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

[Lagoon4us]

Looking good

[catakate]

So far we have two test runs that proof these easily provide more then are rated. One would normally expect less then is promised, how come they promise so little?
Good to know about something that is better then true, though

[T1 Terry]

Originally Posted by FlyingCloud1937

Hi Main...

That begs the question? I see 3 dissimilar metals used to make the jumper connections. It's a number 1 sin to use copper and aluminum connection in the marine market, as the alum... will quick corrode.

Also what concern have you with with the big alum... nut as it expands and contracts at a different rate to the copper.

The Li-Ion bat uses copper foil for the negitive plates with the foil ends clamped internally to the alum... post, and the big alum... nut screwed with case between.

Now we also have copper connected to alum... in an electrolyte inside the battery. What about the corrosion issue, will it over time develop a high resistance? Will each cell develop equal resistance? How will that effect cell balance. And over what period of time?

Lloyd

Hi Lloyd,
The dissimilar is an issue, add current flowing in each direction and it becomes a serious issue. The blocks that make up the the terminal posts are copper for the negative terminal and aluminium for the positive terminal, this matches the foils used to carry the active materials in each cell packet so there usually isn't an internal resistance problem, but certain abuse conditions will trigger a problem. Over current and over voltage are the offenders that will cause the most common problems but dragging a cell to zero volts and holding it there will cause the same problem. Internally generated heat is the killer for these cells, any one of the 3 I just mentioned will generate internal heat. When either copper or aluminium is heated beyond a certain point oxides will form as it cools, these oxides create a corrosive layer in spot you can't clean, erratic cell behaviour is the end result.
Externally, we have a copper link bolted with a stainless fastener in a copper block at one terminal and an aluminium block at the other, they can not be simply bolted on without problems occurring.
A long way back in this thread I recommended that the cell terminals and the copper links should lightly sanded to produce a shiny finish, then coated with Alminox or a similar copper/aluminium corrosion preventative and bolted together, this process needs to be used each time the links are removed and refitted, at least a recoat with corrosion inhibitor.
Do this and you will have very little trouble, skip this step and you will see cells going out of balance with weird patterns forming, not always the same cell goes high or low and often the same cell will be both high and low when charging or discharging, a sure sign you have a high resistance oxide coating developed some where, hopefully on the outside link so you can clean it.
Another tip, if you want long cycle life stay away from cell voltages below 3v and above 3.6v, the upper one is the most important, anything pushed into the cell above 3.4v can be quickly determined as heating in watts, 3.6v - 3.4v = 0.2v, even at 100 amps that's only 20w of heating, but 3.9v -3.4v = 0.5v x 100amps is now a 50w heater inside a sealed plastic case, this will not have a happy ending.
Use a cell monitor, set the alarms for 2.8v and 3.6v and you can virtually do what ever you like with these cells, but best to stay under the 3C limits, 0.5C will see the pack cruising easy.
My 720Ah @12v pack is approaching 2 yrs 24/7 service, every night sees 35% of the capacity drained to run my house stuff, on no solar days another 40% but that only happens if I disconnect the solar, even in the rain some charge current comes in. I can get 3 night and 2 days before a cell hits 2.8v under load, I had to use an alternate supplementary charge maybe 4 times in those 2 yrs, their discharge capabilities are match by their charging capabilities, the voltage will sit steady until they approach fully charged, maybe the last 2% of capacity is when the sharp voltage rise occurs.
Hope that helps some a bit with understanding some of the do's and don't with these cell, what they can do is incredible when compared to lead acid batteries, the lack of requiring constant attention is equally as impressive.

T1 Terry

[deckofficer]

Originally Posted by catakate

So far we have two test runs that proof these easily provide more then are rated. One would normally expect less then is promised, how come they promise so little?
Good to know about something that is better then true, though

Mine should have returned more than rated because I was only pulling 3.7 amps.

Maine Sail's test was the one that proved how good these cells are, pulling a bit over 100 amps for 4 1/4 hours and still didn't get to the knee.

From this point on, I'm using no other battery than LiFePO4 for my scooter, kayak, electric car and truck, and house/inverter bank. It is a no brainer.

[T1 Terry]

Originally Posted by catakate

So far we have two test runs that proof these easily provide more then are rated. One would normally expect less then is promised, how come they promise so little?
Good to know about something that is better then true, though

the quote capacity is at a 1C discharge rate, if you drag it out slower there is more capacity to be had, just like any other battery, jst their charge rates and discharge rates are much higher before Puekert's factor becomes involved.
A chart from Winston Battery Company

T1 Terry

[downunder]

Originally Posted by deckofficer

Mine should have returned more than rated because I was only pulling 3.7 amps.

Maine Sail's test was the one that proved how good these cells are, pulling a bit over 100 amps for 4 1/4 hours and still didn't get to the knee.

From this point on, I'm using no other battery than LiFePO4 for my scooter, kayak, electric car and truck, and house/inverter bank. It is a no brainer.

Bob,

you will be interested in this link.

Battery Power « Current Sunshine

[deckofficer]

Originally Posted by downunder

Bob,

you will be interested in this link.

Battery Power « Current Sunshine

I read it a couple of months ago. Nice fast tri with a good pedigree at a good price. The electric propulsion is just the icing on the cake.

[catakate]

Originally Posted by deckofficer

Maine Sail's test was the one that proved how good these cells are, pulling a bit over 100 amps for 4 1/4 hours and still didn't get to the knee.

From this point on, I'm using no other battery than LiFePO4 for my scooter, kayak, electric car and truck, and house/inverter bank. It is a no brainer.

Besides looking as the beginning of a beautiful friendship, this looks like the beginning of the end for the combustion engines as well. And is greener.

Thanks everyone for sharing!

[deckofficer]

Originally Posted by catakate

Besides looking as the beginning of a beautiful friendship, this looks like the beginning of the end for the combustion engines as well. And is greener.

Thanks everyone for sharing!

My dad is 93 and was around when electrics were more popular than gas. He enjoys my electric Smart Car, as we come full circle during his lifetime.

[hellosailor]

Bob-
Like Terry said, the "rating" is based on a heavy current draw. You haven't exceeded the rating, because you used a VERY light draw, which will indeed provide more amp hours. This is why "everyone" now says to use one big bank instead of alternating between two smaller banks. The one big bank effectively means you are drawing at 1/2 the rate, whatever that may be, compared to twin banks. So you draw at the lower rate and get better power cycles from the battery.

39 hours for a charge...We're gonna start calling you "Patient Bob".

IIRC there are also losses in charging, and charging at too slow a rate increases your losses compared to a more normal (i.e. 10 hour) charge.

[deckofficer]

That was 39 hours for a discharge. But I'll earn that name on the very slow parallel charge, 3.5 amps into a 3.2 volt 400 amp battery. Something over 100 hours.

[diugo]

For those with the inclination and the patience (Bob?), the Peukert exponent n is easily calculated:

n = (log(t2) - log(t1)) / (log(I1) - log(I2))

where I1 is a high current, I2 is a much lower current, and t1 and t2 are how long it takes to discharge a fully charged battery down to the same low voltage, at I1 and I2, respectively.

I suspect n will vary widely with the capacity and age of the cell and across different manufacturers. But it might be fun to get a real handle on it, rather than simply assuming it to be 1.0 (which it obviously is not) or that it only applies at high currents (which it does not).

[FlyingCloud1937]

Originally Posted by T1 Terry

Hi Lloyd,
The dissimilar is an issue, add current flowing in each direction and it becomes a serious issue. The blocks that make up the the terminal posts are copper for the negative terminal and aluminum for the positive terminal, this matches the foils used to carry the active materials in each cell packet so there usually isn't an internal resistance problem, but certain abuse conditions will trigger a problem. Over current and over voltage are the offenders that will cause the most common problems but dragging a cell to zero volts and holding it there will cause the same problem. Internally generated heat is the killer for these cells, any one of the 3 I just mentioned will generate internal heat. When either copper or aluminum is heated beyond a certain point oxides will form as it cools, these oxides create a corrosive layer in spot you can't clean, erratic cell behavior is the end result.
Externally, we have a copper link bolted with a stainless fastener in a copper block at one terminal and an aluminum block at the other, they can not be simply bolted on without problems occurring.
T1 Terry

Terry.

Now that is the meat of the bone I've been looking for.

So what I can infer from the discussion and what I know...Huston We Have A Problem.

In this photo which is not a Winston Cell, it clearly shows the neg. terminal with copper foil being clamped to an Alum...Post inside the battery.
Annotating what is a dissolved clamp. Which we can assume is a high resistance connection to the charge/discharge bus.



In this picture we see a copper based clamp of the foils to what appears to be an Aluminum Post of the negative which says it's a Thunder/Winston LFP, which also shows buldging



Now in this image another bulging Winston which appears to have Alum... post as well as bulging.




To this pic which shows recently received Bolqon/Winston Cells from a fellow form member. Which clearly appears to have a copper neg. post with what appears to an alum..nut



Now this next pic shows another fellow form member balancing his new Bolqon/Winston cells. Using just about any means to connect them. Including SS, copper, and alum... all measured by some LED array that is suppose to indicate cell voltage based on the color of the LED??

Trying to measure mill-ohm voltage differences.



Now we have Mainsail's excellent test bench, which I can't see if his post internal are dissimilar metals, but certainly the pos/neg connections include copper on alum... Brass on alum...



So it begs the question what is a Marinized LFP bat. cell?

Does it make sense to include a temp measurement/controller/alarm at each and every bat post as part of the BMS?

Lloyd

[senormechanico]

Originally Posted by catakate

So far we have two test runs that proof these easily provide more then are rated. One would normally expect less then is promised, how come they promise so little?
Good to know about something that is better then true, though

Maybe we're seeing a repeat of the dawn of fiberglass boats being heavily built because nobody knew how long they would last.
Wouldn't that be great?

[rocksnsand]

PO is for phosphate and I believe in the above case ferric phosphate (en.wikipedia.org/wiki/Lithium_iron_phosphate_battery)