Powerconsumption.PNG

Well done - I see consumption is considerably lower in 2016 versus 2015

Thanks Chris. Yes, I imagine that I would be running at about 15kWh and less in summer. 

hi, i am a bit slow, is the consumption lower from the PV's to charge the new batteries?

or is that output consumption?

reason for my question because i am also running 4 x 105ah Royal batteries and they dont last longer than 4 hours on +- 150W load. 

my setup is very similar. Some inverter and +- same PV input power (900W)

26 minutes ago, Fritz said:

... 4 x 105ah Royal batteries and they dont last longer than 4 hours on +- 150W load.

48v system, 105ah with 150w load should power for about +-14 hours to 50% DOD. 20% DOD would be 5 hours.

Depends what people mean by "last". What they usually mean is that the batteries will dip below the set voltage point in less than 4 hours. In theory however, the important part isn't how long the voltage is kept up, but what your coulomb counter (aka Victron BMV) is telling you. If a 12V battery is sitting at 11.5V under load it might be far from empty.

Of course you cannot completely discount voltage:

1.If it drops under 10.5V per 12V unit, inverters starts turning themselves off.

2. Doing the same work at a lower voltage means compensating with more current, and that comes out of the Ah kitty. In other words, voltage efficiency does matter.

You also have to understand a bit about how  battery works. The amount of energy in it is really proportional to the amount of electrolyte. The rate at which that energy can be liberated is proportional to the amount of lead in it. When the voltage drops, it is an indication that the lead plates are falling behind in terms of charge collecting on them. This might be due to the electrolyte being weak (through discharge), but it could also be that the rate of discharge is just too high and you're not giving the chemical reactions enough time to restore charge to the lead plates.

With all that said, at 150w on a 48V battery, that is below the C20 discharge rate, and a battery that is supposedly fully charged and is discharged at such a low rate really should be able to keep up the voltage. So however you look at it, even when bringing in some mitigating factors as I did above... still sounds like those are crappy batteries.

I have a pretty similar capacity, a 24V 200Ah bank (which in kwh is about the same as a 48V 100Ah bank). With my set point of 24.7, in December, when no TV is being watched and it's pretty much the fridge and freezer and a few lights running from it (similar load), that will go until 5AM the next morning before it hits 24.7V :-) It is a pair of Victron AGMs though, so it is too not surprising.

(Yes, I know an AGM battery isn't the optimum for regular cycling... but they are cheap second-hand ex-UPS, I don't mind if I kill them in 600 cycles).

thank you for responses.

Sorry i forgot to say i dont want to run my 48V battery bank lower than 47V  ( 11.75V = 30% ) and with a non "heavy" load the batteries are not forced to discharge high peaks.

So these batteries slowly but surely creep down to 47V +-5 hours.

i know all fingers point to a proper batt monitor but dont have the moola now. first trying to make sens of it all.

ps: my batteries are standing on the cold garage tile floor.  is that a problem ? 

23 minutes ago, plonkster said:

... still sounds like those are crappy batteries.

Royals are maintenance free, maybe +-250 cycles.

They are VERY good training batteries. 

I suspect they are on their way out. Sorry Fritz.

8 minutes ago, Fritz said:

my batteries are standing on the cold garage tile floor

No, in the old days leaving batts on concrete floors was a problem with acid, today that problem is solved. Casings don't leak anymore as they used to.

But I would still put a plank or some such under them.

2 minutes ago, The Terrible Triplett said:

Royals are maintenance free, maybe +-250 cycles.

They are VERY good training batteries. 

I suspect they are on their way out. Sorry Fritz.

No, in the old day leaving batts on concrete floors was a problem with acid, today that problem is solved. Casing don't leak anymore as they used to.

But I would still put a plank or some such under them.

somebody told me these car like batteries dont like to be to cold. 

my batteries are +-3 weeks old , cant be dead already. 

I have a general purpose battery tester which suggest they are good. 

2 minutes ago, Fritz said:

my batteries are +-3 weeks old , cant be dead already.

Question: What current is the inverter drawing at night? Maybe there is more going out of the batteries than what you calculate.

Being so young still, you need to run them on 20% DOD only as soon as you can.

But as was said a BMV is required to do it properly. Shop hard for one and ask someone here if they want to upgrade and sell you theirs at a bargain. It does happen. Trust me. 

BMV will save you more than they costs ito frequent battery replacements. 

Few lose thoughts:
Next time you buy batts, send me a PM. Who knows, maybe there are Trojans with your name on at but a few 100 more per batt than Royals.
If you batts are shot one day, trade them in for cash. The lead in them is worth a few rands, offsetting the cost of new batts.
I leave these three letters with you for contemplation, IF your constant load is low: 24v

29 minutes ago, The Terrible Triplett said:

Few lose thoughts:

You "lost" me there :-P

30 minutes ago, The Terrible Triplett said:

IF your constant load is low: 24v

Still not always sure I agree with that :-) It's rather a long term commitment to buy an inverter for 15 years+ (or at least 5 years) and to think you're never going to need to do a bit more. If you can afford the 48V one, do yourself a favour and just do it. :-)

8 minutes ago, plonkster said:

41 minutes ago, The Terrible Triplett said:

IF your constant load is low: 24v

Still not always sure I agree with that :-) It's rather a long term commitment to buy an inverter for 15 years+ (or at least 5 years) and to think you're never going to need to do a bit more. If you can afford the 48V one, do yourself a favour and just do it. :-)

Best advice plonkster, why one would even go back down to 24v after running a 48v does not make any sense at all
 

7 minutes ago, viper_za said:

Best advice plonkster, why one would even go back down to 24v after running a 48v does not make any sense at all
 

Don't get me wrong, I can see a case for that, it really depends on what you want to do. I'm also somewhat of a defender of 24V systems, in the sense that the advantages are often overlooked in favour of the mantra that higher is better. Yes, higher (voltage) IS better, and lower (amps) IS better... BUT, TTT has a point. No point in spending 20k on a Trojan battery bank (8 x 225Ah) and then never breaking 800 watts, that would be downright silly.

Nevertheless, I've shed some of my 24V enthusiasm, mostly because I'm jealous of the low currents the 48V people have. I still think it's a good compromise, and specifically for the target I'm aiming for (a hub system with a small backup component) it makes a heck of a lot of sense, BUT... when handing out generic advice I'm a little wary of pushing too hard for 24V... you know, a newbie is going to buy that advice and be terribly upset a few years down the line.

You really have to do your homework :-)

Ditto Plonk, there is a case to look at BOTH 24v and 48v after you know what you want powered, when, why and for how long.

There is the little matter of "Did you know that 24v is an option, on a budget, if you don't have a lot to power like 150w?"

If this forum was just 24v and everyone was  shouting 24v, I would have said "Did you know that with 48v ... ": No jokes.

But there are two bigger issue we miss, being passionate about our solar systems. Cost of batts in cities with Eskom, and whether you are you at home during the day to use the solar power.

Eskom just lost in court, they are going to have to pull back the increase they applied in April. All the i's and t's where not crossed. Maybe it happens and they don't appeal. Then the matter of that political party slowly losing its stranglehold.

We are not going to see regular and / or have ongoing power failures in the near future which means not today and not tomorrow are batteries getting cheaper than Eskom when you factor ALL into the kwh price of batteries, not the calcs on the back of a cigarette box, the financial calcs. A member is busy with that calc, being a financial person.

Off grid on the other hand is a totally different story when you compare gennie running costs to batts, then 48v is a no brainer.

But in cities, with Eskom, you have to be joking with big battery banks. Gooi daai panels for daytime loads If you can use it during the day, which is another little bliksim one can miss.

If you are not home and want to save some real money, last person out switches the DB off with EV tubes for geyser on a solar circulation pompie. 
Then put up a very small solar system for the alarm, garage motor and such.
And if you have a pool, fill it. 

The fact that a person bought 48v does not mean that is the one and only option for everyone else. So give the both sides of the coin, not just ones own preferences.

26 minutes ago, The Terrible Triplett said:

If this forum was just 24v and everyone was  shouting 24v, I would have said "Did you know that with 48v ... ": No jokes.

Agreed... I think that's what I'm sort of trying to do now :-)

26 minutes ago, The Terrible Triplett said:

Cost of batts in cities with Eskom, and are you ate home during the day to use the solar power.

And hence me looking at a hub system. If I can push that energy back into the rest of the house, then all that energy that's just being choked out of the panels right now can be used to run dishwashers and tumble dryers and geysers and stuff like that... without requiring extra batteries. Now I only have to size the battery according to the expected backup load... and bingo. Backup load is <=500W, so I can get away with a 24V system. In addition, all your good deep cycle batteries are larger than 100Ah, perhaps with the exception of the Sonic Gel stuff (that's also available in 50Ah), but generally, the good stuff that will last a long time? 150Ah and up. So I'm EXACTLY in that bracket where a 48V battery bank is just going to cost too much.

But as you can see, I weighed the compromise very carefully in terms of my end goal. My end goal has always been to take everything below 1kw off the grid, not to go completely off the grid. It's actually very easy to explain this using a good old load chart (watts on the vertical, time on the horizontal). The effect I'm talking about is cutting off everything below a certain point, say 500W for a start. In other words, you move the zero on the Y axis up to where 500 is now... and only the bits that stick out above that is now your consumption. The consumption is the AREA covered by these peaks (watts on the vertical, time on the horizontal, hence the area is watts times time, or some multiple of watt hours)... so this exercise is really about trying to draw a dividing line through the chart such that there is more coloured in area below the line than above (ie bill cut in half), and then discovering you can do this with a small inverter and a 24V bat. Very very specific and highly technical goal. I almost lost sight of it until we had that wind turbine discussion a few weeks back...

If your goal is to eventually go completely off-grid, pretty please with knobs on... standardise on 48V, even if it means buying crap batteries in the beginning.

26 minutes ago, The Terrible Triplett said:

Eskom just lost in court, they are going to have to pull back the increase they applied in April

Doesn't mean we're not seeing an increase, also doesn't mean price is going back to the old one immediately. Probably means a rebate will be worked into the price and/or a new slightly lower increase will be negotiated with all the relevant two-stroke letters receiving the full treatment.

Edit: Same thing really for eToll. Yes, court struck it down based on process not being followed correctly, so we won't be getting toll gates in the Western cape YET... but nothing prevents Sanral from going back, actually following the process correctly, and getting there anyway. These efforts to stop someone temporarily are good and all, but sometimes I worry that the underlying problem -- namely that the consumer is seen as a product to be sold, an investment opportunity for those who befriend government officials -- that that is not addressed, or at least not addressed fast enough.

1 hour ago, The Terrible Triplett said:

Ditto Plonk, there is a case to look at BOTH 24v and 48v after you know what you want powered, when, why and for how long.

There is the little matter of "Did you know that 24v is an option, on a budget, if you don't have a lot to power like 150w?"

If this forum was just 24v and everyone was  shouting 24v, I would have said "Did you know that with 48v ... ": No jokes.

But there are two bigger issue we miss, being passionate about our solar systems. Cost of batts in cities with Eskom, and whether you are you at home during the day to use the solar power.

Eskom just lost in court, they are going to have to pull back the increase they applied in April. All the i's and t's where not crossed. Maybe it happens and they don't appeal. Then the matter of that political party slowly losing its stranglehold.

We are not going to see regular and / or have ongoing power failures in the near future which means not today and not tomorrow are batteries getting cheaper than Eskom when you factor ALL into the kwh price of batteries, not the calcs on the back of a cigarette box, the financial calcs. A member is busy with that calc, being a financial person.

Off grid on the other hand is a totally different story when you compare gennie running costs to batts, then 48v is a no brainer.

But in cities, with Eskom, you have to be joking with big battery banks. Gooi daai panels for daytime loads If you can use it during the day, which is another little bliksim one can miss.

If you are not home and want to save some real money, last person out switches the DB off with EV tubes for geyser on a solar circulation pompie. 
Then put up a very small solar system for the alarm, garage motor and such.
And if you have a pool, fill it. 

The fact that a person bought 48v does not mean that is the one and only option for everyone else. So give the both sides of the coin, not just ones own preferences.

You have a very good point. 

My idea with my solar system is not to run big loads after sunset. only 150W for lights. 

I want to run the pool for 2 hours  (750W) for 2 hours and then dump 3hours into a 2kW geyser element.  2 x 750 + 3 x 2000 = 7500W

then 150W load for 6 hours.    daily total = 8400W

how big should my PV array be? at the moment it is only 900W and i struggle with the 150w light load. 

batteries dont make sense for my app if the idea is to save money long term. 

My array is 1800W (6x300W) and I generate and use about 6KWh per day in winter, up to 9kWh in summer. If I wanted to get 8.4kWh, I would have to increase my system size to 2700W which would be 9x300W for winter, summer would be OK. 

Right, now we are getting to the facts.

The pool pump has a high startup current and the 2000w for the geyser, both forcing you to a bigger inverter. To power these things from solar is the way to go! Grid tie being the cheapest possible option, and you will save the most. But, T&C apply and you have a inverter and batts.

The problem is the 150w for evenings. I presume from sunset to sunup? 

To give you an idea. To power 150w for 12 hours = 1800wh per day.
If you run lean and mean, 1 day backup and 50% DOD which is not ideal, you will need 4 x 105ah batteries, as you have.

I presume 900w panels = 300w x 3 panesl? To recharge the batteries in 5.5 hours, nothing else powered, you need 2 x 300w panels.
But you must understand, the Royals will last maybe +-250 cycles / evenings, maybe 1 year if your run 150w for 12 hours.

Now at my rate of R2.28 per kw, 150w watts will cost per day: 150w x 12 hours / 1000 x 2.28 = R4.10 per day x 31 = R127 per month.
4 x Royals are wot, R1500 x 4 = R6000 = roughly 47 months of Eskom.

If you do it right, 1 days backup and not below 20% DOD using Trojan T105RE's, they can last maybe 10 years. You now need 8 of them as they are 6v batteries. Cost about R22k. Cost for Eskom could still be cheaper when you consider you have to replace the bank in 10 years time, so you need save each month, which will earn interest, so the calcs for that becomes very interesting.

EDIT: Bottom line, to power the pump, and later the geyser, you need a lot more panels. Batts, reduce the load and see if you can get away with smaller cheaper batteries.

2 hours ago, plonkster said:

Same thing really for eToll.

And same can be said for solar and batteries, there is no load shedding, why waste money today on batteries. Put the money in an investment account and patiently wait for if it returns, then buy a new fancy cool battery bank.

Go grid tie using the smallest possible cheap 2nd hand bank - but do it legally ne!

Off grid, powering a house, yes, 48v, not negotiable. If it is just lights or a very small load, se dinges, gat vir 12v.  (Had to bring that in.)

And as you said in your post above, as you and I have done, you have to carefully assess your own needs and then decide the path forward, planning as many escape routes in as you can, for who knows what they future holds.

If I have to, no choice, go 48v and have no cash, I will sell my inverter for a bargain and with that money I buy a cheap Axpert ... nee se dinges, a 2nd hand 48v online UPS for even less because my controllers are sorted.  

32 minutes ago, The Terrible Triplett said:

Go grid tie using the smallest possible cheap 2nd hand bank - but do it legally ne!

I'll be ordering the Ziehl anti-islanding and the current sensor soon (ish). A good 10k worth of equipment (but only 700 Euro, which in European terms is pretty average really).

That's the last two parts I need, the current sensor is to avoid tripping the prepaid meter. The Ziehl is to comply with NRS-097-2-1. Then to get the paperwork in order, which is likely going to cost some money too, because you have to pay for someone's time to sign it off. Apparently you need like a registered professional electrical engineer to sign off the NRS compliance... and most engineers I know decided not to bother with the professional certification.

After that, the expansion process is pretty much adding more solar modules and MPPT units.

32 minutes ago, The Terrible Triplett said:

for who knows what they future holds

Indeed. If the future is particularly unstable, that would be another reason to go 48V. If there is a chance that your local municipality is going to be funny with connection fees etc, and you decide to just drop them completely... then you want the big one :-)

32 minutes ago, The Terrible Triplett said:

buy a cheap Axpert

I think that they are 6999 ex vat now. That means a second hand one might go for less than 5. For 3k, I'd take the Axpert. R1/watt for a pure sine wave inverter? Bargain, even if it isn't blue :-)

i am also looking at grid tie. this is the cheapest i can find. but they don't have a grid limiter to prevent me putting power back.  please advise..

http://www.bundupower.co.za/solarongrid-single-single.php?mod=SL-TL5000T

YES, Fritz, grid tie is the best option if you can do it legally. Or at least legal, with all the paperwork if the local council does not have regulations in place yet.

City of Cape Town initiative is a good one to follow.

There are guys here that has done this. Options are:
Victron - Plonkster and Edmund.
And then there is a host of other cheaper options. @Jaco de Jongh the most recent.

That inverter has a RS485 interface. Chances are it speaks SunSpec (over modbus RTU). So a limiter is at least possible in principle, though there is no telling if one already exists. There is a local guy who made a limiter for the Fronius, you might want to chat to him. Somehow I think the Fronius is also sunspec.

http://chtech.co.za/

Secondly, it is possible that that inverter supports GFPR (Grid Frequency-dependent Power Reduction). That means that it delivers full power at 50Hz, and linearly reduces the power to zero if you move the frequency up to 53Hz. This is useful in micro-grids/islands, where the GTI ties to another inverter downstairs, usually a Quattro or Multiplus. Probably not what you want to do, as that will more than double the costs.

The other option would be to not use that inverter, and go with something like an infinisolar, Imeon, or whatever, that has this capability already. I don't know any of these inverters though, can't say I'm a fan of either, but that's the easiest and most cost-effective way of doing it right now.

The one I know best is the Victron, but then you need to add at least some batteries. You can run this as a hub-1 setup and configure it not to push back into the grid. For best results, you'll want to add a CCGX as well, and you need to get one of the newer models with anti-islanding that complies with NRS-097-2-1. Not all their models are compliant, and then you need a Ziehl UFR1001e, which is another 7k. Basically... much as I love their stuff, I always have to remind people that it is NOT cheap.

thanks for all the info guys!