iops

Very good idea. I will intergrate that.
J

Opted for the US3000, it was installed on Wednesday and I'm very happy! Thanks again for everyone's input and advice, these forums are great!
I'm going to add a second in a few months and then probably move into a cabinet to add the third, potentially a fourth, pending how many panels I can add to my roof first.

Opted for the US3000, it was installed on Wednesday and I'm very happy! Thanks again for everyone's input and advice, these forums are great!
I'm going to add a second in a few months and then probably move into a cabinet to add the third, potentially a fourth, pending how many panels I can add to my roof first.

Pylontech US3000 3.5 KW Lithium ion Battery Special R19500.00 Vat Included excl Delivery  Valid While Stocks last 

Low voltage energy storage system 2019.pdf

You only need to purchase 1 set of connection cables for the batteries to the inverter (or breaker/fuse inbetween) and is rated for 120A (meaning you can safely run 3x US3000 on it). The pack is just under R400. Each Pylon comes with the set of cables to connect to another battery
Any standard 19" rack will work for the Pylons. I am using a Samson SRK12 12U rack, which can hold 4x US3000
I personally would rather go with the US3000 than the US2000 (can't really explain why - do not have any experience with the US2000). Even a single US3000 will 'work' with your inverter. You wrote in your initial post that you don't think that you'll draw more than 35A with your loads, so the single US3000 would be able to handle that. I would probably start off with a single US3000 (which is what I also initially did) and then add the 2nd once you have saved up the balance
I have read somewhere that you can also mix US2000 and US3000 in a stack - but do not have any experience with that. Maybe someone else here does and can help out?
I am running 4x US3000
Just keep in mind your 80% DOD, which means that this would give you 7.7kWh of usable capacity

Useful script for sure, maybe it would be easier if you forked the initial project so a diff between the two is easier?
How are you planning on merging changes from the initial project to yours?

I'll add my +1 here, I had it working with my Axpert 3kVa with a Victron MPPT, just no Raspberry Pi - thanks to advice from @wolfandy over on one of my threads. Setting the Axpert to SBU worked like a charm, my inverter seemed happier without having to worry about MPPT. It also allows for greater expandablilty in the future.
Ultimately I opted for the Multiplus II, no looking back.

So this kind of went down a rabbit hole, and with lock down hasn't given me much else to obsess about. With the batteries now finally sorted out and performing as they should, it gave me a good few weeks to look at the system as a whole.
 
What started out as a perfectly decent loadshedding backup solution has turned into a bit of a pet project. While I think the Axpert has been great for the loadshedding solution it hasn't performed very well (and I admit that the installation, battery and solar issues from the installers point of view have soured things a bit) I looked at the longer terms goals and wasn't comfortable with what the Axpert offered:
12 month warranty
Limited expandability
 
After a lot of thought and being very impressed with the 150V/70A MPPT from Victron and BMV-712 on the way, I've decided to take the plunge to the Blue team and get the MultiPlus II 3kVa inverter. I'm hoping it'll be installed some time this weekend - new thread with some pictures coming soon
I decided to upgrade from a 24V system for better lithium battery support in the future, which is the next to upgrade on the list. I'm keeping my 8 AGMs for now and just wiring them up into 2x 48V banks.
Huge thanks again for everyone involved in this thread, it's been hugely helpful. Excited to be part of the forum!
Thanks again for your help on this, it's nudged me in the right direct to get things setup.
You're right, it wouldn't but the VRM monitoring from Victron appealed to me. And I like tinkering with Raspberry Pi's to see what they can achieve Ultimately that, and this thread, has cost me a lot of money. I'll PM everyone invoices who contributed to the Blue movement

So this kind of went down a rabbit hole, and with lock down hasn't given me much else to obsess about. With the batteries now finally sorted out and performing as they should, it gave me a good few weeks to look at the system as a whole.
 
What started out as a perfectly decent loadshedding backup solution has turned into a bit of a pet project. While I think the Axpert has been great for the loadshedding solution it hasn't performed very well (and I admit that the installation, battery and solar issues from the installers point of view have soured things a bit) I looked at the longer terms goals and wasn't comfortable with what the Axpert offered:
12 month warranty
Limited expandability
 
After a lot of thought and being very impressed with the 150V/70A MPPT from Victron and BMV-712 on the way, I've decided to take the plunge to the Blue team and get the MultiPlus II 3kVa inverter. I'm hoping it'll be installed some time this weekend - new thread with some pictures coming soon
I decided to upgrade from a 24V system for better lithium battery support in the future, which is the next to upgrade on the list. I'm keeping my 8 AGMs for now and just wiring them up into 2x 48V banks.
Huge thanks again for everyone involved in this thread, it's been hugely helpful. Excited to be part of the forum!
Thanks again for your help on this, it's nudged me in the right direct to get things setup.
You're right, it wouldn't but the VRM monitoring from Victron appealed to me. And I like tinkering with Raspberry Pi's to see what they can achieve Ultimately that, and this thread, has cost me a lot of money. I'll PM everyone invoices who contributed to the Blue movement

Cool - happy to hear that it's working 🙂
I have my system running in SBU (Solar - Battery - Utility). As there is no PV available for the Axpert, it will use batteries - which is where your Victron MPPT is also connected to. Once your MPPT production drops below your load, the batteries will automatically provide the difference. If you do not have sufficient battery capacity to get you through the night, Axpert will change over to Utility based on your "Back to Grid" setting
One question about this: I do not think that VenusOS will give you any benefits at this point as it cannot control the Axpert. I am running ICC on my Pi, which by now is also able to communicate also with a Victron BMV and MPPT. ICC can then switch your Axpert between SBU and Utility based on the SOC calculated by your BMV rather than the Axpert's guesstimate. If you already have a Pi, then the cost of the software is relatively low. It works great for me

I tried installing something similar for my folks and the experience was pretty poor.
It's far more reliable, potentially even cheaper, to run ethernet. CAT5E and an AP hooked onto that should be a reliable solution.
 
If you're looking for a scalable solution you can check out mesh networks too, essentially uses a bunch of wireless access points that connect to each other.

Quick update, I'm using the opportunity to slowly move things over to Victron. I have a 150/70 MPPT controller on the way along with a BMV-712. I have a spare Pi that'll be used to run VenusOS. A replacement inverter for the Axpert will happen a little down the line.
For now the priority is getting the full potential out of these panels and proper battery monitoring in place. Unfortunately it seems like the full month (plus a few days) of running the system incorrectly seems to have shot the batteries entirely. I can barely squeeze 2 hours out of a 200aH bank and that's with the inverter screaming about low voltage.
 
Fingers crossed the batteries can be replaced.

I've asked the installer how they're installed, it is within the max limitation of my inverter of 1500W though. When PV is in use the inverter reports 70V at max. I'll get back to you on this though.
Also just a heads up, I believe the Must is a clone of the Axpert.
Good question, I believe the Synapse isn't a clone - perhaps someone more knowledable than my can confirm?
I've had a read through another forum thread which shows you how to identify if it is a clone, I've done most of them apart from hooking up that dreadful WatchPower app. I've included a picture of the side labels below.
The fans do run a lot. I think the only time it isn't running is when the batteries are in float and PV isn't being used. I had my panels installed mid-month last month and have noticed that the fans run a lot. It doesn't really bother me since the setup is in a room outside of the main house.
The Synapse has been running as a loadshedding backup solution with batteries connected since December, I haven't had any issues... apart from posting here and wanting to upgrade to Victron gear
 

I agree absolutely with what has been said. I have 8x200Ah Naradas which are going into their fourth year. I had your dilemma and 4yrs on they are on SBU. I think I discharge them around 30-40%. They have served me some money and I am yet to see them start showing signs of age really. I have 12 x 260W panels and they are made to work to recharge them and run the load which is about your load. Hell, during the day the 1.5HP sprinkler pump is used to good use on free solar and the 1HP booster to overhead tank also runs to utilise the panels. Life is good.

When you cycle them you can recover a bit of your money. Even if you keep them charged, they only have a certain chemical design life and they will seldom go past that lifetime. 
Being AGMs, I wont recommend discharging them to much, but letting them die from old age without recovering at least some of your money is also not wise. When there is loadshediing keep them charged, in times like now, cycle them even if its only 10 or 20%. 
I hope this makes sense. 

I agree fully with this. Most batteries of the UPS variety lasts 5 years max, even if you leave them on float permanently. AGM batteries usually give you somewhere between 600 and 1200 cycles to 50%, depending on quality, which is essentially 2 or 3 years.
So I would essentially take the 3-year plan, look at their specs, and cycle the such that they die in around 3 years or so. It's a bit thumbsuck, but in that 3-5 year bracket you're essentially just one cell failure away from replacing the whole thing anyway, so I'd say it is pointless to cycle them so shallowly in an attempt to make 4 years or more.