How do I size panels for SmartSolar MPPT VE.Can 150/70 rev2

[Deliotron (Gab)]

I have this controller integrated in an easysolar 24v 3000va system and has 2 sets of MC4 connectors for the panels.

At the moment I have 3x 240w panels in series on each line, non fused, for a total of 1500W.

From my understandings I can max the controller to 2000W, can I put all my 240w panels on one string in series (I guess fusing them) and have a different wattage panels in series on the other line to maximize my controller?

I am a bit struggling about this and any help clarifying my options be very appreciated.

Cheers

[RoarPowerNZ (RP)]

Can you post a pic? I thought the EasySolar 24/3000 had M6 bolts for the PV input?

The MPPT controller rating is shown on the front like 150/70 or 250/100.

For a 250/70, you can put a lot more than 2000w on it, but you will get ~2kw going to the battery (nominally 24v, at the MPPT max of 70A = 1.68kw)

You don’t say what your panel specs are, but i’d guess ~40v panels, so 6 in series is still fine (6 x 40v = 240v, vs the mppt limit of 250v) but check your panel specs carefully, including calculating your max PV Voc at min temperature )

Victron suggest overpanelling to 130%, so this would mean 2000w x 1.3 = 2600w, divided by your panel wattage (240) = 10.8 panels. Now 10 panels in series is far too much voltage, but 5s2p (5 in series, with another string of 5 in parallel) might be an option.

Definitely fuse an array like that. Your panels sound like about 6A Isc, so your array Isc is 12A, well below the 35A “max pv short circuit current” for that MPPT controller.

Come back with any questions.

[RaySun]

^^^ That mos def ^^^

[Deliotron (Gab)]

Thanks for the reply!
I’m puzzled too as in settings it clearly says it’s a 150/70 and have the MC4 input, can’t find anything like that online… maybe is because is a rev.2?
Can try take better pictures but unit is jammed at the back of the easysolar and can’t access easily.
In pictures you can also find the sticker for the panels, I have as said 3 in series on one pv in and 3 the same in series on the other.
I have the possibility to buy for a good price some suntech panels to add to my system, last picture is all the info.

[RoarPowerNZ (RP)]

Ok - my mistake, you pic of the EasySolar shows it clearly as an EasySolar with the 150/70 mppt.

So now you collect your panel specs; Voc = 37.8, Isc = 9.02.
Now a quick calc says 150v (mppt limit) / 37.8 = 3.97, ie 4 panels would put you over the MPPT controller’s voltage limit, but 3 would be ok.
In practice, you first work out the Voc at the coldest temperature expected, but our headroom here is almost 38v so we will be fine. Our table looks like this now;
3s1p = 113.4v, Isc = 9A
With a 150/70, if we run 2 strings (ie 3s2p) we get this
3s2p = 113.4v, Isc = 18A, and might put out (6 panels x 240w = 1440w, divided by battery voltage = 1440/24=60A. (which is less than 70)
So at 3s2p we are still under-panelled.
If we go to 3s3p;
3s3p = 113.4v, Isc = 27A, and 9 panels = 2160w, / 24v = 90A - over the 70A limit of the MPPT controller so we are overpanelled.
How much? We have 2160, spec sheet says 2000w, 2160/2000 = 1.08, so we are 8% overpanelled.

This is totally fine, so 9 panels is probably your sweet spot. Your controller might clip (limit the battery current) at the best part of the day, but with only 8% its likely you don’t actually make it to clipping anyway (if you redo the calcs using NTC (Normal Test Conditions) instead of STC (Standard Test Conditions) you will probably be not even at 100%.

Lets see what happens at NTC, 3s4p;
Doh! the panel specs only show STC, but we can apply typical curves or just find a similar 240w panel. I think at 800w/m^2 that panel will probably be around 190w, and Isc will be ~7.5A
3s4p = 113.4V, Isc = 30A, and with 12 panels (@ 190w) we are pushing 95A to the battery, but obviously clipping that to 70A due to the controller. Our array at NTC is about 2280w, / 2000w, = 14% overpanelled - also fine.

@Community: Let me know if i’ve made any mistakes
@Deliotron : Come back if you don’t follow

[RoarPowerNZ (RP)]

I forgot to mention the MC4;

Some controllers make it easy for you by having MC4 connectors. I don’t use these, as they derate the unit (MC4 typically require your design to limit the current through any particular MC4 to 30A, so in this case if there are 2 pairs, then 60A is your input current limit.
But as you saw from the previous reply, even at 4 strings in parallel, we only have 30A total.

So to know how to wire this, we need to look at the MPPT wiring diagram - these MC4 could be in series or in parallel.

In the manual this line is important:
“For MC4 models: several of the solar charger’s MC4 pairs may be needed to parallel solar panel strings. Be
aware that the maximum current through a MC4 connection cannot exceed 30A”

So … its clear (to me) that if you put the whole array on one pair of those MC4, you would be asking for trouble (with a theoretical 30A current, you could get higher currents - look up “edge effect” and “high temperature solar panel current”), so we have to split our array.
You could run 3s2p into each MC4 pair, or if you ended up with 3s3p, just put a 3s1p into one, and a 3s2p into the other. Because the voltage of the array is always the same (3s = 113.4v), it doesn’t matter that one pair of MC4 will handle 2 strings while the other MC4 handles 1 string.
Also, internally those MC4 are wired together (in parallel), so we are only doing this to manage the 30A limit on the MC4.

Correction: I just noticed your panels are 265w, not 240w, but really, a panel that old is probably only pushing 240w by now, so the difference in the math is not very consequential

[pwfarnell]

You can run the Victron solar calculator. Input your MPPT, use advanced panel settings to input the panel details if the database does not have your panels.

[Deliotron (Gab)]

Oh man, you are a LEGEND!
I’ll sit down later to absorb it properly but I think it’s pretty clear math you done there.
I’m gonna go 3s2p with my existing ones (fusing 20 amps on the positive before the breaker, correct?) and 3 of the suntech in 3s1p on the other leg non fused.
They are old panels indeed, same as the ones I’m getting, so they not cranking in full force.
Plus I’m in tassie inland, not much sun LOL.
Thanks again, tomorrow I’ll pick the panels up and will wire the lot over the weekend , will let you know how it went.

[Deliotron (Gab)]

Oh cheers, will look into it!

[RoarPowerNZ (RP)]

Ok - you might want to check the numbers.

Those MC4 are wired in parallel, so if you put non identical strings on each, you can have problems.
In this case, with old panels, you will probably be fine.
But consider a case where one string of 3s1p has a Voc of say 120v, another has a string of 3s1p of a different panel that only has a Voc of 90v. The currents will be different too, but in this case the voltages are the issue.

Do you have the option of all one brand/model?

Edit: We are in Stratford, Taranaki, NZ, so I somewhat understand the solar conditions you have. You will be about 800w/m^2 most of the time, so use NTC (if its available) to calc your expected performance, but still use STC to calculate your safe array specs.

[Deliotron (Gab)]

Can’t get them all same brand but I can get 6 of the suntech to put along the other 6 I have? The guy selling them is suggesting the setup in picture.
I’ll sit down and do the maths tomorrow, I think I understood where the limitations are.
Plenty of material to work with, thank you!

[RoarPowerNZ (RP)]

Those isolators (in your diagram) are on the roof.
This used to be the standard in Australia, but not in NZ.
Background: Although NZ and Aus share one set of standards (“AS/NZS”), there are a couple of places where there are separate rules, and this is one of them. Recently, this particular “Aussie only” standard (of isolators required on the roof) was brought into line with NZ, and its due to the large number of fires on roofs caused by faults in the isolators.

So in my opinion (not being an expert on what Aussie inspectors will/wont pass), you should bring the isos down to right next to the SCC (Solar Charge Controller), where the DC circuit breaker currently is.

If you use an SISO-40 isolator like this next to the SCC;

… you won’t need 2 separate roof isolators, as this iso is rated at 40A.

Usually you would use 4->1 connectors on the roof, then one cable down to the SISO40.
For high current panels, you might run two cables down so that the current is spread across twice as many cores. In this case you would have 2->1 connectors at the panels.
In your diagram, there are two cables coming down from the roof - if you are running 4mm cores, this would only be necessary if the run was long - 4mm is fine for 32A over a short run.
If the run is longer, jump to 6mm. There are heaps of voltage drop calcs on the web, just keep that drop under 3%.

That double pole DC breaker is not necessary for AS NZS code if you use inline MC4 fuses at the panel end (which i would recommend) and the SISO-40 at the SCC, but it wont hurt.

For parallel strings, the rules say you must have over current protection, and isolation, and a disconnection device at the panels.

Over-current: MC4 fuses provide this, and each “leg” (subarray) of the array has its own. They are only about $20 each and can have the fuse in them replaced (but i’ve never seen one blow).
Isolation: Your on-roof isolators count, as does the double pole DC breaker, but like i said, the rules have changed and you are probably better to not have the isolators on the roof unless you can be very very sure the housing ingress protection is top-notch and the iso is top quality. You generally don’t get a second chance with these, as when they start smelling and smoking, you can’t see them (as they are on the roof).
Disconnection: Your MC4 connectors count as a disconnection device (they qualify because they require a tool to be disconnected) - you can disconnect to do maintenance/change the array etc. The “isolators on the roof” used to provide this.

If you are intending to get your system inspected, run the plan past your sparkie first - simple things like what must be in conduit will end up being a half day to fix if you have to go back.

Another example : At least in NZ, there is only one grade of conduit available from most electrical suppliers, and its the solar grade (must be marked with “SOLAR” every 1m i think, and is thicker than std with more “spade/hammer resistance”), but again, if you find a supplier of non-solar grade, and install that, then the inspector says rip it out and redo it, you’ve wasted a lot of time.

As mentioned before, making an array with sub-arrays from different panels won’t meet the code unless they are very closely matched (which these are not), but i’ve seen many older arrays have a mismatch of panels due to a smashed panel or a shortage of identical panels, and perform fine.

[Deliotron (Gab)]

Finally had a moment to sit down and read it all.
I wouldn’t mind keep 2 isolators anyways as they not too bad price wise.
What you reckon of my sketch? Did I got your information right?
This is just for me to understand and order the correct parts, I will do the actual job with a sparky but would like to waste less of his time as possible as it will probably doit as a favour.
Not sure 6 mm be enough to avoid drop but will calculate that properly tomorrow using the info you gave me, I need to measure the length if cable once I’ll have the frame down.
20 amp fuse for the panels be enough ye?
Thanks bud, I feel like I’ll have to make it square with you somehow after all this time you taking… lots of ppl charge a lot for this kind of help… will figure something out.

[RoarPowerNZ (RP)]

Looking good.
I would use these fuses - really easy to use, really good at remaining sealed, but still have replacable fuses inside them.

You have a 0.9v difference between subarrays - some inspectors might fail that, but on older panels where the time-based degradation is variable, its probably no problem. I’ve seen brand new identical panels where the difference between panels in the same sunlight was ~5% (which was ~2v), so that is the basis of my opinion that the voltage difference will be fine. More experienced designers; please chip in!
6mm will be fine for this array - although we are aiming for <3% voltage drop, you will only be hitting that max current (and therefore the max voltage drop) on the best of the best days with panels of your spec and age.

With your two isolators - an inspector might want these to be;

• located close(ish) to each other (ie so that you can’t find one iso and be oblivious to the existance of the other)
  AND
• labelled with a warning something like “isolating one subarray DOES NOT ISOLATE ALL SUBARRAYS”
  ie consider the situation where someone unfamiliar with the setup isolates one sub array and then opens a terminal thinking they have killed the whole array, creating an arc and getting splattered with molten copper.
  Talk to your sparkie first - some guidance on the regs will avoid you cutting the cables to one length and then having to re-do the job.

Interleaving; (probably should be its own post)
Most panels come with ~750mm leads. The typical way to draw an array is from left to right as you have drawn it. This calms our brains, but its not optimal to do this physically. In your array, if the left most panel is #1, and the right most is #12, the links (eg the cable between #1 and #2) needs to be clipped up. Your panels will be 20mm apart (mid-clamp width), with the connectors being about 300mm apart (depending on where the + and - block on the back of the panel are), but you have 750mm x 2 = 1500mm of cable! - all that cable has to be clipped up, if it touches the roof, you fail your inspection (for good reason - encouraging birds nests, obstruction of leaves which leads to roof-rot, in a van it will tap tap tap when driving, and could eventually rub through the insulation and short on the roof, etc)

Interleaving is where, instead of the panels going 1,2,3,4,5,6 etc, they go something like
1, 3, 5, 6, 4, 2. Think of it as connecting odd numbers heading away from the SCC, turning at the end, and connecting evens coming back.
The net result is that;

1. you can end up back where you started - this means no single wire running the length of the array
2. all that coiled up cable under the panel gets used
3. cheaper - due to less cable being used
4. tidier - due to no loops of leads when you are connecting adjacent panels.

Now … this is usually more relevant on a 12s array than your 3s subarrays, but it will gain you some cable.

This might not even be possible for you. On modern panels the connectors are almost always half way down the panel, but on older panels they are at the top, as you have drawn. Panel manufactures learnt this a long time ago - things that have rotational and mirror symmetry are easier to work with. You drew your panels with the leads at the top, so this might not work for you, but readers with newer panels, stick with me here.

So imagine your 10s (10 in series) array. Instead of running the positive to panel #1, and your negative ~10m away to panel #10, you cable like the lower half of this diagram shows;
[ Ignore the panel numbers in the 2nd part of the diagram. Diagrams.net is what i use, but it has weird object-numbering rules, so rotated panels are out of order ]

Note that every second panel is rotated 180.
Interleaving can also be done from the center, so where the only option for the cables from the isolator is to pop up in the middle of an array, you can send the positive to the left, turn at the end of the array and head back to the middle, and end up back where you started.

There is also a benefit with interleaving in that you aren’t creating antenna loops, but that’s too deep for this post.

So for you, if its possible you would move your panels so that the positive goes to panel 2, then 1, then 3 (effectively shortening the positive) and then your right most sub array does similar, and
same for the center sub arrays, and you save about 4m of cable.
Let me know if your leads are long enough to go from one panel, skip a panel, and to the next panel - if so i’ll write you a diagram that minimises the cable run.

[Deliotron (Gab)]

I actually just scored 12 panels all same brand and model for 20 bucks each! In the next days I will test them with a multimeter but they look pretty good. I struggle a bit to see what you mean with a different disposition but I’m not stressed if I have to get a bit more cable plus the array will be on the ground. I will definitely put the iso close to each other (safe in the room where the inverter is or better to keep them outside you reckon?) and label them properly. Any suggestion for an isolator, pref on ebay?
I don’t think my panels will allow to cross connect, will check and report back but again I wouldn’t stress doing it differently of what’s in the diagram to spare some cable… you should definitely open a post tho to discuss the possibility for the modern panels as that sounds very interesting tho!
Exciting stuff indeed, now that slowly things start to make sense it’s even more fun (always respecting the element of course… not keen dying for a stupid mistake, I’ll harass a sparkie any day instead ).
Pic of panels stickers, not new but they will do for now (building a house within next 3/4 years and by then I’ll do a proper solar system, this is a test run to learn and run basic stuff).

[RoarPowerNZ (RP)]

Well done getting a set of matching panels.

Checking your panels;
Equipment:

1. a multimeter and a clamp (for measuring current), or a multimeter with a 10A current capability.
2. a pair of MC4 plugs with a breaker between them. Cut the barbs and guards off the MC4 plugs so they are easy to plug/remove.

This is my one, sparkies/installers will have a better one. The current clamp goes where the carabina is. My one runs the risk of catching the breaker on fire because i’m constantly disconnecting (turning off) the breaker “under load” - see YouTube vids of this happening. Also see vids of why AS/NZS no longer allows polarized breakers.

• Choose a sunny day - you don’t want clouds coming over making a panel look bad.
• Lay them all down on the ground, with a brick under the top edge to tilt them towards the sun, and the leads out on the ground where its easy to get to.
• Wait for them to all get to the same temperature - probably only ten minutes, but the ground under them needs to be roughly the same temp (this affects the cell temperature a lot if one panel is on hot concrete and the next is on cool grass)
• As quickly as possible, go from panel to panel and measure the voltage from the leads. Write this on the back of the panel or the frame as “Voc = xx.x V”
• Now go back and connect some kind of current measuring device. If using a multimeter you will need to be quick - they can’t handle too long if the panel does actually produce 8+ amps.
• Write the current on the panel too, “Isc = x.xx A”
• Again, move as quickly as you can so that conditions are as similar as possible.

Now you have a table of panel actual specs, and we can work out which sets of 3 make most sense to put together.
This is where i’m hoping some other forum members jump in and suggest how to best match up panels.
I would expect that you want the highest current panels together - one low-current panel in a string limits the performance of that sub-array, where as a low voltage panel makes almost no difference.
So i think you would rank them by current, split the table into 4 groups of 3, and call those the A, B, C, and D sub arrays, but mostly we are filtering out a dud that looks ok but has suffered from bad hot spots, fractured bus bars, or similar non-visible wear or damage.

[Deliotron (Gab)]

Sounds good, will definitely do that asap, just have to wait for a sunny day as it’s been pouring rain non stop.
Fingers crossed there’s no dud but, if I find one, the guy that sold them to me have another couple sitting there.
Will get back with numbers at hand, thanks champ once again

[RoarPowerNZ (RP)]

[Deliotron (Gab)]

Will 2 of these do or are overkill?

[RoarPowerNZ (RP)]

One of those will do.

There are 8 terminals in them, with 4 contacts.
1 3 5 7
/ / / /
2 4 6 8
The “/” represents the contacts
If you are bringing a cable from each sub array to the isolator, you would combine pairs at the terminals, e.g pos from A and B go into the left most terminal (1) , neg from A and B into the next (3), then neg from C and D into the next (5), and finally pos from C and D into the right most terminal (7). Two wires in each terminal at the top, and just one wire out each terminal at the bottom. At the bottom you would join the 2 positives together (from 2 and 8), and the 2 negatives together (4 and 6).
In this config, pos and neg are broken for all sub arrays at the same time through one contact each.

If you have join two sub arrays with 2->1 connectors, so A and B are already joined and C and D are already joined, then you have the same situation, but just one wire into each terminal - then do the same, A/B positive into terminal 1, A/B negative into terminal 3, C/D negative into terminal 5, C/D positive into terminal 7.

But … if you can, actually go in the bottom of the IP box rather than the top - then the quality of your gland doesn’t matter so much because water can’t track up/down the jacket into the box.

Another option, if you have joined all 4 sub arrays at the arrays (with 4->1 connectors) is to put the A/B/C/D positive into terminal 2 (the bottom), install the links (they come with the SISO-40) to bridge terminals 1 and 3, and then your positive “out” comes out terminal 4.
Negative would go into 6, bridge 5 and 7, and out 8.

In this config, pos and neg are broken for all sub arrays at the same time through two contact each, but those contacts are carrying double the current.
I haven’t done the numbers but if that current is too high (>32A) then make up some more links, and put pos into 1 and 3 (bridged) and out 2 and 4 (bridged), and neg into 5 and 7 (bridged) and out 6 and 8 (bridged).
Your sparkie will check this for you.