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

[T1 Terry]

Originally Posted by FlyingCloud1937

Once we have a thermal runaway...no matter the Li-type cell all we can do is wait for the energy to stop. Firefighting is to the surrounding flammables.

Containment of fire spreed is always number one in firefighting.

Lloyd

Time to look for another bridge Loyd, this one has anti troll protection fitted.

[FlyingCloud1937]

Originally Posted by T1 Terry

That sounds like "Because the Hindenburg caught fire all balloons should be banned or at least in serious doubt.
The Boeing batteries were lithium cobalt oxide, a known thermal runaway battery http://www.gsyuasa-lp.com/aviation-lithium-batteries no one is suggesting to use these cells for house batteries. The nonsense the press is going on with at the moment is quite laughable for anyone who knows anything about these type of batteries, they a prattling on with nonsense and gullible people are taking it as genuine information, now it becomes scary. When people start believing the rubbish they hear on the 6 o'clock news we are in real trouble.

T1 Terry

T1 Terry,

FUD, can be scarey and used to sway...

but reality is, any bat can go into thermal.

So it must be front and center, especially when we are talking the propensity that these bats have.

So finding answers, is the proper course.
Whether, its voltage, amp, or firefighting control.

That is what will bring them to reality.

Lloyd

[T1 Terry]

This from abuse testing by Sandia Laboratory (Batlab) http://www.transportation.anl.gov/ba...atlab_roth.pdf


As you can see, LiFeP04 does not suffer from thermal runaway, if it gets hot enough in a fire it simply burns, it does add to the over all temperature of the fire in any significant way.

The testing does show that abuse overcharging will cause them to explode though, I doubt we would have the available voltages or currents to produce such a reaction but it can be done if someone is determined to do it.

T1 Terry

[mcarling]

Originally Posted by T1 Terry

That sounds like "Because the Hindenburg caught fire all balloons should be banned or at least in serious doubt.

That's a good analogy, in my opinion. The safety difference between the LiCoPO2 batteries in the 787 and the LiFePO4 batteries we're using is, in my opinion, similar to the safety difference between hydrogen balloons and helium balloons. Both can deflate, but only one can explode.

[daddle]

T1 Terry - thanks for the charts. I like real data.

Question: Is stopping charging only on reaching 3.8V reasonable? What if, say, 3.7V is reached and then some load comes online that prevents the voltage from reaching 3.8V but maintains 3.7V for an extended period of time?

[mcarling]

Originally Posted by daddle

Question: Is stopping charging only on reaching 3.8V reasonable? What if, say, 3.7V is reached and then some load comes online that prevents the voltage from reaching 3.8V but maintains 3.7V for an extended period of time?

The range of loads capable of doing that is very narrow. A small deviation in the load will either lower the voltage to about nominal (13.5V or less) or raise it enough to stop charging (13.8V).

[daddle]

Originally Posted by mcarling

The range of loads capable of doing that is very narrow. A small deviation in the load will either lower the voltage to about nominal (13.5V or less) or raise it enough to stop charging (13.8V).

Yeah, but what if? In my long electronics design experience these very things should be considered. Low probability is not good enough because we are designing for long safe economical service.

[FlyingCloud1937]

To Everyone...

I just want to say for the record....I'm not against Li-technology.

Especially LFP...in it's current state., and hopefully better in it's future state.

I'm not stuck on LA, SLA GLA.

As a matter of fact I hope we can find any alternative better, then what we have now.

I just don't want something like Mainsails' experience with AGM's...I was against AGM's when they entered the market place originally.

Why ..? because they were designed for FLOAT Service... then someone said yea they are also good for deep cycle.

I was there in the beginning, and I was an Early Out, bc I knew that marketing was in play as opposed to reality.

I now believe that Li-Tech is a potential reality...until the marketing side convinces the right side that facts don't matter.

Lloyd

[mcarling]

Originally Posted by daddle

Yeah, but what if? In my long electronics design experience these very things should be considered. Low probability is not good enough because we are designing for long safe economical service.

How long are you considering sustaining 13.7V at the battery? Minutes? Hours? Days?

[T1 Terry]

There is nothing to be gained by holding a cell at 3.7v, it will just heat it up internally finally resulting in damaging it. We have used a system where a heavy load is switched on for a few minutes if a cell triggers the 3.6v alarm, we used a water heater element powered by the inverter
this is a 24v battery, 8 cells in series. As you can see, 2 cells are ahead of the others, once these 2 cells were drawn down to the same capacity as the others they all came up close together, eventually one will run away again though if the end voltage is held high enough. Brian still uses this system to burn off excess solar, sort of a dump load idea, solar heated hot water in a round about way :lol:

T1 Terry

[K_V_B]

I've bene following the discussion here, and it has been quite interesting. I'm the kind of person fascinated by cutting edge technology, and I'm learning a lot here.

I just made a few calculations to have an idea how the LA and LFP stack up pricewise, and how they compare with LFP batteries currently offered by the likes of Mastervolt and Victron. And there the result is quite shocking.
LA seems to cost around 180 euro/kWh at both Mastervolt and Victron.
LFP buying cells from evpower.eu comes to about 360 euro/kWh.
That means that from a TOC perspective LFP probably comes out on top.

Both Victron and Mastervolt offer LFP batteries too, with integrated BMS systems. These however are quite expensive. Prices are around 1000-1200 euro/kWh.
Why are they so expensive? I think this picture explains part of it:



It looks like they put a DC/DC convertor on every cell (that's the only reason I can think of for the small transformers). So they have what they call "active balancing".
(I got that from here: Li-Ion BMS - Lithium-Ion Battery Management Systems and large battery packs). This supposedly actively shuffles energy between the cells to keep them at the same SOC, thus enabling getting about 10% more out of your pack.
However, why this would make sense in a marine application escapes me...

[daddle]

T1 Terry indicated that charging was stopped when the voltage reached 3.8V. That combined with other indications that holding the cell at higher than 3.4V could be harmful, perhaps dangerous, made me point out a possible problem with the automatic charging controls. As described the charger could run for an infinite period between 3.4V and 3.8V without ever terminating charging.

The question was if this is a battery technical problem, not its probability. It's probability approaches 1 as the total installed base and hours naturally increases. Would the specifications be violated?

The solution might be to integrate the time and voltage to cause charging to stop. Perhaps so many W-hrs charged after exceeding 3.4V? Not perfect, but far better?

[T1 Terry]

Originally Posted by daddle

T1 Terry indicated that charging was stopped when the voltage reached 3.8V. That combined with other indications that holding the cell at higher than 3.4V could be harmful, perhaps dangerous, made me point out a possible problem with the automatic charging controls. As described the charger could run for an infinite period between 3.4V and 3.8V without ever terminating charging.

The question was if this is a battery technical problem, not its probability. It's probability approaches 1 as the total installed base and hours naturally increases. Would the specifications be violated?

The solution might be to integrate the time and voltage to cause charging to stop. Perhaps so many W-hrs charged after exceeding 3.4V? Not perfect, but far better?

The solution we use is a little less complicated. We can already buy an off the shelf item that will trigger an alarm signal if a cell reaches 3.6v, the Junsi cell logger. If the alarm sounds and it’s generator charging, turn it off at that point. If it's solar charging or charging from a propulsion motor then the over cell voltage alarm triggers a simple timer circuit linked to a relay in the charging circuit. This would stop the charging for however long the timer circuit was programmed for, say 5 mins, the systems normal house load will burn off the high voltage spike on the over charged cell so after 5 mins the charging starts again, if the same cell goes high again, the process is repeated till the other cells catch up or the charging finally stops.
The aim is to have all cells to reach 3.45v at the same time or close enough to it to make a total of 13.8v so the regulator stops charging, if one cell reaches 3.6v before the total voltage reaches 13.8v then the over voltage alarm stops the charging. This way a solar regulator can keep the battery topped up without a cell going into voltage run away and damaging itself.

T1 Terry

[daddle]

Originally Posted by T1 Terry

The solution we use is a little less complicated...

Okay. That sounds reasonable. It however implies that holding at 3.59V is not harmful?

I'm not being negative. Just critical. Especially so after suffering with not very competent designs for LA charging from the popular manufacturers.

[T1 Terry]

It doesn't really hold any cells at 3.6v, as soon as they reach that voltage the charging stops and the voltage drops due to the load.
As mentioned by another poster, heat is the real enemy, at 3.6v these is 0.2v as over charge, the amps would not be very high or the climb from 3.4v to 3.6v would be very short, possibly around 2 amps per 100Ah of capacity. 2 x 0.2 = 0.4w, for the few minutes it takes to climb from 3.4v to 3.6v x 0.4w isn't going to produce much internal heating. If the 3.6v was maintained for a long period the cell would finally start to heat up, but if the over voltage was higher then the watts would be higher and a problem would develop.
Once you head over the 4v per cell mark different things start to happen inside the cell as well, not only is the electrolyte being boiled and lost as vapour through the vent, the lithium starts to plate on to the anode, restricting access for future ion exchange and reducing the amount of free ions available for transfer between the anode and cathode, a loss of capacity.

T1 Terry