Thanks for the responses guys.
I suppose the absolute cheapest way would be to connect my geyser to the AC 2 out port? But this will come at a huge effort as I would have to lay 30m of 10mm2 armoured cable again over challenging terrain.
I guess the easiest would be to upgrade to the 5Kva or add another 3 Kva.
[quote=âfredhen, post:80, topic:310â]
Pretty much, yes. But just remember that the limit is only for DC to AC inverting. When your batteries are charging during the morning you should see your PV put to good use as that is DC to DC. [/quote]
So what you are saying here, is that if my batteries had been drained somewhat, my panels would produce more as there is no inverting involved for charging the batteries?
I think youâre are more or less saying the same thing as Fredhen? Voltage of batteries being low, my panels would work a little harder as a certain portion of PV will go towards charging them?
I get the bit about my MPPTâs amp limit. I have two 150 | 35 MPPTâs so I think it might be worth it to get a larger inverter?
I was wondering why my PVâs thatâs capable of producing 4Kw easily is not doing that and the inverter still calls upon Eskom to sustain 3Kw+ load with panels being in full sun.
It has been pointed out to me that the inverterâs transfer capability is only for loads connected on the AC out 2 port, which is not active in my case.
So my panels are not being put to their full potential and thus my reasoning to get a larger inverter.
Hmmm with two 35As I donât think it is worth it upgrading your inverter. You have a maximum generating capacity of roughly 3700W, regardless of the amount of panels on your roof.
If you buy a 5kVA you should also get another 35 MPPT if you want to see much of a difference.
Thatâs not the case, no. Your inverter is still limited to inverting ~2400W, so you will only be able to provide that much to AC loads regardless of if the load is before or after the inverter.
Correct. Just like @jykenmynie mentions (I am not too clued up on MPPT charging), but from what I understand you will in essence be able to charge your batteries up at either the max Amps of the MPPT or the batteries (whichever is lower)
Following @jykenmynie math here youâll be able to only get ~3700W out of them which should then be what youâll see when the batteries are charging up (if they allow it)
Mmm, it seems then for the time being that I would have to start using the batteries more often during the day then, to also assist with loads when possible? so that at least my panels and MPPTâs are doing what theyâre supposed to.
Maybe sell both 150 | 35âs and try and get one 250 | 100 MPPT?
Any other ideas on what to do? What would I better benefit from now, a larger inverter or rather a lithium battery?
My thought process would be:
Basically, I think your system is size reasonably well. Maybe a bit too much PV on the roof, but other than that alright.
If you want to upgrade, you would probably need to get more batteries, a larger MPPT AND a larger inverter. Otherwise, if you just upgrade one thing, youâll just introduce a new bottleneck.
I agree with jykenmynie. Your system is fairly well size with a bit too much PV, but the extra PV will help you on the cloudy days, so that is a plus (and you already payed the cost).
Donât worry about using the battery during the day as well. If I were you I would look at moving some loads around to better handle your available power, or look at doing a few smart hom things to optimise that (my current project).
Battery life is measured in cycles where the cycle is from empty to full and back. So if you charge and discharge your battery once every day then 6000 cycles means (6000/1/365) 16.44 years. Now say instead of using the 4kWh (assuming a 4kWh battery as an example) from the battery once a day you are using 4.5kWh each day where the 500Wh is the extra bit you discharge and charge in between which is 1.125 times a day. Then the you reduce that to (6000/1.125/365) 14.61 years.
If you look at this from a cost side.
Say your battery cost is R20k and will last the 16.44 years under perfect conditions, then this is R3.33/day. If you use it more and it last 14.61 years this is R3.75/day. That reduction of 667 days cost you R2500.
Nothing you are planning (inverter, mppt) will cost less than that R2500.
Just enjoy it. Itâs money well spent 
Thank you guys, you are giving much to think about.
PV was a bit overkill, but yes, the idea was actually to assist on cloudy days.
To provide some information as to the issues raised. Yes, I run out of capacity with batteries. Iâd like to extend their life as much as possible so I only take them down to 70% SOC during some nights. Theyâre 4 x 12V AGM lead acids, 200 Ah in capacity.
So my thinking is, should I substitute them for a lithium which I can take down to 20% SOC and more regularly, this would be an improvement as they charge faster and will last longer. Maybe a Pylontech US3000? or 2?
So, my point is, if Iâm more comfortable to use my batteries more often, then it wont really matter that much that my PV generating capacity is limited. Because then the panels WILL work to charge them and I will see better figures.
Am I on the right track here?
For my lead acids: 30% usage from 200Ah = 60 usable Ah
For 2 x 3000USâs = 80% usage from 150Ah = 120 usable Ah so I will have doubled my capacity. But yes, at great cost. Roughly R40k it seems.
Ah. Lead Acids. I missed that part. Yes they are limited.
The problem is that as far as I know you cannot mix leads and lithiums in the same system (without lots of trickery and extra cost). So it would be best to use your existing batteries that you have. When they die you can replace them with lithium.
PS. I limit my Lithiums to 25% (on normal days it only goes down to 30%)
Yes yes, no mixing involved. Iâve read enough to know that theyâre vastly different animals and mixing them is not an option.
No Iâd like to replace with lithium âŚ. one day.
In the mean time, Iâve scrutinized my yard this past weekend to see what improvements could be made to help the system. I really despise doing things like this, but it turns out that if I remove just one tree, or at least trim it down to half its size, I will have all panels in sunlight from about 08h00 already as opposed to 12h00
Now that would be a huge improvement. Batteries receiving a proper charge from much earlier, much more pv available to assist with the kettle and geyser early in the morning etc.
But ja, decisions decisions. One of the reasons I bought where I bought is because of astounding number of trees and bushes bringing birdlife etc. My whole back yard is a miniature forrest. And the trees does a lot to keep the house cool.
But for half the price of an extra MPPT, I could have this tree removed. And it is actually big enough to be posing a danger to the house, 2 fences, solar panels, garage afdak etc.
Looks like a bottlebrush, so the other plus would be much less red mess 
Well spotted Louis! Indeed it is.
Jaaa the other thing is the mess it makes. So one more reason to get rid of it.
Sigh. I really hate chopping down trees but this is actually a danger. When moderate winds blow here those two kettie type branches sway heavily and I guess it is a matter of time before they will split in the middle.
Donât know if you noticed the plank hammered in about hallway up and I spun a thick wire around the two stems to try and support each one with the other one.
This is what my panels is doing now, at 09h50 with only 3 strings of 6 panels (one string on a different MPPT) in terms of charging my battery.
I can just imagine if all 12 panels already has full sun on them.
Guys Iâve just stumbled onto a potential seller of unit similar to mine 3Kva 48VMultiplus II
Given my setup and the comments above, would it be worth it to buy if I can get it for a steal?
Would this actually benefit my setup? I suspect it is still well into its warranty period, and I may get it at a very good price.
If I work according to this, my solar generation limit would be (35 + 35) x 54V for my AGMs = 3780W.
It is quite a lot more than my current 2400W
Would it be worth it if I can pick up the unit for say R8k?
Just make sure they are very similar⌠ie need to run the same firmware if they will be paralleled.
Also remember these derate when they get hot (defined as >40C). Not by much, but they may shave 2A off the top when running very hot, a good 200W. So at noon you may well see 3500W or so peak.
TTT has alerted me to this fact as well thanks. Turns out this unit is the Mutli Plus. I have the II
So this is not NRS approved which is a deal breaker for me, even if they had the same firmware/software etc.
