Optimal Batteries SOC

[PaulF007]

Now I assume that this has been asked but just for fun Ill ask it again.

I was always under the impression that to get the best value for money out of your batteries you must run them at 20% soc. But then I had a chat this morning with someone and it I did some calculations after wards by the look of things you should get better value if you run them at  70% soc per cycle. 

I used the T105-RE for the calculation and I did a price per khw. So to start here is my formula that I used 

Battery price / ((V x A x Cycles x(SOC/100)) /1000) = give you the price per kwh ( IS THIS THE CORRECT WAY TO CALCULATE)

Below is the Answers that I got.

Just for fun the Speck on DOD

 

Any thoughts on this?

[___]

29 minutes ago, PaulF007 said:

Any thoughts on this?

12 cents on R2 is 6%. You could easily be off by that much on reading the chart, so I don't think you can say what is best. Your calculations show pretty much what I've also calculated for myself, that there isn't an optimimum DoD that gives you better bang for buck. The price per kwh is within a few percent pretty much the same :-)

[DeepBass9]

Doesn,t make sense. You have paid a fixed amount of money for those batteries. If you choose to wear them out in six months instead of five years then you wasted a lot of money.

Manage your load and make your batteries last as long as possible. I can,t think of any scenario where abusing batteries to 90% DoD will save you any money, or be more cost effective. 

In fact your numbers prove the case that using expensive battery kWh as opposed to cheap solar kwh makes no sense. The optimal DOD is zero, but if you are forced to use batteries, best be kind to them........

[Haggers]

Hi, if trojan is r2000 per battery then less than r2 per kwh which is cheaper than eskom over time period guaranteed. then look at the attached which i took from another site (guy name biltong did it) says omnipower even cheaper. if this the case, then payback better.

So priority of use solar then batteries then eskom?

Thoughts?

[DeepBass9]

Batteries don,t produce power, they store it, so add whatever the cost of the power was in the first place, plus efticiency in and out, plus the battery power cost. The whole premise of the above calculations is flawed. Solar, then eskom, then batteries!!!!!! 

[___]

@DeepBass9 makes a good point. It has to be said that as a lead acid charges above 85% SoC the efficiency comes down significantly, you could be losing as much as 15% of your energy during float charging. However, I think in these calculations the generation and efficiency is often ignored with good cause: If you have the oversized array already and the power would normally just go to waste, you are justified in looking at JUST the cost of storing that energy for later use.

[Guest]

43 minutes ago, plonkster said:

...a lead acid charges above 85% SoC the efficiency comes down significantly, you could be losing as much as 15% of your energy during float charging.

True but as you say, if the panels, load and batts are matched, then you will be in float whilst powering the load, not so?

So there are losses but as you say, you have a panel or so more in any event.

[Jaco De Jongh]

4 hours ago, DeepBass9 said:

I can,t think of any scenario where abusing batteries to 90% DoD will save you any money, or be more cost effective. 

@PaulF007 This is not an answer to your question, it is just my opinion on the topic and  to some of the statements above.

@DeepBass9, i think the use of batteries in a solar setup is a personal choice. You have 2 choices, you can buy  30K bank and keep it charged, waiting for the 3 or 4 power failures a year to use them. But one thing you forget is that a battery has a life span, if you use it or not.

Have you experienced a situation where you buy a ups for your pc, its stands in your office, no power failures for 2 years, and then the day you need it, there is hardly enough backup power to close the document you were busy on. Those batteries packed up, just standing there being kept at 100% soc. The same can happen to your 30k bank, or.......

You can use it, and cover some of the cost of the bank. The Trojans at a 20% discharge should give you 4000 cycles, if you do that on a daily basis that should give you 10.95 years. I take 4-5kwh from my bank on a daily basis, that will leave me with an SOC of 75 to 85 %every morning. Lets say i get 3000 cycles from my bank the total KWH over that period will be 12000-15000kwh @ an average of R1.30/kwh (Prepaid rates, I am not even sure of the rates). Theoretically my R30k bank can save me R15600-R19500 on my electricity bill over that period. (at current rates). I am almost certain with the planned increases the bank will pay for itself over that period. At the current rates it might seem as if it will be cheaper using grid, but over the lifetime of the batteries I think the Battery is the cheaper option.

I hope you see what i am trying to say, Buy the R30k bank, let it stand for 10 to 12 years or until it fails and toss the R30k away, or buy the R30k bank, use it for 8.2 years, save R19500 (plus increases)on your electricity bill, and if it fails after 8 years replace it.

Please correct me if i am wrong, with my small calculator that was the only thing that made scene to me. Buy it and use it, don't let it die of old age unused.

[SilverNodashi]

8 hours ago, PaulF007 said:

Now I assume that this has been asked but just for fun Ill ask it again.

I was always under the impression that to get the best value for money out of your batteries you must run them at 20% soc. But then I had a chat this morning with someone and it I did some calculations after wards by the look of things you should get better value if you run them at  70% soc per cycle. 

I used the T105-RE for the calculation and I did a price per khw. So to start here is my formula that I used 

Battery price / ((V x A x Cycles x(SOC/100)) /1000) = give you the price per kwh ( IS THIS THE CORRECT WAY TO CALCULATE)

Below is the Answers that I got.

Just for fun the Speck on DOD

 

Any thoughts on this?

I did some similar math two years ago, but with different batteries, and the consensus was similar to your findings. Some batteries gave better bang for back, at different levels than others. 

Here's the major issue with all of this: Who ever uses their batteries upto 20%, every night? Without purposefully shutting down your inverter (or switching back to eskom, if you can) it's very difficult to stick with the numbers. 

 

[___]

2 minutes ago, SilverNodashi said:

Here's the major issue with all of this: Who ever uses their batteries upto 20%, every night? Without purposefully shutting down your inverter (or switching back to eskom, if you can) it's very difficult to stick with the numbers. 

With the hub-4 setup I'm playing with you can actually do a 20% DoD very accurately every night. You set the max discharge rate to something between the C10 and C20 rate of the battery and the maximum depth of discharge to what you want, and it will blend that much power into your Eskom mix until it hits the depth of discharge you asked for. But yes, you're right... this is rather an exotic setup, not common at all. Hitting the numbers takes work.

[SilverNodashi]

Just now, plonkster said:

With the hub-4 setup I'm playing with you can actually do a 20% DoD very accurately every night. You set the max discharge rate to something between the C10 and C20 rate of the battery and the maximum depth of discharge to what you want, and it will blend that much power into your Eskom mix until it hits the depth of discharge you asked for. But yes, you're right... this is rather an exotic setup, not common at all. Hitting the numbers takes work.

I know the hub-4 can do this, and other systems can do it as well. The issue lies with the fact that people's livestyle dictates that energy used differs every night. As example I was working in the garage till about 30 minutes ago, and my wife is still watching TV. Last night she went to bed earlier and I only worked on my laptop. That's two exta cups coffee, about 40W extra light, an extra radio, and about 120W longer for the TV. 

The other night, I had the humidifier on in the kids room, all night long, and I think it's a 1Kw unit. So unless one can change this behavior, limiting batteries purely on SOC is almost impossible. Perhaps, rather design your bank with your desired Watt/hour rating and SOC level, and add 40% for variance. If it's not used upto the 40% mark, you score, if it is, you're still within budget and expectations. 

[PaulF007]

You see when I looked at these figures yesterday I thought ok every time we have a question regarding batts and the design of them if you need 4kwh of power per night and using @Jaco de Jongh 's small calculator   4000 / 48 v  = 83ah  now at 20% soc that would be a 83 / 20% = 420ah bank costing x amount of capital layout. But if your cost per kwh on the batts is not that much different , run the bank at 60% then you will need only a 140ah bank to achieve the same goal and replace the bank sooner. Just a thought. 

Then regarding:

7 hours ago, Jaco de Jongh said:

At the current rates it might seem as if it will be cheaper using grid, but over the lifetime of the batteries I think the Battery is the cheaper option.

I see this argument almost every time and my first answer is well , then wait for Eskom to become more expensive than batts and then go and buy your batteries then , then you will have the saving immediately and would not have to wait for three years. The above example , if you go for 20% Soc , the bank will set you back about R 37 000 , give or take.But if you use the grid at R2.00 ( I know some places it's even cheaper)  per kwh at night time this will give you  about R 250 per month for your eskom bill ( R2 x 4kh x 30). Now that is a monthly payment that is full under your control.  If you go away on holidays you use only one unit per day ... 

Now let's look at the scenario. 

1. If you wanted to use your own money to buy the batts but decided to use Eskom and invested your money it shows that you will be able to pay the grid with just the growth earned depending on the investment portfolio.  I have taken a Ultra conservative vs a moderate aggressive portfolio just to compare and that money will take you quite far before you could break even..

Now if you borrowed the money then you can work a the 10.95% and then you will never catch up.Event at double the price of the grid you will be hard pressed to "beat the bank" so to speak if you had to borrow the money to buy the batts.

The only argument that , I personally feel , could be made is that you buy units into the feature so in 10 years time your units will be half the price that you paid for them in the beginning. Whereas the grid will at least stay with inflation. But if you consider the risk involved with batteries and after ten 10 years the batts will also be kapoet, well that the beauty of us people every one unto its own.

 

[viper_za]

7 hours ago, SilverNodashi said:

Here's the major issue with all of this: Who ever uses their batteries upto 20%, every night? Without purposefully shutting down your inverter (or switching back to eskom, if you can) it's very difficult to stick with the numbers. 

Software developed on this forum makes this very easy indeed

[viper_za]

3 hours ago, PaulF007 said:

I see this argument almost every time and my first answer is well , then wait for Eskom to become more expensive than batts and then go and buy your batteries then , then you will have the saving immediately and would not have to wait for three years

If I wait for E̶s̶k̶o̶m̶ Batteries to be cheaper than Eskom the bank's price would have increased due to inflation and then you are back to square one again.
But if you buy it now, it will eventually catch up and be cheaper  than Eskom, so use it in the beginning not cycling too deep just to keep them cycling every now and then and when Eskom after a few years gets more expensive then start cycling deeper.

Just as an example with my own batts now.
Bought them at the beginning of the year for just over 20k.
Last week the price of 1 battery went up with R500 x 8 for my bank so work out how many more years this alone adds to my example

Edited November 17, 2016 by viper_za
Fixed it thx Paul

[DeepBass9]

8 hours ago, Jaco de Jongh said:

@PaulF007 This is not an answer to your question, it is just my opinion on the topic and  to some of the statements above.

@DeepBass9, i think the use of batteries in a solar setup is a personal choice. You have 2 choices, you can buy  30K bank and keep it charged, waiting for the 3 or 4 power failures a year to use them. But one thing you forget is that a battery has a life span, if you use it or not.

Have you experienced a situation where you buy a ups for your pc, its stands in your office, no power failures for 2 years, and then the day you need it, there is hardly enough backup power to close the document you were busy on. Those batteries packed up, just standing there being kept at 100% soc. The same can happen to your 30k bank, or.......

I hope you see what i am trying to say, Buy the R30k bank, let it stand for 10 to 12 years or until it fails and toss the R30k away, or buy the R30k bank, use it for 8.2 years, save R19500 (plus increases)on your electricity bill, and if it fails after 8 years replace it.

 

In a grid tie situation, where the only use for the batteries is power backup for power outages, what the above figures show is that it makes no sense whatsoever to buy a massive battery bank and hardly use it, rather buy a small bank and work it hard on the rare occasion when you do need to use it.

If the power is on, you are making no 'saving' by using an expensive battery bank which costs R2+ for the power coming out of it, when you could be using grid power at R1.  I still don't see how the conclusion is reached that if you discharge your bank to 90% because the kwh is 'cheaper' then you have saved something. If you are replacing a R30K battery bank every year, instead of every 5 years you are losing big time.

Off grid where you use batteries every night, you want them to last as long as possible, and the kwh cost of the power stored is irrelevant, so you cycle them as little as possible.

 

[PaulF007]

1 hour ago, DeepBass9 said:

f you are replacing a R30K battery bank every year, instead of every 5 years you are losing big time.

But that's exactly my point you won't need a R30k bank as you are using a lot more from a much smaller bank. Remember your usage will stay the same it is only your bank size that will reduce not your consumption that will increase.  

I agree with the idea that you use a small bank as backup and when it is necessary you can work it hard for once twice a month and have it on standby. Obviously that is n a Town scenario where you don't have line charges and all  

 

[PaulF007]

1 hour ago, viper_za said:

If I wait for Eskom to be cheaper the bank's price would have increased due to inflation and then you are back to square one again.

But Eskom is already cheaper than your bank so why wait  

Exactly my point batteries will only be cheaper than eskom due to the fact of Inflation , You are effectively buying prepaid power at lets say R 2.35 per kwh for maybe 10 years so yes it must become "cheaper: at some stage . Only problem is your batts will still only last ten years as if it takes 5 years to become cheaper you will only have 5 years left for the "cheaper" rates.

But as showed above if you had to borrow the money to by the batts you will never ( I now that's a bold statement and looking for trouble) catch up ..

[___]

Cape Town Domestic above 600kwh is R2.28. This year the increase was 6.8%, but only because they moved a bunch of people who were on the wrong tariff, otherwise it would have been 8% (ish). So conservatively we can expect above inflation increases for the foreseeable future, between 6% and 8%, but the year before that it was 11% and I think we all remember the 25% three-years-in-a-row thing that for all practical purposes kick-started the domestic solar revolution... :-)

Aaaanyway, so assuming a 7% increase next year I'm looking at R2.42 from next year on. If I can do R2.35 with a battery right now... I'm inclined to agree with the sentiment that it makes sense to pay 5 years or 10 years ahead at next year's price... as I would get 3 to 7 years at a lower price. Even if it DOES cost more this year...

[viper_za]

3 minutes ago, plonkster said:

Aaaanyway, so assuming a 7% increase next year I'm looking at R2.42 from next year on. If I can do R2.35 with a battery right now... I'm inclined to agree with the sentiment that it makes sense to pay 5 years or 10 years ahead at next year's price... as I would get 3 to 7 years at a lower price. Even if it DOES cost more this year...

Finally someone sees it like I do

[DeepBass9]

But, it costs R2.35 on the way out of the battery (thoeretically). You have to put the power in there as well, so either you play eskom R2.28 to charge and R2.35 to withdraw, or work out  the cost that solar is costing you to charge, and add that to R2.35. Plus some efficiency factors. Still doesn't make sense. Grid tie with no batteries is the most cost effective still. Batteries are only for backup in a grid tie situation, where the cost of sitting in the dark, outweighs the cost of batteries.

[PaulF007]

2 minutes ago, DeepBass9 said:

Grid tie with no batteries is the most cost effective still. 

Finally someone sees it like I do

[DeepBass9]

2 minutes ago, PaulF007 said:

Finally someone sees it like I do

 In fact, a small backup generator may be more cost effective than batteries in the grid tie situation.

[viper_za]

7 minutes ago, DeepBass9 said:

But, it costs R2.35 on the way out of the battery (thoeretically). You have to put the power in there as well, so either you play eskom R2.28 to charge and R2.35 to withdraw, or work out  the cost that solar is costing you to charge, and add that to R2.35. Plus some efficiency factors. Still doesn't make sense. Grid tie with no batteries is the most cost effective still. Batteries are only for backup in a grid tie situation, where the cost of sitting in the dark, outweighs the cost of batteries.

DeepBass you should then remember also your PV panels/Inverter mostly pays themselves for your day time consumption.
You can't simply add all the costs to your batteries. It's doing so much more than just charging batteries
(This will give a way more complex calculation)

Edited November 17, 2016 by viper_za

[DeepBass9]

True.

[___]

1 minute ago, DeepBass9 said:

You have to put the power in there as well

The context here is someone with surplus power. Our fictitious home owner has a GTI on the roof and a backup inverter in the garage, and he wonders if he should invest in a battery bank and do a bit of self consumption or not. The business/ROI plan for the GTI is already in place and paid for in daytime hours.

The surplus power is going to waste at the moment, so it is in the same category as the turbo on your engine: reclaim it or not? The energy is essentially "free"... not really of course, Turbos fail, so do batteries... turbos cause a bit of back-pressure, so there is a small cost... can't think of a similar battery analogy. So lets assume that it is free for the sake of argument and revisit that later.

I think the trick is to sit around waiting for specials on batteries. Every now and then you get a Lead Crystal for 6.5k or something, 3000 cycles at 50%...

200*12*3000*0.5 = 3600kwh, 6500/3600 = R1.81.

Okay, so having ascertained that it COULD work if the energy is "free" or at least very cheap. If I now revisit that assumption: can we think of any reason why the surplus energy might not be free? Any reason why the inefficiency (essentially only reclaiming 70% of my surplus) matters? I'm sorry... I don't see it... maybe I should but I don't.

:-)

The problem: If I have to buy now to get it cheaper in 3 years time... too many unknowns in that gamble. Also, big capital outlay. No immediate benefit.

If I have to buy now to get it cheaper next year? Well now we're talking! Less of a gamble. Still big outlay though.

GTI remains the most cost-effective, yes. Just attempting to see if there is a case for batteries. Some good investment guy will quickly point out that putting the money in some investment will way outperform the 7c/kwh next year anyway... :-)