Update on our lithium "test" bank in 07 Roadtrek 190P

[booster]

I have mentioned a couple of times that I decided to do a test installation of three SOK, 206ah, 12v, "drop in" style batteries.


Having mentioned many times in the past that I felt the whole idea of "drop in" lithium swaps was not a great idea because of the differences in care and feeding that lithium require, and I still feel that way. I have watched the lithium market for a long time for pricing, features, claims (outrageous and some believeable ), etc and a few recent things made me decide it might be a feasible for our particular system because of it's programmability.


The things that have changed in the marketplace to convince me.


* Much lower pricing on lithium

* The availability of a removable cover, serviceable, drop in battery

* Ability to monitor individual cell voltages on all cells

* Improved amp hour capacity per footprint space and volume

* Class A cells being offered

* Some changes in specifications and parameters from manufacturers but all are still contradictory and only address some of what "best practices" appear to be.

* The availability of a much better alternator remote regulator that can terminate charge based on shunt measured amps to the batteries.


We had two trips planned for this summer, our yearly spring 16 days to Custer State Park in SD, and a one month long touring type trip with 2-5 day stays along the way going to the U.P. in Michigan, western New York to visit a fellow member, and then the Skyine, Blue Ridge, Smoky parkways and our final stay in Smoky Mtn NP.

We simulated how we intended to test the lithium bank on our trip to Custer, but with our 440ah of AGM batteries. We never used shore power and had the solar shut off. This gave us a comparison of what SOC the AGMs would settle at with less than the very long charge time it takes to get to 100%. As expected, they ran in the 40-85% range on that trip except for the long ride home where they got nearly to 100%.

I put the lithium in with temporary wiring when we got home and got what I felt were safe parameters. I programmed the shore charger and solar controller for them, but we did not intend to use either unless we had to. The Ample Power remote regulator on the parallel alternators was changed to a profile that was closest to decent for lithium, on the safe side and I shut off the charging by just disconnecting them from the alternators based on an ammeter on the dash.


The batteries were not very good out of the box and required a bunch of work at install, which the next post will go over. Long story short is that if they could not have the covers removed, they would not have worked adequately, IMO.

[booster]

I will go over what I found in trying and testing the batteries at final install. Lots of detail will not be there and if interested just PM me or ask here.


Batteries are 3 X SOK 206ah 12v lithium batteries with claimed grade A cells and removable covers for servicing if needed. They are replacing 440ah of Lifeline AGM 6v GC batteries.


Instructions for the batteries said to charge each of the batteries individually to 14.4-14.6v to get them even in SOC before installation and charging/discharging as a group in parallel.


First issue showed up immediately as none of the batteries would go past 13.8v without tripping the internal high voltage trip out. I later figured out this was because of poor cell to cell balance and the trip out looks at the individual cell voltages so if one is high it trips out early.


SOK said to just leave it at 14.4v and let it cycle up and down until the internal balancer fixed the balance, but the trip kicks out the smart charger so it is not ready to charge again when the voltage finally goes down. After half a dozen manual restarts and nearly no improvement I decided to order a 10amp regulated power supply to do the initial charge. They are under $50 on Amazon and other places. Using it the cycling would continue on it's own, but after 48 hours they would barely charge to a higher voltage overall and I decided I couldn't wait a month to get them all balanced. I removed the covers and balanced the individual cells with power supply to 14.4v and 3 amps current. Put them back together and all of them would go to 14.4v without tripping so I let it run until the Bluetooth showed very good cell to cell balance.


I put them in the van and did a discharge test where it ran all the accessories just fine, but the batteries show huge differential between discharge currents of 50% or more which caused the batteries to trip out early. Recharged and saw the same variations. SOK did not offer any useful help so I started looking for resistance variations between the batteries and found none, cable lengths balanced etc.



I then removed all 3 batteries and opened them up, and moved the BMS and cover from the slowest battery to the fastest one and vice versa. The problem moved with the BMS. SOK wouldn't send new BMS to try without sending in the whole battery at my cost, so I did a bit more checking for resistance and found the wire connection from the cells/BMS to the cover connecting point showed high resistance. I took it apart, cleaned it all and put it together and reran the tests.



I had received two batteries with much lower serial numbers than the third one and the newer one was working much better for charge and recharge rate than the other two, although the really low one was good now with the wiring cleaning. I opened them again and noticed the two lower serial number batteries had the BMS/balancing wires (4) placed under the buss bars and wiring while the higher serial number had them on top like I now know is how they should be. I moved the wires on the two batteries and retested them.


Charge and discharge rates were now very close to equal between all three batteries and they charged well and capacity tested very good.


After a few cycles the balance of them got even better to nearly perfect and stayed there even if I changed to lower than the balancer activation voltage of 14.4 to 14.6v as I consider that too high for long battery life.


I finished up the rest of the install to get ready for a 30 day shake down trip that would be done with only alternator charging with the best settings I could get with the Charles/Ample Power regulator at 13.8v but not very stable.


The batteries performed very well on the test trip and actually exceeded expectations. The trip also let me test controlling charge termination by using amps instead of volts alone and I found it was much more consistent that way.


The next post will summarize how I tested and what the results were.

[GallenH]

Good reading. I'm enjoying all of your posts/info. Thanks for taking the time to keep us all informed.

[GeorgeRa]

Originally Posted by booster

.........
We simulated how we intended to test the lithium bank on our trip to Custer, but with our 440ah of AGM batteries. We never used shore power and had the solar shut off. This gave us a comparison of what SOC the AGMs would settle at with less than the very long charge time it takes to get to 100%. As expected, they ran in the 40-85% range on that trip except for the long ride home where they got nearly to 100%.

...........

Are you planning to use shore and solar charging sources. Are your solar charge controller and shore power charger programmable for Lithium profile?

[booster]

Originally Posted by GeorgeRa

Are you planning to use shore and solar charging sources. Are your solar charge controller and shore power charger programmable for Lithium profile?

I think both of those items may qualify for the "yet to be determined" category, but we really enjoyed the no plug in, no wet dirty cord, routine on the test trip which was intended to test only alternator charging.


If I had to guess right now, shore power will probably not be used at all unless we absolutely need air conditioning.


Solar is a bit tougher as it does not have a "lithium" setting, but IMO the lithium settings are generally not what I want anyway. It appears that the newer versions of our Blue Sky controller have a two stage setting which would work well and even if we have to use 3 stage we would only use it if we we want to be offgrid without driving for about a week. Our controller rebulks at the float voltage so pretty much a two stage anyway.


We were totally amazed with how much more convenient it is to not plug in, not from the plug in necessarily, but from the putting away a cold, still, dirty cord early in the morning when it is cold and dark.


I am finding this whole adventure very interesting and educational, plus it looks like we may actually come out of it all with a quite functional system.


l will point out, and probably repeat many times, that our charging equipment is very much more flexible than most systems in class b's so our experience will likely not be able to just repeat for anyone else. It is very obvious to me that the lithium batteries we got are far from being true "drop in" to a lead acid system.

[@Michael]

Originally Posted by booster

It is very obvious to me that the lithium batteries we got are far from being true "drop in" to a lead acid system.

Seems like SOK's assembly process has room for improvement. The issues you've discovered are pretty fundamental to building a lithium battery.

I appreciate your willingness to write up what you've experienced - it'll no doubt help someone someday.

[booster]

Originally Posted by @Michael

Seems like SOK's assembly process has room for improvement. The issues you've discovered are pretty fundamental to building a lithium battery.

I appreciate your willingness to write up what you've experienced - it'll no doubt help someone someday.

I completely agree on your assessment, although for me it was not unexpected. I have read about numerous lithium drop ins of various brands tripping out early so kind of expected it. I probably would not even gone on this adventure if I couldn't get the batteries open to be able to address the predictable problems.


My guess is that very few of the manufacturers top balance the cells the put in their batteries and that causes the issues I, and others, have seen.


When you only have a couple of tenths of a volt between the charge voltage and full shutoff of the battery, there is very little room for inbalance.

[GeorgeRa]

Thank you for your update. It helped me to decide staying with Fullriver AGMs. My 2013 system is designed for low attention required with AGMs. All 3 charging sources don’t have Li profiles (whatever the lithium profile is).

For alternator charging I would need to change from ME-SBC to something else. ME-SBC has voltage adjustment but likely not accurate enough. Charge termination would need to be manual.

For solar charging my Mornigstar TS-MPPT 45 has fixed voltage option but charge termination would need to be manual.

For shore charging my Magnum MMS1012 inverter/charger has custom voltage setup available but again termination of charge would need to be manual.

I could go with Lithium by completely replacing all 3 charge controllers, this would be a big project, too big for my 2013 campervan.

With my current charging system the decision would have two main factors:

Lithium advantages over AGMs (weight, capacity)
vs
practically automatic operation of my current AGM system.

[@Michael]

Originally Posted by booster

... so I did a bit more checking for resistance and found the wire connection from the cells/BMS to the cover connecting point showed high resistance. I took it apart, cleaned it all and put it together and reran the tests.

[...]

I opened them again and noticed the two lower serial number batteries had the BMS/balancing wires (4) placed under the buss bars ...

I also learned (from experience) that low resistance, high-current charge/discharge setups are far less tolerant of marginal wiring, buss bars, etc. My current setup can charge at up to 95A, making the quality & cleanliness of crimps, wiring and connectors, as well as proper wire selection much more relevant.

Seems as though some LiFePo4 vendors are still on that learning curve.

[phantomjock]

...but by then we may all be using hydrogen fuel cells

Then you could repurpose the extra water tank and use as a fresh byproduct tank...

Great summery of the setup. Thanks.

Cheers - Jim

[booster]

Now to what would probably be considered the more controversial part of this whole testing project, the charging control and settings. I am totally aware that what I intend to do is far from what a lot of the "drop in" style batteries list as recommended settings and I will attempt to explain why I think going to a different way is more suitable, at least for our system.


We have the ability to charge to any specific voltage limit and also have the charging terminated based on the amps to the batteries at the set voltage. This is the best way to charge AGM batteries to tail amps so we have it left over from the AGM system we just removed. As I read more and more scholarly papers on lifepo4 battery charging the premise of charging to an amperage level started coming up as preferred very regularly. Consistency of process was the most common reason to choose such a method as voltage changes are too small for that to be used accurately, while the amperage changes are quite large in the changeover zones.


The information I am mostly missing is coming up with what is really the best amperage to terminate charging at any particular voltage.


I decided to try to "back door" that decision by working backwards from what I consider other best practices for charging. I chose a 13.8v max charge voltage and a resting SOC after charging of 13.35v which is in the 95% SOC range.


Those parameters allowed me to conduct on the fly testing on our 30 day shakedown trip by charging at 13.8v until the amperage to the batteries dropped to various levels, and then terminating charging and letting the batteries rest a while to see what voltage they settled at. From those tests it looks like 50-60 amps at 13.8v will probably be a good point to be at, but still needs to confirmed with better controls and more data points.


Once I have these points, setting up to do it with our equipment gets much easier. I just set the solar and shore chargers to 13.8v absorption and the tail amps to 55 and good to go. I think the Wakespeed alternator regulator will be able to do the same but don't know that yet as I it hasn't arrived and been installed to this point.


As it is stated above I have chosen 13.8v max voltage and 95% max SOC for my limits on charging. I believe that these should be the best balance of charging speed and battery life, but we all know lithium best practices change with the wind so could be way off also.


Many batteries charge much higher to 14.6v to get to 100% full, but it appears this may cause life issues per much of the literature. Same is true about going to very deep discharges to near zero SOC. Those recommendations may just not have been updated or could be influenced by marketing and warranty issues, but I have no idea what is going on. They may also be the best, who knows at this point.


The gist I got from lots, and lots, of reading on the topic is that higher temps, maybe even over 95*F, shorten life. Higher voltage, maybe over 14.1v but certainly over 14.4v shorten life. Time spent at either of the above issues adds continuously more damage. Holding at moderate to high voltages when charge current is very low may be a problem but not as certain as the above two. Battery life gets shorter based on how deep the discharges are before charging, similar to what we know about AGM batteries, but maybe even worse. There are a lot of conflicting specs on this issue. lower charge rates in the .2-.4C are recommended for longer battery life as well as lower discharge rates than the 1C and greater previously stated.


So, bottom line is I chose what I chose based on what I felt the best information indicated would work best with our system. It may be right or wrong, but I bet at least I will learn something from it all and that was the purpose of this whole test install.


I hope others take other paths and share how it all works out so we have some comparison data.


I will also be very interested in how Avanti sets up his larger and more sophisticated bank on his new van, as I am sure he has been reading a lot of the same stuff I have.

[GeorgeRa]

Charging termination based on the current set point at the final voltage level will require charging by one charging source only. Selection of one of the three sources, alternator, PV, or shore, would like to be done manually, not a big issue but attention would be required.

Do all of your charging sources have capability for current tail end termination at owner selected voltage?

[booster]

Originally Posted by GeorgeRa

Charging termination based on the current set point at the final voltage level will require charging by one charging source only. Selection of one of the three sources, alternator, PV, or shore, would like to be done manually, not a big issue but attention would be required.

Do all of your charging sources have capability for current tail end termination at owner selected voltage?

We have done charging from multiple sources to tail amps at the same time for many years as they seem to play together just fine.


It was always two sources though, as there isn't really a case where we would have all the sources running at the same time. It was solar/shore or solar/alternator.


Both the solar and the shore charger will automatically and independently charge to the reading of amps from the same shunt. Minor differences make them shut off at slightly different times, but close together if set to be that way.


The alternator charging was done manually as I drove by watching an ammeter on the dash and shutting off the charging when it was done or the solar could produce enough to finish the charge automatically.


Hopefully, the Wakespeed will make the alternator charging also automatic so I don't have to worry about watching it. The lithium amps change very rapidly at the desired end point so if you are in traffic or other attention needing conditions it is easy to miss manually so automatic will be very nice to have if it works.


The solar and shore chargers are fully settable for voltages, tail amps, and/or hold times. The current regulator is the weak spot with only preset voltages, IIRC about 4 options, and the current is set via putting a cap on field % and only limits charge rate. Alternator charging now is not automatic but should be in the future with full amps/volts ect setability.

[GeorgeRa]

Originally Posted by booster

We have done charging from multiple sources to tail amps at the same time for many years as they seem to play together just fine.


It was always two sources though, as there isn't really a case where we would have all the sources running at the same time. It was solar/shore or solar/alternator.


Both the solar and the shore charger will automatically and independently charge to the reading of amps from the same shunt. Minor differences make them shut off at slightly different times, but close together if set to be that way.


The alternator charging was done manually as I drove by watching an ammeter on the dash and shutting off the charging when it was done or the solar could produce enough to finish the charge automatically.


Hopefully, the Wakespeed will make the alternator charging also automatic so I don't have to worry about watching it. The lithium amps change very rapidly at the desired end point so if you are in traffic or other attention needing conditions it is easy to miss manually so automatic will be very nice to have if it works.


The solar and shore chargers are fully settable for voltages, tail amps, and/or hold times. The current regulator is the weak spot with only preset voltages, IIRC about 4 options, and the current is set via putting a cap on field % and only limits charge rate. Alternator charging now is not automatic but should be in the future with full amps/volts ect setability.

Thank you.

Using the same shunt, I think you mentioned it earlier, I missed it. I agree, 3 sources wound never be used simultaneously.

At some point a company will come up with a system capable to:

IN - take 3 power sources
OUT – charge Lithium batteries, AC power
IN/OUT – control/monitor panel.

Perhaps they exist but I am not up to date with the current leading edge.
Thank you for updating your progress.