10 minutes ago, The Terrible Triplett said:

I like.

Have a feeling that more and more people are considering alternatives, coming over to the light side. (pun intended) 

To split it more succinctly, you have a distinct difference between devices that are:
(1) only used at night (lights);
(2) small devices that are on 24/7 like alarm systems;
(3) Then the rest - garage and gate motors tend to have a built in battery backup.
(1) and (2) tend to me small loads that are perfectly suited for 12v batteries and inverters, huge savings versus 48v batteries. And you may have a few spare from a replaced 48v bank.

Have a 'spesmaas' it will become more prominent into the future due to no Eskom failures, with the big banks being monitored making more people aware of their battery use resulting in them starting to eye them batteries thinking: I have to replace these one day. What is the ROI now really?

The penny drops real hard the first time you realise: Wait a minute, I do not have the AH I use to have in the bank.

Unless you are or on your way to go off-grid, batteries for city use does not make sense if you make an effort to save power at night. Even less so if you can feed back to Eskom, legally. Batteries in cities only makes sense if there are regular power failures.

Here's the problem: Most (non technical) folk are overwhelmed by solar systems. A single system is already a big learning curve for them. Now you want to give them 3 or 4 systems to monitor / maintain and worry about. 

Running  a gate motor, garage motor, alarm and electric fence is a no brainer. Use larger batteries to carry you through the night (7Ah is often just not enough) and a 50W panel, and you're "a-away". 

Running a fridge off a 100W panel + battery is also very doable. 

Running your router + laptop off a small panel + battery is also very doable. 

The bigger loads, like pool pump, dish washer, tumble dryer, etc, etc, all require a bigger system though. If you run these items only during the day, on an off grid inverter, batteries are not necessary. Unless you want / need some backup to cover cloudy days, or as a buffer on a windy / rainy day with clouds moving around the whole time. 

 

P.S. How do you think big roof top installations on malls are setup? Pretty much the same way you're trying to do it here. But they use some nice equipment that's built for that purpose. 

 

YES!

And as we all learn more here, or we start to have to replace parts, the more we will move over to smaller systems, for you are right, it is a very steep learning curve.

And therein what we should be focusing on is to tell new people coming on board, to go split systems, one at a time, and motivate out of experience, why we say so.

I think the problem starts to come in when you have large startup currents, loads, and from that point onwards the train picks up speed, all downhill.

I concur. The thinking here in very basic terms is this - utilize the big bank less to promote duration and divert load to smaller subsystems "run" by the pv system when power is in abundance. The easy route is still grid tie but at the end of the day I am a bit of a stickler for legislation. Not wanting a tussle with City of Tshwane I realize there are inefficiencies involved, but in doing a bit of rudementary math I can still score a lot and not bugger my main battery bank in doing so.

It is definitely not the answer for the grandparents though...

Sent from my SM-N900 using Tapatalk

Edit: plus it requires very little capital investment. For me at least😊

2 hours ago, edmundp said:

I realize there are inefficiencies involved,

That's what initially worried me as well. You already have a round trip of around 70% on lead acid + inverter setups, so this will essentially drop the overall efficiency to around 50%. That's worst case of course. If you can rig it up to prioritise solar charging, then even if you charge from the AC line you could potentially do around 65% round trip. Given that PV panels are relatively cheap compared to other components, the inefficiency might just be forgivable.

If in addition, you can charge DC to DC with excess power, now we're beginning to talk. Suppose your downstream power systems are all 12V or 24V systems and the main one is 48V (or in my case, the downstream one will have to be 12V cause the main one is 24), you can charge them using a buck converter and potentially achieve better efficiency that way.

Also, to prioritise solar, you could program the CC to close its on-board relay when it goes to float (that is often a configurable option) and use that to activate downstream chargers.

Man... I might just begin to see _a_ light too...

18 hours ago, The Terrible Triplett said:

YES!

And as we all learn more here, or we start to have to replace parts, the more we will move over to smaller systems, for you are right, it is a very steep learning curve.

And therein what we should be focusing on is to tell new people coming on board, to go split systems, one at a time, and motivate out of experience, why we say so.

I think the problem starts to come in when you have large startup currents, loads, and from that point onwards the train picks up speed, all downhill.

From experience it isn't always possible, or easy or cheap to go split systems if you want valid COC's. Many sparkies will argue that you need to run a separate DB for every feed to the house. Some will say that as long as the cables / CB's are physically split and labelled properly it's fine. But then you also need a seperate isolator and earth leakage for every set of CB's. Lights can probably go without an earth leakage but I know many sparkies will fail the COC if that's the case. 

Very good point.

Points to chat about:
You do not need a CoC for a UPS, so how can one get away with a solar powered UPS?
Solar Fridge / Freezers: http://www.solar-solutions.co.za/solar-fridges?gclid=CJyZqOnu280CFdS7GwodigoHoQ - no Coc?
All koi / pool pumps can run on off-grid solar systems - no CoC.
Separate DB's and wires are still less expensive, and lasts 20+ years, than more batteries.
 

Here is a good one: http://www.solar-solutions.co.za/files/sol/1397038315_solar fridges.pdf

1 minute ago, The Terrible Triplett said:

Very good point.

Points to chat about:
You do not need a CoC for a UPS, so how can one get away with a solar powered UPS?
Solar Fridge / Freezers: http://www.solar-solutions.co.za/solar-fridges?gclid=CJyZqOnu280CFdS7GwodigoHoQ - no Coc?
All koi / pool pumps can run on off-grid solar systems - no CoC.
Separate DB's and wires are still less expensive, and lasts 20+ years, than more batteries.
 

You need a COC as soon as you make changes to your electrical wiring. You could get away with it, when you sell your house and take everything out. But you won't get awat with it if your house burns down and your insurance doesn't pay out cause there's no valid COC. 

SO, if what you're powering isn't connected to your house's wiring, you might get away with it. If it's mobile, you definitely don't need a COC. 

P.S. you don't need a COC for a UPS, since it's not hard wired into the DB board, but the moment you hardwire it into the DB board, a COC needs to be signed / amended. 

To recap and be 100% clear a CoC for a DB connected solar system is a non-negotiable and in most cases the easiest way to do it.

But, to recap, 2-3 years later, as I said before, the reality of having to replace the battery bank on a DB connect system starts becoming very real when there are no power failures. And with the economy getting tougher, not easier year in and year out, in some cases it will be unavoidable luxury.

Panels on the other hand, last +-20 years.

So focusing on this point, using same panels, there are alternatives.

Back to the point. It becomes easier for in the DB board, if you run split systems, you just need a CoC for the lights with 1 or 2 panels, 12v batt bank and a small inverter and you are done for life, bar the batt replacement every 10-15 years and if you bought a Victron sine wave inverter.

Rest are all separate from DB board, removable, if it can be done neatly, for in the end, neatness is second to safety.

27 minutes ago, The Terrible Triplett said:

Here is a good one: http://www.solar-solutions.co.za/files/sol/1397038315_solar fridges.pdf

R25K for a fridge, some panels + MPPT + battery!

I'm sure I could do the same with my R8K fridge. R17K will get me 2 x 250W panels (R5K) + 20A MPPT (R2K) + 105A battery (R2K) + cables (R500) + 1Kw inverter (R4K) + pizza and plenty food to fill the fridge (R3.5K)

Or I could install an Axpert and have Eskom backup as well. For when the cheap-ass battery dies and it's too cloudy for three days to keep the food frozen. 

And then I could add another fridge to the same system, at only the price of the fridge itself. And even run some LED 

On the subject of modified inverters, I know my fridge can run off a 1Kw modified inverter 

Jip, they are rather stupid with their prices.

Just showed the pic for a pic is a 1000 words, and from that pic, you already did the sums to do it even cheaper. 

The point here is to go forth and explore other options and not just revert back to Axpert/48v/4500w panels/500ah batteries (that needs replacing at current usage, probably in 5-8 years). 

Explore new ideas to do it cheaper and more efficiently, for we now know more than before, how would we do it better.

And no cheap-ass battery. I am talking a good battery for as I said, batt replacement every 10-15 years.
But you can, in case of emergency, use a cheap-ass battery.

Idea is to take that which can go off-grid, off-grid completely for as long a period as you can.

Yes, if there is no sun, for days on end like in Cpt lately, it becomes a problem, dressed in overalls.

45 minutes ago, The Terrible Triplett said:

Jip, they are rather stupid with their prices.

Just showed the pic for a pic is a 1000 words, and from that pic, you already did the sums to do it even cheaper. 

 

It's not just that company. Just about anything which is main stream is generally double the price in a 12V version - even though just about everything runs a DC bus in any case. For example, 10W, 200V flood lights can be had for about R100. 12V equivalent is R200!

I won't even go into the dynamic of manufacturing these devices....

A camping fridge cost say R12,000 whereas a bar fridge is R2,000. Ironically though I met a guy some time ago who build those camp fridges and sell them for R4000. He sells the exact same fridges to three big chain stores, who rebrand them and sell them at R12,000! Makro, for example, sells a 250W solar panel for R7K. Builders used to as well but the prices came down to R4.5k

 

45 minutes ago, The Terrible Triplett said:

The point here is to go forth and explore other options and not just revert back to Axpert/48v/4500w panels/500ah batteries (that needs replacing at current usage, probably in 5-8 years). 

Why not? It's cheaper than the blue stuff

45 minutes ago, The Terrible Triplett said:

Explore new ideas to do it cheaper and more efficiently, for we now know more than before, how would we do it better.

Well, here's where it gets interesting, to run everything as efficient as possible. 

45 minutes ago, The Terrible Triplett said:

The point here is to go forth and explore other options and not just revert back to Axpert/48v/4500w panels/500ah batteries (that needs replacing at current usage, probably in 5-8 years). 

Yes, if there is no sun, for days on end like in Cpt lately, it becomes a problem, dressed in overalls.

How do you deal with this conundrum?  

32 minutes ago, SilverNodashi said:

How do you deal with this conundrum?

By doing the sums right so that you have 2 days backup on 20% DOD, at least. 

As long as the panels can recharge the batts in 5.5 hours, at least once a week in winter to float and once every 2-3 months equalise properly, you should be fine.

And in really extended "dark" times, switch off non-essential lights to stretch the batts to go for say a week with no charge, right down to 50% DOD.

To do same with a large system is going to cost a small fortune on batteries.

Use what you have, start by adding a smaller inverter to power the above 24/7 off existing bank, the main system never to use the batts unless there is a power failure. 

But, and here is the little problem over time. As was said on another thread this morning, if your inverter draws 20-50w in standby, that is a substantial waste of your battery investment at night for just sitting there. Not efficient at all. 

 

Just now, The Terrible Triplett said:

By doing the sums right so that you have 2 days backup on 20% DOD, at least. 

As long as the panels can recharge the batts in 5.5 hours, at least once a week in winter to float and once every 2-3 months equalise properly, you should be fine.

And in really extended "dark" times, switch off non-essential lights to stretch the batts to go for say a week with no charge, right down to 50% DOD.

To do same with a large system is going to cost a small fortune on batteries.

Use what you have, start by adding a smaller inverter to power the above 24/7 off existing bank, the main system never to use the batts unless there is a power failure. 

But, and here is the little problem over time. As was said on another thread this morning, if your inverter draws 20-50w in standby, that is a substantial waste of your battery investment at night for just sitting there. Not efficient at all. 

 

So, you're saying I'm wasting money on my bigger system, by using Eskom as a standby? On the farm we have a genset as standy since there is no Eskom. With the 3phase equipment we have it's a bit difficult to split the systems as have been suggested above. And it's a bit difficult to switch non essentials on a game farm which requires 10,000L water pumped from the well for the animals every day. 

All I want is flexibility really. To spend a couple of thou every other month and slowly build a system that makes me independent from Eskom. So far, there is just nothing on the market that does that. End of story.

10 minutes ago, plonkster said:

All I want is flexibility really. To spend a couple of thou every other month and slowly build a system that makes me independent from Eskom. So far, there is just nothing on the market that does that. End of story.

How so Plonkster?

1 hour ago, The Terrible Triplett said:

How so Plonkster?

So way back in 2012 I got very begeesterd when I heard of this thing called a micro-inverter. How awesome is this, I thought! One solar panel at around 2.5k. One micro inverter at around 3.5k. 6k. Rather steep, but as a percentage of the average household income of most middle-class former oppressors, quite doable every second month or so. In a year you can buy like 6 of those, that's 1.5kw for around 36k. Rather steep for what amounts to a small GTI, but all that individual panel optimisation and no issues with shading and just having the freedom of doing it as you have money available was very attractive.

Except that load shedding was looming, I was working from home, and the prospect of at least 5 hours downtime per week (in stage 1) and as much as 20 hours when it gets really bad. There is also cost-inefficiency of doing it this way (a proper large GTI would set you back at least 10k less back then), and then the ever-irritating prepaid meter that can't turn back and micro-inverters that cannot be throttled. So that was the end of that idea.

Then I looked at a small battery system, and due to the threat of load shedding, that's what I went with. But there is just nothing on the battery side that is anywhere as flexible as the micro-GTI idea. The only problem with the Micro-inverter is it cannot be throttled. They will abruptly turn off if you push the frequency up, but none of them supports proper GFPR.

There was another reason I went with the Victron: Their self-consumption white-paper made it sound like Nirvana: Buy an undersized inverter that carries your baseline and essentials, and draw the rest from the grid! Only, it wasn't quite ready for that. Hub-1 was a step closer, Hub-4 is another step closer, we're finally there where it is technically possible to do it... but the bloddy anti-islanding costs another 7k (and you can buy an Axpert for that price).

So I'm perpetually in a bit of price-paralysis. I would like to spend an amount under 10k that will have an immediate effect, but it just doesn't work that way in solar. Whatever you buy, is most often too big or too small, won't work with anything else, has a limited life (batteries), blows up before it should (Microcare charge controller), or your president sends the currency to hell and you cannot afford the second inverter you wanted to parallel.

The real problem, truth be told, is that the kids' doctor's bills were killing me. We moved them to a different daycare, and I kid you not: We haven't been to the doctor since March, where previously we literally went there twice a month. Rant over... :-)

Amen bother, Amen.

Yes, the micro inverters, was also looking at them for a long time. They have improved.

Here is my plan. Because I have the equipment for off-grid in place:
The 2 x 200w panels stay for all off grid stuff, as I said above, the maths is NB on this one.
Maybe add 2 more panels, if I need,

Now I can start looking at this, GTI, with financing:  http://www.solaredge.com/products/pv-inverter/single-phase#/

The supplier need to sort three things:
1)) Be able to use my existing panels, the 3 x 310w ones, as a starting point, to lower costs, connected legally to my DB with no feedback to Eskom, yet have the option.
2)) Must be able to add more panels, for piloting is needed winter being ideal, as I see the usage, after I trained the household when to use the power.
3)) Must be able to switch the panels, during power failures, from GTI to power my off-grid system, and back.

If they can do this, rent to own, their problem to support and maintain, at a price that I can reduce my bill over a year by R1000pm, after I paid them, Bob's your Uncle.

Ps. But, Murphy is Bob's brother, and me and him, we are big pals.

 

Pss. We also learned with our babies, twice it was proven way back, if they get sick constantly, or cry badly every time on the way to daycare or cannot wait to leave every time, change daycare, make a plan if it is a problem. They must want to go, want to stay, and only get sick once in a while.

 

 

5 hours ago, The Terrible Triplett said:

But, to recap, 2-3 years later, as I said before, the reality of having to replace the battery bank on a DB connect system starts becoming very real when there are no power failures. And with the economy getting tougher, not easier year in and year out, in some cases it will be unavoidable luxury.

Why would this be needed if you use your batteries only for loadshedding?
They should last full term when you only float them along.
During the day PV powers your loads, at night you swap to Eskom?

Should be the same if you have 4 small 12v systems or 1 bigger 48v system.
In the end they will all have to be replaced.

I'm with Silver. Small systems just adds way too many complications that's unneeded.
Only scenario that's makes this even a little plausible is if you want to upgrade your bank a year or so down the line of your solar journey.

3 hours ago, The Terrible Triplett said:

And in really extended "dark" times, switch off non-essential lights to stretch the batts to go for say a week with no charge, right down to 50% DOD.

To do same with a large system is going to cost a small fortune on batteries.

Why? I just turn of loads that does not need to run for the long stretches and run only what's needed

 

45 minutes ago, viper_za said:

Why would this be needed if you use your batteries only for loadshedding?

My experiences over many years with UPS'es, on float all the time, the day you really need them batts, they are end of term, having stood the design life on float.

We all bought batteries for load shedding and they worked beautifully at the time. But to have that cash sitting there for a few years not used, is not a good investment.

And if there is no load shedding to use batts daily only to replace them every X years, is not a good investment either. 

I am not sure how many people would have been on solar today if it was not for load shedding. And if they wanted solar, they would probably have gone GTI, not so?

 

45 minutes ago, viper_za said:

Only scenario that's makes this even a little plausible is if you want to upgrade your bank a year or so down the line of your solar journey.

Ditto!

If you have the stuff, and you want, re-purpose them, that is the idea for the chat here.
And if you don't have any yet, just give it some thought to start smaller, if you want.

Having had to replace batteries before, a few times due to learning curves, to see that cash go out is not cool if you cannot really justify the ROI, all done before load shedding took hold.

1 hour ago, plonkster said:

So way back in 2012 I got very begeesterd when I heard of this thing called a micro-inverter. How awesome is this, I thought! One solar panel at around 2.5k. One micro inverter at around 3.5k. 6k. Rather steep, but as a percentage of the average household income of most middle-class former oppressors, quite doable every second month or so. In a year you can buy like 6 of those, that's 1.5kw for around 36k. Rather steep for what amounts to a small GTI, but all that individual panel optimisation and no issues with shading and just having the freedom of doing it as you have money available was very attractive.

Except that load shedding was looming, I was working from home, and the prospect of at least 5 hours downtime per week (in stage 1) and as much as 20 hours when it gets really bad. There is also cost-inefficiency of doing it this way (a proper large GTI would set you back at least 10k less back then), and then the ever-irritating prepaid meter that can't turn back and micro-inverters that cannot be throttled. So that was the end of that idea.

Then I looked at a small battery system, and due to the threat of load shedding, that's what I went with. But there is just nothing on the battery side that is anywhere as flexible as the micro-GTI idea. The only problem with the Micro-inverter is it cannot be throttled. They will abruptly turn off if you push the frequency up, but none of them supports proper GFPR.

There was another reason I went with the Victron: Their self-consumption white-paper made it sound like Nirvana: Buy an undersized inverter that carries your baseline and essentials, and draw the rest from the grid! Only, it wasn't quite ready for that. Hub-1 was a step closer, Hub-4 is another step closer, we're finally there where it is technically possible to do it... but the bloddy anti-islanding costs another 7k (and you can buy an Axpert for that price).

So I'm perpetually in a bit of price-paralysis. I would like to spend an amount under 10k that will have an immediate effect, but it just doesn't work that way in solar. Whatever you buy, is most often too big or too small, won't work with anything else, has a limited life (batteries), blows up before it should (Microcare charge controller), or your president sends the currency to hell and you cannot afford the second inverter you wanted to parallel.

The real problem, truth be told, is that the kids' doctor's bills were killing me. We moved them to a different daycare, and I kid you not: We haven't been to the doctor since March, where previously we literally went there twice a month. Rant over... :-)

I actually want to go the micro inverter route as well but already have what I need. The one nice thing about micro inverters is that you can add as many as you have cash for. It's a GTI setup, with the advantage of easy / cheap growth and much more redundancy. If one or two panel / inverter pop, you still have the rest. If that big 5KW GTI pops, you have nothing. 

The Renesola micro inverters have anti island protection built-in already. 

But you're right at some point a GTI is cheaper. 

17 minutes ago, SilverNodashi said:

The Renesola micro inverters have anti island protection built-in already. 

They all have anti-islanding. Well... all that I've seen. The thing I want is GFPR (grid frequency dependent power reduction). This would allow me to tie them to the output on the multiplus, and when there is excess power I can disconnect from the grid, and control their power delivery using the grid frequency. At 50Hz it pushes 100% power, and at 53Hz 0%, with a linear scale in between.

1 hour ago, plonkster said:

They all have anti-islanding. Well... all that I've seen. The thing I want is GFPR (grid frequency dependent power reduction). This would allow me to tie them to the output on the multiplus, and when there is excess power I can disconnect from the grid, and control their power delivery using the grid frequency. At 50Hz it pushes 100% power, and at 53Hz 0%, with a linear scale in between.

So you want to use the Multiplus 220v/50hz to fool the micro inverters into thinking it is Eskom?

I ask that question a few times with no clear answer to date, from "experts" on why I cannot do that with my Phoenix - no connection to Eskom.

1 hour ago, plonkster said:

They all have anti-islanding. Well... all that I've seen. The thing I want is GFPR (grid frequency dependent power reduction). This would allow me to tie them to the output on the multiplus, and when there is excess power I can disconnect from the grid, and control their power delivery using the grid frequency. At 50Hz it pushes 100% power, and at 53Hz 0%, with a linear scale in between.

I'm sorry but you just lost me. You want to tie an inverter output to another inverter output, to disconnect from the grid? You only need a one inverter and some PV to disconnect from the grid. 

 

4 hours ago, The Terrible Triplett said:

So you want to use the Multiplus 220v/50hz to fool the micro inverters into thinking it is Eskom?

Yup. That is how a hub-2 system works (that's what Victron calls it). You tie the PV-inverter to the output of the Multiplus. If it makes less than you're using, the rest is taken either from batteries or from the grid. If the PV inverter makes more than you are using, it pushes back through the Multiplus/Quatro either into the grid, or the inverter may decide to use it to charge batteries. But what if the grid goes down? Then you need a way to throttle the PV-inverter, and this is done by pushing the frequency up. In this way, a Multiplus can work with a PV-inverter on its output, even when the grid is down.

It only works if all the equipment understands the plan. Micro-inverters do not :-)

Edit: So the point of using this method, is that you can tell the Multiplus to disconnect from the grid at low power levels, thereby avoiding feedback.