Battery cycle- vs design life.

[___]

A question. I was wondering how service life relates to cycle life, because it appears they are not the same thing. I also wonder about the cycle life itself. Here's why: From what I read in various googled places, it appears that the cycle life is the amount of full cycles you can put a battery through before it has lost 20% of it's capacity. Many of these articles also talk about how DoD plays a large role, but there is one question I have not managed to find an answer to. I'll put it as two options.

 

If a battery (for argument's sake) has 800 cycles at 50% DoD, does it mean,

 

A. After you've done 800 cycles it has 80% of its original capacity/performance left, or

 

B. After 800 cycles, it has 50% or less of the original capacity left, it is useless in any application that needs at least 50% of that capacity.

 

In the case of A, it would mean you could likely get 1500 cycles out before it's at 50%, which means a much longer life. Reason I ask, lots of batteries come with 1500/1800 cycles to 70%, but only 600-750 cycles to 50%, yet their design life is 9 years? Is the 9 years the most optimistic number you might get if you keep the batteries on float for 9 years straight...? :-)

[Deonsr]

These calculations are theoretically and only to try an prove the theory..... The life cycles of a battery is exponential to the Dod. For example theoretically a 100A/hr 12v battery is a 1,200w battery @ 100% Dod. (12v x 100A = 1,200 watts) If you would use the SLA (Sealed lead acid) battery at 100% Dod, you should have 1 Cycle and the battery will be toast. It will never recover from thps shock. If the manufacturespecs indicate 1800 cycles and it is normally indicated at 50% Dod, theoretically the following applies:

 

If you however use the battery to 50% Dod, you use 50% of the 1,200w (600w) before recharging and could then expect the 1800 cycles.

If you use the battery to 70% Dod, you use 70% of the 1,200w (840w) before recharging you could then expect probably only 1000 cycles.

If you use the battery to 90% Dod, you use 90% of the 1,200w (1080w) before recharging you could then expect probably only 360 cycles.

 

The specs you quote are not very good and you should probably source better batteries. On a good affordable battery you should expect +- 3000 cycles at 50% Dod.

[SilverNodashi]

 

 

The specs you quote are not very good and you should probably source better batteries. On a good affordable battery you should expect +- 3000 cycles at 50% Dod.

Those batteries are quite costly...

[Wetkit]

Another interesting thing I have seen is that a battery cycle is NOT a full charge/discharge cycle!!!

If you only use 5% and then charge the battery, that is a cycle!

Even worse is a partial charge/discharge.

Say you use 50% and then start charging, but before you at 100% again, say 80%, you start discharging again. Very bad for a normal battery

[___]

Only reason I asked. I can get my hands on a pair of Victron AGMs (not the Gel batteries) at less than half their normal price (a battery that's been in a UPS doing nothing for 6 months basically).

 

Their data sheet says they are good for 600 cycles to 50%, and 1500 cycles to 70%. It also says this battery is best for your holiday cabin (aka not for cyclic use).

 

The Gel batteries can however do 800 cycles to 50% and 1800 to 70%. Which is only _slightly_ better.

 

Now all this sounds rather low for a very expensive battery! At what point does it become a good deal? :-)

[SilverNodashi]

Only reason I asked. I can get my hands on a pair of Victron AGMs (not the Gel batteries) at less than half their normal price (a battery that's been in a UPS doing nothing for 6 months basically).

 

Their data sheet says they are good for 600 cycles to 50%, and 1500 cycles to 70%. It also says this battery is best for your holiday cabin (aka not for cyclic use).

 

The Gel batteries can however do 800 cycles to 50% and 1800 to 70%. Which is only _slightly_ better.

 

Now all this sounds rather low for a very expensive battery! At what point does it become a good deal? :-)

It doesn't... A battery will always be a bad expense. 

[___]

It doesn't... A battery will always be a bad expense. 

Ag come on man... you know what I mean :-) At what point is it cheap enough that it can be treated as a consumable that's not too much of a waste of money (given that it's always a waste of money :-P)

[SilverNodashi]

Ag come on man... you know what I mean :-) At what point is it cheap enough that it can be treated as a consumable that's not too much of a waste of money (given that it's always a waste of money :-P)

I didn't mean it's a waste of money, but IMO batteries are never a good investment simple due to the fact that they seldom last as long as you expected them to, and too many factors can influence a battery's life time for it to have a good ROI. Perhaps, one day, when we use highly effective supercaps, it the scenery might change?

[___]

Agreed. There is never a good ROI. What you need is for your system to make/store enough energy so that over a 7-12 year period (assuming you replace batteries in that time) the batteries pay for themselves PLUS give a return that's more than the interest you'd get if you just put the money in the bank. And that is a tall order :-)

 

In my case, the batteries are really there so that load-shedding doesn't affect production (which for me is writing computer software), so the batteries are paid for in terms of recovered productivity (not in saved electricity costs). So it's a justifiable expense.

 

When buying batteries, however, you at least want some value for money. I suppose the answer is obvious: work out the total round trip Watt-hours (cycles times DoD times capacity) and divide the cost through that number for a R/Wh number. I think I'll do that.

 

(Now that's Duck-splaining :-) )

[Chris Hobson]

6 month old batteries at less than half their new price sounds like a bargain. If you can get more than 1000 cycles out of them then that is a bargain. So it depends how hard you are going to work them since they do not seem to be solar batteries.

 

I may be wrong but I feel battery technology is changing so fast that one should use the smallest bank that will give you reasonable autonomy and use them until they no longer can carry your household draw overnight. South Africa is a sunny place and we have inherited (without question) Northern Hemisphere recommendations in terms of battery bank size. Batteries are also the one part of your solar installation that you know you will need to replace. The longer one can delay replacement the greater one's options will be. The rest of the installation could last 20 to 25 years and by then there definitely will be new technology.

 

.

[SilverNodashi]

Agreed. There is never a good ROI. What you need is for your system to make/store enough energy so that over a 7-12 year period (assuming you replace batteries in that time) the batteries pay for themselves PLUS give a return that's more than the interest you'd get if you just put the money in the bank. And that is a tall order :-)

Some say you buy the cheapest you can get and replace them often. Some say buy the most expensive set you can lay your hand on and pay off the debt ;

In my case, the batteries are really there so that load-shedding doesn't affect production (which for me is writing computer software), so the batteries are paid for in terms of recovered productivity (not in saved electricity costs). So it's a justifiable expense.

That, in itself, is a different kind of return on your investment. In this case it doesn't really matter what you spend on batteries, nor how often you replace then (3 years / 7 years / 10 years), but rather the fact that you have guaranteed uptime (maybe not, if you only have one MPPT and one inverter). This is my case as well. We operate a couple of businessed from home and uptime is more important to me than the money I spent on the battery bank. I used to spend that money on a generator, petrol and services in any case. But now I have zero downtime when eskom fails. And I don't pay for electricity during office hours.

When buying batteries, however, you at least want some value for money. I suppose the answer is obvious: work out the total round trip Watt-hours (cycles times DoD times capacity) and divide the cost through that number for a R/Wh number. I think I'll do that.

 

(Now that's Duck-splaining :-) )

Check this out: http://www.4x4community.co.za/forum/showthread.php?t=222455&highlight=omnipower

[___]

So these are the options on my horison. Two new, two used (but in good condition). Looks like the UPS battery will fill the bill even if I'm going to kill it in about two years :-) Quick round-trip cost per stored kwh spreadsheet.