LiFePO4 Batteries: Discussion Thread for Those Using Them as House Banks

[john61ct]

Originally Posted by nebster

Huh? Not in CC. The battery is constraining the voltage, not the configured CV threshold.

The amount of charge current delivered in CC should have nothing to do with the CV threshold.

I thought the post I was responding to was referencing the actual ("negotiated") voltage as measured at the battery posts, or at least on the wire.

Yes of course the setpoint in the regulator is just theoretical until the above V gets there.

[john61ct]

Originally Posted by Delfin

My question is whether that would have any material impact on longevity compared to not drawing the batts down to 20% SOC.

Likely never get data on those very different aspects, too many variables.

By following the gentler charging on the top end, lifetime is **so** bloody long afaic may as well go deep when needed, sacrificing 30-40% of such expensive capacity to (maybe) get even more cycles makes no sense, ROI that long is already very risky.

[Delfin]

Originally Posted by nebster

Okay, thanks for the extra detail.

Are you able to see on the Balmar whether it is in CC or has reached CV? I'm assuming you can do that on the Trace unit for sure.

Cannot on the Balmar, but since the absorption time on the Trace is 0:00, one would assume you go directly from CC to float. It appears that the Balmar dominates; why I am not sure.

Originally Posted by nebster

Your description of the behavior doesn't make sense to me. If a charger is truly in CC, it should be putting out its rated current (modulo thermal limiting, etc.).

Since I am measuring the voltage at the battery, the charging sources may well be putting out their rated current, but I am only going to see what the battery voltage is at any moment. Look at the chart I posted above. You can see the typical behavior of a charging battery, or at least a lithium one. The battery moves slowly from its depleted voltage to around 27.1 volts, then ramps up fairly quickly as it nears capacity, then rapidly as the CAR drops.

Originally Posted by nebster

How many times did you test this charge strategy? What was the SOC at the start of the charge, if you recall?

Only once, and as noted, both the CAR was lower with a lower bulk setting and the system reached the lower voltage fairly quickly, then went to float. I'll run them down over the next couple of days and try again.

Originally Posted by nebster

How are the two charging devices sensing voltage? Are they doing it on the charging wires directly, or do you have independent voltage sensing set up?

As noted, I am measuring voltage at the positive terminal of the battery.

Originally Posted by nebster

This is also fascinating, and weird. You're saying that, at 3.60v CV threshold, your system never reaches that threshold and thus stays in CC, but yet it manages to taper anyway?

Almost. What I am saying is that as the battery nears the end of the charge cycle, as defined by an acceptance rate < 5%, I disconnect the batteries from charging. If I fall off the boat, they would go to the 28.8 volts before going to float per the programmed values of the regulator and/or charger. What I don't understand is that the Balmar is set at a bulk voltage of 28.8 volts. I will disconnect charging around 28.3, then power off the genset. Once would think the Balmar would still be in bulk mode seeking that 28.8 volts, but nope, when the genset goes off, the Balmar is in float at 26.8 volts. The only guesses I have is that the Balmar is temperature compensating, although that doesn't seem likely since I have nothing connected to the temperature sense terminal on the regulator, or one of the programmable values that I have no clue what they mean are impacting things. The ones I find incomprehensible and are totally unexplained in the Balmar manual are the Field Threshold for Bulk to Absorption and the Field Threshold Float to Absorption (both set to the default value of 65%). This is Lithionics recommendations. The Balmar manual states that this setting affects the "field output required to maintain calculated bulk charging mode", but that doesn't sound like it would have an effect. When we return from this cruise, I will call Balmar and see if they can interpret.

Originally Posted by nebster

From my perspective, the good news is that your system has something else going on that is more than just a simple CC/CV charger connected to an array of LFP cells. A basic setup definitely does not behave like this, and so I think I can stand by my refutation of your item 4; that is, in my experience it is perfectly viable to lower the CV threshold as a "poor man's" way of limiting the SOC.

Perhaps. I assume you have a bank of these cells installed? Power or sail? One thing to remember, is that when I am charging, I am also running a watermaker sucking down 12 amps @ 120 volts, and usually a 1/3 hp electric motor polishing fuel. I suspect the significant demands I am making while charging may have an effect on behaviors observed.

[nebster]

Originally Posted by Delfin

Cannot on the Balmar, but since the absorption time on the Trace is 0:00, one would assume you go directly from CC to float. It appears that the Balmar dominates; why I am not sure.

If they are independent of each other, they may each be doing different things. In fact, it is likely, given your description of the behavior at the other end of the wires (the battery).

Since I am measuring the voltage at the battery, the charging sources may well be putting out their rated current, but I am only going to see what the battery voltage is at any moment.

No, current doesn't work that way. The current at the battery will be the same as the current coming out of the chargers (combined). It's the voltage that changes, but the voltage also drives the current-supplying logic inside the chargers.

Look at the chart I posted above. You can see the typical behavior of a charging battery, or at least a lithium one. The battery moves slowly from its depleted voltage to around 27.1 volts, then ramps up fairly quickly as it nears capacity, then rapidly as the CAR drops.

Yes, that is the expected voltage behavior versus charge. However, when a charger is in bulk, it will provide continuous current. Your charging system is reducing its current before it (supposedly) gets out of bulk, which doesn't make sense. But what I think is happening is that one of your chargers is going into absorption, you just don't know it.

Chargers in absorption (CV) behave like your current tapering in your chart.

Only once, and as noted, both the CAR was lower with a lower bulk setting and the system reached the lower voltage fairly quickly, then went to float. I'll run them down over the next couple of days and try again.

As noted, I am measuring voltage at the positive terminal of the battery.

When you have a chance to test it, please measure the voltage at the chargers if you can, in addition to at the battery. I believe you will see a higher voltage at the chargers, and that is probly what is making it behave that way.

Almost. What I am saying is that as the battery nears the end of the charge cycle, as defined by an acceptance rate < 5%, I disconnect the batteries from charging. If I fall off the boat, they would go to the 28.8 volts before going to float per the programmed values of the regulator and/or charger.

And what I am saying is that, more likely what is really happening is that one or more of the chargers already thinks it is at 28.8v well before you stopped making your chart. Somehow it is in constant-voltage charging, and staying there for some time.

The easy way to confirm that is to measure the voltages at the chargers, but I can guarantee you that if you are not using sense-based voltages as inputs to your charge logic, then you will have to "futz" with the voltages to achieve whatever charge goal you desire. It looks to me like your effective CV threshold is somewhat lower than 28.8v already.

What I don't understand is that the Balmar is set at a bulk voltage of 28.8 volts. I will disconnect charging around 28.3, then power off the genset. Once would think the Balmar would still be in bulk mode seeking that 28.8 volts, but nope, when the genset goes off, the Balmar is in float at 26.8 volts.

Based on the current taper in your chart, I would expect both chargers to be in either of absorb or float. I would expect neither are in bulk.

And when you did the experiment at the lower setting, I suspect one of the two went into float immediately, effectively turning itself off. Do you know the maximum charge currents for the Balmar and Trace chargers?

The only guesses I have is that the Balmar is temperature compensating, although that doesn't seem likely since I have nothing connected to the temperature sense terminal on the regulator, or one of the programmable values that I have no clue what they mean are impacting things. The ones I find incomprehensible and are totally unexplained in the Balmar manual are the Field Threshold for Bulk to Absorption and the Field Threshold Float to Absorption (both set to the default value of 65%). This is Lithionics recommendations. The Balmar manual states that this setting affects the "field output required to maintain calculated bulk charging mode", but that doesn't sound like it would have an effect. When we return from this cruise, I will call Balmar and see if they can interpret.

My vague impressions of Balmar regulators is that they are somewhat opaque and hard to get programmed right. I have not had to mess with one yet, so I can't help you with those details.

I assume you have a bank of these cells installed? Power or sail? One thing to remember, is that when I am charging, I am also running a watermaker sucking down 12 amps @ 120 volts, and usually a 1/3 hp electric motor polishing fuel. I suspect the significant demands I am making while charging may have an effect on behaviors observed.

I have a bank of 600Ah at 51.2v. They are in my RV, and they charge from shore, genset, and/or solar.

I place substantial loads on my system during charging, and as long as my inverter/charger still has excess capacity from its power source during charging sufficient to supply the load, the charge will stay at max current in bulk and the load will be supplied.

Should the loads plus the charge current exceed the configured supply current, the charger will reduce the charge current to keep the total draw at the maximum allowed.

If you were experiencing this latter scenario, I believe you would see your charge current changing minute by minute -- not simply pegged at a single lower value.

[evm1024]

As I recall the Balmar MC612 and MC614 both (and likely other models) do give the mode they are in. The 5th element of the rotating display is the state of charge. It is in the form of 5xx codes.

Than manual tells what each code indicates. On my MC612 for example stage 6 (code 506) is absorption charge, Holds voltage at bulk voltage for the minimum time.

Regards

[Hoopla]

Originally Posted by nebster

I can barely tell what you're talking about, but I think it's unlikely I'm wrong and much more likely that you're misunderstanding me.

A bottom-balanced string of cells will not diverge at 30% SOC or 0% SOC.

That is what I said. Nothing more.



You probably know what you're talking about, but you're not expressing it clearly. I think you should say what you mean instead of "cats," because after this last post, I have no idea what you mean.

At very low SOC in a top-balanced series string of cells, the individual cell voltages will diverge. It is possible that one of them will end up very low even though the whole string's combined voltage is still okay. A system that does not monitor individual cell voltages will struggle to protect all the cells in this scenario.

This can be the case at 0.3C or 0C, so I am confused about what you are trying to get it, since it seems that you think the loads matter.



I'm sure you do, but more than one of us reading also have a pretty good understanding of cell behavior, and it's very hard to understand what you are writing right now.

To facilitate an informative reply simply advise as follows based on your experience;.

1. Expected and acceptable level of cell drift or divergence measured in cell mV at 25% SOC for "top balanced" cell packs (industry accepted norm for sailboat fractional use applications), not "bottom balanced".

2. That drift/divergence measured in mV under different loads say extreme 0.3C or not a lot.

3. The above using out of the box and brand new cells compared to say cells having a few thousand cycles on board and or say more than just a few years old.

Your posts indicate you have all the data at hand to answer the above in a heartbeat, I know I can. It is slso a simple Q/A 1,2 & 3, no novel required on your behalf.

With that knowledge of your thinking and experience I'm quite happy to respond and in the context of my OP (which made no mention of 30% or 0% SOC) that you are struggling with understanding.

[john61ct]

OP means #1 post in a thread, or the member that wrote it.

[Hoopla]

Thanks

[Hoopla]

Originally Posted by nebster

I agree. However, there are no USA-made LFP storage cells anyway, so it's a moot point.

....

Actualy there are but not necessarily LFP chemistry and so appear mostly EV market and a sort of here one day gone the next business. Nissan closed their Tennessee plant and now source from Korea. One has just opened in California by ex Musk engineer. You could have had you own factory with this gear up for sale earllier in the year for around $5m so no doubt someone somewhere is putting it to work.

https://www.hilcoind.com/sale/alevo/featured

Pic is of a plastics factory in Florida punching out Lithionics cases. Have no idea where their LFP cells come from, they say USA.

[nebster]

Originally Posted by Hoopla

To facilitate an informative reply simply advise as follows based on your experience;.

I'm not sure why you have decided to have the conversation this way, but I'm happy to talk about things further. I will not limit myself to the premises embedded in your questions, however. Please understand that some of us look at the problem differently than others (and perhaps you) do.

1. Expected and acceptable level of cell drift or divergence measured in cell mV at 25% SOC for "top balanced" cell packs (industry accepted norm for sailboat fractional use applications), not "bottom balanced".

You're forcing me into a box here a bit unfairly. (Were you on the debate team in high school?)

My attempt to clarify your obscure phraseology ("cats") was very specific: I said that you were trying to explain that a string of LFP cells will experience voltage divergence at low SOC. Since your statement about the cats is only true for top-balanced packs, I added that clarifier.

Since I have a bottom-balanced pack, my cell voltages will NOT turn into "cats" at low SOC. But conversely I have the "cats" at high SOC, of course.

I do not agree that top-balance is the best way to run a lithium pack, but it is one way. It has advantages and disadvantages. I do agree that the industry mobile house power LFP offerings all top-balance, and there's nothing intrinsically wrong with it. I'll start another thread on this so we don't get off track.

Anyway, here is my opinion/answer to your question:

I expect and find acceptable that 100 quality LFP cells should exhibit a capacity spread of <2%. In my own case, my pallet of 112 fairly low-cost cells shipped direct from Shenzhen came with 2 DOA and a best-to-worst capacity spread of 1.7%. I have observed on other recent DIY fairly similar capacity spreads.

I would expect an engineered LFP pack to have even tighter tolerances, both because I'm paying them a steep premium to filter them, and because they can mix and match across a large pool of raw cells.

I would expect even a DIY pack of some size to exhibit tighter tolerances, because it makes sense to capacity-balance a large number of cells when building them up into larger batteries.

I am sure some engineered packs don't meet those tolerances, and likewise I'm sure some DIY packs aren't properly qualified prior to putting the cells together.

So, assuming 2% spread, in the worst-case with the two outlier cells wired in series, the slope of the charge and discharge curve for LFP chemistry is ~2.5mV per 1% SOC at 25% SOC. In a top-balanced pack, then, I would expect a spread of <5mV (and probably <4mV, since we're only ~75% discharged).

And in a bottom-balanced pack at 25% SOC, I would expect a spread of <1.5mV, since we're closing in on the balance point. Conversely, we can look at 75% SOC for bottom-balance. As I was trying to say before, in the linear regime, the charge curve is very, uh, linear. It is also pretty symmetric! Whether discharging or charging at 0.2C, the slope is again ~2.5mV/1% SOC. So I would expect roughly the same maximum delta at 75% SOC from best to worst cell in the same series string.


In my random first 16s string with a manufacturer-supplied 80% SOC "mid balance," I have data in my notes showing delta 2mV at 25% SOC and 9mV at 14% SOC under 0.05C discharge. This roughly matches the theory above. (Shortly after receiving the cells, I switched to bottom-balance, so I don't have any more interesting divergence data at low SOC to share.)

2. That drift/divergence measured in mV under different loads say extreme 0.3C or not a lot.

Well, first, all the cells are going to drop under substantial load.

The amount of the drop will vary depending on the internal resistance of the cells. Ri is a complex topic and a both dynamic and "synthetic" measurement that is complex to measure, and I won't pretend to be an electrochemist. However, I suspect what you are hinting at -- and if so, why you can't just write it explicitly instead of playing this survey game, I'm not sure -- is that cell Ri can diverge over time as well, further exacerbating voltage divergence under load.

I agree, but I don't think it matters in real life in a healthy system.

I don't have access to tons of data, but anecdotally there are a handful of substantial, large LFP packs that have cycle counts at or beyond 1000, some deployed as traction packs and some for house/ESS, some bottom-balanced and some top-balanced, where the owners document stable Ri and voltage balance over an extended period.

Moreover, I can say for my own cells that they show an Ri spread of 20%, worst cell to best cell, at 50% SOC (where Ri is lowest but perhaps also most indicative of "health").

I'll spare you the math for the expected mV delta under 0.3C, but suffice it to say, I don't think it is a first-order concern for well-managed packs. It should definitely not be a concern at 25% SOC, because even a meaningful Ri imbalance shouldn't send the worst cell below the LVD. (And a good, modern BMS should be doing current-compensated LVD, as well.)

I do think lots of packs have accidentally accelerated cell damage and then both Ri and capacity fade happen, way faster than necessary. I bet this is what you have observed and are getting at.

3. The above using out of the box and brand new cells compared to say cells having a few thousand cycles on board and or say more than just a few years old.

I have only anecdotal data from DIYers on technical forums who occasionally check in, years later, with "yep, pack's stable, still balanced even, no problems." Very few installs have >2k cycles, and honestly I think beyond that we are in the noise. If your new boat lithium house power system gives you 2k cycles to 75% rated capacity, that's a ton of full-cycle charges and most of us will have another boat, or we will have a successor to LFP that we all like more, or we will have fusion reactors on board, or we'll be dead.

Your posts indicate you have all the data at hand to answer the above in a heartbeat, I know I can. It is slso a simple Q/A 1,2 & 3, no novel required on your behalf.

I don't think that's fair, and it's almost like you're taunting. My statement was really simple and clear.

With that knowledge of your thinking and experience I'm quite happy to respond and in the context of my OP (which made no mention of 30% or 0% SOC) that you are struggling with understanding.

Sorry, I think I confused 0.3C and your 25% SOC at one point. That's probably why I wrote 30% SOC at one point. My point still stands: if a top-balanced pack is struggling at 75% DOD, or a bottom-balanced pack is struggling at 75% SOC, something is seriously wrong and probably has been for a while.

I'm going to start a fresh thread with a clean and clear description of my setup in hopes that it helps a few folks look at LFP a little differently. I've done a lot of stuff differently than the "norm," and it's helpful to see both sides of an engineering choice sometimes.

Cheers (and stop trolling me and be nice! ).

[nebster]

Originally Posted by Hoopla

Actualy there are but not necessarily LFP chemistry and so appear mostly EV market and a sort of here one day gone the next business.

I know you probably just want to have a fun conversation and stir up debate on this web forum, but after beating me up on your last post, now you post this?

I mean, the title of this ginormous thread has the chemical formula "LiFePO4" in it, my own post you quoted specifically says "LFP" and is in reference to discussion about procuring LFP cells, and now you start your sentence with the corrective "actualy" (sic) and then proceed to explain that you're maybe talking about other chemistries that are totally irrelevant?

Pic is of a plastics factory in Florida punching out Lithionics cases. Have no idea where their LFP cells come from, they say USA.

Where do they say that? It would be unlike them to misrepresent something like that, and dollars to donuts they are not sourcing LFP prismatics from a secret USA factory.

[Hoopla]

Originally Posted by nebster

I'm not sure why you have decided to have the conversation this way, but I'm happy to talk about things further. I will not limit myself to the premises embedded in your questions, however. Please understand that some of us look at the problem differently than others (and perhaps you) do.


You're forcing me into a box here a bit unfairly. (Were you on the debate team in high school?)

Nebster thanks for that and not my intention to troll just try and extract as much info on experiences of others who do collect data to bolt onto my own. Even though LFP in its current commercial form is 10 years old there is not an abundance of reliable data around specific to fractional use, in fact this thread is probably one of the premier sources.

It will take me a while to respond in a sharing tone I hope Thanks again.

[nebster]

Originally Posted by Hoopla

Nebster thanks for that and not my intention to troll just try and extract as much info on experiences of others who do collect data to bolt onto my own. Even though LFP in its current commercial form is 10 years old there is not an abundance of reliable data around specific to fractional use, in fact this thread is probably one of the premier sources.

It will take me a while to respond in a sharing tone I hope Thanks again.

Totally fair -- I probably misinterpreted your tone! But I have thick skin, and I hope you can tell that I am not bothered either way.

Let me know your thoughts when you have time.

[Hoopla]

Originally Posted by nebster

I know you probably just want to have a fun conversation and stir up debate on this web forum,....

Where do they say that? It would be unlike them to misrepresent something like that, and dollars to donuts they are not sourcing LFP prismatics from a secret USA factory.

Hoping someone here may know of an existing or pending US supplier interested in our market as reliance on Chinese is becoming tiresome, particularly in countries where they have no distributers or if they do they either don't carry sufficient stock and or mark the price up by a unsatisfactory margin.

They say,;

By leveraging over 30 years of factory automation and custom fabrication experience, our company produces batteries that deliver a longer life, a faster recharge, are safer and environmental friendly. Best of all, these batteries are designed and built inside the same country – the United States of America!


http://lithionicsbattery.com/about-us/

[Hoopla]

Originally Posted by nebster

Totally fair -- I probably misinterpreted your tone! But I have thick skin, and I hope you can tell that I am not bothered either way.

Let me know your thoughts when you have time.

Excellent. I won't get to it until the weekend. You and john clearly have experience and kind understanding of those with Internet Asperger's :-)