DIY Lithium

[Derekoak]

I am hoping that being at least a month behind Harry and Phil I can learn and just follow the leaders.
However there is one area where I am installing differently. My battery will be underslung beside the lpg tank and diesel tank. That is where the existing AGM battery has been with no problem. So heating in cold will be more of a problem and an enclosure to protect against cold, knocks, stones, water spray and perhaps immersion will have to work harder.
On the other hand getting hot should be less of a problem.
So far I have decided that my 8 cells will be in 2 batteries of 4 cells in parallel, with 2 BMS, but all in one insulated waterproof enclosure, with the opening sealed lid at the top just under the car floor . They can then share heat. Apparently my maximum expected charge rate of 30 amp will only be 0.15C. At the moment even with the fridge small inverter, lights and everything on I will not discharge at more than 40amp = 0.2C. That scenario has never happened yet. Under these circumstances my research shows the cells produce very little heat. They may only be a degree or 2 above ambient. I think the BMS will produce more heat but I have no direct experience. They apparently get warm when they are discharging a cell, during charging, to balance the cells. The BMS have an aluminium finned heat sink but no fan I think.
My problem: Can I put them inside the waterproof enclosure to warm the batteries and be safe and use the wire lengths provided or do they need a different enclosure?

 

[Trotter]

I am hoping that being at least a month behind Harry and Phil I can learn and just follow the leaders.
However there is one area where I am installing differently. My battery will be underslung beside the lpg tank and diesel tank. That is where the existing AGM battery has been with no problem. So heating in cold will be more of a problem and an enclosure to protect against cold, knocks, stones, water spray and perhaps immersion will have to work harder.
On the other hand getting hot should be less of a problem.
So far I have decided that my 8 cells will be in 2 batteries of 4 cells in parallel, with 2 BMS, but all in one insulated waterproof enclosure, with the opening sealed lid at the top just under the car floor . They can then share heat. Apparently my maximum expected charge rate of 30 amp will only be 0.15C. At the moment even with the fridge small inverter, lights and everything on I will not discharge at more than 40amp = 0.2C. That scenario has never happened yet. Under these circumstances my research shows the cells produce very little heat. They may only be a degree or 2 above ambient. I think the BMS will produce more heat but I have no direct experience. They apparently get warm when they are discharging a cell, during charging, to balance the cells. The BMS have an aluminium finned heat sink but no fan I think.
My problem: Can I put them inside the waterproof enclosure to warm the batteries and be safe and use the wire lengths provided or do they need a different enclosure?

Following this thread with interest. Those bits I understand, that is.
The enclosure you're putting the cells and possibly the BMS x2 into, would there be room to add a small 10cm computer fan as well as any heater mats?

 

[Derekoak]

The space is possible, a fan would help cool if that is a problem for my situation. However an input hole and a fan mounted in an output hole would make waterproofing difficult and immersion worse. If the BMS are the heat problem they perhaps should be in a different enclosure but still holes for fans are difficult.
There is a sort of heat thermostat in our camper. If we are hot we tend to drive up a mountain. Unless the battery produces its own heat above 30C I do not think It will ever be hot for long under our car. The link I posted a few posts ago said that for longevity lifepo4 should live between 15 and 30 degree C as much as possible. We are much more below 15 than above 30. I suppose global heating might change that calculation in the future.

 

[Derekoak]

If you meant a fan to circulate the BMS heat within the enclosure, almost using the cells as a heat sink might be an idea. But first experience. Is heat a problem? The bluetooth will tell me something.

 

[Trotter]

If you meant a fan to circulate the BMS heat within the enclosure, almost using the cells as a heat sink might be an idea. But first experience. Is heat a problem? The bluetooth will tell me something.

I'd like to say, YES . But I never thought of that.

 

[xsilvergs]

Having considered the replies to my question on float charge levels for LiFePO4 batteries I found this video,

.

I'm still trying to get my brain out of 1980 and into 2021 (Ashes to Ashes was good but I thought Life on Mars was better), and both my Victron MPPT and B2B are adjustable for Absorption and Float I found the video interesting.

Hopefully my Ecotree lithium's will turn up tomorrow and once shoe horned in I can start my own testing.

Your thoughts would be appreciated.

 

[wildebus]

There is a setup I have mused on for ages as it made a great deal of sense to me and will almost certainly be doing this in the fairly near future once I sort out the right deal.
You see it mentioned occasionally but doesn't seem to really get discussed.

What it it? This chap presents it very nicely with good graphics to illustrate how it works ...

 

[mark61]

There is a setup I have mused on for ages as it made a great deal of sense to me and will almost certainly be doing this in the fairly near future once I sort out the right deal.
You see it mentioned occasionally but doesn't seem to really get discussed.

What it it? This chap presents it very nicely with good graphics to illustrate how it works ...

What a great video.
Cheers for linking.

 

[xsilvergs]

There is a setup I have mused on for ages as it made a great deal of sense to me and will almost certainly be doing this in the fairly near future once I sort out the right deal.
You see it mentioned occasionally but doesn't seem to really get discussed.

What it it? This chap presents it very nicely with good graphics to illustrate how it works ...

I'd never considered mixing lead and lithium!

I look forward to hearing your experience when you try 'the mix'.

 

[Derekoak]

Having considered the replies to my question on float charge levels for LiFePO4 batteries I found this video,

.

I'm still trying to get my brain out of 1980 and into 2021 (Ashes to Ashes was good but I thought Life on Mars was better), and both my Victron MPPT and B2B are adjustable for Absorption and Float I found the video interesting.

Hopefully my Ecotree lithium's will turn up tomorrow and once shoe horned in I can start my own testing.

Your thoughts would be appreciated.

It seems to me that as the man says it comes down to your particular charger. It seemed a very clear video. When I get my lithium I can test what my ring b2b actually does .

 

[Derekoak]

There is a setup I have mused on for ages as it made a great deal of sense to me and will almost certainly be doing this in the fairly near future once I sort out the right deal.
You see it mentioned occasionally but doesn't seem to really get discussed.

What it it? This chap presents it very nicely with good graphics to illustrate how it works ...

That looks as if it would suit you fine. I have a suspect Agm leisure battery so just lithium for me.
My new lithium should have about 4-5 times the usable capacity, less weight and fits in the same space. My Agm has to go!
He says lithium can do high C rate discharge but harms itself. Phil claimed that one advantage of lithium was it could run a bigger inverter than a similar lead battery. Probably how high is high is the issue. This yacht owner has, I suspect, a really really big lead acid bank and was imagining replacing with a smaller lithium bank due to cost.

 

[wildebus]

That looks as if it would suit you fine. I have a suspect Agm leisure battery so just lithium for me.
My new lithium should have about 4-5 times the usable capacity, less weight and fits in the same space. My Agm has to go!
He says lithium can do high C rate discharge but harms itself. Phil claimed that one advantage of lithium was it could run a bigger inverter than a similar lead battery. Probably how high is high is the issue. This yacht owner has, I suspect, a really really big lead acid bank and was imagining replacing with a smaller lithium bank due to cost.

His lead bank is 1500AH!

Ref the C-Rate, some cheaper lithium batteries BMSes have a pretty poor maximum current and just cannot run a high C rate and will either cut off, fail, or cook the cable depending on the design. whereas a Lead battery is not limited by that factor, but by the voltage sag - higher the current, greater the sag.
A newer, well designed/spec'ed 100Ah Lithium could deliver 100A or more no problem. You may not want to do that with a 200Ah lead bank (using 100 Vs 200 as "common wisdom" is 1Ah of lithium is worth 2Ah of lead) but you could for short periods. But a poorer 100Ah lithium might not even let you do it at all.
Plenty of people might want to run a hairdryer in their motorhome for say 5 minutes. That could easily take over 100Amps (a 1500W dryer is on the small side)
Running for 5 minutes will not take an enormous charge out the batteries and is perfectly ok, if inefficient, to so on a 200Ah Lead bank.
If you had certain lithium batteries, you could not even have the option to do it! The 100Ah Victron Smart LiFePO4 battery for example is 100A max - and that is not a cheap battery by anyone's calculation. Other lithiums will be lower (many 100Ah lithiums have a 50A limit).
Lithium IS better at running Inverters once you have a battery/bank that can cope with the maximum load you have - which can often mean multiple lithium to get to that point, not just a single battery.

 

[Trotter]

1500ah of lead acid provides a lot of ballast in a boat. Take that out, and he’d be bouncing around like a cork

 

[Philip Tomlinson]

Hymer have been installing hybrid LiFePO4/AGM systems for a few years now.

 

[Trotter]

Think I’ve found my next big spend

 

[SquirrellCook]

Let me get this straight in my head, combined lithium and lead. Parallel connection? Short high current demands taken for the lead and long low current demands from the lithium? I guess you need some kind of overload detector to disconnect the lithium for the high demands? The other way I imagine the lithium would look after the lead? Charging, I imagine one or both would need to be out of circuit to do this. Some clever electrickery required.

 

[Derekoak]

Heat pads to warm lithium. is the freely roaming (Dan) video missing something

Short version
I have had to relearn a lot of A level physics (50 years ago for me) to get this far . I may be wrong, I hope so. I have based by calculations on a video linked to earlier in this thread by Dan of freely roaming. He uses the same battery cells as Harry and Phil. In the third part he tested his heat pads in his camper at just above freezing. He tested how long it took to get his near frozen batteries to 7 C which he had decided was warm enough to start charging them. He did not test the temperature of his heat pads during the experiment he did not test the temperature at the top of his cells, they were hidden in his wooden enclosure box. I think he may have got some parts of his battery too hot for longevity whilst some parts were still too cold to charge . What do others think?
WHAT FOLLOWS IS COMPLICATED HEAT TRANSFER PHYSICS. DO NOT READ UNLESS YOU WILL BE GETTING DIY LITHIUM TO USE IN A SOMETIMES COLD CLIMATE. OR JUST LOVE MATHS!
I have been looking into heat movement formulae and other sources.
I found on google from several sources that the specific heat of a lifepo4 cell was 900 to 1100 j/kg/degree K. So if you want to know how long a certain weight of LiFePo4 cells will take to warm up from a cold temperature to a warmer. You can use the formula.
Time in seconds = specific heat as above x the required temperature difference between existing and a good charge temperature in degree C x the weight in kg divided by the watts.
Putting Dans video experiment with 69 watts and 21kg of cells into the formula . (He has the same size battery as Harry and one of Phil's). I took his example as 4 deg difference . Time = 1100 x 4 x 21/69 = 1339 seconds that is 22 minutes, which was about what he saw. This is not the end of it because his battery near the heat pads will be hotter than target, the temperature sensor got to target and the higher half of the battery will still be cold. So read on.
I also found a formula for heat transfer and what difference in temperature there is between pad and the coldest point. Q/t = KA dT/s
Q is the input energy in joules (a watt is a joule/second).
t is the time taken, so Q/t is the watts.
K is the coefficient of thermal conductivity for a lifepo4 cell
dT is the difference in temperature,
s is the distance from hot to cold in m.
A is the area of the cell side to be heated in m. squared.
I found this link https://www.researchgate.net/public...on_models_for_characterization_and_simulation.
This says that when a lifepo4 battery is heated from within by self discharge it looses the heat mostly via the X axis that is out of the biggest face. The core of the battery is much slower by at least (100times) to pass heat than the Aluminium casing. Remember however the casing has a much thinner cross section.
It gives a resistance of the core out through the X axis of 0.8 Degree C/ watt. Convert that to the coefficient of thermal conductivity K and you get 1.62 W/m/degreeC. The link does not give resistances for the Y and Z directions it says they are higher.
So lets try to use Dans heating experiment to see if we can predict his battery temperature gradient.
First the Area. Dan used 4 different sized heat pads and a thin aluminium plate to try to spread the heat evenly. He used a silicon trivet with ribs to insulate to avoid some loss of heat downwards. The battery will absorb heat faster than the trivet but some heat will be lost let us assume 20% of his watts are lost downwards. Otherwise lets assume his plate spreads the heat. His battery base with 3 big cell spacers 0.174m x 0.3 = 0.0522 m sq.
Then the coefficient lets assume the x axis coefficient for the Z axis . Its the best I can find. The coefficient will be higher than that. 1.62 then.
Q/t the watts 80% of 69 is 55 watts
S the distance. The distance to the sensors about halfway up is 0.105m. To the highest point is 0.205m
dT I forget I think his experiment started at 3.7 and finished at 7 degree C dT would be 3.3 deg C
dT= Q/t x s all divided by K x A
= 55 x .105/1.62 x 0.0522
dT = 68 deg C. Starting ambient 3.7 deg C at the beginning his heat pads and aluminium plate warm up to 71.7 degree C. They then slowly produce a heat gradient that warms the cells until the sensors reach 7 C. The pads are now at 75C. Some regions of the cells will be nearly as hot. Meanwhile the temperature gradient has not started warming most of the top part of the cells.

 

[wildebus]

Let me get this straight in my head, combined lithium and lead. Parallel connection? Short high current demands taken for the lead and long low current demands from the lithium? I guess you need some kind of overload detector to disconnect the lithium for the high demands? The other way I imagine the lithium would look after the lead? Charging, I imagine one or both would need to be out of circuit to do this. Some clever electrickery required.

The way I think it would work - and pretty well 'confirmed' by Clark, the video man - is that the normal operating voltage range of the Lithium is in essence above the normal 'resting' voltage of the Lead.
So as you call on the Hybrid (as Phil says Hymer refer to it) Bank, it tends to be the Lithium battery that delivers it. (even though the batteries are parallel-connected, they are unbalanced when working, which would be a bad thing usually, but it is what makes this hybrid system work).
And it is only when the SOC of the Lithium Battery is low enough (around 20% to 30% SOC depending on load) that the voltage drops to the level where the Lead starts to kick in, contribute and start to take over.
I also think that for this to work optimally is that you need a battery bank that is sized with a degree of over-capacity, and not one you take down everyday to a low level and recharge.
(In my case, I have 300AH of Lead Carbon Battery and I have that sized in order to reliably deliver up to around 240Ah if I need to. But generally don't expect to use more than say 75Ah in a day. If I added a 100Ah Lithium Battery, that would take care of the typical daily use but still have that extra 240Ah from the Lead Bank in my back pocket - and at a cost of less than 2/3rds of that one 100Ah Lithium)

For this to work, I think you have to have a lithium battery/s that are able to deliver the maximum current your installation would demand - otherwise if the lithium kept disconnecting and reconnecting the setup would get very messed up. I don't believe an overload detector to disconnect would be a good idea as it would stop it being a Hybrid setup consistently.

What I have thought is good about this kind of setup - and the video (and some of his replies within the comments) put this into words much better than I was able to - was that it results in a setup where you end up with generally the Lithium doing most of the day to day work, but not dropping below 20% DoD (which typically means it will last upto twice as many cycles compared to full 100% DoD); And because the Lead Acid batteries are not being used as much and dropping as much, they are not using up their much lesser charge cycle count anywhere near as fast.
End result being that both types of battery last longer than if they were on their own.

Plus another benefit .... because it is the Lithium that is being depleted generally, and lithium charges more efficiently, you are getting better and faster charging overall.
The charging profiles are not that different really. Using the one charger and the two technologies paralleled together would mean, I think, that a level of compromise for charge levels is need to accomodate both, but the ranges are such that I don't reckon that should be an issue (and if you have Lead Carbon, the voltages for those tend to be lower than standard Lead Acid which is an advantage).

Looking forward to trying this out hopefully in a couple of weeks time - and will likely fit a Victron SmartShunt on the -ve from the Lithium Battery to the general Battery Bank -ve. Using this to monitor the Lithium specifically along with the BMV to monitor the entire bank of Lead + Lithium will reveal just how well the Hybrid setup is working and if it actually does what I am expecting

 

[Derekoak]

Maybe the heat travels so fast through the aluminium casing that there is much less heat gradient?
If it were so the thermal conductivity of aluminium is about 230 w/m x degC. The cross section area of all the aluminium sides is very low depending on the casing thickness. At 1mm thick it is only 0.00194 m sq area.
But if it were so the heat would run round the casings and start to enter the core of the cells from all sides? If that were so the temperature gradient would be only about 3 degree. There would be no problem. Which is it?

 

[Petes]

Having considered the replies to my question on float charge levels for LiFePO4 batteries I found this video,

.

I'm still trying to get my brain out of 1980 and into 2021 (Ashes to Ashes was good but I thought Life on Mars was better), and both my Victron MPPT and B2B are adjustable for Absorption and Float I found the video interesting.

Hopefully my Ecotree lithium's will turn up tomorrow and once shoe horned in I can start my own testing.

Your thoughts would be appreciated.

Interesting video but I was unsure what his BMS was doing, it suggested to me that he didnt have a BMS but I agree with everything he said.

I have the Victron Solar controller and have my absorption at 14.2 and float at 13.5V, the BMS in the battery is controlling the charge, it Bulks to around 13.79V and then the current starts to tail off until it reaches 14.2 at which point it might be carrying out some cell balancing as there are current spikes. These reduce over around 15 minutes then current falls to 0, after an hour it falls to float and the battery never takes anymore current.

So he is right you cannot float a lithium but if you are running a BMS, mines Daly it will do the battery management, the settings on the controller just facilitate that, they dont manage the battery.
I really should get my act together and give you some graphs from the VRM but I'm not that organised!!