LiFePO4 Batteries - Okay Tear Me Apart ;-)

[bvosloo]

Guys

Seems to me that the device to invest in is the intelligent regulator, in a way this becomes the bms and charger:

With an intelligent regulator:

1) Altenator charging is covered
2) AC charging on shore is covered
3) Solar and wind AC is covered
4) HVC is catered for
5) LVC alarm is sensed

No further chargers/devices are neccesary. The batteries I am looking at is Lifebatt which includes a VMS ensuring balancing, not much 'management' beyond that.

All I am not sure of is which regulator is sufficiently custom programmable to charge optimally for LiFepo4? An added advantage is whichever AC device is charging, the device has to deliver only the required current based on the current battery charge state.

Am I looking at this too simplistically?

[electric1]

Originally Posted by bvosloo

The batteries I am looking at is Lifebatt which includes a VMS ensuring balancing, not much 'management' beyond that.

Lifebatt is like 3-4 times more expensive than large prismatic cells from Thundersky/SkyEnergy/HiPower. They use many small cylindrical cells in parallel inside a big plastic box with BMS. Lifebatt design is somewhat similar to what Tesla has in their $100K roadster. It certainly would work, but at very premium cost.

[hellosailor]

bv-
What you call "the intelligent regulator" is way more than what that term has traditionally been used to mean. A regulator normally regulates *one* source of power, intelligent or otherwise.
What folks sometimes look for today is what I have taken to calling "One ring to rule them all", that is, one regulator capable of *multiple* inputs and outputs in multiple *modes* as well, since there is dump regulation versus alternator foldback, two very different ways to regulate power.
Eventually I figured out how this could be done--but of course it wouldn't be cheap, and it would have to be planned with limits (# inputs, # outputs, wattage limits) that ensure one size couldn't fit all. Wouldn't be cheap, either, do there's something to be said for multiple systems, each charging different batteries or used at different times. Kinda like keeping the mariachi music and the opera company apart and not playing both at the same time as the polka band.<G>

[goboatingnow]

At around 200 dollars for approx 100 a/h sets , these are interesting batteries, charge controllers should be easy to build ( actually individual cell charging is easier to control then massive big chargers.

I'm a EE and hence it shouldnt be differnt to arrange the HVC and LVC controls ( though I understandn that LIFEPO4 is far more tollerant to LVC events) controllers, intrgrate this with cell level chargers ( ie those that accept DC feeds) , Hence source feed is handled,

have a read of the diy electric forums, lots of real life experience and knowledge

must talk to thundersky

[s/v Jedi]

Yes, I also think that individual cell charging is the final answer. The problem is the thicker wiring to each cell but that isn't a big problem at all as it can be short and part of the "cell modules" of the complete system. These cell chargers can be simple single stage with halt on a settable max. voltage level and a feedback to the central controller. It could be designed to do all charging or just a finishing stage.... but with the high charge current all the way to full charge it'll make more sense to do all charging on cell level.

The difference between boats and EV's is that EV's need much more cells so cost is multiplied more.

cheers,
Nick.

[electric1]

Individual cell charging is not the answer. It gives you virtually no benefits, but quadruples the cost and complexity of the charger and adds more risk of failures. It also quadruples electrical losses due to multiple DC-DC conversion stages. This has been discussed to death in EV forums and many people tried, but no one has been truly successful.

Why would anyone want to reinvent the wheel when you have 100 years of proven product lines made for Lead Acid?

[ctsbillc]

Different technologies need different appproaches, what worked for lead acid does not work for other technology types.

You do have to monitor each cell (or cells in parallel) individually, to prevent over and under discharge. Over charging of a cell is normally avoided by limiting the applied voltage to each cell by shunting current away when the max cell voltage is reached.

Bill

[electric1]

Originally Posted by ctsbillc

Different technologies need different appproaches, what worked for lead acid does not work for other technology types.

You do have to monitor each cell (or cells in parallel) individually, to prevent over and under discharge. Over charging of a cell is normally avoided by limiting the applied voltage to each cell by shunting current away when the max cell voltage is reached.

Bill

Hence the need for BMS, which is much more appropriate and cost effective than redesigning the charger and retrofitting existing systems.

[ctsbillc]

A BMS designed for NiMH or Li technology should monitor on a cell by cell basis. A lead acid system cannot, you get six cells in series for a 12V battery, and that's it, you just monotor across the whoile stack. Mind you, that's all you need for lead acid. The Mastervolt Li batteries have all this individual cell monitoring built into the battery, I suspect all the big 12 or 24 volt Lithiums do - would be crazy not to, a little bit of over voltage kills the cell.

Bill

[CharlieJ]

A lead acid system cannot, you get six cells in series for a 12V battery, and that's it, you just monotor across the whoile stack.

Not really true in all cases. There are large FLA and AGM battery banks comprised of the proper number of nominal 2 VDC cells to provide system voltage.

[ctsbillc]

You're right Charlie, on larger capacity batteries they are often independant cells, and you can treat them separately. In addition, the more expensive the battery, the more important it is to use them properly. I was really aiming this at general, smaller, installations. When I was involved in large battery installations with 2.2V independant cells, we did monitor their voltage independantly, but only during routine maintenance periods, this was a few years back, before we had microcontrollers in everything. These were on big 100 kW and 1 MW back-up generator installations.

Bill

[s/v Jedi]

Originally Posted by electric1

Individual cell charging is not the answer. It gives you virtually no benefits, but quadruples the cost and complexity of the charger and adds more risk of failures. It also quadruples electrical losses due to multiple DC-DC conversion stages. This has been discussed to death in EV forums and many people tried, but no one has been truly successful.

Why would anyone want to reinvent the wheel when you have 100 years of proven product lines made for Lead Acid?

Because LFP is different than LA. Also, individual cell charging is considered the best option by experts but it is complex and high cost for big packs, like used in electric cars. But when we use 4 or max. 8 cells, it is very different from the 48 cells or more that the cars use.

On risk I think there's no difference because this factor is determined by the sensor that determines cell voltage and both systems need that. Cost will be more but could still be very manageable because we only have few cells. I don't see why it is less efficient because efficiency of DC-DC converters is a percentage and in principle not about how many converters are in use, although a big converter is often a bit more efficient than a small one. It will be much more efficient than shunt charging where a significant amount of the charge current is shunted.

I was unable to find a discussion on this in any forum but would love to read them. Do you have any pointers?

I did find this on the DIY forum:

Li-Ion Charge Management:
A Li-Ion battery has stringent charge management requirements because the battery may ignite if overcharged. This problem is especially serious for the big packs in EVs and HEVs where it could cause a fatal accident. To ensure safety, the voltage of each Li-Ion cell must be measured very accurately since this is the best indicator of the SOC.
Balancing the cell voltages (equalization) also is more difficult because the simple method of overcharge with a small current (trickle charge) cannot be used. Ideally, the Li-Ion cells should be charged individually during charge. (See "Balancing" section below as an alternative to individual cell charging)

Abnormal conditions for the Li-Ion cell voltage, battery current and temperature must activate an alarm and be handled promptly.
Since the safety of the battery pack is dependent on the management system, the reliability of the management system becomes very critical.

Li-ion Group Balancing vs. Individual Cell Charging
There appears to be two mechanims used for lithium pack charging.
Individual cell charging is the best mechanism to obtain complete charge, but is technically difficult to do in a long series of small lithium batteries.

On cell balancing I found good info here: Cell Balancing and Battery Equalisation

There's a lot of info on that site (Electropaedia)

cheers,
Nick.

[goboatingnow]

Individual cell charging is not the answer. It gives you virtually no benefits, but quadruples the cost and complexity of the charger and adds more risk of failures. It also quadruples electrical losses due to multiple DC-DC conversion stages. This has been discussed to death in EV forums and many people tried, but no one has been truly successful.

This is ill informed, firstly multiple chargers increases redundancy and removes the need to generate huge individual chargers ( and by chargers I mean full battery management systems). In practice, given the ability of LI tech to accept 10C-35C charging, such chargers on a complete bank basis would be huge and massively expensive. Charger Efficiency can be maintained at least as high as other methods.LifePo4 batteries are actually easy to charge as they exhibit definable characteristics at low and high charge states , unlike LA which is actaully very difficult to determine electronically when full charge is reached. In LiFePo4 technology terminal voltage is a close approximation to SOC , whereas thats not the case in LA technology. Hence LI technology actually suits modern electronics.

Since cell balencing is an issue though much less for LifePo4 then say LiPo. individual cell charging is best.

Why would anyone want to reinvent the wheel when you have 100 years of proven product lines made for Lead Acid?

Because the power density, charge acceptance, and safety of LiFePo4 cells simply blows the doors off LA technology. Thats why

If you read this thread people get all worked over controlling charging sources, such as wind generators, alternators and solar panels. Nothing could actually be simpler, Solar is simple. wind sometimes needs a dump load, and alternators can be controlled via their field current.

Sure we need to develop and invent some new products , but hey that makes the world go around. ( and keeps people like us in jobs)

The fact is that LA technology is running out of steam. Boats need bigger and bigger battery banks to cope with the enivitable demand for power. ( and dont say it will go away). KISS is fine in principle, untill it meets your wife and two daugthers!.

Large LA banks, are heavy, inefficient, dangerous, hard to charge fast, and very sensitive to deep discharge cycles. All the things we actually need on a boat.

LifePo4 is not the answer to everything , but its a better setup then LA.

A Li-Ion battery has stringent charge management requirements because the battery may ignite if overcharged. This problem is especially serious for the big packs in EVs and HEVs where it could cause a fatal accident. To ensure safety, the voltage of each Li-Ion cell must be measured very accurately since this is the best indicator of the SOC

Lets not develop any "urban myths" here, LiPo techology has thermal runaway issues, LifePo4 does not and is much much safer. Safer that LA technology which also exhibits thermal runnaway

[imcbride]

We've considered how to set up several charging sources but not really DC Gensets. These can't just be "stopped" on a dime like an externally regulated alternator. Even if you can remote stop them, they still generate current even if the battery has hit HVC. So perhaps a "dump load" is simply becoming a requirement?

The "one BMS device to rule them" controller would manage all charging inputs and battery banks. Once all the batteries hit HVC (or current flow stops/end of charge), the controller would have a dump load that can be switched into the charging DC circuit at the same time as a "STOP" signal is sent (Alt. Field Discon., generator stop) to the charging devices that can be remotely controlled. This would eat up any available current without frying anything, handle multiple battery banks, and automatically shut down charging sources.

When the controller detects a drop on a battery's voltage, wait a few seconds (voltage sag), remove the dump load from the charging circuit and send current to the battery bank(s) again. Maybe even start the genset again, if so equipped.

[hellosailor]

Not necessarily. There are somethings (like wind generators) that need a dump load because they will still be running. While others may "need" to be dumped, but can safely withstand being simply shut down and allowed to go open circuit for the time it takes them to stop. Or dumped to a heat sink, which will simply eat the surge--but not being adequate for a constant dump load. A matter of load and time, not just dumping.
For instance, most batteries even when fully charged wouldn't care if they were used as dump loads for five seconds longer.