1400 Ah Lithium from cells

[john61ct]

Originally Posted by OceanSeaSpray

The effect of trickle charging is overcharging and depositing lithium where it should not be, like on the edges of the plates, and can't be recovered afterwards ever. Trickle charging will erase the memory first if there is one, and then damage the cells.

Please define trickle charging as you use it above.

To me it just means charging at a low rate.

The normal-cycling stop-charging profiles I define for day-to-day use do not involve any Absorb/CV hold time, simply stop charging when a V setpoint is reached.

To my mind, that means the same setpoint can be used with any rate of charge. If I use 3.45-3.50V I can't see any harmful overcharging taking place.

Yes, at very different C rates the bank will end up at different SoC, but in the context of the memory effect, I see that as a Good Thing.

If by trickle charging you mean endlessly charging even after .025C endAmps has been passed, then yes I agree that is harmful at least to longevity, and I for now I plan to do never do that.

[CatNewBee]

Trickle is usually float + after some time a short cycle to full (shorter adsorption phase) or some charge impulses to prevent layering of the electrolyte in a wet battery or a short reconditioning cycle introduced once a week or so. It is something invented to implement a "battery pulser" functionality into a charger when used to continuously charge the battery in storage mode.

I doubt it is helpful on a LFP at all.

[john61ct]

No, I call that type of **charger** "too smart", and wouldn't use on deep cycle lead either.

If a charger advertises anything more than CC/CV/Float, I make sure at least those bells & whistles can be disabled, especially equalizing/conditioning.

But in this case I was asking about the term's usage as a type of charge current profile, specifically wrt the effects mentioned.

[nebster]

Originally Posted by john61ct

Sorry I'm confused. Which two "approaches"? To me that implies the attempted solutions rather than causes.

I was talking about the two charging strategies, labeled "Approach 1" and "Approach 2", in my earlier post. Neither were attempts to address this nefarious memory effect. I was trying to describe two different charging schemes that arrived at the same chemical potential and wondering if one or the other would be more susceptible to the memory effect.

> CALBs. Presumably they vary a little bit.

Not that I'm aware, chemistry is chemistry, all the large LFP prismatics should have the same behaviour afaic. Winston/Thundersky adding Yttrium doping may change things a bit, I believe they can charge at a lower temp.

I thought so as well, but numerous others do report slight differences. At a minimum, though, there will be Ri and thus thermal and thus second-order variation in how cells charge.

Not sure what you mean.

How low do you let it go, in voltage (light vs heavy loading) and estimated SoC?

I would think in that situation just vary the top/stop setpoints using whole bank or string/pack/block aggregate-average.

Are you trying to keep everything automated? Are these setpoints difficult to adjust?

I mean that I want my pack to be bottom-balanced so that the cells can survive a deep, accidental discharge. I don't intend to let it go that low, but I will have to leave the system running but unattended for stretches of time, at times.

My observation is that in my 16s strings, if I bring it up close to 56V, I hit 3.6V on one or more cells before others even get to 3.40V. Without some form of active balancing, I see no easy way to ensure all the cells are brought into the supposed memory-erasing regime.

Here is a typical set of stopping voltages on one string, plucked out of my log:

{3.393, 3.420, 3.538, 3.401, 3.412, 3.388, 3.421, 3.390, 3.426, 3.387, 3.400, 3.392, 3.610, 3.539, 3.408, 3.401}

I'm not (yet) willing to add in all the complexity and consequences of an active BMS just to counter this memory effect. Not enough evidence to suggest the trade is a good one.

[nebster]

Originally Posted by john61ct

My point is that the SoC% are not objective, the user defines them, so saying "your Full is at 85%" makes no sense, your Full is by definition your 100%, there is in practice no "real 85%", SoC can't be used to define anything, it is the measure that each, user must decide how to define for herself.

I don't understand how this got so twisted up in your and OSS's heads, but I must be a poor communicator.

First, if someone casually says to me "I have an LFP cell at about 85% SOC", I know exactly what that means. I know what that cell's OCV 12 hours after discharge will be, pretty much. I know how much chemical potential is left in that cell (once they tell me how many Ah it is, of course). I know about how that cell will perform a year from now, if it is held at that SOC.

If your point is that every cell is unique or that state estimation is hard, well, sure. It's always an estimate, and the underlying chemistry is generally drifting away from the starting point. ???

My example, wherein I said "charge to 85% SOC", was an attempt to describe briefly two charge strategies that each reached the same chemical potential on a given cell at their conclusion. That's something that's not hard to concoct in real life, if you wanted to, but I tried to preface it by saying it was a thought experiment!

So stay in CC until Vpc climbs above 3.5 once in a while. Even at low charge rates, you just keep charging longer right? Are you trying to do this on solar, saying the sun goes down too soon?

Depending on charge rate, I do stay in CC, yeah. I can easily start the generator and reach CV. But on solar, not so much.

And with either source, I still have the capacity imbalance problem that makes achieving 3.5V on every cell really hard, unless I want to take a few cells to 3.7 or 3.8V or implement a balancing BMS.

[OceanSeaSpray]

Originally Posted by john61ct

Please define trickle charging as you use it above.

To me it just means charging at a low rate.

Charging with no end to it, as LA battery chargers do. They hold the voltage up to keep the battery full. The current is low only because the voltage differential is small then.

Originally Posted by john61ct

The normal-cycling stop-charging profiles I define for day-to-day use do not involve any Absorb/CV hold time, simply stop charging when a V setpoint is reached.

If you charge at very, very low C-rates, this can fully charge the battery, if you actually manage to get there. With a faster charge, it results in systematic partial charge cycles and fails to wipe the memory, which is why absorption and current taper are important.

Originally Posted by john61ct

To my mind, that means the same setpoint can be used with any rate of charge. If I use 3.45-3.50V I can't see any harmful overcharging taking place.

Yes, at very different C rates the bank will end up at different SoC, but in the context of the memory effect, I see that as a Good Thing.

It won't and can't realistically overcharge this way, but it also systematically fails to charge properly. By failing to shift the last of the charge carriers, you keep strengthening the memory effect and you lose access to the capacity just the same. What you describe here is essentially what the 100-RVs guy had been doing.
There is no alternative or substitute to charging properly and it must happen from time to time.

[OceanSeaSpray]

Originally Posted by nebster

First, if someone casually says to me "I have an LFP cell at about 85% SOC", I know exactly what that means. I know what that cell's OCV 12 hours after discharge will be, pretty much. I know how much chemical potential is left in that cell (once they tell me how many Ah it is, of course).

Correct. SOC is a chemical state, percentage of carriers on each side. You don't make it up to suit yourself.

Originally Posted by nebster

My example, wherein I said "charge to 85% SOC", was an attempt to describe briefly two charge strategies that each reached the same chemical potential on a given cell at their conclusion. That's something that's not hard to concoct in real life, if you wanted to, but I tried to preface it by saying it was a thought experiment!

Well, one was not actually feasible, so you have an issue of your own to work on.

Originally Posted by nebster

Depending on charge rate, I do stay in CC, yeah. I can easily start the generator and reach CV. But on solar, not so much.

You are really staying in Bulk. Few sources on boats are able to deliver CC. At low currents, bulk lasts much longer and absorption gets much shorter. If your charging current was 0.03C, then by definition there would be no absorption left.

Originally Posted by nebster

And with either source, I still have the capacity imbalance problem that makes achieving 3.5V on every cell really hard, unless I want to take a few cells to 3.7 or 3.8V or implement a balancing BMS.

If you are unable to take all your cells to 3.500V each +/- say 10mV initially, you are in serious trouble. Your cells are not balanced and you should not be operating the battery. Assembling a balanced pack is a prerequisite before anything else with lithium batteries.
Alternatively, they are going out of balance all the time now because you have damaged cells: that means a trip to the rubbish skip.

[arsenelupiga]

Originally Posted by CatNewBee

Zhe BMS I use

REC d.o.o. Manufacturer web site

Installation Video:

http://www.rec-bms.com/images/A_BMS/ABMS_connection.mp4

Configuring settings:

Uploading custom Firmware - if you need one:

Here a large install using many cells and the big brother of the ABMS:

excellent , thanks that where I will start. Will also get my german language refresher as once upon the time, I was fluent in german. And in french as well. Who would think i would use that in my retirement, apart from touristing.

[nebster]

Originally Posted by OceanSeaSpray

Well, one was not actually feasible, so you have an issue of your own to work on.

Sure it's feasible. Put a coulomb counter on. Use the other approach. Measure how much energy went in.

Discharge the cell.

Use the coulomb counter to charge with the other approach to the same calculated SOC/Ah.

Modulo minor counting inaccuracies, both methods will bring the same cell back to the exact same chemical potential at rest.

You are really staying in Bulk. Few sources on boats are able to deliver CC. At low currents, bulk lasts much longer and absorption gets much shorter. If your charging current was 0.03C, then by definition there would be no absorption left.

I don't know what "bulk" is, except some term somebody made up a long time for lead acid. I do know what constant current is, and if you prefer to call it "current-limited charging", fine by me. But we both know what I am talking about, so again I feel you are elaborating in way that serves only to obfuscate.

And, yes, I am at 0.02 to 0.05C on my large pack with my (relatively) small solar array for large periods of unattended charging. That's part of why I posed my hypothetical question about the dreaded memory effect and whether it forms regardless of charge strategy.

If you are unable to take all your cells to 3.500V each +/- say 10mV initially, you are in serious trouble. Your cells are not balanced and you should not be operating the battery. Assembling a balanced pack is a prerequisite before anything else with lithium batteries.
Alternatively, they are going out of balance all the time now because you have damaged cells: that means a trip to the rubbish skip.

I completely disagree with this paragraph. Your site is great, but it is most certainly not an authoritative treatment of all pack management strategies.

There are plenty of bottom-balanced LFP packs being run perfectly happily. Top-balancing is an alternate (and popular) approach replete with its own tradeoffs. Your statements only make sense in the context of the latter.

Your proclivity for declaring things to be "catastrophic" or in "serious trouble" is troubling. I'm running out of energy to have impassioned debates about this this week, so, in fair warning, I'm going to dial back my writing and let you and some of the others hash this out further if you all want to.

What I can promise is that I'll come back and update the forum with real data as I glean it. That is something that we're terribly short of here.

Moreover, I'll gladly capitulate and redesign my entire pack if a memory bump prevents it from working well over time... and acknowledging that discovery won't hurt my feelings at all. But, on balance, given the anecdotal information we have today and the other constraints I have imposed on my design, my assessment is still that this is likely not to be enough of an issue to justify a totally different approach.

Cheers, everyone.

[arsenelupiga]

Originally Posted by CatNewBee

- GEL regimes charge in bulk up to 14.4V and stay there for 4 and more hours for absorption in GEL. Then switch to float at 13.6V this is the closest what you can use for LFP if you do not have a programmable charge controller. Not optimal, but doable..

you wealth of info cat !

speaking high level here:
what re you saying that as I have currently gel charging regime in my L 400, could basically reuse it for lithium batteries ?!

So in principle, need to build LIFEPO 12 V battery, say 400 AH , add active balancer, replace my 480 AH exide and off I go

disclaimar: there are some technical details i have to address, which i am sure can do to satisfaction and decent safety level.

[CatNewBee]

As I said, it is possible, but not optimal. What charger will you use mainly for your batteries?

If you think about solar, chose a MPPT Controler that can be programmed for a LFP bank - like the Victron MPPT's with lower voltages.

If you are going the same road with REC BMS for the Lithiums you can ask also for a custom programming (as I have) to have the relay outputs separate from the opto-coupler outputs and use the relay as your "circuit breaker" for prevention of overcharging, while the opto-coupler signals battery full - without disconnecting all charge sources, that can be set to a lower cell voltage. You can then use this signal to stop charging from your legacy GEL-programmed sources when the desired cell voltage is reached, the programmed solar controller and all other sources - if any - that have a Lithium profile will remain connected.

The shore power charger will be only disconnected from the Lithium bank but continue its charging of the start batteries.

[arsenelupiga]

Originally Posted by CatNewBee

As I said, it is possible, but not optimal. What charger will you use mainly for your batteries?

If you think about solar, chose a MPPT Controler that can be programmed for a LFP bank - like the Victron MPPT's with lower voltages.

If you are going the same road with REC BMS for the Lithiums you can ask also for a custom programming (as I have) to have the relay outputs separate from the opto-coupler outputs and use the relay as your "circuit breaker" for prevention of overcharging, while the opto-coupler signals battery full - without disconnecting all charge sources, that can be set to a lower cell voltage. You can then use this signal to stop charging from your legacy GEL-programmed sources when the desired cell voltage is reached, the programmed solar controller and all other sources - if any - that have a Lithium profile will remain connected.

The shore power charger will be only disconnected from the Lithium bank but continue its charging of the start batteries.

i have 2 x victron mppt 75/15, attached to 2 x Victron 180 W solar, so total 360W solar. Mppt i suspect may be able to be programmed for Lithium. Then have 2 alternators and 2 shore chargers, guess same as you.

Not really concerned about getting optimal charging at this point. From observations, I think 400 AH lithium would be possible to be charged by 360 W panels often enough. We are modest power users. If needed can double solar wattage up later or beef up alternators. I am not fan of large solar arrays.

Now you got me really interested. So I will look into this in more detail.

[CatNewBee]

OK, if you have the same configuration as I have on my L400S2 (from 2013),
there are Crystrec chargers for shore power and battery isolator diodes for the alternator charging installed.

If you have no other sources - besides the Victron MPPT's, that usually can be set up properly - and if you decide to go with the REC BMS with custom programming - this could be your solution:

The REC BMS is custom-programmed to split opto-coupler and relay outputs.

The relays of the BMS switch via an interface to ML-RBS latching relays (or similar relays) all charging sources in case of over-voltage (MAX Cell Voltage - e.g. 3.65V) off and by a second relay in case of low-voltage (MIN Cell Voltage e.g. 2.9V) all loads off. This is what regular BMS should do.

The optocoupler for over voltage are programmed to react on the value END Charging (e.g. set to 3.5V), that can be separately set in the BMS. This will switch over an Interface a relay off and on that controls all incomming current from your Cristec chargers and from the battery isolator / charge distribution diodes.

In fact, the BMS will enforce a LFP charge regime using your legacy gear.
It will turn off this chargers as soon as the condition is met and re-enable them after a hysteresis, that can also be programmed (Charge Hysteresis, usually preset to 0.25V - so charging will be re-enabled below cell voltage of 3.25V, but you can configure any suitable value).

You can also use the second opto coupler to disable heavy loads - like an inverter earlier by a separate parameter - like I do, but this is optional.

What are the benefits?

1.) Well the disconnect of the alternators is not a problem with our lagoons, because the start batteries remain connected via the second path of the diode isolators, so no harm on this side, the power then goes only to the start battery (AGM)

2.) the disconnect of the Crystec 220V charger is no problem at all. They have 3 separate outputs that can be used to charge up to 3 batteries in parallel, one goes to the start battery, the second to the house battery and the third - optional generator start battery. There are at least two of them - one per engine. It is not critical to just disconnect them from the house LFP.
When the LFP is dropped, they continue to charge the start batteries alone.

3.) The solar charger remain on, they are programmed for LFP, so they do not exceed the allowed voltage, also they are programmed for a very short absorption (to allow some balancing if necessary) and will go to float at a low voltage afterwards. It is low enough, that charging will stop, but they will provide power to the house when loads kick in and battery voltage drops below the float setting, they will try keep up voltage at this level.

About the interface mentioned above - it is pretty simple, I can provide you my solution for this. The reason I add it is, the relay outputs are NO/NC relay contacts that provide a static output. (one side on / the other off and vice-versa). But I need a set / reset impulse for the ML-RBS latching relays, so I added 1 capacitor, 2 resistors and 1 LED per channel to create the impulses and have a visual feedback of the state of the BMS relays and a protection of the BMS relay contacts from over-current. I use latching relays insted of regular solenoids, because they draw zero power after the switching. They are expensive, but have a great advantage over other relays - they can be switched manually on and off. So if something breaks, you always have an option to tutn off the BMS and override manually the bms logic. This is a great safety feature when something happens on high seas and you have no time to figure it out and fix it.

I also have built an interface for the opto-coupler outputs, that switch a small signal relay with potential free contacts, so I can isolate more circuits and protect the opto-coupler outputs from over current (Interface consists of a relay, npn transistor, LED and a resistor that protects the relay contacts from over-current)

The circuitry mentioned above has been tested with my cells and works great.

If you use different relays, you may need a different interface or can contact them directly to the BMS outputs - the relay contacts can switch up to 1A according to the product sheet.

[CatNewBee]

Here is the principal diagram of a single relay interface:



The first resistor (15 Ohm) limits the total current over the relay contacts in case of a shortcut to < 1A.

When the relay flips and provides power (12V+) to the output, the empty capacitor starts charging and drags the input of the ML-RBS R/S input to high and flips the relay, when the capacitor is full, the voltage at the R/S Input drops to low and the positive impulse finishes. This happens within 200ms. The output on the BMS remains high and the LED is on on this channel, the resistor limits the current to 10mA - if you use a 1.5 kOhm resistor.

When the BMS switches to the other side, the capacitor discharges over the LED and the resistor quickly (within less than 1 second) and is then able to take the next impulse if necessary. Without the LED/resistor it would take a long time to discharge the capacitor, so this is important to have the LED/Resistor in place there.

The other side does the same as above, charge C2, flip the relay other LED on...

You need 2 of these interfaces, one for over voltage protection and one for under voltage protection if you intent to use the latching relays too. A normal solenoid can be probably driven either directly (if the coil does not draw more than 1 A) or over a second stronger relay, or over a power-transistor.

[arsenelupiga]

ok thanks for info. Will digest what you said and get back to you, if you do not mind.

My Lagoon is vintage 2013, ID 306.