Lithium Batteries (for the rest of us)

[ebaugh]

Originally Posted by nimblemotors

might be better to invest in general fire control system than fuses..

A fire control system puts out a fire. That's a little too late. I'd rather prevent one.

Still, I'm not proposing this as necessary, but it is at least an interesting mental exercise for consideration.

I improved the circuit and have a way to monitor the condition of the fuses. See attached. It would take 2 more AA batteries and 2 more analog inputs on the BMS, in addition to the one shown for the other 2 parallel strings. Plus you would need even more very small fuses for the sense lead, also not shown. Or maybe just use fusible link wire.

[alanapar]

Originally Posted by ebaugh

Is this the fusing arrangement you suggested? On first blush, I like it. It essentially duplicates what Genasun does with a single serial fuse in their builds. And duplicates the parallel fusing at the cell level instead of the bank level like Mastervolt recommends.

Indeed that is what I was suggesting. Sorry if the original post was not quite clear.

As I pointed out, in this design, when certain of the batteries short, the fault current will be shared by multiple fuses, and several fuses might have to open to clear the fault.

Also, when using fuses for paralleling, you need to be aware that their resistance can be significant. A 10A fuse might have resistance of 10mOhm or greater. This is significantly larger I expect that the internal resistance of the individual cells. So you will lose to some extent the benefits of paralleling. It would be rather complicated to analyze this design as to the effects!

[ebaugh]

Originally Posted by alanapar

Indeed that is what I was suggesting. Sorry if the original post was not quite clear.

As I pointed out, in this design, when certain of the batteries short, the fault current will be shared by multiple fuses, and several fuses might have to open to clear the fault.

Also, when using fuses for paralleling, you need to be aware that their resistance can be significant. A 10A fuse might have resistance of 10mOhm or greater. This is significantly larger I expect that the internal resistance of the individual cells. So you will lose to some extent the benefits of paralleling. It would be rather complicated to analyze this design as to the effects!

Thanks for the suggestion! Glad I got it right, so far it looks like the easiest to implement since 1/2 of it is already there. I've got another couple of months to consider it as we work our way back to Florida.

The parallel fuses don't probably hurt anything, there should be only small amounts of current flowing there. Is there an energy efficiency penalty moving 15A through a 30A fuse? Can that be calculated? The voltage drop alone is not a huge issue unless it translates to heat, and wasted energy.

Some one smarter than me calculated a rough internal resistance on a measured .12V rise when switching from 20A discharge to 150A charge at 13V using the 170A difference. I forget the value, I was never sure exactly what it meant from a practical perspective.

[sgtPluck]

I'm wondering if any of the DIY implementers here would care to comment on Smart Battery's Lithium batteries:

Smart Battery FAQ's

They appear to suggest that they are a drop in replacement (due to built-in balancing and protection systems I suppose.) but the FAQ doesn't seem to address all the issues raised in this thread above, such as large draw on the alternator. What would you suggest would remain to be done to make sure the whole system was set up correctly outside of these batteries?


They are very expensive. But I would love to get lithium batteries. Space and capacity are big concerns for me.

[hellosailor]

I feel that the "large draw on alternator" is a myth. Alternators pretty much always are built so that they will cut back on their output as they heat up. So if the battery draws "too much" the alternator heats up faster and then reduces power output sooner. Net result? No harm to the alternator, it knows how to protect itself.

Now, if you've put in an external regulator, and you've removed the thermal protection from the loop (by cheaping out and not installing a temperature sensor for the alternator) that might be something else, but that would apply to any alternator and any battery, unless the alternator was happily and coincidentally oversized.

If you're going to spent $1000+++ for a Group 24/27 size battery, surely, you can afford the best of external regulators, larger alternators, and temperature sensors to match. <G>

At that price they should at least throw in a leather jacket with a big logo on it, to let everyone else know how much money someone spent on a battery.

[ebaugh]

I think considering something like the Smart Battery, on an economic scale depends on the capacity required to meet your needs. I wanted 1200 Ah, IMO that's way too much. For small banks, it may be a reasonable consideration.

There is good evidence showing no need to balance often. In fact, until you have essentially a cell failure, I'm not sure there is going to be any reason for that beyond installation.

There is a case to make sure all cells remain between say 2.5V and 3.6V give or take a little. Exceeding limits will cause reduced life to cell failure. However, many modern charging systems can provide first level protection for overcharge based on bank voltage. But unless you have a big solar and/or wind array that always exceeds your usage, under discharge can still represent a problem.

The level of sophistication must reflect:

1) Your ability and willingness to keep an eye on things
2) The size (or cost) of the bank, if you have a 1000+ Ah bank, it's different perhaps that a 400Ah one.

Outside of Balqon, the market cost of cells is about $5 per Ah at 12V, plus the cost of battery management needed, which is specific to you, your vessels power requirements and charging systems, and your cruising plans.

[pbiJim]

Originally Posted by skipmac

Well same here. Methinks the LiFePO batteriies are just not quite ready for prime time. Long ago and far away I was an EE and even with that background the care and feeding of LiFePO batteries seems like a lot of hassle and lot's of potential for screw-ups and trashing a big dollar investment.

I'm thinking pretty much the same thing.

There is a lot of research going on right now & I expect substantial advances in the next few years. I like the idea of the fast charge rate, but I think that I'm going to sit on the sidelines for a few more years & see what else comes down the pike before I dump big bucks into this type of battery.

On the flip side, I've seen some little AA cells that work off this same technology. They put out about twice the voltage of a normal AA, so you need to use half as many of them & then fill in the blank spots with dummy cells, but the fast charge time is a big plus. The flat discharge voltage is icing on the cake.

[T1 Terry]

If you have a specially built alternator and a regulator that temp senses and cuts back the current if it gets too hot, then there will not be an issue. If you are just using an automitive alternator, the bigger the capacity the greater the problem. a 40 amp alternator isn't going to build up that much heat, but a 200 amp one certainly will. In my Mazda motorhome the poor alternator copped such a beatting it first destroyed multiple fanbelts due to drive issues, yet it had never had this problem with the AGM batteries, then the alternator simply stopped charging any more. My fix was to drive an inverter from the start battery once the engine was running, then power a battery charger, that limited the alternator ouput and stopped the over heating and drive belt issues.

T1 Terry

[Lagoon4us]

On the 23/04/2013 i activated our 400 amp LiFeYPo4 bank of Winston cells and that bank has operated day and night since. The following facts apply.

Usage:-
1) We live on board for 8 months at a time and am currently cruising the Greek Islands, the installation is our 'house batteries' the start batteries are Lead acids.
2) On average we deplete 25% of our bank during the night.

Charging:-
1) We have 750 watts of solar at 36volts. Solar is ideal as it rests the batteries at night i.e. 'cycles' them.
2) Our Solar Controller is a GSL MPPT60L designed for Lithium. It is set at a 'long life' charge rate at 14.30 ( 3.57 per cell) charge and a 'maintenance charge' of 14.10 volts (3.52 per cell)
3) Our port side main engines alternator charges it's own LA start battery but i have a switch i employ to boost the Lithiums from it should i desire. There have been no modifications to the alternators regulator, it and the alternator, are all OEM.
4) We have a 'Dolphin' 80 amp charger powered by shore or 6Kva Genset set on 'Position 08' which effectively is a flat 13.6 volt charge i.e. 3.40 volts per cell.

The cells:-
Winstons brand new never used 5 weeks old when i purchased direct from China, they were supplied precisely charged ready to use NO BALANCING required from the factory.
The batteries are consecutively numbered ie built on the same date/same production run.

Other things of interest.
I have an individual 6 volt digital read out in the saloon with 4 switches that allows me to check each voltage whenever i wish. I did this because Cell loggers readouts are minuscule.

I have the cell logger monitoring voltages high and low.

I have two GWL 80 amp relay switches that allow me to cut power both to and from the bank.

So nearly 4 months and this pack is running sweet, voltages 20 minutes ago were:-
Cell (1) 3.38.
(2) 3.37.
(3) 3.38.
(4) 3.37.

FANTASTIC is all i can say, no more sulphation smells, gas build up etc etc....

Easy peasy.
Cheers Frank

[Zanshin]

Lagon4us - do you have any individual cell balancing installed at all? I'm putting together the hardware and programming a microcontroller for a Winston bank aboard my boat but am having trouble figuring out how much balancing (in terms of Watts or Amps) I really need to have. How do you determine when to stop charging, when the highest voltage cell reaches the cutoff point? Do you have a low-voltage cutout?

[Lagoon4us]

No balancing at all we only have four cells as against some installations that have many cells. ie the stand alone 12 volt Lithium batterys are 'STAND ALONE' they have no BMS.

The solar controller simply controls throttling back as required as does the 'Dolphin' charger. So i don't determine when to stop it's regulated. Now we also have a BEP Battery monitor that is very effective at telling me the state of charge, voltage etc.

When i charge the cells with the alternator they are charging a lead acid at the same time, i simply monitor it by looking at the 'state of charge' on my BEP Monitor and flicking a switch to cut the charge to the Lithiums, any excess spike is absorbed by the lead acid which then deals with the regulator on the alternator.

In all cases in four months of constant use the cells have always charged equally at least within a few thousandths of a volt. Once load goes on these minutely 'imbalanced' cells they quickly align to precisely the same volts.

The cell logger alarms on both high or low individual voltages, in addition i set the alarm on the BEP Monitor as well.

I could and probably will eventually wire the cell logger to the relays but not yet.

I've built this system taking info from posts from 3 or 4 of the guy's on the 'house bank' thread.

The critical thing in my mind is that the cells are unused, not mistreated preferably brand new consecutive builds. I was cruising Croatia when i bought them from China so it's not difficult. Seems a lot of American stock is second hand clearance crap with all the mis-treatment thrown in.

Cheers works for us! Frank

[ebaugh]

Originally Posted by Zanshin

Lagon4us - do you have any individual cell balancing installed at all? I'm putting together the hardware and programming a microcontroller for a Winston bank aboard my boat but am having trouble figuring out how much balancing (in terms of Watts or Amps) I really need to have. How do you determine when to stop charging, when the highest voltage cell reaches the cutoff point? Do you have a low-voltage cutout?

I've programmed a PLC that monitors voltage and has a balancing routine that can shed about 10A per cell. Its a .33 ohm 50 W resistor used at a maximum 66% duty cycle at the moment. The balancing is only needed at installation, for maintenance when you add or remove cells, and I thought maybe periodically say once a year. But I am beginning to believe that short of a problem, once initially installed, there will be no periodic need.

I'd suggest you "stop charging" when the highest cell reaches 3.45V. I would remove all loads when the lowest cell reaches 2.8-3.0V. If you need the micro controller as a safety to prevent overvoltage, set that at about 3.6V.

What may be difficult, if the micro controller has to manage it, is once "charged" to 3.45, or full charge, then you must reduce voltage to below 3.4, I suggest 3.3-3.35 to "float".

[Lagoon4us]

Upon delivery the 4 cells were resting on around 3.365 volts and i made the connections for the bank letting them rest connected for 48 hours, i then checked the voltages = 'still the same. I then introduced load and charge as per our house demands as indicated in 159.
Upper and lower balancing is only recommended by the manufacturer to align 'damaged' batteries.
I believe Bob will be correct regarding periodic balancing and more so for 4 cell installations.

[pbiJim]

Originally Posted by Lagoon4us

Now we also have a BEP Battery monitor that is very effective at telling me the state of charge, voltage etc.

Does that particular monitor determine state of charge by reading voltage? Or is it some high tech gizmo that measures current flow over time & does a calculation?

[Maine Sail]

Originally Posted by Zanshin

Lagon4us - do you have any individual cell balancing installed at all? I'm putting together the hardware and programming a microcontroller for a Winston bank aboard my boat but am having trouble figuring out how much balancing (in terms of Watts or Amps) I really need to have. How do you determine when to stop charging, when the highest voltage cell reaches the cutoff point? Do you have a low-voltage cutout?

There are a number of commercially available BMS products out there. On our bank I am using a Clean Power Auto House BMS and the cell modules. The alt & solar cut can be made from simple Cole Herse DPDT relays and the load cut from a Tyco EV500 contactor. We also have a manual alt cut switch that disconnects the B+ of the MC-614 regulator and a manual on/off for solar. I also wired in a Cell Log 8 to monitor cell level voltage.

I would recommend charging the Winston cells to no more than 14V then dropping back to 13.8V then to 13.5-13.6... Programming the reg for your use with the bank will take the most time...

Be 100% sure you have all battery temp sensing for all charging sources turned off and that the alt reg pos volt sense and neg wire connect directly to the banks posts...

I would strongly urge you to consider a Balmar AT series hairpin wound alternator. These banks will eat wimpy alts for lunch. As it is now our alt is limited to about 110A to stay out of alt temp limiting range. Once you start bouncing in and out of temp sensing the charging performance sucks. I've found the overall charging performance far superior by limiting the alts output by using "belt manager". It may take a while to find the sweet spot where she will run at 190-200f all day long.. She's already physically melted the heat shrink off one alt temp sensor.....