-- THE SINGLE CELL CHARGER POST --

[Deepkimchi]

When I get thru taking my zapped charger apart and analyzing the pieces I'll bypass this black box crap and use single components...

This sucker took an hour to dig out without damaging the circuit board..

DK

 

[Deepkimchi]

I still don't believe single cell chargers are any better than the Flintstone charger (when it works).

The principle behind the FS charger is providing 3.65V to each cell, only it adds 3.65V to each output.

3.65 - 7.3 - 10.95 - 14.6 etc.

Whether a single cell charger or the above method, an exact voltage would depend on the precision (tolerance) of each systems parts (resistors, diodes, etc).

DK

 

[efreak]

actually that seems to be the easiest and least cumbersome even in rc circles cause of it not relying on a bms every so often then they check for relative balance which is then more managable

efreak

 

[ramboman]

Using Artesyn BXB150 seems to be the best approach to the problem.
I use two 12s2p packs on a model airplane, charging in the field remains an issue.
I found the link on e-bay and the specification sheet, but I don't understand how to adapt the voltage to 3.6V instead of 5V.
More, is there a way to limit the current for low speed charging (a123 recommands 3A)?
Thanks

 

[flip_normal]

Using Artesyn BXB150 seems to be the best approach to the problem.
I use two 12s2p packs on a model airplane, charging in the field remains an issue.
I found the link on e-bay and the specification sheet, but I don't understand how to adapt the voltage to 3.6V instead of 5V.
More, is there a way to limit the current for low speed charging (a123 recommands 3A)?
Thanks

read this thread

http://endless-sphere.com/forums/viewtopic.php?f=14&t=2824

Oh and welcome to the forum!

 

[ramboman]

Very easy indead, just add a multiturn 10k pot to set the voltage...
In my case, input is provided by a string of four 38Ah gel batteries;
I need only the twelve "artesyn" with add'l pots and fuses...
But I will have to rewire my packs as 12s3p with 30A "balancing" taps (presently 3A).
13 poles are needed, I guess Anderson would be appropriate. What's your recommandation?
Merci!

 

[Doctorbass]

Very easy indead, just add a multiturn 10k pot to set the voltage...
In my case, input is provided by a string of four 38Ah gel batteries;
I need only the twelve "artesyn" with add'l pots and fuses...
But I will have to rewire my packs as 12s3p with 30A "balancing" taps (presently 3A).
13 poles are needed, I guess Anderson would be appropriate. What's your recommandation?
Merci!

You will need to add a pot to EACH DC-DC.. they need to be isolated from each others.

after seting the voltage to 3.6V, you will need to calculate the current max you will allo to the battery.

one artesyn modules of 150W 5V can give 39A continuous in short circuit mode. so you need to calculate the resistance of the wires from the Dc-DC to the cells.

When a single cell is fully discharged, if you put the max allowed current to it (10A) the voltage will start to around 3.0V.. so between the DC-DC and the cell, you will have a total resistance that will allow 0.6V (3.6-3.0=0.6V) that will be lost in the wires.

The resistor of the wire if defined by ( V loss / current)= 0.6/10 = 0.06ohm (60miliohms)

10A at 0.6V is 6W.. so you will dissipate 6Watts in the cable bewteen each cell and each DC-DC

Refering to this chart, (http://www.powerstream.com/Wire_Size.htm

a 16 AWG wire should be ok (it have 4 miliohm per foot)

so with 15 foot of this wire you (a pair of 7.5foot) that should be ok to limit the current..

If you need more current to cahrge more than one cell, you need to choose the wire that will be heavier and calculate the resistance of the wire to fit with that.


But with this approach,there is a little probelm:
The current will only stay to 10A at the begining of the charging process dur to the voltage difference between the DC-DC and the cell that will decrease as the cell become charged.

i'm working on a circuit that will directly limit the current on the dc-dc and not using resistor.. just stay tuned on my post above :wink:

Doc

 

[ramboman]

So, there in no inherent current limitation!
I charge presently my 6-packs (coupled as 12s2p) either with a Mastech 5020 or with a Robbe Infinity III with 12s balancer or directly from a string of 2 lead batteries thru a current limiting resistance of 0.5ohms 50watts.
That's working pretty well, starting at 14A but stabilizing very fast at 8 to 9A; without balancing.
If the Artesyn had a constant current function (to charge as CCCV)...

 

[Doctorbass]

So, there in no inherent current limitation!
I charge presently my 6-packs (coupled as 12s2p) either with a Mastech 5020 or with a Robbe Infinity III with 12s balancer or directly from a string of 2 lead batteries thru a current limiting resistance of 0.5ohms 50watts.
That's working pretty well, starting at 14A but stabilizing very fast at 8 to 9A; without balancing.
If the Artesyn had a constant current function (to charge as CCCV)...

Some DC-DC have this fonction. with a input pin feedback

the BXB150 serie have an internal I limit for protection only.

 

[ramboman]

Do you know some ?

 

[ramboman]

Hi Doc,
this is probably a stupid question...
Why don't you charge directly from 48V?
You should need no more than a resistance (box) to limit the current.
Typically:
- 4 lead batteries makes +- 48V
- a123 average voltage during charge is 3.4V i.e. 40.8V for a 12s pack
- voltage drop is then 7.2V
- if you set current to 30A, R=E/I, i.e. 0.25 ohms
- then P=I²R, i.e. 225W
- charging at 1440W, efficiency is the same: 84%...
I did it under 24V with a 6s1p pack and a 0.5 ohm 50W resistance, it works pretty well.
But I don't know how to upload the graph...

 

[safe]

That sounds like a great idea...

Why not?

If you need a big voltage supply and have a few big SLA's laying around to provide the voltage then why not use them?

How about this:

You connect your SLA charger to your SLA batteries and keep the charge topped off all the time.

You then connect your SLA batteries to a 12 volt PWM controller and connect the output of that to your LiFePO4 cells.

http://kingsolar.com/catalog/mfg/bzproducts/bz240.html

$65

Now you set the throttle to a level so that the voltage the LiFePO4 cells see is equal to the desired level and when the cells become full you set up some sort of HVC (high voltage cutoff) to pull down the throttle when appropriate.

You could add some LED that signals the charge is done when the throttle is completely pulled down to zero.

The SLA battery ends up being just a power source and the logic for charging is simply moved to the HVC circuit and throttle.

Maybe a DC to DC step down converter from 12V to 4V would work? The main point is you need some sort of throttle to control and reduce the PWM as the cell charges.

 

[safe]

IS PWM SAFE FOR LIFEPO4?

The core question is whether PWM is "good" or "bad" for LiFePO4. If it's a good thing (which I would think it would be since it would allow the cell to rest between pulses) then charging using PWM is a great idea. If not, then you can always add large capacitors into the charging so that the pulses are smoothed out.

Resistors are fine on the charging side anyway, so why not for that too?

 

[Doctorbass]

Hi Doc,
this is probably a stupid question...
Why don't you charge directly from 48V?
You should need no more than a resistance (box) to limit the current.
Typically:
- 4 lead batteries makes +- 48V
- a123 average voltage during charge is 3.4V i.e. 40.8V for a 12s pack
- voltage drop is then 7.2V
- if you set current to 30A, R=E/I, i.e. 0.25 ohms
- then P=I²R, i.e. 225W
- charging at 1440W, efficiency is the same: 84%...
I did it under 24V with a 6s1p pack and a 0.5 ohm 50W resistance, it works pretty well.
But I don't know how to upload the graph...

ramboman,
Please carefully read the first page of that post and you will understand why i prefer to charge each cell parallel group with an individual charger .. and to why i prefer to avoid charging a serie of cells using one charger.

The goal is to provide an independent toping charge voltage to each parallel group. AND TO AVOID OVERCHARGING a parallel group of a string that have different capacity.
When the capacity vary from a cell just a little bit to another in the string , the voltage will increase on the lower capacity cell more than the rest of cells in the string, creating imbalance.

My single cahrger cell desing id bulletproof and avoid ANY problem with balance of each cells and will cahrge each parallel group to 100% of it's SOC.

In a single charger for a cell string, that is not the case.

Dewalt charger have this problem and the imbalance between cells increase in time and amplify the problem du e to the poor voltage protection monitoring circuit and the poor balancer acuracy.

My statement is clare: a single 3.6V CC-CV power supply, DC-DC or charger for each parallel group of 3.3V of LiFePo4 cells

The only way i would encourage a single charger for a cells string would be if using it with a Voltage clamper circuit on each parallel group. That would have effect on balancing the pack and protecting it from overvoltage into each parallel group cells.

Doc

 

[ramboman]

I understand, but:
1. it's not too difficult to check the voltage of a cell and to activate a shunt when necessary;
2. with 6 cells in parallel, what happens if one of them fails?
Ideally, we should have one CCCV charger per cell...

 

[Doctorbass]

I understand, but:
1. it's not too difficult to check the voltage of a cell and to activate a shunt when necessary;

Some simple circuit can do that using shunt regolator like the TL431 and some mosfet or TIP transistors

2. with 6 cells in parallel, what happens if one of them fails?...

About a cell that could fail in parallel with other, i say pretty impossible!

Just think about that question: Why a cell usually fail?

I say:
-it could be due to temperature that could be outside the recommand range
-it could be due to over or undervoltage that happened
-it could be due to mecanical dammage .. anyway if it happen the pack need to be revised...

now, if temperature problem occure with a cell in a parallel group, that could be impossible that other cells still ok... probably the rest of the cells around will be affected too!

if a over or under voltage occur in a parallel cell group, ALL the other cells will be certainly affected too!

if a mecanical dammage occure to a cell in a paralle group, .. the pack need to be revised... and rebuilt for sure!.. at least for safety!

A CELL CAN NOT FAIL ITSELF AND ONLY ITSELF IN A PARALLEL CELL GROUP !

I HAVE NEVER SEEN THAT AND I HAVE NEVER KNOWN SOMEONE WITH THAT CELL PROBLEM.

ALL THE VOLTAGE OF THE CELLS IN A PARALLEL GROUP ARE THE SAME .. so ALL THE SOC ARE THE SAME!!

Ideally, we should have one CCCV charger per cell

Ramboman, read this:http://endless-sphere.com/forums/viewtopic.php?f=14&t=3437


Doc

 

[ramboman]

Very interesting.
But why do you not use CCCV chargers?
Artesyn are not!
btw, how do you assemble the cells?
I am learning a lot ))
Merci

 

[Doctorbass]

CC-CV IS what is required to charge Lithium ion battery.

li-ion have a very close disharging curve and parameter than SLA, but the charging process IS different.

I think you should read some great articles on batteryuniversity site.
http://www.batteryuniversity.com/partone-12.htm

You should find alot of great answer and explanations!

I am working on a way to control the Dc-DC artesyn to limit their current.

Their voltage IS very stable (0.001V+/-) when using the sense input feedback.

I assemble the cells un xS * xP mode. and will sporweld the string using my big flux capacitor

Doc

 

[efreak]

Good Job Doc,

keep up the good work, I am just reading and trying order and gather all the neccesaryy parts these modules are so small , But I ,ll ned heat sinks for sure . and a low noise power supply I guess the same ones i mentioned ealier . i got the lvc boards from Gary thanks Gary great production. noticed no good deals on dewalts anymore

this thread and dc-dc thread should merege
the one meanwell power supply/charger you mentioned is really a very easy solution for anyone out there with multiple out put and very versetile.

still waiting on my mega power and still searching for a suitable power supply for it. I read that they perform uptimally being fed with a 20 amp capable power supply 12volts 20 amps any suggestions?

and the two 48 volt ps's on ebay the mastech would cost me as least $340 ouch.

happy friday everyone.

efreak

 

[ramboman]

"sporweld the string using my big flux capacitor"
How do you do that?

 

[kbarrett]

"sporweld the string using my big flux capacitor"
How do you do that?

Use lots of beer.

 

[Doctorbass]

and some cooper electrode and nickel or cooper strip!

Doc

 

[ramboman]

and some cooper electrode and nickel or cooper strip!

Doc

I got beer and streep, but it does not work