How to Calculate Your Battery Capacity?

[Itinerant1]

Originally Posted by 757driver

My Magnum inverters (MS2812 and the MS2012) both react the same. When a charge is first applied it looks at the voltage as displayed on the ME-RC remote. If the voltage is above 12.6v the Magnum chargers go straight to a Float charge. My LFP bank will never see a voltage that low so no Bulk/Absorb phase. For the most part I use solar to keep the bank maintained which will go into Bulk at a much higher voltage or if I need a quick charge back to 100% I will use Equalize set at 14.0v which is just a manual Bulk phase and monitor as it approaches 100% SOC to terminate.

I guess I understand that you're using the one button to trigger a charge at a higher voltage.
I've never used mine like that (equalize button), never thought it was really needed. With the small generator giving a "hold over" charge of 100ah with the Magnum (MSH3012M) doing a "float charge" 13.6v and letting it go for a couple hours that's enough for the solar to take over when it's there. That's why I put those recordings in the other post. Isn't one of the benefits of lfp not having/ needing to go 14v +. Looking at my records, the last time the bank was at 14.1v (my set bulk) was 11/22, 11/3 prior to that. This time of year from late fall to early spring the max charge is 13.6-13.8v from solar with what seems like no ill effects. 11/22 also would of been the last time the batteries were at "my" 100% SoC, I've been bouncing between 30-80% SoC during this period. Once again another reason lfp being used. Last time the system was plugged in other than occasional generator hold over charge was 2 years ago and this is continuous boondocking and before that it was 200 days of boondocking in 2016. System has never been turned off or not in use since April 2016.

Edit after thought..or do you need the higher voltage for your dropins so it syncs your SoC with actual battery capacity?

[757driver]

Originally Posted by Itinerant1

I guess I understand that you're using the one button to trigger a charge at a higher voltage.
I've never used mine like that (equalize button), never thought it was really needed. With the small generator giving a "hold over" charge of 100ah with the Magnum (MSH3012M) doing a "float charge" 13.6v and letting it go for a couple hours that's enough for the solar to take over when it's there. That's why I put those recordings in the other post. Isn't one of the benefits of lfp not having/ needing to go 14v +. Looking at my records, the last time the bank was at 14.1v (my set bulk) was 11/22, 11/3 prior to that. This time of year from late fall to early spring the max charge is 13.6-13.8v from solar with what seems like no ill effects. 11/22 also would of been the last time the batteries were at "my" 100% SoC, I've been bouncing between 30-80% SoC during this period. Once again another reason lfp being used. Last time the system was plugged in other than occasional generator hold over charge was 2 years ago and this is continuous boondocking and before that it was 200 days of boondocking in 2016. System has never been turned off or not in use since April 2016.

Edit after thought..or do you need the higher voltage for your dropins so it syncs your SoC with actual battery capacity?

The newer MHS3012 and ARC remote may react a little differently than my older setup because I have found if I need the generator out in the desert Floating at 13.6-13.8v will give a slightly lower total amps going to the batteries after a few minutes as compared to using Equalize. Perhaps it is the way the two inverters interact when both are charging, hard to say.

Also I keep the Float setting on the Magnum chargers down at 13.5v instead of turning them up and down because if there is a shore power interruption both chargers default to ON once power is restored. Normally I have both chargers turned OFF and if it went any length of time on charge and went unnoticed I would be sitting at 100% SOC continuously, not something I would desire.

With regards to charging I could not agree more, I very seldom take the pack up to 14.0v and/or 100% SOC. To get an accurate reading on the Trimetric battery monitor and the SOC meters in the LifeBlues I will bring the SOC up to full as needed from 50%, probably every 3-4 weeks. While out in the desert with sunny
days I am usually bouncing between 70-95% SOC.

[757driver]

Originally Posted by hclarkx

Got it. Thanks. But, he's wrong about what his curves (the ones you referenced) show; … that being that the battery is full at 13.8V (4x3.45V) because the curve is "vertical after 3.45V" per cell. His curve is anything but vertical where it's rising beyond 3.45V. I wouldn't argue with 90-95% but it's not 100%. If he was charging at a lower current, he would be closer to correct as his curves would rise later and more quickly beyond 13.8V and thus be putting in less energy after hitting 13.8V. Below is a link to a typical set of LFP charge curves for bulk currents from low to modest (.05 to 0.2C). Even at 0.2C, likely lower than in his example, the charge is only 87% when bulk voltage hits 13.8V.

In my case 0.05C is about 15Amps. On days when solar isn't great, my charge current can be 15A or less in late afternoon as the battery approaches 100%. As such I may be very close to 100% when the voltage is still down around 13.8V. In the linked example, 0.05C gets 98% charge at 13.8V. Also note that at 0.05C the voltage does rise very quickly in the last minute or two of charging. This is in part why I've never caught this "spike" even though I know it occurs because my controllers do switch from bulk to float. My best case is about 36 amps from the solar and if my LFP reaches 100% while the current is that high (around 1 or 2 pm, for instance) I'm barely over 0.1C and would be at only about 94% at 13.8V. My SOC monitor confirms this (when it's been resent recently) in that it usually reads 98 or at most 99% with my limited charging current.

In summary, stopping the charge at 13.8V will get close to 100% only if the charge rate is quite low and thus taking many hours. At higher currents the LFP will be well below 100% when voltage hits 13.8V.

If one wants to limit charge to 90%, one needs to set the bulk stage ending voltage in accordance with the charge current. That's a moving target given that solar charging current is highly variable. I recall posting my frustration on IRV2 over a year ago when my $2700 LFP was new and I was trying to stop at 90%. I eventually gave up realizing that the bulk-ending voltage to do that varied widely from day to day so doing so was just not practical. I.e., use 14V +/- and let the chips fall where they may.

https://www.bing.com/images/search?v...x=0&ajaxhist=0

Splitting hairs to a certain extent. Every different LFP battery make will produce a slightly different graph depending on the charge rate and chemistry. In the verbiage below that graph Rod mentions that the battery is essentially full at 13.9v (99.5-100% SOC) because he has many times taken a capacity test right after and found it to be the case.

The other graph (with the white background) shows a full charge at just 3.45vpc or 13.8v where the charge line goes vertical so each one can be ever so slightly different.

[757driver]

Originally Posted by 757driver

The newer MHS3012 and ARC remote may react a little differently than my older setup because I have found if I need the generator out in the desert Floating at 13.6-13.8v will give a slightly lower total amps going to the batteries after a few minutes as compared to using Equalize. Perhaps it is the way the two inverters interact when both are charging, hard to say.

Also I keep the Float setting on the Magnum chargers down at 13.5v instead of turning them up and down because if there is a shore power interruption both chargers default to ON once power is restored. Normally I have both chargers turned OFF and if it went any length of time on charge and went unnoticed I would be sitting at 100% SOC continuously, not something I would desire.

With regards to charging I could not agree more, I very seldom take the pack up to 14.0v and/or 100% SOC. To get an accurate reading on the Trimetric battery monitor and the SOC meters in the LifeBlues I will bring the SOC up to full as needed from 50%, probably every 3-4 weeks. While out in the desert with sunny
days I am usually bouncing between 70-95% SOC.

Just tried an experiment with Float set to 13.5v and Equalize set to 14.0v. Float charge amps settled in at 80 amps total, and Equalize settled in at 170 amps total. I am sure a Float of 13.8v would be much higher but still not as high as Equalize of 14.0v. When running the big 12.5kw generator I want to make the most of the run time.

[hclarkx]

Originally Posted by 757driver

I am somewhat surprised that you are not able to get more amps going in at 50% SOC and 13.8v.

Sorry I did not catch this sooner. As you may have since deduced from my earlier replay, I was up around 90% or more, not 50%, when I was seeing just a few amps. Effectively I was on the tail of what would have been a long exponential current decay had I started charging at 13.8V sooner. Mostly I was indicating a lot of hours to get to or close to 100% with 13.8 in that the last 5 or 10% would be 10's of hours from what I saw.

Earlier I mentioned having only 17 amps, 8 from the generator (current limited to 8) and 9 from my 35 Amp converter in absorb or float mode ... probably float as I could not get it into bulk mode so it probably would skip absorb as well. That experience was indeed with the LFP down below 50%. But the low current was due to not having proper charging equipment ... i.e., so much of "drop-in" and use your existing converter!

[757driver]

Originally Posted by hclarkx

Sorry I did not catch this sooner. As you may have since deduced from my earlier replay, I was up around 90% or more, not 50%, when I was seeing just a few amps. Effectively I was on the tail of what would have been a long exponential current decay had I started charging at 13.8V sooner. Mostly I was indicating a lot of hours to get to or close to 100% with 13.8 in that the last 5 or 10% would be 10's of hours from what I saw.

Earlier I mentioned having only 17 amps, 8 from the generator (current limited to 8) and 9 from my 35 Amp converter in absorb or float mode ... probably float as I could not get it into bulk mode so it probably would skip absorb as well. That experience was indeed with the LFP down below 50%. But the low current was due to not having proper charging equipment ... i.e., so much of "drop-in" and use your existing converter!

Ahhh. That explains it, poor charging equipment would make a big difference in the amount of charging amps.

[Itinerant1]

Originally Posted by 757driver

Just tried an experiment with Float set to 13.5v and Equalize set to 14.0v. Float charge amps settled in at 80 amps total, and Equalize settled in at 170 amps total. I am sure a Float of 13.8v would be much higher but still not as high as Equalize of 14.0v. When running the big 12.5kw generator I want to make the most of the run time.

170a sure does beat 80a on a 1,200ah bank, that speeds up the process nicely. Understood about gen run time, 2 hours is my limit with my "quiet" Honda...yea sure they are. I'll take the noisy solar all day long.

[C.Martin]

It would be nice if Battery Monitoring numbers had the accuracy to calculate “capacity”.

I compared two systems this morning...each having 928aH Interstate GC-2 banks. The first calculation resulted in 2158aH the second bank yielded 1457aH. This is definitely outside the realm of possibility. So...until they manufacture a Battery Monitor with pinpoint accuracy...this is a waste of time. The second bank...at less than C/100 load is reading below 12.1v...definetly NOT 93% SOC. I would guess that bank has less than 50% of it’s specified capacity.

SOC is a quick at a glance way to look at charge and it helps compensate to a degree for loads...but you have to consider inherent degrees of accuracy/inaccuracy. In my experience...not all that accurate. So, apply a buffer to avoid exceeding your desired limits.

As far as LFP systems, specifically drop-ins...Top Balanced systems need to get to and hold for a long enough period...the voltage that enables cell balancing. It would be nice if the manufacturers would publish the BMS specs, or at least see the blinking lights like some of the prismatic cell BMS systems have...so you know what you’re working with. If you have an EV and are trying to get every bit of range out of the vehicle...and to heck with battery life. By all means...charge them to 14.XX volts each and every time. In a motorhome...especially one with a luxurious overhead of LFP, I would rather establish a charge routine that charges to the first part of the knee...and set a rebulk voltage aimed at keeping the battery in the mid SOC range. I can hit “rebulk” feature on the ME-ARC if desired to pump up for a known dry spell. Then adjust the amps on the solar to raise the SOC each day...but not to 100%. Once a week...use the equalization feature set to go to the BMS required voltage to enable periodic cell balancing...for a reasonable time period. But again...one would need to know how far the cells are spreading, and what rate the balancing feature of the BMS system is capable of. I think Battle Borns have four parallel strings of 30 26650 LFP cells...for the 100aH models. No way on earth for the end user to see individual cell voltage. So...without knowing...my battery management technique would be perhaps foolish...perhaps brilliant. No way of knowing really. I can’t be the only person to wish to get 8000 cycles vs 2000 cycles...if that could be achieved.

[757driver]

Charlie,

Much of what I have read about top balancing and balance boards is if the individual cells are fairly well balanced to start with balancing them further just ends up requiring more balancing which becomes a vicious circle. I think most balance boards bleed off the excess current when the trigger voltage is reached BUT it is a small amount (0.5 - 1.0 amp). If the charge rate for four cells is much above 4-5 amps then the balancing can’t keep up and is being overwhelmed. In theory it sounds great but practically I don’t think it works that well. If one has the ability to separate the individual cells and manually rebalance each one would be much better off.

[hclarkx]

Originally Posted by Itinerant1

. . . . . . This time of year from late fall to early spring the max charge is 13.6-13.8v from solar with what seems like no ill effects. 11/22 also would of been the last time the batteries were at "my" 100% SoC, I've been bouncing between 30-80% SoC during this period. Once again another reason lfp being used. Last time the system was plugged in other than occasional generator hold over charge was 2 years ago and this is continuous boondocking and before that it was 200 days of boondocking in 2016. System has never been turned off or not in use since April 2016.

Edit after thought..or do you need the higher voltage for your dropins so it syncs your SoC with actual battery capacity?

You seem to have confidence in your external SOC monitor. Do you know how it maintains accuracy? How does it get reset?

Drop-ins with internal SOC monitoring do need an occasional true 100% SOC to reset; at least mine does (mine was reading high by 20% on one occasion). An external SOC monitor would have the same problem drop-in or otherwise. How do external SOC monitors maintain accuracy in the face of varying internal losses (varying with load and temperature)?

[hclarkx]

Originally Posted by C.Martin

. . . . . .

SOC is a quick at a glance way to look at charge and it helps compensate to a degree for loads...but you have to consider inherent degrees of accuracy/inaccuracy. In my experience...not all that accurate. So, apply a buffer to avoid exceeding your desired limits.
. . . . . .

Well said. Though no SOC monitor would be useful after a while if it did not have some way to reset after a period of highly variable and difficult to predict LFP internal losses. It would be very desirable to know how they work. I know how my internal Lifeblue one works, but what about the external ones? Knowing how they work would allow users to keep them accurate. Does the manual cover this?

You mention ranging from 30-80% SOC ... in my case 30% was actually 10% after a month of SOC not getting over 90%.

[hclarkx]

Originally Posted by 757driver

Splitting hairs to a certain extent. Every different LFP battery make will produce a slightly different graph depending on the charge rate and chemistry. In the verbiage below that graph Rod mentions that the battery is essentially full at 13.9v (99.5-100% SOC) because he has many times taken a capacity test right after and found it to be the case.

The other graph (with the white background) shows a full charge at just 3.45vpc or 13.8v where the charge line goes vertical so each one can be ever so slightly different.

Sure he confirms his charge witha discharge but that doesn't mean he got to 1005 with a short charge period which was likely 0.5C or higher. He only got out what he put in. I have to wonder if he was re-calibrating his SOC monitor or just trusting it. A very big error as we know.

But, I wouldn't call from less than 87% to 100% splitting hairs when our goal is longer life and we really want to know what we are doing. Google LFP charging curves ... there are dozens and they are all very similar to the one I linked to. Rod's curves would be similar too if he expanded them for a closer look. Most curves you find show 0.5C and higher charging where stopping at 13.8V can be as low as 70%. Here's an example; charge at 0.5C and stop at 13.8V. SOC is about 71%. Rod's batteries can't be that different from others.

https://www.batteryspace.com/prod-specs/9055.pdf

Of course, holding at 13.8 will eventually get one much closer to 100% but the current will be lower and dropping over many hours. In our case, we don't have those hours because we are cycling daily.

[Itinerant1]

Originally Posted by hclarkx

You seem to have confidence in your external SOC monitor. Do you know how it maintains accuracy? How does it get reset?

Drop-ins with internal SOC monitoring do need an occasional true 100% SOC to reset; at least mine does (mine was reading high by 20% on one occasion). An external SOC monitor would have the same problem drop-in or otherwise. How do external SOC monitors maintain accuracy in the face of varying internal losses (varying with load and temperature)?

No, I do not know how it maintains its accuracy. What I do know after 32+ months of living off of the lfp/ solar it seems to maintain it pretty well. From looking at readings during summer daily SoC/ voltage they compared to be about the same now, meaning they mirror each other pretty close. Could it be off by a couple % sure it could be and if it is I'm not noticing it. If it where off like yours was of 10-20% then I would see a considerable difference in the SoC/ voltage readings.
Example of one;
6/19/18_06:36_ 36% SoC, 13.11v Battery pack display -2.9a draw, inverter (me-arc display) 13.0v -7a (controller Sleeping)
12/22/18_07:00_ 36% SoC, 13.10v Battery pack display -7.9a draw, inverter (me-arc display) 12.9v -10a (controller Sleeping)

Misc pic of displays...


This one when using the microwave...


Maybe being over solared helps some 1,280w for 500ah bank?
Even now with the panels flat mounted most days it's showing 500-700w, 2.5 - 3.5 kwh harvested, 190 - 250ah to batteries. Days will start getting longer now, better production coming.

How does reset...
The EMS system has a minimum recorded resolution of +/- 0.5A so any current below that level is not accounted for. The system could be setup for a finer resolution, but the shunt value would have to be reduced which reduces the systems maximum current capability.*

The system will reset the capacity to 100% whenever the pack voltage reaches 3.49 to 3.52 V/cell on average, depending on how it was programmed, and the pack current must be negative (charging) and typically under 35A and then it will set the capacity to 100%. Again, these are all firmware variables* so it's possible your system was programmed differently. You should continue charging after this point until the battery truly is full to have the reference point of 100% be correct.*

If you have many small loads that are not accounted for then you would occasionally need to re-sync the pack with 100%. The frequency at which you have to do this will vary depending on your application and usage.

cpu/ ems

[PeterFTH]

Batteries are funny things - a fully charged 12v battery sits at 12.8vdc with no load and no charge. Below 12.2 is considered discharged, although some equipment will continue to operate at levels even lower. Trouble is, the recharge time is exponentially more when you drop below 12.2 Real dead is about 11volts - at that level you probably cannot even start your genset. I have solar which keeps me well charged all day, but by evening I run the generator until the charger output falls below 50 amps ... might be half an hour, might be 2 hours. Skipping this boost will come back to haunt you after a few days when your charger output (75 amps in my case) will take several hours to rejuvenate the batteries. Try not to let them get below 12.2. Peter