DIY Remote Solar Power System

[Hiro]

From post #55

For GT panels to work in a off grid scenario you use MPPT controllers and series wire panels to a higher voltage, (1 1/2 to 2X) than your nominal battery voltage. When you are pushing this a long distance to your controller you can use higher voltage, up to your controllers limit, to keep amperage low and minimize voltage drop over long distances.

From post #57


Your proposed battery bank, at 420 Ah would like to see from 5 to 13 % charge rate. 5%, or 21 amps, being only good for intermittent ie; weekend use. Any full time off grid scenario wants to see 10-13% so 42 to 55 amps would work best. Charge voltage needs to be set accordingly for flooded or VRLA, (AGM) batteries. Stay away from Gel batteries. They don't work well in off grid situations.

I think these are what you are asking for. I realize this thread is getting pretty long winded but I cant change the chronological order of the postings.

From Post #48 as well
Not sure I understand your PV wire sketch. Text at top says 48V 24.7A
Is that the rating for each panel, or each string?
For a MPPT controller to work well with a 48V battery bank, you need to supply at least 1.5x battery voltage, up to 2x battery. So 73V of array would be the least, up to 96 for the best efficiency in the Controller. Going higher voltage may be desired if you have long runs ........

I stand corrected.

[Hiro]

No you are just not understanding.



You tell me.

MPPT Output Current = Panel Wattage / Battery Voltage.

2400 watts / 50 volts = ??????

I get 48 amps. So a 60 amp MPPT will work.

Yep got that when I said; I need the MPPT to convert it to 48 amps @ 50 volts (2400W) in Post #71

OK so what is your point. It takes what it takes. The sky is the limit when it comes to off-grid budget. That is the choice you are making.

No point, just a whinge...

This is getting old, and I am not going to go back and keep running numbers for you. Everything starts with determining how many watt hours you need in a day. You cannot pass GO until you do that.

With further measuring, I decided to work on 5.8kWh/day in my next diagram. Measuring the Wh was not accurate enough for me, where as with the kWh usage I have much more accurate data through the powermeters I bought.

So let's say you need 4 Kwh per day. Immediately you know you need a 20 Kwh battery. At 48 volts is 20,000 Kwh / 48 volts = 416 AH. So how did you come up with 48 volts @ 420 AH battery? Only way is if you need 4 Kwh per day. It is what it is.

Next you must have enough panel wattage to generate 6 Kwh per day to replace the 4 Kwh per day you use. To find the Panel Wattage you must know the Sun Hours received in the WORSE CASE MOTNTHS. For you that is the Months of June and July. To come up with 2400 watts can only mean you must have used 2.5 Sun Hours.

Again Panel Wattage = [Daily Kwh x 1.5] / Sun Hours

So [4000 wh x 1.5] / 2.5 Sun Hours = 2400 watts. It is what it is.

But I can tell you made a mistake on your website. It is optimized for Grid Tied systems, or maximum Yearly Average. You cannot do that with off-grid, Well you can, it just cost a lot more money. If you change the Tilt Angle to optimize for Winter months, your Sun Hours will be higher than 2.5 hours. That means less panel wattage and less current out of the MPPT. Example if you change the tilt angle and it gives you 4 Sun Hours worse case, then your panel wattage and MPPT go to.

Panel Wattage = [4 Kwh x 1.5] / 4 h = 1500 watts
MPPT Amps = 1500 watts / 50 volts = 30 amps.

If I were you I would look into optimizing for winter months vs yearly average. Could save you several thousand dollars. But it is not my problem, it is yours to figure out. I bet you can probably get at least 3.5 Sun Hours in winter by changing panel tilt. Its your money, not mine.

Working it all out tonight.
Thanks heaps Sunking

[Hiro]

Did I get this right?
Screenshot 2017-03-01 23.02.14.png

[littleharbor]

Getting closer... Your Voc. at STC is 152 volts, Depending on your expected lowest temperature this Voc. could go much higher. Not going to fly with a 130 volt CC limit. You would need a higher Voc. controller

Your CC and load connections need to be on the same terminals and at opposite, dia
gonally opposed "corners" of the bank. Refer to the bottom left example here
a Battery MultipleVoltWiringDiagram.jpg

.

[Mike90250]

Additinally, with 3 strings of panels, you need a Combiner Box with fuses or Circuit Breakers, I expect them to be in the 12 -14 amp range, the sticker on the back of the panels should have a Series Fuse xyzA spec on it.

[Sunking]

Did I get this right?

Better but still some issues.

With 4 panels wired in series Voc exceeds the 150 Voc limit you show on the controller. If you add up the 4 panels Voc (38 volts x 4) you get 152 volts, By the time you apply Temperature Correction by multiplying by 1.25 you come up with 190 volts. You have a couple of options to fix that.

1. 3S4P strings. With three panels in series you Temp Corrected Voc = 3 x 38 Voc x 1.25 = 142.5 Voc which is less than 150 Voc limit of the controller..

2. The option I like better is to use a Controller with at least 200 Voc input limit. They go as high as 600 volts which means you could use 6S2P and not have to worry about using using fuses or breakers between the panels and controller.

Lastly I do not see any fuses or breakers in your drawing for the batteries. Look at this thread part 3.

[Hiro]

Thanks for that littleharbor I was a little unclear on the input and output terminals... Batteries facing each other will have terminals at opposites in reality so I want to keep the perspective realistic.

Thanks Sunking
Won't 6s2p make the system 237.5 VoC 17.27 Imp and make the amperage too low? Does the CC need a certain amperage into it to work efficiently?

I didn't put fuses in Batt. Bank cos the drawing didn't have space. I have spread it out to include them now.

Be patient people. It may get long winded, but some of you have been working with these figures, parameters & specs for years. It's ingrained in your brains. I'm working with them for 2 weeks and not put it into practice much, yet...
I'm a Master drywall plasterer who hasn't worked in plastering for 10yrs, but I can rattle of systems, appropriate materials for almost any job still. It's ingrained...
I appreciate your patience in having to repeat yourselves sometimes... it WILL sink in... Screenshot 2017-03-02 11.08.25.png

[ButchDeal]

your battery bank is not wired up right. you can tell because you have one battery with nothing on the negative terminal.

[Mike90250]

i think when you price the 600V 80A controller, you may want to revise the PV layout to use a much less expensive controller. Or not.

[Hiro]

ButchDeal Thanks mate. Fixed it up.

Can we have a consensus which layout is better 6s2p or 4s3p?

I did check prices for the 200V 80A and 600V 80A with an approx. difference of $500. Where as with the 4s3p I need to buy a combiner $220, fuses $10.
Screenshot 2017-03-02 21.04.54.pngScreenshot 2017-03-02 21.04.38.png

[Sunking]

Won't 6s2p make the system 237.5 VoC 17.27 Imp and make the amperage too low? Does the CC need a certain amperage into it to work efficiently?

No Sir, not at all. Again get yopur head wrapped around this.

MPPT Output Current = Panel Wattage / Battery Voltage.

The whole point of using a HV Controller is to get the current way down so you have less loss on wiring, longer runs on cheap small wire. As you have also noticed it can save you big bucks on Combiners and Breakers. If low voltage was a good thing Grid Tied Systems, Electric Vehicles, Industry, and utilities would use low voltage. Fact is low voltage suks. A grid tied solar system runns 300 to 500 volts.

As Mike cautioned your cost needs to be considered and balanced out. In your case the higher cost of the HV Controller is offset by the savings gained in using smaller wire and not having to use Combiners and Breakers.

As for your drawings, they need work.

[Hiro]

My thought process with this scenario is although a Schneider 600V MPPT is $1900au.
A Midnite 200V MPPT is $1500au + eBay Combiner $200au + inline Fuses $100au + thicker cable $50au = $1850.

6s2p has less parts to the system, less connections and from that perspective easier maintenance, installation & neatness...

I updated the Drawing by improving the Battery wiring. It may need work but that is vague Sunking. Which area: PV, Battery, Wiring, fusing?

I'm thinking the Battery bank connections to the system are still incorrect. I have one battery managing all the split load. Instinctively it still doesn't look right to me and think the red should be elsewhere..? Screenshot 2017-03-03 11.09.12.png

[Sunking]

I think you need to stop and go back to the drawing boards. I am not going to do it for you, but go to a website and see what the optimum Panel Tilt is for the months of June and July. Your first trip told you the optimum tilt angle for yearly average. You do not want that. You want optimum for Winter, not year round. If you do that, then you can likely cut your panel wattage from 1/3 to 1/2 a huge chunk of change. Your first go around in post 22 gave you 2.36 Sun Hours. Change the tilt angle and I bet you can get that up to 4 or more hours. Get it to 4 hours and that 5000 watt panel drops to 3000 watts. About $4000 savings just in panel cost.

Now get to work.

DO NOT USE PARALLEL BATTERIES.

[cebury]

This is why many people pay someone to design the right system for their needs. Lots of nook cranny details in there to overlook as a newbie trying to learn it all from books and online volunteers, neither of which will get injured or waste money if you make a serious mistake. I'm a DIYer and consider myself above average intelligence, but my first step to build my own system would've been to read a couple starter books/guides before I hit the line diagrams. Just sayin... IMO, folks have been pretty patient in this thread compared to some historical ones.

[Mike90250]

While you might use #3 ga wire for the charge controller, you will need something at least 1/O or even 2/O for the inverter wiring,
5Kw peak surge @ 80% eff, will be 125A