diy solar

diy solar

Looking to fix my neighbor's place... long story ahead!

Rednecktek

Solar Wizard
Joined
Sep 8, 2021
Messages
5,444
Location
On a boat usually.
OK guys, looking for a sanity check here. My neighbors in NE Washington are totally off grid and, let's be honest, their system is a nightmare. Since the guy looks after my place the 48 weeks a year I'm at sea, and doesn't mind me doing a little experimentation, we're looking to completely re-do his solar setup.

Right now it consists of a couple of old 100w? panels he got from a guy, and a couple of maybe 60w? panels he got from a drinking buddy, and a generator, and he's trying to get a micro-turbine on his spring, and a stack of old batteries, and a cheapie inverter, and... well.. you get the idea.

TL;DR - Total Clusterf$%k

The biggest challenges at the moment is money. He works a side job about a month a year and I have a real home and mortgage so we're working with a very limited budget. It also needs to be simple and easy to repair.

So, here's my thoughts:

I'm looking at doing a bulk buy of 20 used 200w panels (certified for 170w) running about 30VoC each. We'll run these in a 3s/4P setup to however much the controller lets us have, maybe a little light over-paneling since it is almost Canada and Idaho so sunny days are hit or miss. I figure if we use the 200w settings for all the math we shouldn't have any worries of frying anything. Whatever we can't use at his place will go to my camp and my garden shed at home.

I'm looking at the MPP 3048LV 3kw inverter/controller running at 48v nominal.

I'm looking at 8 of the 120ah Marine Deep Cycle batteries from WallyWorld for a pair of 48v banks.

The micro-turbine will come into play later when we figure out how to get it running reliably, find a 3phAC -> 48vDC rectifier, and another charge controller for it paralleled into the system. Currently a non-factor.

I know everyone here is rolling their eyes at the batteries. Go ahead, I'll give you a minute. :rolleyes:

I'm choosing those batteries for a couple of reasons. First: they're cheap and I'm already spending a lot on the panels and controller, plus costs for the wire and lugs and fuses and and and. Second: If a battery dies or he needs to add more, it's only a 3 hour round trip to get to WallyWorld, buy more batteries (they usually have a dozen or so in stock), and drive home. Waiting on deliveries to get to his friends place (nobody delivers to his house up the hill and down the dirt road) and then getting them up to his shed is a real PITA. Yes, I understand that 120AH FLA's are only good to 50%, which is why I'm spending just under $800 for 8 batteries. That's 5.7KWh (60Ah usable X 48v X 2banks) of usable batteries for $800 and another 2.8Kwh for only $400. If there's a better way that is in a comperable price range I'd love to know about it. Unfortunately that much Lith-Ion would require 7 very special (lotto) numbers and the Platinum Black credit card from my LearJet.

So does that sound like a reasonable setup? Is there a better battery option that doesn't require me having kids so I can sell them? Am I missing something in my math?

Thanks guys!
 
sounds like a plan..

do make sure you take care of the cold, but given the circumstances i can completely see your point.

what i would suggest, once the clusterf.. turns into a stable system you tell me to save ( or try to work a bit more) to start building a lifepo4 bank.

this setup should, if take care of, give him a 3 - 4 years to save...

for a relative small system, you could consider going 24v , as this will allow you to make due with less lifepo4 cells
 
OK, next question is a Check-My-Math question.

The MPP inverter I'm wanting to use lists optimum charging range from 60-115v, with a max voltage of 145v for its 80a controller on a 48v system.

The panels I'm looking at getting are listed as VoC: 36.3v, VMP: 28.71v, and Imp: 7.25a

By my math, using the VoC numbers, I can get 3 in series per set which gets me to 109v which is right in the optimum charging range, but 4 in series would get me 145.2v which is just over the limit and even worse when winter hits in the mountains and it gets stupid cold.

Calculating the 7.25a gives me 3 sets of panels, or 65.25a max, but if I do 4 sets I get 87a which puts me over the 80a limit on the charge controller.

With a 3000w inverter coming from a 48v system I *think* I need an 80a fuse? That's (3000/48)*1.25 for the math, or should I go bigger for surges?

The batteries are FLA deep cycle 120Ah batteries in a 4s/2p setup for 48v. Assuming a maximum of 50% usability that equates out to 5,760Wh or 120Ah@48v. Is that right? At half capacity on the inverter (1500w) that gives me less than 4 hours of runtime?

So, am I stuck at a 3s/3p panel maximum with an 80a fuse? Or am I miscalculating? It seems like I can only get about 2.5Kw worth of panel on this setup. Is that correct?
 
Last edited:
The panels I'm looking at getting are listed as VoC: 36.3v
Don't know how cold it gets there (pretty cold I bet) but you probably need to allow an increase of ~20% on Voc when doing your calcs. Voc goes up as temps fall.

36.3V x 3 panels in series x 120% = 131V, so still within the AIO inverter's Voc limit.

But check the specs on the panels - they will have a Voc temperature coefficient on the panel spec label, something like -0.3% per °C above/below 25°C. e.g. at 0°C that means Voc goes up by 25 x 0.3% = 7.5%. Substitute the spec for your panel but it'll be something like that.

Calculating the 7.25a gives me 3 sets of panels, or 65.25a max, but if I do 4 sets I get 87a which puts me over the 80a limit on the charge controller.

So, am I stuck at a 3s/3p panel maximum
No.

3 sets of panels in parallel = 7.25A x 3 = 21.75A
4 sets of panels = 29.00A
5 sets of panels = 36.25A
6 sets of panels = 43.50A

So you can have 9 panels in a 3S3P arrangement, 12 panels 3S4P, 15 panels 3S5P or even 18 panels 3S6P.

The more arrays in parallel = higher currents and wiring to suit. And a combiner box (or two) to fuse each string.


with an 80a fuse
As far as the battery fuse goes your maths is right. Surge current might need to be allowed for. That model is rated for 2 x continuous power rating for 5-seconds.

However it's not just the amp rating of the fuse but the very type of fuse that matters. Different fuse types operate differently and some are more suitable for DC battery set ups. Also make sure they are rated to a high enough voltage to make sure they don't keep arcing even if the fuse blows. I'm going to be using a high rupture capacity fuse on each bank of batteries.

Some fuse info:
 
Rednecktek said:
Calculating the 7.25a gives me 3 sets of panels, or 65.25a max, but if I do 4 sets I get 87a which puts me over the 80a limit on the charge controller.

Rednecktek said:
So, am I stuck at a 3s/3p panel maximum
No.

3 sets of panels in parallel = 7.25A x 3 = 21.75A
4 sets of panels = 29.00A
5 sets of panels = 36.25A
6 sets of panels = 43.50A

Uuuhhhh... I'm not following the math there. 3 panels in series gives me 109v and 21.75a per panel "set", right?

Wouldn't 2 sets give me 109v and 43.5a when paralleled with the first set? My math told me that 3 sets was going to max me out at 109v and 62.5a... or am I completely wacked on my math? If I put 4 panels in series together my VoC fries the charge controller.

Panels in series = volts X $n panels and amps X $n panels
Panels in parallel = Volts of each panel and amps X $n panels
Right?
 
Panels in series = volts X $n panels and amps X $n panels
No.

Panels in series: add the volts of each panel. No change to current.
Panels in parallel: add the currents. No change to voltage.

e.g.
36V / 7A panels
2 in Series = 72V and 7A
2 in Parallel = 36V and 14A
2 Parallel strings of 2 panels in Series = 72V and 14A
 
It's the same principle as with battery voltage and amp hour rating.
2 x 12V 100Ah batteries in series = 24V 100Ah battery
2 x 12V 100Ah batteries in parallel = 12V 200Ah battery
 
Aaaahhh, THAT's what I was missing. So assuming I can only series up 3 at a time to keep my VoC under control, I could set up 30 panels to get 109v and 72.5a total for 7.9Kw of solar.

OK, next question is the manual on the unit says the charge controller is good for 4000w of paneling. Does this mean that my best bet would be 6 strings? That maths out to 3.6kw, 109VoC, 43.5a in my head.

What would happen to the other 4kw if I only focused on the VoC and amps?

Also, there's a chance I may be going 100ft or more between the panels and the combiner box. At only 7a per set, 10AWG should be more than plenty for that length of run, right? Or can I go smaller realistically?
 
Last edited:
The extra watts would just sit there, as the sun rises and falls you would get more full power for a longer time .
technically you could add as Manny Watts Of solar to your system as you want , but I would go more then 20%
I’m thinking it can take 4000watt ? I think .
This makes sense for its size .
I would try to buy sams club golf cart battery’s they are 100 bucks each ant will hold up
don’t add new battery’s to old .
It seams like 4000watts is almost double what a230 ah battery pack will need .
I use 16gc battery’s and 4500 watts of solar .
 
there are 2 mppt chargers if memory serves me well.

The website indicates 1 x MPPT with a 4kW rating:
80A MPPT Charge controller , up to 4Kw of Solar input (60 to 115V DC Vmp)

Not sure why they say 80A for that voltage range and wattage limit.

The manual doesn't give the specs in a clear manner. Seems to be missing the array sizing section I have in the manual for my (similar) inverter.

I would be contacting MPP Solar to clarify the solar input limit for that unit because I think you'll be limited to 4kW.
 
10AWG should be more than plenty for that length of run, right? Or can I go smaller realistically?
I didn’t do the math but I’d never go smaller than 10ga
I’m <20ft and I use 8ga solar wire to the controller. It’s a pizza or two difference-ish dollarwise
 
I'm hoping the budget will allow me to set up a 3rd battery bank so he'll have 180Ah of battery at 48v. With it only being 7a and the fact that I have a 500ft spool of 10AWG sitting in the shed, I figure it ought to be fine. :p

No go on the golf cart batteries, it would double the battery cost for less amp hours, plus there isn't a Sam's on that side of the state and Wally World is only an hour away. Being as that I'm at sea 8 months a year, I'm trying to make this as low maintenance as possible within a budget.

I'm glad I'm not the only one confused about the wattage rating on a controller that beefy. ?

Thanks guys! Next step is to figure out when I'll be home and drop the hammer on the panels and inverter and tools stuff. I've already got a layout plan and when I get some quality time in Visio I'll throw it up here for review. I think I might stick with either 3 or 4 strings. I'll find out when I can get shoreside and we can go walk around with a tape measure and figure out how many strings we can physically fit. 18 panels sounds great on paper, but there's only so many places to stick 'em.
 
Last edited:
RNT your amp hours # look sketchy better check them out
the Costco GC battery’s are 100 bucks each 800 for 8 230 ah @ 48 volts
the Walmart marine batteries suck they are not deep cycle any thing with a cca rateing
is not deep cycle ? these will last . They are the best battery for thin money .
Lots of guys use them
6699D2AD-E52C-4413-AE5A-29CDF1291156.jpeg
 
It's a 48V model - 4kW/48V = 80A. The 80A is battery side.
The voltage range is solar side.
I understand that except the listed specification is ambiguous.

It should clearly distinguish the MPPT input specs from the battery connection and charge specs - it's mixing both in the same line.

Meanwhile, apart from Voc limit and MPPT operating range, the actual user manual is completely devoid of any information on the capacity of PV array which may be connected, so we are left with an ambiguous statement on a website ad.

There's a whole section devoted to PV arrays in the manual for my AIO but is missing from the manual for this model.
 
Back
Top