any comments on this LiFePO4 pack & charging plan?

voicecoils

voicecoils

1 MW
Joined
Sep 16, 2007
Messages
2,173
Location
Sydney, NSW, Australia
72v pack of 3.3v, 10ah LiFePO4 cells: 24s1p = 24 cells total. Pack capabilities : 140A 18sec drain, 120A continuous drain. Power output capability 11kW hot off the charger :shock:

Charging plan: Use twelve voltphreak 3.2v 2a chargers on a 1s2p pair of cells each. Charger link: http://www.voltphreaks.com/ssl/catalog/product_info.php?products_id=60

BMS: NONE. Simply set LVC on the CycleAnalyst. All cells would be fully charged and balanced each time by the voltphreak chargers.

Does this sound do-able or am I missing something critical? Anyone able to estimate how long it would take the voltphreak charger to charge up a paralleled pair of these cells? Looks like it puts out around 8W charing power. Assuming the cells have 33wh of energy, that's ~4hr/cell or ~8hr/pair, so the set of 12 chargers could charge the pack in ~8hr if my math and reasoning are correct...
 
doh! :oops: 20ah per pair for charging, yes.

Do you think BMS and heavy expensive chargers can safely be avoided by using this single cell charger approach? I know it's not a new idea, but I haven't read about many doing it...

Thanks Ypedal
 
voicecoils said:
BMS: NONE. Simply set LVC on the CycleAnalyst. All cells would be fully charged and balanced each time by the voltphreak chargers.

Does this sound do-able or am I missing something critical?
Click to expand...

Cell level LVC is something critical with LiFePO4, since the voltage drops so quickly at the end of discharge and thus can damage/kill cells.
 
i have done a 48 volts 6.9 ah a123 pack without bms cms or lvc ...bulk charging them with soneil 48 volts. and so far i am very pleased i did however monitored each cell to see how they behave and they seem to do exactly the same thing over and over and i even went discharging them once by being far from home and not wanting to pedal all the way i think it was down to 30 volts and they recovered right back to their initial voltage after i bulk charged them again ...i have over 20 cycles already. and the soneil cuts them they never go overcharged.
 
The Dewalt pack BMS is widely regarded as not being very good and has reportedly resulted in damaged/dead cells thus demonstrating the need for a good cell level LVC even for a123 cells. Voicecoils' cells are another make (what you got there voicecoils?) which won't be as tolerant as a123s. In the long-run (If you want a long-run!) a cell level LVC on LiFePO4 is needed IMHO. Gary sells 'em or the circuit diagram is around here somewhere IIRC. :D
 
Yes absolutly the bms is the cause for damaging the cells where it was the fault i don t know ...and you re right about voicecoils having another kind of batteries and i am not telling him what to do but i am sharing my setup.

i read from this forums that a123 can and does recover from low voltage if recharged soon after dischaged...
 
The Dewalt BMS must have LVC issues since it has allowed a123 cells to drop below a voltage which has caused them damage and reduced cycle life or even killed them. If you want to get the most out of LiFePO4 cells you should avoid letting any of them get overdischarged.

slayer, you clearly monitor your cells closely but without cell level LVC and periodic balancing I think you may well be risking the longevity of your pack thats all. Best of luck tho', interesting experiment. :wink:
 
yes i knew all that information before assembling the^pack ... but waiting for the bms and looking at my batteries i decided to go ahead and try it ...since i know that 500 cycles is a lot cause i have a 36 volts nimh 15 ah that there are many life in ...so if my a123 give me even 200 i ll be happy but personnaly i think i ll get a lot more ...and i did not pay that much for them...half price of my nimh
 
voicecoils said:
Charging plan: Use twelve voltphreak 3.2v 2a chargers on a 1s2p pair of cells each. Charger link: http://www.voltphreaks.com/ssl/catalog/product_info.php?products_id=60
Click to expand...

Voicecoils you need to charge to 3.6V for LiFePO4 not 3.2V.

At this stage I think it would be a good idea to hold out for the New 16-cell Battery Management System (BMS) which is reported to be nearly ready and easily scaleable. It may involve some fabrication and soldering tho', but I'm sure someone here would do that for you, for a fee, if it's not something you wanted to have a go at.
 
flip_normal
I'm learning as I go along here, thanks flip. I hadn't thought about LVC of individual cells, is it the end of discharge when a bad/dying cell is most likely to take a voltage dive and get freaky?

Don't have the cells yet, looking at the LiFeBATT group buy. I was hoping to keep the system lightweight and simple with minimum numbers of cells, no BMS, no heavy chargers.

My cells would be monitored, fully balanced (through charging) and would have a conservative overall pack LVC set. (maybe to correspond to 8ah of 10ah available from pack, I don't know what voltage that would correspond to without seeing a discharge characteristic curve).

As Ypedal mentioned, the single charger does charge at 3.65v then float to 3.8/3.9v

@ Slayer
good to know it can be done. thanks for sharing your experience.
 
voicecoils said:
...is it the end of discharge when a bad/dying cell is most likely to take a voltage dive and get freaky?
Click to expand...
Letting a LifeBatt cell drop below 2.1V is not recomended and will probably weaken that cell and therefore reduce both the overall pack capacity and the number of cycles you can expect to get.

There are no published discharge curves for LifeBatt cells as far as I'm aware :? !! But the voltage will be fairly flat till the end when it takes the dive, so I doubt you'll be able to accurately judge remaining pack capacity by overtall pack voltage. If you monitor Ah used with the CycleAnalyst and cut off at 8Ah then you should be OK but you'll have to always keep an eye on it.
Spending a bit of extra cash now on cell level LVC is well worth it to get the most out of $700+ invested in cells, IMHO.
 
any luck finding that ypedal?

Ok, I've ordered 12 voltphreak chargers. Now, the voltfreaks site says that the chargers are isolated so that they can be used while cells are connected in series.

If I have 24 cells in series and 12 chargers, I could break the pack into two 12 cell series strings and first charge one, then charge the other. This would probably be quite time consuming.

What's the best way of configuring the pack so that the charger "sees" a 1s2p pair of cells and charges/balances them in parallel. I'm trying to think through the connection and can't visualize it.

In essence I'd like to use the cells in a 24s1p string, charge as 1s2p groups, with the minimal wiring complications. Any ideas?
 

Attachments

  • 24_lifebatts.jpg
    86.3 KB · Views: 1,106
There are a few ways to do this..

-12 more chargers :shock: :D
- Charge one set of 12 at a time. then switch to the 2nd pack. ( 2amp chargers.. takes some time if you drain heavy )
- Get an sla charger to bring the bulk of the charge up quickly .. then connect to the VP chargers to top off/ballance them.

- rig the pack with a complicated wire harness to switch paralell/series for charge/discharge.. Not sure how specific voltage of each cell would matter when you first hook them up.. would need those darn things called diodes hehe.
 
It would be hard to reconfigure the pack to charge all the cells at the same time. You'd need some huge anderson block with a contact for each cell and a jumper plug.

It would be much easier to charge each half sequentially. The charging plug could have much smaller contacts since it doesn't have to pass the full discharge current.
 
Ypedal said:
There are a few ways to do this..
-12 more chargers :shock: :D
- Charge one set of 12 at a time. then switch to the 2nd pack. ( 2amp chargers.. takes some time if you drain heavy )
- rig the pack with a complicated wire harness to switch paralell/series for charge/discharge.. Not sure how specific voltage of each cell would matter when you first hook them up.. would need those darn things called diodes hehe.
Click to expand...

Great feedback, thanks. For now I'm going to skip going the extra cell charger route, I'm already in ~$160 for the single cell chargers :shock:

Breaking the big pack into two 12s1p packs requires just one connector, I like that! I'm trying to maintain simplicity (and low weight) so diodes and complicated harnesses are out. Ok, so to charge, break 72v pack into two 36v packs. Connect each single cell charger to + & - of each cell (hopefully set up so I can plug the voltphreak end connectors as one big block into a block of connectors terminating at the cell taps). Do the same for the second pack.

- Get an sla charger to bring the bulk of the charge up quickly .. then connect to the VP chargers to top off/ballance them.
Click to expand...

This is an appealing idea. A question though, what voltage does the bulk charing need to be done at? Would an SLA charger have the appropriate voltage?

It's CV/CC right? So basically I want to charge at the correct bulk charge voltage at a current as high as possible (up to 30A). Use the charger for ~30min to dump in heaps of current, then connect the voltphreaks to quickly massage the pack up to full SOC. :D

I could keep the bulk chargers at home, and keep the lighter weight voltphreaks on the bike too so I could at least do some emergency charging if necessary away from home.
 
fechter said:
It would be hard to reconfigure the pack to charge all the cells at the same time. You'd need some huge anderson block with a contact for each cell and a jumper plug.
It would be much easier to charge each half sequentially. The charging plug could have much smaller contacts since it doesn't have to pass the full discharge current.
Click to expand...

Thanks fecher, charging plug doesn't need to pass full discharge current true, but the plug would actually have to be a big block of 24 connections for the 12s1p pack, to connect to the voltphreaks, right? The chargers only put out ~8 watts each so I think fairly thin wiring could be used for charging.
 
For a 12s string, you really only need 13 contacts. All the in-between connections can be shared.

To estimate the necessary charging voltage, try 12 x 3.65v or so. That's 43.8v I don't think a 36v supply will tweak that much easily. It might work better to get a 48v supply and tweak it down.
 
fechter said:
For a 12s string, you really only need 13 contacts. All the in-between connections can be shared.
Click to expand...

13 contacts cool, I can't visualise what shared in-between connections means. I understand the basic concepts of series & parallel pack building from the voltage and amps/ah perspective but large pack building and efficient layout methodology alludes me. I've googled pack building but haven't found a good explanation.

]To estimate the necessary charging voltage, try 12 x 3.65v or so. That's 43.8v I don't think a 36v supply will tweak that much easily. It might work better to get a 48v supply and tweak it down.
Click to expand...

cool 48v chargers with 10% trim would do the trick then. If I used a 48v supply on plugged into a wall timer for example, this should be a safe way to quickly bulk charge without constant monitoring?

thanks fetch, your patience and help is invaluable! :D
 
You must log in or register to reply here.

Similar threads

electroneg
New to LiFePo4 and New Member - Battery, charging and BMS Questions.
Replies
24
Views
4,993
john61ct
J
W
Some LifePO4 BMS and charging questions mixing 12 and 24v
Replies
13
Views
1,632
john61ct
J
R
54.6V charger on a 58.4V lifepo4 pack
Replies
8
Views
2,087
redwing0001
R
noah6624
48v 20ah 3.2v 16s Battery LiFePo4 - I'm New at This
Replies
13
Views
4,568
999zip999
999zip999
S
DeWalt 20V Max 3.0Ah Battery Pack Teardown & Analysis
Replies
10
Views
34,502
qwertyES
Q
Back
Top