the last grease you’ll ever use? Sorry I couldn’t resist.

 

[wildebus]

Fair bit of cloud in the sky today so the solar harvesting is coming and going a little bit, but pretty happy with the setup as it now stands and looks like everything is working correctly
So far today I have hit a peak of 371W Solar Yield, and a current peaking at over 28A into the batteries which is pretty funky for April in Scotland