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

[KASHMIR]

Originally Posted by downunder

There are at least 4 australian providers of engineered Lithium systems for boat applications.

I would be interested in any feedback by and of the experts here on CF on the relative merits of these systems.

Lightwave have now 5 such installs using the Outback Marine system. According to the independent electrician who does the install for Lightwave there have been no issues whatsoever.

1. Outback Marine Energy Storage Lithium Battery Systems

2. Enerdrive The Lithium Battery - The Ultimate Energy Storage System

3. EV power EV Power - Australian Electric Vehicle Specialists — EV Power - Australian Electric Vehicles sells Electric Bikes and car conversions, Electric Bicycles, Electric Vehicles, Conversion Kits. Sail/surf/roam has one of these systems and also IMPI has just installed an 800AH system

4. EVO Lithium https://www.evolithium.com.au/

Cheers

There is also us and we have our own system in use on our own 12m yacht, it is fantastic. Located in Melbourne and shipping Aust wide.

[OceanSeaSpray]

Originally Posted by Maine Sail

On the good side the BMS saved the BMS from itself and also saved my cell.. If I had not yet fully convinced myself that balancing is unnecessary, at the charge voltages I use, I am now... I plan to talk to the designer and see if the BMS's other parameters will be impacted if the resistors are physically removed from the cell boards. I personally have zero need for automated balancing as evidenced by the previous 861 cycles.

If a cell board shorts out and you don't catch it in time, it will eventually kill the cell no matter what. It is beyond the control of the BMS. This being said, I would refrain from generalising from a single data point coming from a cheap and nasty cell board. Simple things can be done to hugely increase reliability from an electronic design point of view.

The shunting resistor is in series with a MOSFET acting as a switch (which likely failed on your board). Removing the resistor has the same effect as operating the board below the shunting voltage with the transistor off. Since your shunts never engage anyway, you can just as well do that. Otherwise you could use a more robust (and expensive) transistor.

Incidentally, the more powerful the shunt balancer, the quicker it will destroy the cell if it shorts out.

Originally Posted by Maine Sail

The hard part is now to re-balance the pack on the boat on a mooring. More just tedious than difficult though...

I have adjusted pre-existing poorly balanced banks "in service" simply by shorting the high cells with a few metres of 10-12AWG wire while they were charging towards 100% at a modest current. It saves dismantling everything, but the result isn't quite the lab-grade baseline you might want. Still within a few millivolts at the end with a bit of care.

Originally Posted by Maine Sail

A new feature on my dream BMS is a manual lock-out for cell balancing so this type of failure mode can't occur..

It is a thought. The question is always how much are you really prepared to pay to go from a perfectly functional design to another equally perfectly functional design that is superior in some very unlikely circumstance?
Add a small switch in series with the balancing resistor and it will do that - as well as require "human management". Over time, the contacts might oxidise a little and result in "selective" balancing when you want it.

The most likely failure is from the switching MOSFET. Use a more expensive, higher-rated one. Not good enough? It can be doubled-up and then both would need to short out before the cell gets discharged uncontrollably. Much better. Still not happy? Then also double up the control circuitry for each transistor as well, etc. It just means more and more components and larger PCBs for the same functionality.

This thread has often suggested that people are happy to splash out thousands on lithium cells, chargers and gadgets while cringing bitterly about every dollar related to protection: that is the cell board market segment.
My feeling is that this might gradually evolve as experience and maturity keep increasing. We were wondering whether LFP cells could last 10 years on boats. It now looks like they might well last anywhere between 10 and 20 years in (some) house bank applications. What this really means is that the BMS must also last the distance.

[downunder]

Originally Posted by KASHMIR

There is also us and we have our own system in use on our own 12m yacht, it is fantastic. Located in Melbourne and shipping Aust wide.

Presumably this is the website. Have you designed your BMS? EV power I note.

Lithium power

[KASHMIR]

Originally Posted by downunder

Presumably this is the website. Have you designed your BMS?

Lithium power

Downunder, yes thats us. We are in the middle of trials of our own BMS. Currently we use another Australian designed and manufactured unit that has proven to be very reliable. Our own unit is still some months away as we are only just starting trials.

[john61ct]

Interesting, Chevy VOLT "LiMn2O4 + LiNiO2" 100AH modules, made in the US by LG Chemical, also the same cells used in Nissan, Ford and Renault.

Originally 24v, rewired for 12v output

but only 3S? presumably requires the proprietary BMS?

http://www.ebay.com/itm/182636171979

[OceanSeaSpray]

Originally Posted by Maine Sail

This can't physically happen with our system. HVC is purposely programmed before the voltage where cell boards begin shunting. I set it up this way so balancing could be a "manual & attended" event like equalizing a lead acid battery. I quickly discovered however that shunt balancing keeps cells at a high voltage for far longer than a manual parallel balance does so decided that I would not use shunt balancing.

I actually ran a head to head test where I purposely unbalanced a cell then let shunt balancing have a go at it. After more than 20 hours at 3.55V + the cells were still not in balance. I then repeated the same imbalance with the same cells only this time wired them in parallel and pushed them until they were no longer taking any current. Parallel was multiples less duration spent at a high voltage than shunt balancing required.

The whole discussion about shunting or not shunting is not that relevant to your case for the simple reason that you have never used the shunt balancers. You can just as well have none!

The cells will rebalance if you keep charging at low current until all of your charging current is running through the shunts. Think about it this way: the shunts then regulate each cell voltage at whatever value they operate at, maybe 3.6V, so it is exactly equivalent to charging in parallel at a common voltage.
The problem may well have been that, by limiting the voltage to "3.55V/cell", you actually prevented the shunts from passing current and... nothing happened of course.
And if you charge with more current than what the shunts can bypass, they will all shunt full current and no balancing will occur either, at least at first!

That business of shunting as the voltage comes up and the charge finishes is a flawed strategy. To make it work, your maximum charging voltage must be carefully set to match the operation of the shunts... it is usually higher than what you would care having on a regular basis AND you must charge all the way to the top every time. As I wrote earlier, voltage-based shunting cell board schemes don't work for the kind of application we have here. It is for EVs and RC models.
Balancing properly requires a central processor to manage it and it needs to know the battery current as well as all the cell voltages.

Originally Posted by Maine Sail

You talk a good game so why not put your money where your mouth is and we can all move forward... With Dimitri off at Lithionics & out of the game it seems you'd be rolling in the dough....

Haha... time has been more of an issue than money so far, arguably because there is no one to pay me when I work on the BMS! It is well on its way and will get there however.
You are welcome to have one once I am happy with the hardware. I can't think of too many better places to send one actually.

[exMaggieDrum]

MaineSail, what is the shunt voltage for your HousePower cell boards?

And how does one "take out or disconnect" the shunt resistors specifically on these boards if one wants to?

Thanks.

I have to admit I am starting to get a bit leary of using the HousePower BMS and cell boards - no company/designer/producer to back them up, no source to replace any failed parts, and some reported issues with them.

And to replace a one-man operation with another one-man operation does not give me a lot of comfort. I am going to look in to the other BMS providers, like Outback. I did not know they were in the biz now.

[exMaggieDrum]

MaineSail, what is the shunt voltage for your HousePower cell boards?

And how does one "take out or disconnect" the shunt resistors specifically on these boards if one wants to?

Thanks.

I have to admit I am starting to get a bit leary of using the HousePower BMS and cell boards - no company/designer/producer to back them up, no source to replace any failed parts, and some reported issues with them.

And to replace a one-man operation with another one-man operation does not give me a lot of comfort.

I just looked at the Outback Marine website and sent them a query for more information. They do not provide any technical detail and their systems are fairly expensive and would require a significant rewire of any existing system if you use their DC distribution control boards (large bus and Blue Sea remote switches). The do have a BMS monitor/display but they do not provide any detail at all to allow you to see if it could be used as a replacement for standalone BMS's like the HousePower one.

[evm1024]

A quick answer for you while we wait for MainSail to come back online.

You just un-solder the shunt resistors (there are 2 on my cell boards). The shunts are the 2 large black rectangles by the "red wire".

Regards

[senormechanico]

I must have an earlier version of the Houspower BMS on my boat.
The shunt resistors are rectangular 5 watt resistors, white in color.
I am not on the boat right now so don't know the resistance, but it's quite low like 5 ohms or so.

The ones in my home's grid tie system have two black rectangles side by side as evm1024 stated.

[exMaggieDrum]

Thanks evm and senormechanico!

[exMaggieDrum]

This is based on a quick review:

Waiting for info on the Outback gear. I don't think it will work for me though and is very expensive.

The "rec" gear is made in Slovenia and only sold in Europe at the moment. It doesn't look like a good fit either.

The Orion literature isn't very complete but is for sure requires heavy PC integration and I don't want to rely on a PC to operate it. May be I misread it but don't think so. It is designed very specifically for the EV market, as are the others too but even more so.

The Manzanita looks the most promising so I am going to get in to a deeper review of it. Only $200 for the most likely unit. Also for EV but looks more suitable for marine than all but the Outback.

It would be great to hear from anyone who has experience with these, especially (for me) the Manzanita.

[Colin A]

Interesting. News of a Lithium fire and a tech bulletin from yanmar on lithium batts.

Yanmar Provides LION Battery Bulletin - Ed's Boat Tips | Ed's Boat Tips

[john61ct]

This is not an inherent LFP issue, but one of overall system design and specifically limitations of their alt/VR design.

Lots of what they say is spot on, but the main "solution" of the BMS limiting the amps demanded seems odd to me.

The wiring infrastructure should of course be beefed up to handle the high current rates safely, and stress-testing looking for hot spots is a good idea.

IMO with LFP, all charge sources should have the ability built-in to limit current output, both user selected static setpoints and automatic scaling based on high temperature.

wrt an alternator, the MC-614 will handle this, and seems to me a better solution.

Please let me know if I'm off base here.

[Paul L]

Originally Posted by Colin A

Interesting. News of a Lithium fire and a tech bulletin from yanmar on lithium batts.

Yanmar Provides LION Battery Bulletin - Ed's Boat Tips | Ed's Boat Tips

So is this the first documented LiFePo fire causing a boat loss?