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

[T1 Terry]

Originally Posted by darylat8750

Are there any other opinions whether temperature delta between cells could give warning before permenant damage would be done?

As stated by mcarling, you would need a method of measuring the electrolyte, not the case of the terminals. This gets even harder as the cells are electrolyte starved, it isn't sloshing around like a flooded cell lead acid battery. The terminal temp measurement won't work as it they are just a block of aluminium or copper and the thin foil plates bolt to this block. By the time they transmit the heat through the foil plate and heat the mass of the block to a measurable increase the cell internal temp is already high enough to boil off some of the organic solvents in the electrolyte. The heated plastic case becomes rather soft and simply stretches, by the time a sensor picked up this thermal increase the pressure is there and the plastic is soft. About all the good it would do is to let you know how hot the cell was when it developed a pregnancy bulge.

Why is the swelling a problem? As I mentioned earlier, the cels are electrolyte starved, they evacuate the cell and then allow a controlled quantity of electrolyte to be drawn in and this saturates the material in the plates. Remember it is only the ion transfer within the material on each plate being conducted to the plate the provides the electrical energy. The ion transfer requires a shuttle effect of ion into the electrolyte on one side of the separator to push an ion into the material on the other plate.
If the cell swells the plates can move apart, there is not enough electrolyte to fill the gap from top to bottom of the plate so the reactive surface area is reduced, therefore the capacity is reduced.
The plus side to all of this, push the cell back into shape so the close contact returns and the electrolyte is again spread across a greater area of plate and the capacity returns.
It is still one of the great unkowns if an aged cell that has simply lost electrolyte resulting in lost capacity was to be refilled would the capacity return? A few yrs still to go before we have cells with lost capacity purely due to age and not abuse, even the rather abused ones seem to have recovered a lot of their capacity once they were squeezed back together, the resistance is much higher in them though

T1 Terry

[estarzinger]

Question: have any of the major commercial deep cycle applications (like golf carts and fork lifts) switched to LiFePO4?

[T1 Terry]

Most motorised disability scooters are lithium powered these days to give the performance and range. Golf buggies and forklifts use the battery weight as part of their stability but the walk behind type forklifts are mostly lithium now. There are a lot of other battery powered devices like cherry pickers, elevator platforms, fruit tree pruning vehicles, parcel delivery trolleys, that sort of thing that were once lead acid battery powered are now commonly being converted to lithium batteries to increase their operating range, so I guess golf buggies will be the next most likely to switch.


T1 Terry

[exMaggieDrum]

Originally Posted by estarzinger

Question: have any of the major commercial deep cycle applications (like golf carts and fork lifts) switched to LiFePO4?

See Canadian Electric Vehicles Ltd.

[estarzinger]

Thanks. So the scooters and forklifts that are using li . . . Are they using a "drop in" system/product (cells and bms etc) or home brew systems? If drop in, what is the approach/product of choice?

In lead batteries, they had the solution well solved in "golf cart batteries" (T105's and related) . . . So if they have the Li solution similarity worked out .. . . Why reinvent the wheel?

[transmitterdan]

All Li commercial solutions I have seen utilize complete charge solution custom designed for the specific machine and chosen battery. There is no "generic" Li battery that can be swapped from one machine to another except within a given brand/model.

There is no Li equivalent to the T-105. Li batteries typically outlive the machine. They do not require replacement every few years like LA. If they need replacement they must use same chemistry as the OEM. There is no universal chemistry thus no universal aftermarket battery.

[exMaggieDrum]

Originally Posted by estarzinger

Thanks. So the scooters and forklifts that are using li . . . Are they using a "drop in" system/product (cells and bms etc) or home brew systems? If drop in, what is the approach/product of choice?

In lead batteries, they had the solution well solved in "golf cart batteries" (T105's and related) . . . So if they have the Li solution similarity worked out .. . . Why reinvent the wheel?

I didn't really grill them on their ev systems they sell as all I wanted was to get some Sinoply batteries. Another poster (can't remember who) said he got his there. I've been dealing with Randy. Only way to get cells from them is to preorder and then wait for the next boat. No stock on hand.

I believe they custom build their systems. They sell the cells in 12v units of four with connectors and end plates and lifting straps. I don't think they sell custom units for marine or other end user applications nor do they cell BMS's etc. I would not have known they would provide cells until I saw the post.

Nice guy there, but I may go with CALB batteries that are in stock in the US. I really don't have anything concrete on the quality of CALBs vs. Sinoply other than some discussions on various ev sites that they may have thinner plates and that this was a drawback. They have a new generation of cells out now. I guess all this is pioneering stuff for boaters, and even for DIY auto and home users in that not much good recent info on websites/blogs etc.. I've found the info on CF to be the best but hard to siff through for some specifics.

[mcarling]

Originally Posted by exMaggieDrum

I really don't have anything concrete on the quality of CALBs vs. Sinoply other than some discussions on various ev sites that they may have thinner plates and that this was a drawback.

The advantage of thinner plates is that the cells are better able to accommodate high C charging and discharging (the Li ions don't have to move as far to saturate the full depth of the anode or the cathode and are not inhibited by a saturation layer that might quickly build up on the surface). The disadvantage of thinner plates is that the cost and weight per KWh increase. Thinner plates are great for applications like drag racing, but not good for fractional C applications like a house bank. If one were to make the plates very, very thin, > 60C charging and discharging (full charge or discharge in less than one minute) would be possible and would not damage the cells. Theoretically ideal LiFePO4 cells for a marine house bank would have plates so thick that they could not be used in an electric car. Which leads to ...

The idea that whatever the EV folks are doing with cars, forklifts, golf carts, etc. should be mimicked with boats must be rejected. As illustrated above with anode and cathode thickness, the engineering compromises that make sense for applications in the 1C to 20C charge/discharge rate range generally do not make sense in the 0.1C to 0.5C charge/discharge rate range that is prevalent with marine house banks. Perhaps an even bigger difference is that electric cars, forklifts, golf carts, etc. are never charging and discharging at the same time and their circuitry is positively designed to prevent any attempt to operate them while charging. However, operating from a marine house bank while charging is a perfectly normal operation.

[exMaggieDrum]

mcarling - thanks for your feedback. Your explanation makes sense to this amateur on Li batteries. I found that the max charge rate recommended for the Sinoply's was 2C and it is 3C for the CALBs.

It is becoming a big problem to get suitable cells in the US. No one has told me of any supplier. I have found GBS cells offered but they are too tall for any good spots on my boat. Winston, Sinoply, and CALB cells would fit if I use their 100Ah prismatics but no luck finding a US supplier of any but the CALBs and GBSs.

[mcarling]

I'm now using CALB CAM72 cells. They are prismatic but have an aluminum rather than plastic case. They are smaller and lighter per KWh than the prismatic cells with plastic cases, but slightly more expensive. I suspect the aluminum case may be more resistant than the plastic cases to slight bulging pressures, but certainly would not protect against extreme bulging pressures (which should never occur with a marine house bank that isn't abused). I'm compressing them, but it may not be necessary.

[estarzinger]

Originally Posted by mcarling

The idea that whatever the EV folks are doing with cars, forklifts, golf carts, etc. should be mimicked with boats must be rejected. As illustrated above with anode and cathode thickness, the engineering compromises that make sense for applications in the 1C to 20C charge/discharge rate range generally do not make sense in the 0.1C to 0.5C charge/discharge rate range that is prevalent with marine house banks. Perhaps an even bigger difference is that electric cars, forklifts, golf carts, etc. are never charging and discharging at the same time and their circuitry is positively designed to prevent any attempt to operate them while charging. However, operating from a marine house bank while charging is a perfectly normal operation.

Despite your "theory" about the differences, in actual practice, the T105 battery, the classic "golf cart" battery, was also a very excellent house battery. So if there is an li replacement for the T105, it "might" equally make a good house solution (yea, it also might not, but I at least would consider it worth taking a look at). No-one is going to design a cell just specifically for cruising boat house use . . . So we are in fact going to have to use a cell originally designed primarily for some other application.

[mcarling]

Originally Posted by estarzinger

Despite your "theory" about the differences, in actual practice, the T105 battery, the classic "golf cart" battery, was also a very excellent house battery. So if there is an li replacement for the T105, it "might" equally make a good house solution (yea, it also might not, but I at least would consider it worth taking a look at).

I do not share the opinion that golf cart batteries were ever "very excellent" or even ordinarily excellent house batteries. In my opinion, golf cart batteries were simply better (for the price) than the other crappy alternatives until LiFePO4 batteries became affordable about five years ago. Still crap compared to LiFePO4, but better than most of the other crap. Again, that's my opinion. You're welcome to disagree.

Originally Posted by estarzinger

No-one is going to design a cell just specifically for cruising boat house use . . . So we are in fact going to have to use a cell originally designed primarily for some other application.

All the LiFePO4 cells that are commercially available work fine as marine house batteries. They are just slightly more expensive and heavy (by perhaps about 10%) compared to a theoretical ideal marine house bank (which as you correctly point out will never be commercially available).

[estarzinger]

Originally Posted by mcarling

I do not share the opinion that golf cart batteries were ever "very excellent" or even ordinarily excellent house batteries. In my opinion, golf cart batteries were simply better (for the price) than the other crappy alternatives until LiFePO4 batteries became affordable about five years ago. Still crap compared to LiFePO4, but better than most of the other crap. Again, that's my opinion. You're welcome to disagree.

This is not worth debating, just a mater of semantics. It is a simple fact that the "golf cart battery" T105's were the battery of choice for many of us. And I was interested in what they were doing with Li.

All the LiFePO4 cells that are commercially available work fine as marine house batteries. They are just slightly more expensive and heavy (by perhaps about 10%) compared to a theoretical ideal marine house bank (which as you correctly point out will never be commercially available).

So, I am puzzled then, please clarify your comment above that we should not look to the golf cart Li solution because its application is so different than ours. You are saying it is some subset of the electronics that are not applicable? And are you really meaning to say that ALL commercial Li cells are near equally good as house bank cell?or might some cells perhaps be noticsble better for our application that some other cells and might those better cells also be used in deep cycle commercial applications (like golf carts)?

......

[mcarling]

Yes, I am saying that all the commercially available LiFePO4 cells of appropriate capacity from reputable brands are suitable for use in marine house banks. Probably the no name cells at Alibaba are fine too. The house bank application is extremely low-stress for LiFePO4 cells. The various advantages and disadvantages among them are minor in fractional C applications. In my opinion, size, price, and availability are the main factors to consider.

Then balance them properly before installation, don't use a charging voltage above 3.40V to 3.45V per cell, and don't drain them below 2.8V to 2.9V per cell. If leaving them unused for a long time (longer than a few weeks), 30-50% charge is the best state.

How one should go about ensuring that LiFePO4 cells are not overcharged or over discharged has consumed many, many pages in this thread without consensus. MPPT controllers can simply be set to always output 3.40V per cell and you can leave them connected indefinitely. I also set shore power chargers to 3.40V per cell. Alternators and wind generators may be set to 3.45V per cell but then should be switched off when current flow tapers off. They are trickier because suddenly disconnecting them from everything is bad and leaving them connected to the load bus with the battery disconnected is also bad.

In general, if set to 3.40V per cell, charging sources can be left on but, if set to 3.45V per cell should be switched off once current flow tapers off. Before anyone starts screaming, I'm not recommending leaving constant charging sources on indefinitely. Even with intermittent charging sources like MPPT controllers, think about whether or not they need to be left on. In general, it's better to switch off loads that would drain the batteries, but obviously bilge pumps, alarms, security cameras, etc. probably should be left on.

LiFePO4 cells are not rocket science and, used as marine house banks, do not demand complexity. However, they are not drop in replacement for lead-acid batteries. The charging regimes are substantially different. The appropriate charging regime for lead-acid will kill LiFePO4 cells and vice versa.

[GoingWalkabout]

Perhaps the home solar battery market is working out the drop in place solar battery solution. Here is a write up of a German company just entering the USA market. Looks like overkill for a sailboat but could fit on a large cat. Thoughts?

Unlike Tesla, however, sonnen has a significant track record with residential storage. The company is the leading supplier in Europe, where it says it has installed more than 10,000 systems since 2011. Part of this is because the German residential electricity market provides rate incentives for home storage that are much less common in the U.S., though some markets*such as California are shifting in that direction.

Sonnen is introducing what it calls the sonnenBatterie eco, a battery unit about the size and dimensions of a small refrigerator (Figure 1). The eco is available in capacities of 4 kWh up to 16 kWh, and has an integrated smart inverter and control system allowing it to function as a home microgrid. Within that range, the eco units can be sized in 2-kWh steps.

Housebroken.*Sonnen’s eco home battery is designed for the U.S. market, and the company is positioning it for use in tandem with a solar photovoltaic system. Source:*POWER/Tom Overton

Sonnen says they are guaranteed for 10,000 cycles or 10 years of use. The batteries are priced at $10,000 for the 4 kWh model up to $24,000 for the 16 kWh model. That’s more than Tesla’s PowerWall home battery, but Greg Smith, sonnen’s senior technical trainer, sought to contrast the eco against the PowerWall by saying the eco is capable of deeper cycling and more total cycles over its lifetime.

A larger Pro version designed for commercial use is available in capacities of 18 kW/24 kWh up to 90 kW/240 kWh, and is intended mainly for peak shaving.