LFP's Charging Voltage and State of Charge

[Cpt Pat]

I'd like to float an idea here. Since LFP batteries don't have an issue with sulfation, users are free to set any arbitrary SoC as their personal maximum. No harm will come to the battery from maintaining it in a partial SoC below the battery maximum attainable charge.

But then discussions between us become a bit confusing and belabored by differences between a self-defined maximum and the measured absolute maximum, i.e., "my 100% is different than your 100%."

When measuring gas pressures, there are two common terms: absolute and gauge. Absolute is a measurement referencing a vacuum (an unchangeable reference) as "zero." Gauge is a pressure above or below ambient atmospheric, which is nominally 14.7 PSI or 1013.1 millibars - but can vary significantly from absolute measurement with temperature, altitude, and local barometric pressure.

When measuring how close a pressure is to the bursting threshold of a tank, gauge is the best specification because it represents how much internal outward pressure is in the tank measured against the opposing and compensating inward pressure of the atmosphere. When measuring how many molecules of gas are in the tank, absolute is the best measurement because a gauge measurement will vary with atmospheric pressure while the number of molecules inside remain the same.

So... Expressing 100% SoC using the maximum possible charge a LFP battery can hold -- how many electrons are in the "tank" referencing how many electrons the "tank" can hold, I'd call that 100% absolute (at 25 degrees C). When expressing the maximum SoC used in my practical operation of my battery, I'd call that 100% gauge measured against my maximum charge set-point. Finally, there's a third number: the 100% factory rating, which may be above (ideally) or below 100% absolute. No special nomenclature is needed for that number since that number can be stated explicitly or implicit in context as the "factory rating." Saying: "I have a 100 A/H battery," it's implied that I am referring to the factory rating.


Example: "My 100 A/H (factory rating implied) battery has an absolute capacity of 110 A/H. I have 100% gauge set at 80% absolute."

The purpose of any technical nomenclature is to simplify the language by removing the need and tedium of continually explaining how the number is derived. The reference doesn't need to be consistently stated when it should be clear in the context what is meant, but if there is a risk of confusion, it can be used as a convenience to avoid tedious explanation and unneeded arguments over semantics.

What does the community here think of that?

[tanglewood]

Originally Posted by Cpt Pat

I'd like to float an idea here. Since LFP batteries don't have an issue with sulfation, users are free to set any arbitrary SoC as their personal maximum. No harm will come to the battery from maintaining it in a partial SoC below the battery maximum attainable charge.

But then discussions between us become a bit confusing and belabored by differences between a self-defined maximum and the measured absolute maximum, i.e., "my 100% is different than your 100%."

When measuring gas pressures, there are two common terms: absolute and gauge. Absolute is a measurement referencing a vacuum (an unchangeable reference) as "zero." Gauge is a pressure above or below ambient atmospheric, which is nominally 14.7 PSI or 1013.1 millibars - but can vary significantly from absolute measurement with temperature, altitude, and local barometric pressure.

When measuring how close a pressure is to the bursting threshold of a tank, gauge is the best specification because it represents how much internal outward pressure is in the tank measured against the opposing and compensating inward pressure of the atmosphere. When measuring how many molecules of gas are in the tank, absolute is the best measurement because a gauge measurement will vary with atmospheric pressure while the number of molecules inside remain the same.

So... Expressing 100% SoC using the maximum possible charge a LFP battery can hold -- how many electrons are in the "tank" referencing how many electrons the "tank" can hold, I'd call that 100% absolute (at 25 degrees C). When expressing the maximum SoC used in my practical operation of my battery, I'd call that 100% gauge measured against my maximum charge set-point. Finally, there's a third number: the 100% factory rating, which may be above (ideally) or below 100% absolute. No special nomenclature is needed for that number since that number can be stated explicitly as the "factory rating." Saying: "I have a 100 A/H battery," it's implied that I am referring to the factory rating.

The purpose of any technical nomenclature is to simplify the language by removing the need and tedium of continually explaining the reference. The reference doesn't need to be consistently stated when it should be clear in the context what is meant, but if there is a risk of confusion, it can be used as a convenience to avoid tedious explanation and unneeded arguments over semantics.

What does the community here think of that?

I like it, but if Absolute and Manufacturer spec numbers aren't the same, how do you define Absolute?


It seems to me we need two terms; The manufacturer, rated, or nameplate values. And the applications working or operational values.

[john61ct]

Originally Posted by Cpt Pat

discussions between us become a bit confusing and belabored by differences between a self-defined maximum and the measured absolute maximum, i.e., "my 100% is different than your 100%."

Even more so when it hasn't been defined.

Used to be people claimed "I only cycle between 20 and 80% for longevity".

They thought stopping at 13.8V was sacrificing 20% off the top, and with stopping at 12V sacrificing 40% of capacity altogether!

So of course many members say with a $4000 investment, heck with that, and throw out the longevity baby with the bathwater.

_______
Hence my A, B & C examples, in line exactly with what you're talking about, and demonstrating that "avoiding the shoulders" for longevity

not only does not in fact require sacrificing significant capacity,

but serves no useful purpose, other than the marketing agenda of the "drop in" myth.

_______
So, for those that understand these finer points, there is no "canon definition" for 100%,

when people start throwing capacity and SoC numbers around, it's appropriate to ask for how they actually measure their units.

What I recently learned from Maine Sail, is that

later stages of charging, the BM is recording AH accepted that are not appreciably raising SoC,

in order to see "how Full" a given charge profile gets your bank compared to rated AH or the vendor spec'd Full,

you must do load testing.

Fortunately with LFP it doesn't require 20 hours as with lead.

[john61ct]

Originally Posted by tanglewood

how do you define Absolute?

I think using the mfg spec Voltage, and holding that until Zero amps accepted, at say a .1C current rate, might work for "Absolute Full", in my post above that's "Benchmark B", and personally would not inflict that IRL on a bank I bought.

If the mfg publishes an endAmps spec, then that is different, lower, could call it "Vendor Full", or maybe default to .03C, but personally I see no need to add a fourth (!) benchmark.

Finally, what the user / owner considers "100% Full" for BM purposes, since the above are harmful to longevity.

Voltages, matched to a given current level, then not-too-low endAmps when needed for precision, "my C definition".

Then you have AH Rating (my "A" benchmark) which is back to arbitrary, from the vendor and often subject to Marketing's input.

To me, only C is relevant, once I see it's just a few % from damaging Absolute B becomes irrelevant.

I'd much rather say "I top balance at 105% SoC"

If I have to work with a BM that never goes above 96%, how will I keep it accurate?

(ref OT note wrt SG200 above)

[newhaul]

Originally Posted by Cpt Pat

I'd like to float an idea here. Since LFP batteries don't have an issue with sulfation, users are free to set any arbitrary SoC as their personal maximum. No harm will come to the battery from maintaining it in a partial SoC below the battery maximum attainable charge.

But then discussions between us become a bit confusing and belabored by differences between a self-defined maximum and the measured absolute maximum, i.e., "my 100% is different than your 100%."

When measuring gas pressures, there are two common terms: absolute and gauge. Absolute is a measurement referencing a vacuum (an unchangeable reference) as "zero." Gauge is a pressure above or below ambient atmospheric, which is nominally 14.7 PSI or 1013.1 millibars - but can vary significantly from absolute measurement with temperature, altitude, and local barometric pressure.

When measuring how close a pressure is to the bursting threshold of a tank, gauge is the best specification because it represents how much internal outward pressure is in the tank measured against the opposing and compensating inward pressure of the atmosphere. When measuring how many molecules of gas are in the tank, absolute is the best measurement because a gauge measurement will vary with atmospheric pressure while the number of molecules inside remain the same.

So... Expressing 100% SoC using the maximum possible charge a LFP battery can hold -- how many electrons are in the "tank" referencing how many electrons the "tank" can hold, I'd call that 100% absolute (at 25 degrees C). When expressing the maximum SoC used in my practical operation of my battery, I'd call that 100% gauge measured against my maximum charge set-point. Finally, there's a third number: the 100% factory rating, which may be above (ideally) or below 100% absolute. No special nomenclature is needed for that number since that number can be stated explicitly or implicit in context as the "factory rating." Saying: "I have a 100 A/H battery," it's implied that I am referring to the factory rating.


Example: "My 100 A/H (factory rating implied) battery has an absolute capacity of 110 A/H. I have 100% gauge set at 80% absolute."

The purpose of any technical nomenclature is to simplify the language by removing the need and tedium of continually explaining how the number is derived. The reference doesn't need to be consistently stated when it should be clear in the context what is meant, but if there is a risk of confusion, it can be used as a convenience to avoid tedious explanation and unneeded arguments over semantics.

What does the community here think of that?

nicely written however I still have the question where is the white paper to back all of the assertions you have made throughout this and other threads?

[john61ct]

It's been made clear several times:

Originally Posted by Cpt Pat

Please see my link in post #3 above for a practical discussion.

Also see https://arxiv.org/ftp/arxiv/papers/1803/1803.10654.pdf as regards the need to do coulomb counting. That paper has a number of cross-references.

Even if that, **plus** Maine Sail's statements here, **and** the other links I provided above, is as Peggy Lee sang "is that all there is?",

and that means **you think** it's all a pile of hooey,

do please refrain from this badgering in my thread as well as others.

As stated many times, your rig your choice!

From my boilerplate:


_______
Nearly **every** cell vendor, also selling ancillary hardware touted as "LFP ready", gives charging voltages **way too high** for longevity.

Optimizing for longevity is just not high on their list of priorities. All their focus is on their main customers, the military and EV propulsion use cases, where range/per-use capacity is everything, and expected lifespans short. Long-term "House bank storage" use is not even on their radar in economic terms, at most .0001% of the cell manufacturer's market.

Therefore industry and academic papers are not targeted toward maximizing longevity, but oriented around those EV & other high C-rate use cases, so **very** different from our much gentler House bank cycling.

And thus no testing / different specs are developed by industry for the House bank use case; that has been left to the end users, and the more objective/technical vendors like*Maine*Sail.

Consider also, that the auto industry could easily have developed utilitarian cars and*parts*supply chains, so that their customers could have the option of buying vehicles that routinely last 50-100 years. Not talking any tinfoil conspiracy here, but that has never happened, and for purely rational free-market economic reasons.

So, following those vendor charge specs is fine **if** you only want the cycle lifetimes they advertise. You can however get **much** longer life by looking at the vendor's SoC vs Voltage chart, and "avoiding the shoulders" at both ends; stay well within the smooth, low sloped parts of the curve.

[newhaul]

Originally Posted by john61ct

It's been made clear several times:Even if that, **plus** Maine Sail's statements here, **and** the other links I provided above, is as Peggy Lee sang "is that all there is?",

and that means **you think** it's all a pile of hooey,

do please refrain from this badgering in my thread as well as others.

As stated many times, your rig your choice!

From my boilerplate:


_______
Nearly **every** cell vendor, also selling ancillary hardware touted as "LFP ready", gives charging voltages **way too high** for longevity.

Optimizing for longevity is just not high on their list of priorities. All their focus is on their main customers, the military and EV propulsion use cases, where range/per-use capacity is everything, and expected lifespans short. Long-term "House bank storage" use is not even on their radar in economic terms, at most .0001% of the cell manufacturer's market.

Therefore industry and academic papers are not targeted toward maximizing longevity, but oriented around those EV & other high C-rate use cases, so **very** different from our much gentler House bank cycling.

And thus no testing / different specs are developed by industry for the House bank use case; that has been left to the end users, and the more objective/technical vendors like*Maine*Sail.

Consider also, that the auto industry could easily have developed utilitarian cars and*parts*supply chains, so that their customers could have the option of buying vehicles that routinely last 50-100 years. Not talking any tinfoil conspiracy here, but that has never happened, and for purely rational free-market economic reasons.

So, following those vendor charge specs is fine **if** you only want the cycle lifetimes they advertise. You can however get **much** longer life by looking at the vendor's SoC vs Voltage chart, and "avoiding the shoulders" at both ends; stay well within the smooth, low sloped parts of the curve.

I don't even question Rod's ( mainesail )data he shows all of the data he used to derive the information he has put out .
However cpt pat has made several assertions without providing any white paper or links to the data from which he makes his assertions. Data which would be extremely useful wrt my specific setup.
Some of his other assertions go directly against what Rod has stated numerous times so I am just trying to collect the data. ( perhaps it is from a more recent study )

[CatNewBee]

First of all, capacity is in Ah, not A/H, it is an abbreviation to Wh, you just divide them by nominal voltage.

Second most assertions are wrong here, mainly based on speculations and non-experience with a real life LFP battery connected to real life charger and loads.

It is simply WRONG, that charge will continue infinitely when charge switches to FLOAT, this is not the case, the battery discharges and does not accept infinite current, she discharges about 5% and then stays simply at a high SOC of 95% not being charged or accepting a single Amp.

Real life usage does not discharge the battery to zero and very rarely would hit 20% SOC, simply because you plan you system with some reserve capacity for bad weather or higher energy demands in the future and you not want to run out of power on the water, you use every sun ray to charge if possible, you not disable charge sources when sailing.

My battery stays and operates most of the time between 60 and 95% SOC and is very happy about it. I have up to now only 8 cycles counted on the BMV, but have charged and used around 40000 Ah, 28000 since last BMV reset - that would translate to 40 times the battery capacity.

The battery just works, charges and delivers energy, mainly charged and kept at high SOC by solar, I do not disconnet it for storage when leaving the boat, the bilge pumps can use infinite power if necessary.

cycle counts are ridiculous measures steming from FLA, where you can pimp random marketing figures like 300 cycles at 50% SOC, other claim 600 cycles at 70% SOC, etc, what means nothing, the first figure is in fact usable energy over the life cycle of 150 times the nominal capacity and on the second 180 times the nominal capacity, so marginally better, but far from twice the life expectancy.

On my use patern, the battery is charged daily, sometimes twice the day, sometimes it takes several days until he is full again, so I could claim, I have done 360 cycles already, because she is one year old now, but this would be also nonsense.

So what really counts is the energy stored and used off the battery, manufacturer claiming 5000 cycles at DOD of 80% or down to SOC 20% means 4000 times the nominal capacity, of 1000Ah so for my battery it would be 4000000Ah or 52MWh, at home I use around 5200kWh per year, so I could run my home on that battery for 10 years, my cat is much less demanding, so 20 years of usage living all year long on board are possible if no major desaster occurs.

I strongly recommend to built a LFP battery and test it before posting or philosophing about longuetivity, trailing amps and manual balancing preventions.

IF you want most peace of mind and layd back joy on the water without watching all day at the gauges and flipping switches of your battery, invest in a good BMS, that takes care of balancing and cell protection, set the charging sources accordingly for BULK, ABSORPTION and FLOAT, and let them do the magic, it works very well.

[Q Xopa]

This guy is an RVer but seems to be singing pretty much the same tune.

He has a bunch of interesting graphs talking about heat heat, Rate of Charge (C's), SOC levels etc effect on Cycle life. I think he got mostly form Battery University. It's not just referring to LFP but all Lithium chemistry.

https://www.youtube.com/watch?v=Yf9N9zBgyB8

Not that any of this stuff should be considered as gospel as it is rapidly developing technology. But being that different people from different fields are coming up with many similar findings it is adding to the weight of evidence. There is probably some listening to the same Cool Aide stories in there too.

[john61ct]

Very relevant discussion here from this post on http://www.cruisersforum.com/forums/....php?p=2790242

At this post http://www.cruisersforum.com/forums/....php?p=2794268 I thought of Al Thomason's Viking Star open source Very Smart Regulator / "Smart Alternator Regulator" project.

From my email to him:

Could your SAR unit start out at high Amps current as Voltage remains low,
then scale amps down as voltage rises,
Stop lowering current at a floor setpoint, then stop charging (or open a remote solenoid isolating that bank from the charge buss) based on a voltage ceiling setpoint
?

Sample:
Start out at 200A, hold as long as V < 13.72, then drop to 150A
when V < 13.77, drop to 100A
at 13.82, 50A
at 13.85, 20A
Stop charging (or open the remote solenoid) at 13.95V.

[john61ct]

The discussion I linked to above is **really** worth parsing for anyone coming across this topic via googling in the future.

[Q Xopa]

Originally Posted by john61ct

Very relevant discussion here from this post on http://www.cruisersforum.com/forums/....php?p=2790242

At this post http://www.cruisersforum.com/forums/....php?p=2794268 I thought of Al Thomason's Viking Star open source Very Smart Regulator / "Smart Alternator Regulator" project.

From my email to him:

Could your SAR unit start out at high Amps current as Voltage remains low,
then scale amps down as voltage rises,
Stop lowering current at a floor setpoint, then stop charging (or open a remote solenoid isolating that bank from the charge buss) based on a voltage ceiling setpoint
?

Sample:
Start out at 200A, hold as long as V < 13.72, then drop to 150A
when V < 13.77, drop to 100A
at 13.82, 50A
at 13.85, 20A
Stop charging (or open the remote solenoid) at 13.95V.

I certainly applaude your efforts and lateral thinking to approach a problem from a different angle.

However this is what to me sounds a little difficult to practically impliment-

"13.6 for low rates, up to 13.8V for high".

Then having Charging Current limited at different levels at different voltage steps.

I realise you are looking at the SAR.

Is this the same person that was suggesting the CVSR is too convoluted?

[CatNewBee]

Or in other words set absortion to 13.95V....
Amps will drop when voltage rises anyway, at 13.95 no amps will flow.

Set float to a low value like 13.4V or less if you want a discharge by some percent...

[john61ct]

Originally Posted by Q Xopa

"13.6 for low rates, up to 13.8V for high"

Sorry, crossed wires there, in that case just talking about having different set points for different charge sources.

A very low amps source like solar, say .05C, gets a LFP bank up to **much** higher SoC for a given voltage.

While a .4C rate like a 200A alternator, can get up to a higher voltage and still be at a much lower SoC, and therefore still be safe wrt longevity.

Therefore different charge sources should get different voltage setpoints,

**if** your goal is to get their finish point to a similar SoC end point.

Which MS has reminded us, is really not a relevant goal with LFP, more reminiscent of lead thinking.

[john61ct]

Originally Posted by CatNewBee

Or in other words set absortion to 13.95V....
Amps will drop when voltage rises anyway, at 13.95 no amps will flow.

Not in my experience, even at 14.6V amps will flow.

But going to that high a definition of 100% Full is harmful to longevity.

At a high current rate, just hitting 13.95V and stopping is likely no problem.

But holding it as CV waiting for amps to taper is IMO getting into a grey area.

And using a very low current source, would definitely be pushing too high.