Alternator to LI batteries

[v1nd]

Hi,

I'm working on a way to charge some A123 Li-Ion batteries (ANR26650M1) with a Honda TRX450 alternator. So far, I'm thinking of just setting up some dummy loads to pull CC/CV from the alternator and then supply CC with varying voltage as the batteries like to be charged. Any comments/suggestions on how to get a circuit started?

Attached is also a sample circuit diagram my friend gave me, it's to charge the batteries with an external charger, I was told a couple of adjustments to said circuit would be enough to get it working w/ an alternator instead.

Any replies appreciated,

Thanks,

[NiHaoMike]

It's just a matter of modifying the voltage regulator to output the proper voltage for the batteries. Almost every automotive alternator is a 3 phase wound rotor synchronous machine with integrated output rectifier and many have integrated voltage regulators.

[Simon]

Are you saying modify the alternator output voltage or the charging circuits ?

[NiHaoMike]

Modify the charging voltage.

[v1nd]

I believe the A123's like to be charged through constant current varying voltage until about 85-90% charge, so wouldn't using constant voltage be the wrong thing to do?

[NiHaoMike]

That part is handled by the current limit.

[v1nd]

OK, so the datasheet for the batteries detail a recommended charge current "10A to 3.6V CCCV 15min". Would that mean by supplying 3.6V, the batteries would pull 10A by themselves, or does the "to" specify some sort of  gradual change?

Also, would I still require a circuit to pull maximum power from the rectifier (alternator), or could I just dump the output of the rectifier to the batteries through a voltage regulator?

Lastly, do the batteries stop drawing power on their own, or will they likely require an overcharge protection circuit?

Appreciate the help,

[scrat]

OK, so the datasheet for the batteries detail a recommended charge current "10A to 3.6V CCCV 15min". Would that mean by supplying 3.6V, the batteries would pull 10A by themselves, or does the "to" specify some sort of  gradual change?
[...]
Lastly, do the batteries stop drawing power on their own, or will they likely require an overcharge protection circuit?

As far as I know, that sentence on the datasheet says you have to control current at 10A. You still need an overcharge protection, or a circuit that can interrupt charge at the proper time. Looking at the datasheet of BQ2954 is very useful, in my opinion. It shows typical charge diagrams, the IC is meant to manage charge cycle. Of course, settings (passive external components values) have to be choosen the right way.

Also, would I still require a circuit to pull maximum power from the rectifier (alternator), or could I just dump the output of the rectifier to the batteries through a voltage regulator?

In car alternators, the regulator (excitation winding controller) will adapt to the external load applied and speed, providing the needed power at constant voltage. Troubles may exist during transients, since excitation winding control could be slower than battery charger transient, so speed (and then voltage) could drop at charge starting. If rectifier output remains well above battery voltage (which is likely to happen, since 3.6V is well below 24V), it should continue working.
When I looked for Honda TRX450 alternator, I found this:
http://www.cheapcycleparts.com/model_years/1518-honda-2007-TRX450R/assemblies/18802
The figure shows something that is more likely a synchronous alternator, so output voltage depends on speed, unless it has a quite complex controller or a regulator on the rectified voltage. Just measure rectifier output at various speeds to see if it is constant.

[v1nd]

The cells actually total 24V over 2 packs of 12V; the first pack has 4 banks (to reach 13.2V) with 3 cells per bank (to reach ~7Ah); the second is just 4 cells in series to reach 24V for a starter motor (through testing, we determined that ~24V were required for reliable startup).

However, the rectifier will only be hooked up across the first pack, as a result, the voltage will range from 8.0V (2.0/cell -- recommended discharge cut-off) to 15.2V (3.8/cell -- maximum recommended charge); is the difference in voltage high enough for the transients to not matter in this case?

I took some time to read more of the BQ2954 datasheet, I'm surprised I didn't read more of it earlier; I realize now that most of what I'm asking is probably detailed at least roughly throughout it. I'll spend a couple days reading through it to try to understand more of the circuit layout before asking anymore questions, although I'm still pretty clueless about the other elements in the circuit such as transistors, transformers, zener/schottky diodes, not to mention the actual design element of why things are where they are, but I suppose this is the best way to start learning.

Thanks for all the help so far,

[scrat]

The figure shows something that is more likely a synchronous alternator, so output voltage depends on speed, unless it has a quite complex controller or a regulator on the rectified voltage. Just measure rectifier output at various speeds to see if it is constant.

Excuse me if I quote myself, but I'm quite curious about how this alternator works. So, could you please take a measure of its rectfied output voltage? It will also help answering the question about the voltage margin... Thanks

[v1nd]

Yea, I was planning to do as such, unfortunately I don't have access to the car until at least the first week of September. I'll get the values up when it's possible though.