DIY portable, useful LiFePO4

[@Michael]

So when there are no charge sources the house will draw down the lead acid's, and the lithium will replenish them from a .3v higher state?

What stops the lead acids from draining the lithium below its minimum voltage? Does the lithium battery BMS take care of that, or do you need a low-voltage disconnect somewhere?

[markopolo]

Originally Posted by @Michael

So when there are no charge sources the house will draw down the lead acid's, and the lithium will replenish them from a .3v higher state?

I don't think it'll work exactly like that. I'd expect power to come from a source that can support the highest voltage available in the circuit. As support falls, other sources join in. Examples: With engine running and a 1,000 watt load, the alternator could supply the needed power in my van. At 1500 watts, the batteries join in contributing 100W+. Input from solar panels has a related effect. When the only power source is batteries then I'd expect power to come from the LiFePO4 pack until it's voltage falls to where there's real support from the lead acid batteries starting at around 12.8V. The LiFePO4 pack would be around 13.1V then.

Originally Posted by @Michael

What stops the lead acids from draining the lithium below its minimum voltage? Does the lithium battery BMS take care of that, or do you need a low-voltage disconnect somewhere?

For automatic everyday type use, I would be relying on the BMS there. I've set that to 2.9VPC or 11.6V pack voltage whichever is breached first. The van had approximately 400Ah lead acid batteries at some point. I say had because age and use would have reduced that somewhat by now. Still, it's likely that several hundred amp hours would have to be used before seeing such a low voltage. Manually, there will be a physical switch to cutoff BMS output and also the ability to disable BMS output via phone app.

Note: I don't suggest that anyone copy any of what I've done or plan to do. I'm just trying to add to the knowledge base through trial & error. I have to figure out heat dissipation for the diode pack for example - just one of the things that still needs to be done.

I really appreciate the questions. I was so bummed out when testing showed that the previous plan wasn't good. Then I got overly excited with the new plan coming so quickly and just want to go with it but I do need to find the errors and weaknesses.

[@Michael]

Originally Posted by markopolo

When the only power source is batteries then I'd expect power to come from the LiFePO4 pack until it's voltage falls to where there's real support from the lead acid batteries starting at around 12.8V. The LiFePO4 pack would be around 13.1V then.

Is that because the LiFePo4 pack has lower internal resistance and/or higher voltage?

One thing I see is that you really don't have a way of separating the two packs if you need to for any reason, I.E. if you want to charge or discharge only one of the packs and not the other. But solid state relays in the right spots would make that possible.

[markopolo]

Originally Posted by @Michael

Is that because the LiFePo4 pack has lower internal resistance and/or higher voltage?

My assumption is that higher voltage (with sufficient support) is the cause. A small capacity LiFePO4 pack will realize greater voltage drop under load than a large capacity LiFePO4 pack. With the 122A load while running a coffee maker and a toaster simultaneously - https://www.classbforum.com/forums/f...tml#post106709 - voltage of this smaller pack dropped to 12.47V. If paralleled with lead acid then there would be load sharing and less voltage drop at the LiFePO4 pack.

Originally Posted by @Michael

One thing I see is that you really don't have a way of separating the two packs if you need to for any reason, I.E. if you want to charge or discharge only one of the packs and not the other. But solid state relays in the right spots would make that possible.

That's correct. My intention is to keep them in parallel when using the LiFePO4 pack in the van. That avoids potential problems associated with inrush and outrush current.

Edit to add: There will be fuses or breakers where necessary. That is critically important.

[JohnnyFry]

150 A/H using 4 Prizmic (sp?) cells. I am using the Xlaoxlang BMS with low temp cut out. I am getting the DC-DC charger wired in today.

[Winston]

Why not 'simply' replace those aging AGMs with lithium?

[JohnnyFry]

As I may have mentioned earlier, I have no intension of keeping the battery packs in parallel except under one specific situation. The lead battery in my system is the engine battery and charges as normal from the alternator. The LiFePO4 pack is charged when the engine is running through a DC-DC charger which is enabled when the alternator begins producing voltage. When boondocking, the Li pack provides power. When either shore power is connected or the generator starts, a 120VAC coil DPDT relay energized and does 2 things: it disconnects the enable line to the DC- DC charger and activates a separator relaywhich parallels the two batteries. This accomplishes two things: it allows theTrippLite to charge both batteries, and it provides a stabilizing reference for the TrippLite when the Li pack reaches full charge and goes off line.

[markopolo]

Originally Posted by JohnnyFry

150 A/H using 4 Prizmic (sp?) cells. I am using the Xlaoxlang BMS with low temp cut out. I am getting the DC-DC charger wired in today.

The low temp cutoff appeared to work as it should on mine (150A version). I only did one test. Will Prowse indicated that he struggled with 150A BMS. Could have been that the temperature wasn't calibrated, app version or the temp setting was too low. Mine came set -5C to trigger the cutoff so I changed it.

The enterprise app on his site is version 2.0.1006.4 with the target SDK being Android Jelly Bean 4.3 - so very old. Current app version is 3.1.1015 and target SDK is Android 9.0 Pie and has a minimum SDK of Android 5.0 Lollipop. Probably a lot has changed over the years.

Did you opt for the Bluetooth module and or PC module?

Is your BMS the 120A version? If so, it appears to have a row of TVS diodes that aren't on the 150A version that I have. I wonder if those are there for motor use?

The 120A version actually has more of the same MOSFETs than mine but are much more tightly packed so heat dissipation allowance probably causes the lower amp rating.

Is yours approximately 5.5" x 4"?

The 150A BMS measures approximately 12.5" x 4.75"

We can use the same apps and PC programs and help each other with settings etc.

[markopolo]

Originally Posted by Winston

Why not 'simply' replace those aging AGMs with lithium?

I am planning to run a capacity test of the van's AGM batteries when the temperatures get warmer here.

I don't have any interest in going only LiFePO4 due to lack of need, cost, temperature issues, 2nd alternator etc. It is fun having the portable LiFePO4 pack(s) though. $150 of cells in Pack 1 likely can supply the energy to power the vans toaster and a coffee maker simultaneously 10,000 or 20,000 times. (guessing 5 to 6 per cycle x 2k or 3k cycles)

I do understand that some folks who use their van a lot more than me can utilize a greater capacity LiFePO4 battery bank.

Keeping it portable or somewhat portable means you get to keep the LiFePO4 pack(s) when you sell your van. Take a pack out of your van and use it on your boat as needed. Use it in your house when needed.

The question might very well become "Why not parallel LiFePO4 batteries with the existing batteries in a touring Class B van?" The idea might catch on if a competitively priced high output plug & play module becomes commercially available in North America.

[hbn7hj]

Originally Posted by markopolo

The question might very well become "Why not parallel LiFePO4 batteries with the existing batteries in a touring Class B van?" The idea might catch on if a competitively priced high output plug & play module becomes commercially available in North America.

All it takes is a lithium battery, a Blue Sea battery switch, and a way to charge the lithium battery from AC power and/or the present alternator with a DC to DC charger. You need to have a plan to deal with winter storage. Converter stays in the lead acid mode and does not need replacement.

Minimum cost for 100 AH would be around $1000-$1500.

The portable version is interesting, handy, and a lot cheaper.

[JohnnyFry]

Mine is the smaller version. I am using the iOS version of the software. The original download url from the sellers site worked but I was prompted to upgrade to the “professional “ version for $5.99. I am not convinced that I got anything different. I was able to go into the setup mode and input various parameters. My starting battery was down so I put the batteries in parallel, I got a current flow of about 20 amps toward the starting battery. After waiting 10 minutes or so I started the vehicle engine. Later the iOS app showed that I had drawn 140 amps from the Li pack.

[Winston]

Originally Posted by markopolo

Keeping it portable or somewhat portable means you get to keep the LiFePO4 pack(s) when you sell your van.

You brought a smile to our face - - having spent 4 months of, what seemed like, 24/7 effort to build-out this DIY ProMaster, there is no way we're ever going to sell it. Once was more than enough!

[markopolo]

Johnny - I think the paid iOS version was a good choice to get access to settings. It'll likely prove it's usefulness as you get things dialed in.

Are you thinking that the ammeter or coulomb counting feature is inaccurate?

I know it can be calibrated via PC program but not 100% certain that it can be calibrated using any of the apps. I suspect the "galvanometer" value on the app I use will affect current measurement but I haven't adjusted it.

[markopolo]

With my latest paralleling plan, current from the LiFePO4 will be wasted as it goes through the automatic charge relay and into the starting battery and also supplies any GM related parasitic loads.

I can turn the automatic charge relay off by interrupting the small gauge ground wire using a switch but I'm stumped as to how to automate that. Automatic turn on is 13.3V and turn off is 12.8V now.

Probably just have to live with using the switch, maybe one with a led and move it to a more visible location.

[JohnnyFry]

Just finished my install. 150 AH Li Battery is in heated space, DC/DC charger, 300 Watt Sine Wave inverter to suppliment 750 Watt TrippLite. When on shore power or genny relay cuts off DC/DC Charger and brings Pb starter battery on line to keep it charged and stabalize the 12V from the TrippLite when the Li BMS cuts off charging.

I have a photo but can insert it on this site.

[markopolo]

It's worth mentioning the difference between Power Capacity and Power Capability in this topic.

I see real value in the Power Capability of LiFePO4. That is how large of a load it can power. You could almost say a DIY small pack is a bargain if the large loads are infrequent and short duration like microwave oven use or coffee maker use. This small pack can easily power any load in the van.

In terms of Power Capacity, meaning energy stored, then it's still expensive, IMO. That's worsened by the low Power Capability of many drop-in LiFePO4 offered for sale now.

Renogy's 50Ah LFP battery sells for $425 today - https://www.renogy.com/lithium-iron-...12-volt-50-ah/ - but it's 50A max discharge current rating means that you'd have to buy three ($1,275.00) just to get 1800W capability. Or $1,800 for two Renogy 100Ah LFP batteries just to get 1,800W Power Capability (today's pricing). Three 50Ah LiFePO4 batteries from GoWesty would cost around $1,000.00 - https://www.gowesty.com/product/elec...x_battery-50ah - that makes for a nice blend of Power Capacity and Power Capability but no control of parameter settings or monitoring via your phone like you get with DIY. And it would be a permanent type installation so a DC to DC charger for sure and maybe heating required.

A lot of Class B's with absorption fridges have enough lead acid battery power capacity to meet the user's general needs but lack the power capability to run appliances like a microwave oven from an inverter when off grid. That's where a small, relatively inexpensive LiFePO4 add-on could be really useful.

The cells I used have a 6C max discharge rating. Getting 1800W from the 36 cells in the pack is just above a 2.5C discharge rate so they're not being maxed out.

I came across this 60Ah capacity, 150A capable, Bluetooth controlled, portable LiFePO4 pack yesterday - https://www.aliexpress.com/item/4000133471806.html - that's ready to go at a DIY price. I'd strongly suggest seeing photos of how it's assembled and details of the cells etc. before considering to buy it. It probably has the same 150A BMS that I'm using but get photos of it and the rest of the internals to see what's in there.

[booster]

Very interesting, Marko. A lot of us had to make sure we got enough AGM battery bank to easily run the microwave or a hair dryer as needed, so very familiar with trying to balance the capability with capacity without too many or too few batteries.


It has been mentioned several times that the trend in lithium systems seems to be to dropping the maximum output capability with many of the drop ins down to about .5C now, which is in AGM territory. Some have thought that it was the BMS limitations or maybe to reduce the cost of relays and such, but I don't think anyone really knows yet. IIRC Batteborn's low limit is fairly new. Perhaps they have been seeing issues with high discharge/cahrge rates causing issues.


I think a lot of it may be the "normal" evolution of the products as they are getting used more in the real world. We have seen it go from claims of discharge and charge at 5C, 7000 100% discharge cycles to much more conservative (realistic?) claims. 10+% SOC cushions are not unusual now which put them at a similar .8C usable capacity like AGM. Cycle life of more in the 2000 cycle range. Discharge and charge rates at .5C.


All we can compare to much, I think, are things like the phones we all have, even though they are different chemistry, and the power tool lithium battery packs. IIRC, the last numbers someone here quoted were something like 700 cycles for an Iphone. Phones tend to use a lot of their capacity, it appears, and charge fast. If running games they discharge quickly, but not at huge rates like we are talking about. 700 cycles is not much compared to the claims we see for RV use lithiums. I don't recall seeing any cycle life claims lately on the power tool ones, but the durability seems to vary a lot between brands. I have two Bosch drills and an impact driver with one pair of batteries at about 8 years old, another at about 5, and one set at about 6 months. They all still work well, and I tend to use them a lot for a DIYer. About the only thing I don't use them for is driving a buzzilion screws like if you are putting down surface boards of a deck as that seems to be bad in general as both the drill and batteries get too hot. I previously had a Toschiba drill and batteries that got used less but lasted less the 3 years before the batteries failed. The Bosch system does seem to limit charging amount as it is not blazing fast and they seem to have decent cushion on the bottom. Won't charge is battery is hot and will shut down when drilling if it gets hot.


I think it possible that we will see a lot more of the adjustment of discharge and charge and other specs in the future. Could include charge and discharge rates, top and bottom cushion space, cycle life, etc. It is possible that because the early systems were also almost all very large capacity, they masked the issues of high charge and discharge rates. 300 amp discharge or charge rate is not even .5C on an 800ah bank. It is 1.5C on a getting more common 200ah bank, though and 3C on a 100ah battery.


The big, somewhat still unanswered completely IMO, is that we know the lithium systems CAN discharge and charge at high rates (3C maybe) but the question comes in as to whether they SHOULD be run that hard and if and how much it may shorten their life.


Once they get less expensive it may turn out to just be a personal choice of accepting shorter life to be able to have less batteries even if it is less cost effective, similar to the how deep to discharge lead acid battery decisions we all do. I think that is the way it is with the phones now.

[Winston]

Originally Posted by booster

A lot of us had to make sure we got enough AGM battery bank to easily run the . . . hair dryer . . .

We'll have to visit Bosley first . . .

[markopolo]

Pack 2 finished and working great. It's identical to Pack 1. There's no movement inside the case even if turned upside down.

Pack 2.JPG

These Packs weigh 18 lbs. 12 lbs for just the LiFePO4 cell pack and 6 lbs for everything else.

SOC accuracy looks pretty good so far without doing any calibration. After 30% of capacity discharge it showed 41.22Ah remaining and after 40% of capacity discharge it showed 36.6Ah remaining, 62% left. The SOC metering resets when fully charged.

capacity accuracy.jpg

Voltage alone is not a good way to estimate SOC. This image shows the voltage rebound after discharging. This was at the 50% DOD point with previous rest periods every 10% SOC. You'll measure different voltages depending on how long the battery has been at rest and it does this for many hours. It's hours:minutes:seconds on the chart.

voltage recovery.jpg

[markopolo]

Just some interesting stuff:

Soldering Lithium Batteries by Panasonic:
https://industrial.panasonic.com/cdb...A4000COL21.pdf

Memory effect: - idling for a sufficiently long period of time under certain conditions being used to erase the memory effect is interesting - https://www.psi.ch/en/media/our-rese...-ion-batteries

Supplementary information PDF: https://web.archive.org/web/20200416...mat3623-s1.pdf

Interesting info here: -
Accelerated aging and degradation mechanism of LiFePO4/graphite batteries cycled at high discharge rates: https://pubs.rsc.org/en/content/arti.../ra/c8ra04074e

The cycles were: charge to 3.7V then discharged to 2.5V - so 14.8V down to 10V, way more extreme than typical RV use. That's repeated over and over. I'm not sure how to apply that to very occasional, very short duration 1C or 2C or 3C use in an RV.