So what really counts as a "cycle"?

Re: So what really counts as a "cycle"?

Difference between a "cycle" and a "micro-cycle":

Cycle is to a given DOD then recharge to a given SOC. Specific levels are not necessary; only the operational action is. The levels do not have to be the same each time either.

Micro-cycle is when a cycle is interrupted either on the charge or discharge with the opposite action for a short period of time. An example of this is when a heavy load comes on during charge cycle and PV is not able to supply it and continue charging (even at a reduce level) so for the time of the load the batteries are drain somewhat, then charging resumes after the load is off.

For practical purposes there's no good way to figure micro-cycles much less predict them. We know they have an effect on battery life, but it's just one of the things you have to suck up. Otherwise you are dealing with averaged DOD cycles and recharging to near enough to 100% that it makes no difference and trying to avoid several days of partial recharging below 90%.

For promoted cycle life purposes companies have specific parameters used in laboratory tests to determine this. It will involved controlled temperatures, fixed current discharge, and grid-based recharging ability to precisely meet the best performance characteristics for their batteries. It isn't exactly "real world" stuff.

BTW "Going from 100%SOC to 20%SOC and back to 100%SOC" is called "killing the batteries as fast as possible"; you really shouldn't go to 20% SOC even with forklift batteries.

Re: So what really counts as a "cycle"?

CALLD wrote: »

Cariboocoot: thank you for answering most of my questions, just two more though:

As for micro cycles how significant are they? I may have quite a few of them on cloudy days as the PV output goes up and down with every passing cloud. Let's say there is a steady load of 120watts on the battery, PV output can be over 700watts between clouds and as low as 50watts when a dark cumulus cloud passes. This can happen dozens of times a day!

If it is still charging the battery, even at a reduced rate, it doesn't count as a micro-cycle because the operation hasn't changed from 'charge' to 'discharge'. When it does you have the micro-cycle, and trying to determine its effect on battery life is like trying to pick the right lottery numbers; good luck. Even if you could determine the depth and length of the micro-cycle I'll bet no manufacturer has ever even tried to equate this to a full-cycle effect because it is something that really only occurs in RE applications so they simply would not consider it important.

If you find you have a regular occurrence of noticeable micro-cycling you should consider adding more PV to try and prevent the loss of charging during the load time.

Then of course there is Port Elizabeth's oceanic climate - in a given month there will be 10 days of full sunshine, 12 days of interrupted sunshine where production is up to 70% and 8 to 9 days that are pretty much overcast for most of the day. In the latter scenario production can be lower than 30% of full sunshine days. How does a climate like this affect battery life if there is no way to quickly bring them up to 100% other than wait for sunshine? I have a small charger capable of adding a mere 6 amps to the batteries, but it's nothing compared to the 27-28 amps the PV can supply on a good day..

What you have to watch out for is how many days below what percentage of SOC you experience and how often. If you find you have regular bouts of <75% SOC for more than two days then you definitely have a condition which will affect the battery life. Again the first choice is upping the PV to improve charging on marginal days, but it is sometimes cheaper to just run the generator. Or the standard advice of "25% DOD one day one, 50% DOD on day two, start the gen on day three".

This is also one of the situations where LiFePo's SOC tolerance can be advantageous. But you'll have to think long and hard about whether you want to put the money into the different (and as yet largely unused in RE) chemistry or merely up your PV and possibly battery capacity to give yourself more tolerance to the variable conditions.

Re: So what really counts as a "cycle"?

Typical of company promotions it is somewhat exaggerated: most batteries that are used in RE do not require 100-150 cycles to reach full capacity. Large capacity ones (2V cells, forklift, and those produced by companies not interested in quality control or customer service) will (and it would be nice if they told you this up front) but the smaller ones will all be in-line within 30 days of normal use and you'll never even notice (unless you have a battery monitor which can be instantly wrong due to this initial increase in capacity).

BTW I'm not impugning the company, just the standard "our are better because we say so" claim. Notice the other battery shown appears to be a Trojan, and could well be an industrial model rather than a small motive-force type (i.e. T105).