Zero Sum vs. Sliding Window Charging

[evm1024]

Zero Sum vs. Sliding Window Charging Strategies.

Zero Sum charging is a charging strategy where by any charge used is fully replaced in the cycle time. Typical cycle times are one day and the bank is sized to be 1.5 to 2 times the cycle (days) use.

So if you were using 200 AH per day you would size solar to fully replace your usage in 1 day or around 200 AH worth of solar. With 1.5x to 2x daily usage we end up with a bank size of 400 AH.

This is a pretty normal setup. When we exceed the capacity of solar charging we run our engines or generator to bring us back to fully charged (e.g. 80% SOC). With 2x bank capacity we are able to “slide” out of zero sum for a day or 2.

But in the Sliding Window mode This sliding is built into the design by choice.

In that case we size our house bank to be 5 or 10 times the daily usage. With 200 AH daily usage our bank becomes 1000AH to 2000AH in size (cost, cost, yes we know. Just ignore it for now. This is theory).

Solar (or other prime charge sources) can be sized at 100% of daily usage more or less.

The “window” is 200 AH and is 1/5th of a 1000 AH bank. With a Min and Max bank SOC of 20% and 80% we have 600 AH for the 200 AH daily use window to “slide” in.

When the daily charge equals the daily use the window stays put. When the daily charge is greater than the daily use the window slides up to 80 SOC and we stop charging for the day.

When the daily use is greater than the daily charge the window slides closer to 20%SOC.

When the window slides to 20% SOC we bring on additional charging capability e.g. we run the engine or generator.

So if we were to use some FRN (Farcical Random Numbers) to show the min-max window sliding we could get something like (starting with 80% SOC)

80% - 20% usage = 60% SOC + 20% charge = 80% (Day 1)
80% - 20% usage = 60% SOC + 15% charge = 75% (Day 2, a touch cloudy)
75% - 20% usage = 55% SOC + 15% charge = 70% (Day 3, still cloudy)
70% - 20% usage = 50% SOC + 10% charge = 60% (Day 4, really cloudy)
60% - 5% usage = 55% SOC +20% charge = 75% (Day 5, low use with good sun)
75% - 25% usage = 50% SOC + 5% charge = 55% (Day 6, high use, low charge)
55% - 25% usage = 30% SOC + 5% charge = 35% (Day 7, new beer in fridge, towel on solar array)
35% -25% usage = 10% SOC + 40% charge = 50% (Day 8, run generator)
50% - 25% usage = 25% SOC + 40% charge = 65% (Day 9, ran generator again)
65% - 25% usage = 40% SOC + 40% charge = 80% (Day 10, ran gen, resynced SOC)

I would venture to say that most of us are in the zero sum charging mindset but in fact use some form of sliding window. This is more or less a default.

But to actually use sliding window to its advantage we need to change our mindset and slid the window by design with an increase the total house bank capacity to cover more than 2x daily usage. 5x to 10x being the ballpark size.

Food for thought….

[smac999]

if you use 200ah a day. you want a bank to last 2-3 days (clound, rain, not motoring etc). and a bank is only drawn to 50%

that is 800-1000ah. that is pretty simple...

[evm1024]

Lead Acid thinking - 50% DOD.

LiFePO4 thinking 80%DOD, 90%DOD.

I thought it was obvious that this example is for LiFePO4.

It is simple but we are talking about a shift in thinking. A shift from Lead to Lithium, a shift from zero sum to sliding windows.

As many have said we need to toss all that we know about batteries when dealing with LiFePO4. Sometimes what we know slips in (as in 50% DOD for example) and trips us up.

Lead Acid by the limits of its chemistry must be zero sum. Failure to fully charge (often enough) greatly reduces a LA battery's life.

Whereas LiFePO4 does not even yawn if you do not reach fully charged. Preferred storage is at around 50% SOC.

I think that picking 2 x daily usage for LiFePO4 batteries works fine (that is where the 400 AH came from). But, that is still thinking in zero sum (with a good reserve) IMHO. By using sliding windows as a major part of the design we end up with a much larger bank. Thus 800 or 1000 plus AH. The downside is cost.

With a nice balance of charge capacity, usage and Bank capacity we end up with a daily net charge that mimics hailstone numbers for weeks if not longer.

[valhalla360]

Originally Posted by evm1024

Lead Acid thinking - 50% DOD.

LiFePO4 thinking 80%DOD, 90%DOD.

I thought it was obvious that this example is for LiFePO4.

Nope, you never indicated battery type, so the logical assumption is to match what 95% of boats use...lead acid...in which case you would kill the batteries in under 6 months (if that long). With no knowledge of your background, you could easily be the typical newbie that assumes if you have a 400 amp-hr bank, you can use 400 amp-hr.

For this analysis, really it's easier to calculate the "usable" amp-hr needed and then adjust for the battery type at the end.

You have lots of writing but basically, how many cloudy days do you want to be able to handle before you crank up the motor...multiply that times your average daily amp-hr estimate and that's the "usable" amp-hr you need...then simply divide by the percentage of amp-hr that you believe a battery technology can handle, and you have the battery size you need.

The solar array should typically be able to put out more than your average daily amp-hr usage unless you run the motor or connect to shore power regularly, in which case sizing for average may be viable as that will top you up every few days. If away from other charging sources, if you set up your array to match daily usage on a sunny day, every day you collect less, you will never recover until you crank up the motor or plug into shore power. 30-50% more is probably a good rough starting point if you are mostly in sunny areas. If you tend to be in areas that get a lot of cloud cover, you might want double or more of your daily use.

[rgleason]

Of course there is another level, "Daily Capacity"
Max Daily Use: Cruise: 100ah Sea: 140ah Use 140ah
Bank: 140ah/75% LFP available estimated = 186ah Battery needed. > 200ah LFP bank.
Solar: Sized to replace 140ah in one day, at normal cruising grounds, in in normal use, during spring or fall.
Alternator: Sized to replace 140ah at a C-rate of .5C (occasional faster charge, normally .3C or .35C) with continuous duty output in the specific boat of .5C x Estimated Max Daily Use = .5C x 140ah = 70ah
Alternator alternative: However it would provide more margin to use .5C x LFP bank = .5C x 200ah = 100a Alternator continuous duty output. (See Maine Sail's MarineHowto.com the Alternator article for how to account for small frame alternators heating up and providing anemic charge.) Normal Charging rate would be .3C or .35C for LFP longevity.
Charging: Two good charge source SolarPV and Alternator should make up for not having a bigger bank

This would provide roughly the equivalent of a 400ah FLA (Lead Acid) Bank without the need to bring the batteries to 100%SOC all the time (4-5 hours charging time) compared to LFP alternator charging at (140ah daily load / .3C x 200ah) = 2.333 hours to provide the full 140ah daily load. SolarPV would reduce house loads during the day, so with good solar insolation there might be a 2 day period between engine operation, or the engine can be run for less time.

Occasional fast charging at 140 ah daily load / .5C x 200ah = 1.4 hours charging.

[evm1024]

OF course I should have specified LiFePO4 or any technology that could be safely discharged to 20% SOC or greater.

My oversight.

you could easily be the typical newbie

evm1024
Marine Service Provider

Join Date: May 2011
Location: PDX
Boat: Gulfstar 50
Posts: 1,368

[AKA-None]

Remember that anyone could be reading this who may assume you meant lead acid

[evm1024]

Originally Posted by AKA-None

Remember that anyone could be reading this who may assume you meant lead acid

Quite true - I am remiss. And no slight was intended. I stand corrected.

LA is just not part of my thinking - It must be Zero Sum charging or you will kill it in short order.

[valhalla360]

Originally Posted by evm1024

It must be Zero Sum charging or you will kill it in short order.

No. You just need to stay within the "usable" amp-hr range.

Only difference is you have more "usable" amp-hr for the same "rated' amp-hr with lithium.

If you are worried about not getting to 100% charge on lead acid ever, do your bulk alternator charging in the morning so the solar system has hours to top up the last 10%. But assuming your solar system averages 120-150% of daily average, it likely gets to 100% (or at least pretty close) most clear sunny days.

[evm1024]

I think that we are way off topic (Sliding Windows in LiFePO4 or other battery type that can handle PSOC cycling).

Question: What is the usable AH range for FLA (expressed as SOC)?

What is the lifespan of FLA that is not charged to near 100% SOC.

The goal of zero sum is to get back to 100% SOC every cycle. Not always possible of course.

The goal of sliding windows is to "float" the SOC window inside of the usable charge envelope. Getting to 100% SOC is not a goal.

[noelex 77]

Originally Posted by evm1024

LA is just not part of my thinking - It must be Zero Sum charging or you will kill it in short order.

Charging lead acid batteries to 100% or near 100% occasionally is certainly helpful, but I am not sure where the impression was created that this needs to be done for most cycles to have a reasonable life. Perhaps the use of AGM batteries, which are very sensitve to requiring a frequent 100% charge has created this impression, but it is not true for most types of lead acid batteries.

Back in the days before solar, long distance cruising boats rarely raised their battery bank SOC above 80%. The last 20% is very slow to add and it was uncommon to motor long enough to add the last 20%. In those days sailboats sailed .

My impression was that battery life was shorter in the days before solar, but not drastically so, despite the rare attainment of 100% SOC. Before solar the average battery bank SOC was considerably lower and I think this factor is more influential on lead acid battery life.

The important message is that lithium should not be managed anything like lead. The best SOC for lithium is not entirely clear but the evidence suggests around 60% or perhaps lower, whereas for lead it is 100%. Neither of the targets is practical all the time if you are using the batteries, but charging practices should keep these ideal numbers in mind.

[evm1024]

You are right in that the focus was for LiFePO4 batteries.... and not to discuss lead acid charging.

It is over stating to day that FLA needs to be fully charged every cycle. That is the ideal of course. Rolls has a good description of some life factors in FLA here:

Battery Sulfation : Technical Support Desk

But let's do return to the premise of this thread, to wit:

- The sliding windows charge/discharge regime requires a bank size greater than 3 or 4 time daily use.

- The sliding windows charge/discharge regime with a bank sized accordingly has significant advantages in terms of projected lifespan (the SOC stays much closer to the above quoted 60%) and "reserve capacity".

- It is desirable to use sliding windows by design rather than happenstance.

[valhalla360]

Short of day sailors who plug in each night...pretty much standard to use sliding window, for longer term at anchor...battery type is irrelevant.

Never heard it suggested that you must recharge every day (zero sum)...it's nice if you can and realistically, you need more charging capability than the average use, so by default, you may wind up most days being zero sum because otherwise, you eventually will need to crank up the engine or plug in if you don't have charge capability that at last equals your typical usage.

[tanglewood]

I think every battery system operates in a sliding window, and it's just a question of how often the edge of the window comes up against 100% SOC. With a lead acid based system, it needs to max out on a regular basis. With an LFP system the window can move around with lots of flexibility.


A stand-by system is about the only application where you want zero-sum, parking the SOC at 100% until backup power is required, then going back to 100% SOC and parking there when power is restored.

[evm1024]

My reason for posting was not to address how our battery systems end up operating in the real world. I was pointing out that in the design/sizing phase we can set the parameters with a specific mode in mind. And that there are advantages to doing so.

There are many examples on my boat where I have battery systems operating by design in zero sum or total loss modes.

As an example a start battery is operated in zero sum. My rechargable
VHF radio is operated in zero sum. Keep them fully charged, use it and then recharge to full. My cell phone too.

Flashlights as an example and cordless power tools are total loss systems. Use them till they stop working and then recharge (or replace) the battery.




I was pointing out not how a battery system was operated in practice but
rather what mode the system was designed to operate in. And more
specifically designing to a specific mode can cause suttle changes in