thomasow' SAR and regulating charge current w/ SoC

[john61ct]

Interesting, you may well be on the right track about that flattening of the curve as you get toward the right.

Of course that chart's definition of 100% SoC is likely quite higher than I would want an LFP bank to get even occasionally, and certainly compared to my normal cycling usage.

You state that .5C is a "healthy" current. Does healthy mean, "as fast as possible without hurting longevity too much"?

Higher would be faster of course.

If time is not an issue, isn't .3C **less** harmful to longevity? I suspect just like average DoD vs lifetime, there is no B&W hard line, but a greyscale trade-off.

I personally measure lifespan in cycles and years, never grokked why "greater lifetime throughput" is relevant to anything.

My goal here is the ability to charge at a much **higher** rate than I would usually do, so certainly the overall charge time is **faster** as a result of the regulator scaling current down to gentler levels only as the target SoC / resting voltage is approached.

This is how some existing charge regulators **do actually work** in practice, and apparently the upcoming "new SAR" will be able to operate along the same lines.

So along with your I think relevant explanation, it is hard to see anything "impossible because of any inescapable relationships between these variables".

As I said, plotting Volts against SoC is I think is less relevant in this scenario,

since **resting** volts is the target, that is in effect equivalent to SoC for a given bank.

Perhaps the ideal would be not so stepped a function, as the target charging voltage is reached earlier as a result of higher current, then amps is more continuously ratcheted down to keep the voltage from climbing much over,

again not as a result of battery resistance, but under the control of the regulator.

so in effect keeping your curve closer to flat all the way to the defined-Full stop-charge point,

which is when the precise target **resting** voltage has been reached, without hitting points on the way where the battery resistance kicks in.

[tanglewood]

Originally Posted by john61ct

Interesting, you may well be on the right track about that flattening of the curve as you get toward the right.

Of course that chart's definition of 100% SoC is likely quite higher than I would want an LFP bank to get even occasionally, and certainly compared to my normal cycling usage.

You state that .5C is a "healthy" current. Does healthy mean, "as fast as possible without hurting longevity too much"?

Higher would be faster of course.

If time is not an issue, isn't .3C **less** harmful to longevity? I suspect just like average DoD vs lifetime, there is no B&W hard line, but a greyscale trade-off.

I personally measure lifespan in cycles and years, never grokked why "greater lifetime throughput" is relevant to anything.

My goal here is the ability to charge at a much **higher** rate than I would usually do, so certainly the overall charge time is **faster** as a result of the regulator scaling current down to gentler levels only as the target SoC / resting voltage is approached.

This is how some existing charge regulators **do actually work** in practice, and apparently the upcoming "new SAR" will be able to operate along the same lines.

So along with your I think relevant explanation, it is hard to see anything "impossible because of any inescapable relationships between these variables".

As I said, plotting Volts against SoC is I think is less relevant in this scenario,

since **resting** volts is the target, that is in effect equivalent to SoC for a given bank.

Perhaps the ideal would be not so stepped a function, as the target charging voltage is reached earlier as a result of higher current, then amps is more continuously ratcheted down to keep the voltage from climbing much over,

again not as a result of battery resistance, but under the control of the regulator.

so in effect keeping your curve closer to flat all the way to the defined-Full stop-charge point,

which is when the precise target **resting** voltage has been reached, without hitting points on the way where the battery resistance kicks in.

I think what you are evolving to is what's shown in the attached.


- Charge at .2C to some target voltage


- Taper current to prevent running over voltage


- Stop when you reach the desired SOC, which in this case I have arbitrarily shown as 93% of name plate.


But in doing so, you have reinvented CC CV charging.


You can of course charge at rather other than .2C. I use that value only because there is a real curve for it. But you can guess pretty well where higher and lower charge rate curves will land on the chart.

[tanglewood]

Re some of your other questions, the only research I've seen that shows loss of battery life due to higher charge rates looked at greater than 1C rates. I've seen nothing revealing delving into the differences in sub 1C rates, other than the implications of mfg spec sheets that talk about max charge rates of 1C, and "normal" charge rates of .5C or .3C. What that exactly means is anyone's guess, as are the implications, as are the actual mechanisms that occur in the battery. Now based on the research and the mechanisms that have been explored, two come to mind as possibilities.


1) It's been shown that heat speeds decomposition of the electrolyte, leading to lost capacity. So maybe the C rate limited are related to internal heat generation? Maybe?


2) It's been shown that too much internally developed voltage (I can't remember exactly where in the battery structure) triggers Li plating of the SEI later, and subsequent loss of capacity. The higher the charge rate, the larger voltages will develop, and maybe this is a factor?


But I'm no electrochemist so really have no idea, other than putting two and two together from papers to come up with what might be sensible questions or theories.


So given all this unknown, my own preference is to stay at or below the .5C range per manufacturer's recommendations. That yields a 2hr generator re-charge time, which is heaven compared to equivalent LA capacity recharge. So I'll take that win and move on to the next pressing issue.


And speaking of the unknowns, you might be right that variations of charge cycles will have an impact. But it's just as likely it will have none since nobody really knows. There are lots of beliefs floating around, but very limited actual data. To really answer your questions would require a test with a baseline or two, then variations of charge profiles, and probably 2000-4000 cycles. Then you might actually have some evidence of something, but nothing to explain why. Or you could hire the guy from U Halifax with his super accurate calorimeter machine to do accelerated testing. It's a very cool (pun intended) approach to accelerated testing, looking for tiny differences in heat generated that are caused by side reaction, all of which are detrimental in one way or another. If there are side reactions, something bad is happening.


You also brought up lifetime through put vs cycles & years. I agree that any one of us cares about cycles and years of life for our system. But our systems are all different, primarily in DOD per cycle, so it's impossible to compare or extrapolate since you are only comparing part of the problem. It's like comparing GPH fuel burn with out also talking about displacement and, much more importantly, speed. Without all the variables, it's a meaningless discussion.


Lifetime power throughput, though imperfect, is still the best way to normalize the data to have an objective discussion. Now that's still a hard discussion to have because nobody measures it. But nobody measures DOD per cycle and records it in any meaningful way either, and if they did, well they could derive throughput data from it too.


But it's inescapable how flat the curve becomes when you take a Cycles vs DOD chart and transform it to show lifetime throughput vs DOD. Again, it's not perfect, especially towards the extremes of the DOD range, but it's a way more complete story, and one that can be compared battery to battery.

[john61ct]

Originally Posted by tanglewood

max charge rates of 1C, and "normal" charge rates of .5C or .3C. What that exactly means is anyone's guess, as are the implications, as are the actual mechanisms that occur in the battery

I agree, and usually my default has been to err on the side of being conservative, i.e. prioritizing the longevity factors higher than most v/v other values.

I feel pretty confident (admittedly only at an intuitive level), that going into the .7-1C range is **less** harmful in the middle SoC range than allowing it to continue as you approach Full.

And that even .5C is more harmful than .3C in those upper SoC areas where batt resistance is starting to climb.

When considering voltage only, low charge rates means no significant sacrifice of capacity.

However when minimizing ICE source run-times, that capacity loss grows, to the point that the challenge of stop-charging as per endAmps seems more necessary.

Originally Posted by tanglewood

you have reinvented CC CV charging

Sure, that is the point. Lead's much higher resistance makes it self-limiting wrt damage from high charge rates or holding absorb too long. LFP needs more explicit and precise control in order to optimize the charge profile.

LFP's extremely high CAR is indeed a blessing, and yes in many use cases a 2-hour recharge runtime is fast enough, but

if that can be safely reduced I think it's worth exploring how,

for when you're only burning dino juice for charging purposes.

Again, note this is not "my" idea, it has already been in apparently wide use with those lithium chargers designed to let the user just set a max current rate and target resting finish voltage, no eggtimers no endAmps involved.

If a charge source just lowers the current rate as Full is approached, there is no sacrifice of capacity from using a CC-only profile.

As well as reducing the risk of harm from a too-high (starting) charge rate. Where to draw the lines becomes the next important issue.

Originally Posted by tanglewood

I think what you are evolving to is what's shown in the attached.

Again, you may be right but

I think those are the wrong variables for illustrating this concept.

The flatter that curve is, the less meaningful the graph as a whole;

resting V is such a good proxy for SoC, you end up with the same info on both axes.

Its purpose is to illustrate the effect of bank resistance while CC/CV charging. This concept eliminates the CV part, its goal is to avoid the impact of increased resistance in the first place.

______
Assuming infinite current is "available" upstream, the regulator is explicitly in control of dynamically varying the current rate over the charge cycle.

So the other "usual graph" showing Current overlaid with Voltage on the Y axes, and time passing on X,

would at least illustrate the various possible rates of change and SoC points where current gets reduced,

from a say 1C starting rate down to .1A.

If you could come up with a formula or even just a labeling scheme for representing the different profiles with a single number, then Time Spent getting from 20% to functional-Full SoC would be a useful metric.

Plotting internal temp changes from base 77°F would also be ideal if possible, as a proxy for speculative "loss of longevity".

______
> Without all the variables, it's a meaningless discussion.

I disagree. Gather enough anecdotal reports over many years, filtering those sources you trust more than others, to me that's enough basis to inform my choices.

The resources required for rigorous hard science will (I fear) never be allocated towards the goal of making batteries last many decades.

______
To me, lifetime power throughput doesn't work conceptually, at all. The lab charts' deviation from real-life results is **so** much more extreme as DoD% increases, especially with lead banks when coupled with inevitable PSOC abuse.

I'm not saying a useless concept for others, just not part of my paradigm.

[john61ct]

related http://www.cruisersforum.com/forums/...ct-179240.html

[tanglewood]

OK, so your whole premise is that the harmfulness of any given charge rate varies based on the SOC. And in particular, that higher rates are less harmful at lower SOC, and more harmful at higher SOC. And as you said, that's "intuition", with nothing to actually demonstrate it, either empirically or theoretically.


I guess at this point the only thing I can suggest is that you might want to prove the theory first, then come up with a solution. Otherwise you are solving a problem that might not be there, or might be different that you imagine. Unless of course you are good at selling oil and gas treatments to people, in which case you might become rich of your invention...

[tanglewood]

I'd go back to what MaineSail said early on. Get a variable bench supply, hook up a battery, try some of these ideas, and see how the whole circuit responds.

[john61ct]

For the ?th time, not my idea, no new "invention" involved, not selling anything.

I have seen no "hard proof" that longevity is extended by staying below vendor spec'd charge voltages, nor by avoiding charging rates over .5C.

Many users are also following those guidelines nonetheless.

And as stated, even if the longevity benefit proves false, using such a charger gives a known-precise ending SoC without using endAmps to stop-charge.

[john61ct]

Another real-world CC-only LFP charging example from Texas Instruments.

Battery charger IC BQ 25070, designed by TI specifically for fast-charging LFP

From the data sheet:

The LiFePO4 charging algorithm removes the constant voltage mode control usually present in Li-Ion battery charge cycles

Instead, the battery is fast charged to the overcharge voltage and then allowed to relax to a lower float charge voltage threshold

** The removal of the constant voltage control reduces charge time significantly**

Data sheet here: http://www.ti.com/lit/ds/symlink/bq25070.pdf

Note their voltages are "a bit" high, including the drop-to Float of 3.5V, but the principle holds, go a bit higher as a finish voltage, just briefly, to get that bit extra SoC in quickly, then the regulator drops current rather than waiting for CV /resistance to do so.

The engineers at TI are pretty well respected, and not prone to embedding crazy ideas into their silicon dies.

[Q Xopa]

Originally Posted by tanglewood

OK, so your whole premise is that the harmfulness of any given charge rate varies based on the SOC. And in particular, that higher rates are less harmful at lower SOC, and more harmful at higher SOC. And as you said, that's "intuition", with nothing to actually demonstrate it, either empirically or theoretically.


I guess at this point the only thing I can suggest is that you might want to prove the theory first, then come up with a solution. Otherwise you are solving a problem that might not be there, or might be different that you imagine. Unless of course you are good at selling oil and gas treatments to people, in which case you might become rich of your invention...

Its not a far stretch to extrapolate that the cell manufacturers specified 'nominal' max charge isnt a magic number. Where above that rate its damaging and below that rate no damage occurs.

The same may be said for some other factors like range of SOC per cycle. Probably for similar reasons, heat etc.

The question is if its worth worrying about?

[rgleason]

I believe this is a different routine than you suggested earlier for Alternators, I guess it is for top balancing to 3.55vpc:


Summary
After normal charging, each cell is isolated and charged again with regulation of current output, voltage and temperature.

• Resting Voltage Current Temp Hours
• 3.50vpc starts at 7a 45°C 2hrs max
• 3.52vpc drops to 3a
• 3.54vpc drops to 1A
• 3.55V setpoint holds at .1A (below .0003C) what most "float advocates" here would call close enough to zero.

Also how does this logic get automated?

"However, it only does so for a few minutes, and then switches charging off. If the voltage then drops at all during the max time set, the regulator scales current back up to 3A for a few minutes, loop / repeat.

"When it senses the voltage does **not** drop any more, then it starts charging at between 3.55 and 3.65Vpc at 3A for under a minute each, stopping to check resting volts in between, until each cell **rests** at precisely 3.55Vpc.

"Of course, any EoL cells will never get there, which is what your total maximum charge time setting is for.

"The other cells' charging outputs are just Off during that time as long as they continue to sit at your target voltage.

It seems to me this is what a good BMS should be responsible for, keeping the cells top or bottom balanced.

Originally Posted by john61ct

I've been spending time in various battery-specialist forums.

There are several chargers I've come across that use the topic idea here:
the charge regulator explicitly controlling the current output, rather than just letting battery resistance do that.

**Please** let's keep to that topic here, post commentary or questions on side issues like example voltage setpoints, longevity, specific hardware etc in other threads,

or start a new one - they're Free!

I am particularly interested in models designed to charge each cell, each output is isolated, voltage & current regulated independently.

You just set the target **resting** Voltage, maximum current Amps allowed, and optionally, a total max charge time.

Some include a temperature sensor, so you can also set a max temp allowed; overtemp lowers current, not voltage.

As the regulator approaches the target Voltage, it lowers the Amps current level, to compensate for any sensing voltage drop.

So as an example, a 360A 48V nominal LFP bank that has been bulk-charged in parallel to its normal cycling 55V stop-charge; next, you want to get all the component cells to a perfectly balanced 3.55Vpc, or bank voltage of 56.80V

Just set the charger for a resting target voltage of 3.55V per cell, maximum current at 7A per cell, max temp 45°C, maybe a max charge time of 2 hours to deal with sub-par cells.

As each cell gets to 3.52V, the regulator drops current to 3A.

When they hit 3.54V, current drops down to 1A, and

at the 3.55V setpoint it holds at .1A. Note this is below .0003C, what most "float advocates" here would call close enough to zero.

However, it only does so for a few minutes, and then switches charging off. If the voltage then drops at all during the max time set, the regulator scales current back up to 3A for a few minutes, loop / repeat.

When it senses the voltage does **not** drop any more, then it starts charging at between 3.55 and 3.65Vpc at 3A for under a minute each, stopping to check resting volts in between, until each cell **rests** at precisely 3.55Vpc.

Of course, any EoL cells will never get there, which is what your total maximum charge time setting is for.

The other cells' charging outputs are just Off during that time as long as they continue to sit at your target voltage.

Note there is no endAmps setpoint, nor is the charge continues until current stops completely, the whole regulation algorithm is based on CC / Bulk, voltage setpoint only.

Since these chargers completely isolate the per-cell outputs from each other, the intra-bank connections (in this example 4S2P packs to get to 360AH @48V) could in theory remain in place, but you'd want to ensure the bank's isolation from other charge sources, and ideally any loads for consistency, and getting to your setpoint as quickly as possible.

With a bank design based on a pair of pack-strings in parallel for redundancy, one half could be topped up isolated, while the other remains in production. Such a design also allows for KISS balance midpoint bank voltage monitoring with Victron's 712-BMV.

With a healthy bank, the whole process should take under an hour, even less if the initial bulk charge were tweaked closer to the target balancing voltage.

The fact such chargers are already in common use, using "my" **CC-only with no Absorb** profiles, and the future-SAR algorithm where **the regulator** drops current as SoC climbs, further shows these are not new outlandish ideas.

And with that level of per-cell precision, using load testing to measure the end results of different "Charge To" voltage setpoints, comparing **actual** AH added to SoC,

rather than just "AH supposedly in" measured by a coulomb counter,

should become much easier.

[rgleason]

New Definitions in this post

AHT= AmpHours Trailing

John, without having LiFePo4 experience, I do understand your points:

• No, I am here exploring stop-charging at precise SoC levels without using endAmps at all, Bulk / CC-only, no Absorb.
• Yes, my approach departs from the usual charts showing dynamic charging / discharging cycles.
• And since I'm using CC-only profiles, trailing amps isn't even available as a component to define the variable, AHT is now fixed as a constant - at zero! and trailing amps a moot issue.
  But for a given bank, **resting** voltage (surface charge anomalies removed at the high end) can be calibrated to the above, then serve as a (good-enough for my purposes) proxy for indicating SoC.

You also have some preferences about the "chargers" that excludes use of a BMS

With a per-cell CC-only **charger**, the regulator just "Charges To-and-stops" independently for each cell, and at higher amp rates getting them all to the same SoC point **much** more quickly.

And, at whatever user-customized voltage setpoint you like, in my case likely lower than most. I have yet to see a reasonably priced OTS BMS allow for that.

I also prefer to not keep that per-cell functionality hooked up "live" in normal operations.

So this proposal appears to lend itself to a Manual Top Balancing system where you just connect a small adjustable charger to each cell and follow your routine.

It would be interesting to actually do that on a cell to see what happens, and to determine if the Resting Voltage / Current pairs are very different for your variable CC proposal.

What are the chargers that you have in mind? I suppose they are 120vdc.

[tanglewood]

Originally Posted by rgleason

I believe this is a different routine than you suggested earlier for Alternators, I guess it is for top balancing to 3.55vpc:


Summary
After normal charging, each cell is isolated and charged again with regulation of current output, voltage and temperature.

• Resting Voltage Current Temp Hours
• 3.50vpc starts at 7a 45°C 2hrs max
• 3.52vpc drops to 3a
• 3.54vpc drops to 1A
• 3.55V setpoint holds at .1A (below .0003C) what most "float advocates" here would call close enough to zero.

I'm trying to understand what you are doing at each step, so let's drill down on one or two of them.


The first charge step is "3.50vpc starts at 7a 45°C 2hrs max". The real part I want to focus on is the voltage/current pair in this step and subsequent steps.



- You can't apply 7a at 3.50v. You get to pick one or the other, but not both. What you could do is apply current up to 7a, and until the voltage reaches 3.50v. That's a classic CC charge up to some cutoff voltage. Alternately, you could apply whatever current is required to achieve and maintain 3.50v. Depending on SOC, that current could be anywhere from multiple C to zero. This is classic CV charging, with current limited by the chargers output capacity.


- Then there are similar questions when you reach the next step. First, what triggers it? Reaching 3.50V, which is CC with a terminal voltage? Or reaching 3.50V with <=7a, which is CV with terminal acceptance current? Then what? Raise the voltage and let current rise accordingly? Or drop current to 3a and run until 3.52V?


And it all begs the question of why not run at 7a until voltage hits 3.55, and wait until acceptance current drops to .1A? It places you at the same end point.


I think you are guessing that reducing current at 3.50V vs 3.55V will have a measurable impact on battery life. I haven't seen any research to suggest that, so if it's the belief, then a good starting point would be to figure out if it's true or not. Otherwise you have a solution hoping for a problem rather than a problem looking for a solution.

[rgleason]

John wrote:

• The key concept to me (at this point) is to be able to start off at a much higher charge rate at low SoCs, and for the regulator to actively step that current down as SoC approaches full.
• Perhaps the ideal would be not so stepped a function, as the target charging voltage is reached earlier as a result of higher current, then amps is more continuously ratcheted down to keep the voltage from climbing much over,

I think this graph of Tanglewood's represents the concept very well.
You are taking advantage of fast charging early in the game and slowing down towards the end to be able to take advantage of the the full capacity of the battery. I think it is a very good concept.

Also Tanglewood thank you for finding a chart that shows the result of different continuous charging rates (.1C, .2C etc) It would be nice to have a similar chart for .35C, .4C and .5C.

[rgleason]

Originally Posted by tanglewood

The first charge step is "3.50vpc starts at 7a 45°C 2hrs max". The real part I want to focus on is the voltage/current pair in this step and subsequent steps.

• You can't apply 7a at 3.50v. You get to pick one or the other, but not both.
  • What you could do is apply current up to 7a, and until the voltage reaches 3.50v. That's a classic CC charge up to some cutoff voltage. [Yes, I believe this is what John is suggesting]
  • Alternately, you could apply whatever current is required to achieve and maintain 3.50v. Depending on SOC, that current could be anywhere from multiple C to zero. This is classic CV charging, with current limited by the chargers output capacity. [No, not what John proposes, I should have put "CC"]
• Then there are similar questions when you reach the next step. First, what triggers it?
  • Reaching 3.50V, which is CC with a terminal voltage? [Yes]
  • Or reaching 3.50V with <=7a, which is CV with terminal acceptance current? Then what? [No]
    • Raise the voltage and let current rise accordingly? [No]
    • Or drop current to 3a and run until 3.52V? [Yes, I believe this is what John has suggested.
• And it all begs the question of why not run at 7a until voltage hits 3.55, and wait until acceptance current drops to .1A? It places you at the same end point. [Does it really? We have not tested that.]

I think you are guessing that reducing current at 3.50V vs 3.55V will have a measurable impact on battery life. I haven't seen any research to suggest that, so if it's the belief, then a good starting point would be to figure out if it's true or not.

Otherwise you have a solution hoping for a problem rather than a problem looking for a solution.

Well, I have a slightly different reason for thinking about this routine. I would like to charge at faster rates at lower SOC% to reduce charging time without impacting the battery life (cycles) and capacity adversely.

If I could charge at .35C, .45C or .5C initially to get the battery up towards 67%SOC, without stressing or overheating the battery,

and then reduce the C-rate gradually as the battery gets fuller, such that the battery does not get hotter due to increased resistance, and so that the ending 13.88v point is at a greater %SOC than if I were to simply charge at some fixed CC (ah) up to 13.88v, I would do that.

For me, controlling the alternator accordingly, while protecting the bank and alternator is the problem. I could have a manual variable field control and use my clamp meter at the batteries, but why isn't a user programmable regulator available for this? It's not rocket science, but I am not equipped to make it.


BTW Al Thompson's regulator appears to be gone now. The VSR Alternator Regulator Schematics and CAD files appear to be gone, the onedrive is empty, so how does on get a PCB made?