Depth of Discharge Myth?

[evm1024]

Originally Posted by rom

It seems to me you will reach 100% SOC in about the same time for both batteries. With 150A charging capacity they will both reach absorption very quickly and stay there for a while.


I could read here and there some cruisers advising a smaller bank (LA) as more cost effective. Why not, I understand the idea. But that doesn't work for everyone. Some of us, probably more & more, need high ouput, and therefore higher capacity, even though they don't really need that capacity in terms of Ah. One more vote for LFP...

You would think so but take a look at the diagram below.

It shows that the charge acceptance is quite high in the 50% SOC region (95% or so) and tapers off above 80% SOC.

In terms of area under the curve the overall charge acceptance is near 81% for a FLA bank discharged to 50% SOC and 63% for the bank discharged to 80% SOC.

So the bank with the higher charge acceptance will charge at a greater rate that the bank with the lower overall acceptance rate.

Of course, the last few % will take forever for both banks.

If we replace 150 AH with a charge current of 150 amps at 81% average acceptance we can expect to get to near 100% in around 1.23 hours, at 63% average acceptance we can expect around 1.6 hours.

[ramblinrod]

Originally Posted by evm1024

One of the problems with back of the napkin calculations is that they only give ballpark results. It is really in this case a matter of the time step in the calculation.

Our time-step has been 24 hours. So we say 100 ah dray per day and pretend that that draw is the same each hour. But of course it is not.

Solar output peaks midday. Heat load peaks around 3 or 4 pm. Amp hours consumed jumps all over the place. Running lights at night, Radar, AP etc when on a passage (hardly a constant draw). Domestic loads when at anchor.

Basically you cannot predict when you will run out of power with the 24 hour average calculations. You will know that if you use more than you make you will run out but predicting when (2:30 am? Let's cherry pick shall we)

I've been putting together a model that uses a 1 hour time-step to take a look at usage patterns. Of course it will make some significant WAGS for some of its forcings but others are based on real data (solar irradiance vs time of day for example).

It is not quite ready for showing yet but I am getting some interesting results.

The most interesting results is that system which have less solar/wind than the daily usage will kill their FLA banks in not much time at all... unless ICE is run to 100% SOC when needed to be run. This is not true for LiFePO4 which is happy to sit and cycle back and forth at PSOC.

I'll keep working on it a bit and then present in a new thread when it is more refined.

So as I have stated more than a few times now, to design a properly balanced system one needs to perform an energy needs assessment.

Of course the instant hourly consumption will vary from the average daily consumption/24, as I have also previously indicated many times.

My previous statements regarding running out of capacity were based on 24 hour consumption divided by 24 for a direct comparison of FLA vs LFP at 1/2 capacity as this is a reasonable approximation.

When utilizing solar charging, it is common for the batteries to be at or near 100%, at or near sunset, and then begin the overnight discharge. Wind (if any) will be a wild card variable that could be anywhere, however their could be a trend for wind to be up at 11 and down at 7, of course this rule of thumb is no where near as accurate as solar being up after sunrise and down by sunset.

Because the 24 hour consumption is likely variable by +/- 50% +/- variation by individual use, the time of day of capacity comsumption and charging can certainly vary so wildly, that paralysis by analysis is virtually worthless.

The actual time has little bearing.

The key point is when one reaches 50% SOC in an FLA bank, it is not an emergency to get ICE charging going immediately whereas when LFP reaches 10% it is.

So you can perform all of the WAG you wish as required to skew numbers however you wish, it will not change the fact the LFP systems without reserve capacity are generally bad design.

A possible exception is off-shore racing vessels where some crew should always be monitoring and able to intervene at any time of day.

These people are slave to ICE charging to meet their mission. My experience is that many cruisers don’t wish to be and LFP reserve capacity will help alleviate ICE charging urgency.

[mbartosch]

Originally Posted by evm1024

So for you folks on this thread who switched to LiFePO4, did you stay the same or go up in capacity or did you reduce your capacity?

Replaced 400 Ah of LA with 400
Ah LiFePO4 which is totally sufficient for our cruising needs (12 V system; 100 - 120 Ah on the hook, 250 - 350 Ah on the move).

In addition to this there are two 40 Ah portable LiFePO4 battery packs complete with BMS which are primarily used to power the electric dinghy motor. Both packs can individually be connected to the charge bus for recharging. They can also be configured to extend the house bank (effectively yielding a total of 480 Ah).
Finally, the battery packs could be used to power the boat electrics in the unlikely case the primary bank fails. The BMS of both packs could also be scavenged to easily replace a defective house bank BMS.

[rgleason]

Bruce Schwaub has written:

Regarding Li, I have a simple rule of thumb: If you have enough solar to cover your loads and condition your batteries, Pb will be fine. If your loads are such that you have to charge with fuel, then you should consider Li.

Not quite sure where that puts running eng/alt to charge 50-80% SOC for an hour or so as being acceptable or not, but it is one knowledgeable person's guide about Li.

[evm1024]

Originally Posted by mbartosch

Replaced 400 Ah of LA with 400
Ah LiFePO4 which is totally sufficient for our cruising needs (12 V system; 100 - 120 Ah on the hook, 250 - 350 Ah on the move).

In addition to this there are two 40 Ah portable LiFePO4 battery packs complete with BMS which are primarily used to power the electric dinghy motor. Both packs can individually be connected to the charge bus for recharging. They can also be configured to extend the house bank (effectively yielding a total of 480 Ah).
Finally, the battery packs could be used to power the boat electrics in the unlikely case the primary bank fails. The BMS of both packs could also be scavenged to easily replace a defective house bank BMS.

Elegant solution, I like it.

So far the sample size is 3 with a 100% equal or greater LiFePO4 bank replacing the FLA bank.

[newhaul]

Originally Posted by ramblinrod

So as I have stated more than a few times now, to design a properly balanced system one needs to perform an energy needs assessment.

Of course the instant hourly consumption will vary from the average daily consumption/24, as I have also previously indicated many times.

My previous statements regarding running out of capacity were based on 24 hour consumption divided by 24 for a direct comparison of FLA vs LFP at 1/2 capacity as this is a reasonable approximation.

When utilizing solar charging, it is common for the batteries to be at or near 100%, at or near sunset, and then begin the overnight discharge. Wind (if any) will be a wild card variable that could be anywhere, however their could be a trend for wind to be up at 11 and down at 7, of course this rule of thumb is no where near as accurate as solar being up after sunrise and down by sunset.

Because the 24 hour consumption is likely variable by +/- 50% +/- variation by individual use, the time of day of capacity comsumption and charging can certainly vary so wildly, that paralysis by analysis is virtually worthless.

The actual time has little bearing.

The key point is when one reaches 50% SOC in an FLA bank, it is not an emergency to get ICE charging going immediately whereas when LFP reaches 10% it is.

So you can perform all of the WAG you wish as required to skew numbers however you wish, it will not change the fact the LFP systems without reserve capacity are generally bad design.

A possible exception is off-shore racing vessels where some crew should always be monitoring and able to intervene at any time of day.

These people are slave to ICE charging to meet their mission. My experience is that many cruisers don’t wish to be and LFP reserve capacity will help alleviate ICE charging urgency.

most of this has been proven to be incorrect but there is also the DOD damage to consider .
Every time you go below 50% on Fla you are irreparably shortening the life significantly as had been shown with many real world examples .
The same is true with Lfp as well just not to the degree that it erodes the life of Fla.
DOD loss of life is what this thread is about after all . Not an infomercial.

[newhaul]

Originally Posted by evm1024

Elegant solution, I like it.

So far the sample size is 3 with a 100% equal or greater LiFePO4 bank replacing the FLA bank.

so for now it appears that the 50% less lfp installed is an internet myth.
installing like ah capacity actually seems to increase the usable ah by at least 50% and still saving a bit for the oh crap situation .

[ramblinrod]

Originally Posted by evm1024

I'll grant you everything you said.

But I will also note that:

1) 35' cruising boats are not going to put in 200 AH of LiFePO4. So a bit of a red herring there. Yes, I know that some have said that you need half as much LFP as you do FLA but that is not in the boundry conditions. You want to do this apples to apples? Be sure that the daily consumption is not nearing the Batteries capacity.

I don’t know how you can predict what others are likely to do or not do on their 35 ‘ cruising sailboat. It seems we have at least one cruiser here with it.

I am glad to see you affirm my statements that LFP without about the same reserve capacity as inherent with FLA will lead to problems.

2) also as noted with the FLA at 50% you have run them flat in terms of their usable capacity. Any further use is in the damage zone. As has been noted. See the graph below.

IMHO this is a fabrication. There is no single number above which is not damaging and below which is damaging. FLA are not “FLAT” at 50% SOC. There is lots of usable capacity left. The only caveat, the deeper the discharge the more urgent not to leave them there longer than necessary / practical.

3) at 100 ah per day in one day your 200 ah LiFePO4 bank will be have 50% left, 100 AH. It can go more than 12 hours before it gets to LVD.

Not with the instant daily consumption 50% greater than design, as indicated.

And even with that 12 hours more use it will be sitting at 25%. Plus at 25% SOC it will be around 13.0 volts with more than enough oomph to start your engine. The FLA bank will be around 12 volts and sag quite a bit under load at 50% DOD.

Good luck with your calcs. I look forward to reviewing them.

[evm1024]

No need to be afraid of doing the state of charge calculations 24 time a day (hardly instantaneous).

Let's look at it a different way. So you are sailing to Hawaii. You do shoot the moon and a few stars as darkness falls and you have a fix. Once a day you know where you are. You apply a correction and do it all again.

Or you let you GPS tell you where you are . You log it hourly and make corrections hourly.

Which would you rather do?

Your 24 hour time step give a rough estimate.

Speaking of skewed numbers. You skew the results by cutting the LiFePO4 bank capacity in half. This of course lets you get your 10% SOC figure.

From our limited sample of people who have switched over to LiFePO4 we are running 100% kept the same bank size or make it larger. Let's do compare apples to apples and not skew things in FLA's favor.

The key point is when one reaches 50% SOC in an FLA bank, it is not an emergency to get ICE charging going immediately whereas when LFP reaches 10% it is.

So you can perform all of the WAG you wish as required to skew numbers however you wish, it will not change the fact the LFP systems without reserve capacity are generally bad design.

So it appears that you basically admit that you have presented a case where you posted a defective LiFePO4 system vs you ideal FLA system.

Respectfully, apples to apples.

[evm1024]

Originally Posted by ramblinrod

I don’t know how you can predict what others are likely to do or not do on their 35 ‘ cruising sailboat. It seems we have at least one cruiser here with it.

I am glad to see you affirm my statements that LFP without about the same reserve capacity as inherent with FLA will lead to problems.





IMHO this is a fabrication. There is no single number above which is not damaging and below which is damaging. FLA are not “FLAT” at 50% SOC. There is lots of usable capacity left. The only caveat, the deeper the discharge the more urgent not to leave them there longer than necessary / practical.



Not with the instant daily consumption 50% greater than design, as indicated.



Good luck with your calcs. I look forward to reviewing them.

Not a fabrication, it is a working rule that is supported by quite a lot of research. (did you just call me a liar? I will give you the shadow of the doubt)

I think that anyone you ask will say that at 50% SOC their FLA bank is in need of a recharge. It's usable capacity used up. To go any further is to dip into no man's land. OK, don't like 50% lets use 45%. Don't like 45% OK, pick a number, show us the study where you got that number. Otherwise let's do agree on 50%. We know that it is a rule of thumb.

Oh, then again there was that chart I posted (posting 618) that shows the effects on cycle count of a 50% DOD.

And again no one gets a smaller LiFePO4 bank. To your 200 AH bank is bogus. It is just cherry picking. Apples to apples.....

[ramblinrod]

Originally Posted by evm1024

No need to be afraid of doing the state of charge calculations 24 time a day (hardly instantaneous).

Let's look at it a different way. So you are sailing to Hawaii. You do shoot the moon and a few stars as darkness falls and you have a fix. Once a day you know where you are. You apply a correction and do it all again.

Or you let you GPS tell you where you are . You log it hourly and make corrections hourly.

Which would you rather do?

Your 24 hour time step give a rough estimate.

Speaking of skewed numbers. You skew the results by cutting the LiFePO4 bank capacity in half. This of course lets you get your 10% SOC figure.

From our limited sample of people who have switched over to LiFePO4 we are running 100% kept the same bank size or make it larger. Let's do compare apples to apples and not skew things in FLA's favor.



So it appears that you basically admit that you have presented a case where you posted a defective LiFePO4 system vs you ideal FLA system.

Respectfully, apples to apples.

Stating”respectfully” does not make a post “repsectful”.

In a cas like this, it actually can have the exact opposite result.

Did you know this before you posted?

And no, I presented a possible and likely normal use case, where one elected not to include “reserve capacity” for an LFP bank, compared to the inherent reserve capacity in FLA with a 50% normal design usage cycle.

[evm1024]

Originally Posted by ramblinrod

Stating”respectfully” does not make a post “repsectful”.

In a cas like this, it actually can have the exact opposite result.

Did you know this before you posted?

And no, I presented a possible and likely normal use case, where one elected not to include “reserve capacity” for an LFP bank, compared to the inherent reserve capacity in FLA with a 50% normal design usage cycle.

I said respectfully and mean it respectfully. Apples to apples. 400 AH FLA to 400 AH LiFePO4.

You did present a case where you elected to compare a 200 AH LiFePO4 boat to a 400 AH FLA boat in order to "prove" that FLA was a better choice.

Not apples to apples was it.

So far 100% of the LiFePO4 conversions self reported have equal or greater capacity. That make the half size LiFePO4 bank you proposed appear to be stacking the deck.

My numbers and opinions stand on their own merit. As do yours. Let's take the high road and deal with numbers and facts.

[newhaul]

Ok back on track depth of discharge and life cycles
First is Lfp second is Fla

[evm1024]

Originally Posted by newhaul

Ok back on track depth of discharge and life cycles
First is Lfp second is Fla

Great!

The model that I'm working on will be able to explore the SOC at any given hour of the day. One of the exciting things that it will show (is currently showing) is how much time the battery is spending at PSOC. It indicated the lowest DOD reached and has provision for random solar output (read clouds).

Of course anytime there is not enough solar/wind you will be forced to run a generator. This model will allow for some guesses as to when it is best run and how often based on your energy needs predictions.

It is a rather crude model and I'm hacking it together so don't expect much.

[john61ct]

Originally Posted by evm1024

You would think so but take a look at the diagram below.
…
If we replace 150 AH with a charge current of 150 amps at 81% average acceptance we can expect to get to near 100% in around 1.23 hours, at 63% average acceptance we can expect around 1.6 hours.

Sorry does not work that way.

And forget "near 100%", the ideal is actual 100% Full and

it is the **only** point (other than 0%) where you actually know with precision what your SoC is.

"Average acceptance" is meaningless without getting to true Full and also without a known bottom point.

Any reasonably depleted lead bank - say below 60% SoC - will require 5-7 hours to get to Full.

It is true when high amps are available - say .2C for FLA and .8C for AGM - most of that time will be spent with the bank only accepting a tiny fraction of that potential level.

And with solar or other low current source - say under .05C - then current will remain pretty level for most of the total time.