Lithium Batteries (for the rest of us)

[noelex 77]

Thank or the reply Terry.

As you say the construction of these cells is different, but I have used a lot of cells overt the last 5 years in reasonably undemanding applications with low charge and discharge rates. The cells never get warm.

They only have a useful life of 3-4 years. The RC modellers who push the cells much harder report a shorter life.

I am sure we will get much longer life out of prismatic cells in boat applications. Importantly it will be significantly better than the lead acid alternative.

However I do cringe with some of the very long life some users are expecting.

There was a similar degree of hype when AGM batteries were released. Some of the claims for these batteries have proven to be true, but time has proven many of the initial claims about cycle life aboard a boat were optimistic.

I hope to get some boat cells soon and will happy to be proved wrong

[hellosailor]

Terry-
When you say "If you charge these batteries to the end voltage of the average lead acid designed alternator you will not get a long service life, Li cells charge at a lower voltage than lead acid cells" how do you mean that?
The typical automotive alternator/regulator puts out 14.4 volts. The sloppy older designs, 13.8-14.4 volts. Still, for a four-cell Li-x-battery that works out to 3.6 volts per cell, which should fit nicely in the performance range for any lithium chemistry. Not a full charge, but a good charge, well within the safe operating range.
So are you saying they won't get a good service life because they will never be fully charged? One would have thought they'd get a fairly long life, never being over-charged, never being at risk of over-charging. And yet apparently they still need a BMS in that situation too?

[goboatingnow]

I think this thread has become confused

LiFePO4 in my experience can be summarised as follows

(a) Does not require charging to 100% and can be discharged to 20%. IN fact there is evidence that life cycle is improved by charging to 85-90%

(b) Charging is terminated by simple voltage detection. This is in stark contrast to charge termination of LA, where in fact no real direct method of charge termination detection exists.

( c) in low C applications like boats, where typically we are engaged in fractional C charge and discharge , there is little need for continuous balancing as the evidence is that once the initial balance on installation, cells do not go out of balance and in fact converge.

(d) LiFerrous does not exhibit thermal runaway , though cells can be damaged by overcharging ( as can LAs). Cells also loose capacity from being drawn down too far.

(e) Negible self discharge, virtually zero Peukerts, low slope discharge curve, no need to bring back to 100%. Lightweight ( relative)

Where it not for the costs, I don't think a single argument would be raised against them in boats. They would be near ubiquitous as they possess almost ideal characteristics for boat use. ( The main disadvantage is they don't like float charging or trickle charging )

Lets deal with BMS ( aka battery Monitoring or Battery Management).

Firstly HVC events , i.e. overcharging. IN fact on a boat this can be very difficult to do . outside a charger fault condition at a knee voltage of 3.65V , that requires 14.6or more. Often nothing onboard can certainly generate charging voltage close to 4V per 4 cells that really can do damage.

Mention is made of fault issues that allow individual cells in a series string to experience over the knee voltage. Yes , where charging regimes try and return close to 100% this risk is real and no doubt BMS monitoring of each cell is useful.

However in practice these conditions are not likely as cells do stay converged. That is not to say BMS rant needed, again given the price of the cells , the BMS is a small fraction so generally its added to monitor the cells.

Note that the same faults can occur in out of balance LA cells, especially Gels and AGMS with damaging effects, yet we rarely monitor at cell level.

LVC, is a good idea, and is easy to implement. But again in a boat context by the time the voltage had dropped well short of the 8V, every alarm is screaming in the boat and all the lights have gone out !!. where the boat is left unattended and with Li , the float charger should be disconnected , LVC is useful. Either done via the BMS or specific LVC disconnect ( Mastervolt do a range) .

In my view Li Ferrous are longer lasting, MORE rugged then LA and capable of absorbing much poorer and less controlled charging regimes. All that is against them is cost.


dave

[hellosailor]

Dave-
"(b) Charging is terminated by simple voltage detection. This is in stark contrast to charge termination of LA, where in fact no real direct method of charge termination detection exists."
How do you figure that? AFAIK, charge "termination" for lead acid batteries is always based on either the absorption rate or the voltage of the battery, depending on which logic the charger maker decides to implement. They may simply stop when the absorption current is low enough, or when the voltage is high enough. Or instead of terminating the charge, they switch to afloat charge--which is a common option.
So there may be three options in use, and arguments to be made for or against each, but there still are "terminations" based on objective criteria.
No?

[goboatingnow]

Originally Posted by hellosailor

Dave-
"(b) Charging is terminated by simple voltage detection. This is in stark contrast to charge termination of LA, where in fact no real direct method of charge termination detection exists."
How do you figure that? AFAIK, charge "termination" for lead acid batteries is always based on either the absorption rate or the voltage of the battery, depending on which logic the charger maker decides to implement. They may simply stop when the absorption current is low enough, or when the voltage is high enough. Or instead of terminating the charge, they switch to afloat charge--which is a common option.
So there may be three options in use, and arguments to be made for or against each, but there still are "terminations" based on objective criteria.
No?

Li has the advantage , shared with some other technologies , that terminal voltage closely corresponds to SOC under charging. LA has no such simple indication and various " sleights of hand " are used , typically a time and current profile to determine the nominal end of absorption. In this regards chargers for Li can be very simple affairs, unlike the complex system deployed to try and charge LA batteries.

Dave

[OceanSeaSpray]

Originally Posted by T1 Terry

LiFeP04 cells don't catch fire if they are over charged and believe me, I've tested that one out, a continuous 40 amps into 4 fully charged cells in parallel for I'm guessing 5 hrs. It turned them into a kind of bulged out box, flattened out the ribs in the casing and even the top and bottom bulged, and it was firmly strapped, till it snapped a weld and alerted me to the fact there was a problem.

Well, that's a test not many are likely to perform, so very interesting. What was the capacity of the cell?

What the EV guys are doing is not irrelevant. Remember that they charge quite slowly at constant current, unlike what can happen on boats where there is no current control as such, and they deal with things like 16 cells or more in series, so they are at the forefront of problems there. Having just 4 cells is really great in comparison.

Originally Posted by T1 Terry

If you charge these batteries to the end voltage of the average lead acid designed alternator you will not get a long service life, Li cells charge at a lower voltage than lead acid cells.

No, what I meant is that 4 x LiFePO4 in series have a charge voltage equivalent to 6 x Lead-Acid cells. However, there are little technical niggles all around. If your charging system is temperature compensated and you go to a cold climate, your charging voltage will suddenly get too high by the look of it because temperature compensation for LiFePO4 cells seems to be minimal to negligible - but I am still researching all the ins and outs.
What is interesting is that a stock standard alternator with built-in staged regulator that is at best poor in a cycling application with LA batteries is in fact remarkably adequate when used to bulk charge a LiFePO4 bank. The only real issue can be higher than reasonable current with small banks, and this is easy to fix.

This chip is a charge controller for this chemistry. You can see the features and functionality implemented. See what it can do with over-discharged cells for example.

Regarding the chemistry itself I found an interesting paper here. It contains valuable info. For example, you can't determine state of charge under any load (same for lead acid), but you need to disconnect the bank completely for a surprisingly long time before it stabilises. Open circuit voltage only reflects the state of charge in some areas of the discharge curve; there are zones where you can't tell much.

I am going to design a little board that will do high/low voltage and probably temperature monitoring for individual cells for a 4-cell arrangement with outputs for charge cut-off and load disconnect and a low voltage pre-alarm. If something ever goes wrong, it is likely to involve heat somehow and adding temperature will only cost a few dollars more. Best practice in the industry also seems to be keeping an eye on it for each cell.
I want a very simple, robust, purpose-built solution. This means there won't be a need for a costly monitoring system or rigging up something from a number of components that weren't quite intended for that.

Eric

[hellosailor]

"In this regards chargers for Li can be very simple affairs, unlike the complex system deployed to try and charge LA batteries. "
Quickly, move the dust bunnies to safety, I feel the need to ROFLMAO!

OK, sure, the Li charger can be simple. Now, offset that simplicity with the need for a BMS that can double the cost of the batteries themselves and...ooops, simplicity just went out the window. Mind the broken glass.

I really do look forward to the promise of Li but I feel like I have seen this movie before, over and over again. In the 1950's and early 60's it was the US nuclear & electric industry, with "Reddy Kilowatt" promising us all water-heater sized home nuclear piles to supply all our domestic power needs. So much for promises.

Until the "lithium industry" can get their act together and figure out whether a BMS is necessary, or how to make one cheaply, or even what range to charge the batteries through? It is very kind of so many people to volunteer to be the pioneers, and take the arrows in their backs. If I wanted to have fun gambled that kind of money, I'd just go to Vegas, where at least I'm guaranteed free drinks, free food, and great floor shows.

Easy to charge, sure. Ignoring the great big BMS in the parlor.

[OceanSeaSpray]

Originally Posted by goboatingnow

Li has the advantage , shared with some other technologies , that terminal voltage closely corresponds to SOC under charging. LA has no such simple indication and various " sleights of hand " are used , typically a time and current profile to determine the nominal end of absorption. In this regards chargers for Li can be very simple affairs, unlike the complex system deployed to try and charge LA batteries.

You are completely wrong. First of all, charge termination on LiFePO4 cells is determined by current dwindling down to a very low value under constant voltage, just like for lead acid cells. When maximum charge voltage is reached, this is only the end of the bulk phase. Fortunately, it already represents a very large fraction of the capacity and the battery won't get damaged by not being charged all the way. Just look at a LiFePO4 charge voltage/current curve.

When it comes to lead acid, charge termination seems to be correctly handled only by "proper" charge controllers that also use external current shunts (large stationary installations) and the magic little alternator control boxes sold for the marine market lack external shunts and instead offer a myriad of "parameters" to hand the problem over to the owner. It makes little sense and they can't do a proper job because they have no idea of how much current is actually going into the battery.
If you have decent charge control with SLAs, AGMs etc, they are not as bad as what we see happening on boats. Problem is having to keep them charged up to prevent sulfation, there is no getting away from this.

[goboatingnow]

Originally Posted by OceanSeaSpray

You are completely wrong. First of all, charge termination on LiFePO4 cells is determined by current dwindling down to a very low value under constant voltage, just like for lead acid cells. When maximum charge voltage is reached, this is only the end of the bulk phase. Fortunately, it already represents a very large fraction of the capacity and the battery won't get damaged by not being charged all the way. Just look at a LiFePO4 charge voltage/current curve.

When it comes to lead acid, charge termination seems to be correctly handled only by "proper" charge controllers that also use external current shunts (large stationary installations) and the magic little alternator control boxes sold for the marine market lack external shunts and instead offer a myriad of "parameters" to hand the problem over to the owner. It makes little sense and they can't do a proper job because they have no idea of how much current is actually going into the battery.
If you have decent charge control with SLAs, AGMs etc, they are not as bad as what we see happening on boats. Problem is having to keep them charged up to prevent sulfation, there is no getting away from this.

Nope I'm not wrong nor are all the integrated circuits that do Li charging.

Dave

[T1 Terry]

Originally Posted by noelex 77

Thank or the reply Terry.

As you say the construction of these cells is different, but I have used a lot of cells overt the last 5 years in reasonably undemanding applications with low charge and discharge rates. The cells never get warm.

They only have a useful life of 3-4 years. The RC modellers who push the cells much harder report a shorter life.

I am sure we will get much longer life out of prismatic cells in boat applications. Importantly it will be significantly better than the lead acid alternative.

However I do cringe with some of the very long life some users are expecting.

There was a similar degree of hype when AGM batteries were released. Some of the claims for these batteries have proven to be true, but time has proven many of the initial claims about cycle life aboard a boat were optimistic.

I hope to get some boat cells soon and will happy to be proved wrong

This would indicate the batteries will cycle for at least 10 yrs http://en.wikipedia.org/wiki/Opportunity_(rover)#2013 and it's still functioning. It's twin, Spirit functioned up till 2010, one front wheel stopped driving so the operated it backward dragging that wheel wheel , then the other front wheel locked up so was no longer able to move to safer ground during the Martian winter, they lost communication contact, that doesn't mean it isn't still alive, they just can't communicate with it any more.
So one is still operating after 10 yrs, the other is an unknown but not a battery failure as such, so harsh conditions and calendar life doesn't appear to be an issue. These battery packs and the whole unit was designed to last for 126 days

T1 Terry

[T1 Terry]

Originally Posted by hellosailor

Terry-
When you say "If you charge these batteries to the end voltage of the average lead acid designed alternator you will not get a long service life, Li cells charge at a lower voltage than lead acid cells" how do you mean that?
The typical automotive alternator/regulator puts out 14.4 volts. The sloppy older designs, 13.8-14.4 volts. Still, for a four-cell Li-x-battery that works out to 3.6 volts per cell, which should fit nicely in the performance range for any lithium chemistry. Not a full charge, but a good charge, well within the safe operating range.
So are you saying they won't get a good service life because they will never be fully charged? One would have thought they'd get a fairly long life, never being over-charged, never being at risk of over-charging. And yet apparently they still need a BMS in that situation too?

Li cells are fully charged at 3.4v, anything more is over charged and continually damages the cell every time it happens.
The confusion lies with people reading and not understanding the terminology so simply skipping over that bit and holding on the bits they do understand. To properly understand what is being said in the technical data, highlight the first bit you don't understand, google it, when you know what that bit means, start again and read it with that new bit of understanding, when you reach the next bit you don't understand, repeat the process. You may end up reading some of it a dozen times, but you will understand what it says at the end.
There is no quick path to understanding, the quick path leads to jumping to wrong conclusions and that always ends in disaster.

T1 Terry

[T1 Terry]

Originally Posted by hellosailor

"In this regards chargers for Li can be very simple affairs, unlike the complex system deployed to try and charge LA batteries. "
Quickly, move the dust bunnies to safety, I feel the need to ROFLMAO!

OK, sure, the Li charger can be simple. Now, offset that simplicity with the need for a BMS that can double the cost of the batteries themselves and...ooops, simplicity just went out the window. Mind the broken glass.

I really do look forward to the promise of Li but I feel like I have seen this movie before, over and over again. In the 1950's and early 60's it was the US nuclear & electric industry, with "Reddy Kilowatt" promising us all water-heater sized home nuclear piles to supply all our domestic power
needs. So much for promises.

Until the "lithium industry" can get their act together and figure out whether a BMS is necessary, or how to make one cheaply, or even what range to charge the batteries through? It is very kind of so many people to volunteer to be the pioneers, and take the arrows in their backs. If I wanted to have fun gambled that kind of money, I'd just go to Vegas, where at least I'm guaranteed free drinks, free food, and great floor shows.

Easy to charge, sure. Ignoring the great big BMS in the parlor.

“Quickly, move the dust bunnies to safety, I feel the need to ROFLMAO!”
What is a dust bunny, I hope I didn’t damage mine after reading that post.

Buying a BMS for a 4 cell 12v battery reminds me of the days when television first came out and the scare campaign about picture tube failure was launched. You had to buy picture tube insurance because when the picture tube failed the household budget would be bankrupted and you would be out living on the streets or something. Looking back, how often did the TV picture tube actually fail… the insurance didn’t cover dropping the tele or putting your foot through it either :lol:

If the BMS system really worked like they claim, you could just bolt it on and use the battery, right from the word go. Ask the next salesman if that is how it works and ask for the guarantee in writing that they will replace the battery if the BMS system doesn’t do that, a recording of the conversation would be priceless, especially a picture of the face when you asked for the guarantee in writing.
The facts are, all the cells must be fully charged, none under charged, before you start using the battery. Whether you bolt a BMS system on or not at this point it doesn’t matter. Monitor the cell voltages and stop charging if a cell reaches 3.6v before the regulator stops charging at 13.8v. If the same cell keep reaching 3.6v before the 13.8v cut off is reached, charge the lowest cell up to 3.6v, the problem will go away. $14 for a Junsi cell logger, wire a smoke alarm to it if you want a louder warning of when to stop charging…. It doesn’t sound very complicated or expensive to me and it won’t stuff up and damage your batteries, ask for that in the guarantee as well before you buy any BMS system.

The key to success at anything, understand every thing about what you are reading, find out what the bits mean that you don’t understand, don’t just glide over them. Only then will you be able to weed out the B/S from the useful stuff and really understand.

I think I’ll keep my picture tube insurance in my pocket.

T1 Terry

[familycruisers]

So, anyone want to go on on a group buy from overseas?

[Singleprop]

I'm interested - and also have good contacts to the factories - some are just around the corner from Hong Kong. Last time I checked the price was just around $ 1.2 pr A/h for a small order. That was Winston, also have contact to Seiden and CAB.
All of them deliver with connectors for the batteries - in the desired setup.

[familycruisers]

Think I found cells for $1/ah but twenty cents more for connector sounds good. Shipping, what is the shipping on small orders?