Monday, October 29, 2012

LiFePO4 Battery Charging Method - Don't Ruin your Expensive Cells !

and don't Burn your house !!

I learned this saturday that the new Vectrix Li and Vectrix Li+ have battery problems: These Electric Scooters here cost 11,000 and 12,500 EUR, which is a lot of money; If at this price you have problems, it is not possible ! As a result the dealer I met told me he was going to stop selling them and close his Electric Scooter business... He also told me that Vectrix may not be imported in France anymore because it is not selling ! I am very disappointed to hear that this great scooter might not be available anymore and even more knowing that it is only a Battery Management Problem Again !!

The same day I decided to write this post to try spread the word among the EV community and maybe Vectrix USA will read this too, along with other EV makers

The Method to Charge and Discharge a LiFePO4 battery pack is Very Simple :

A lot of people think they need a BMS along with a LiFePO4 battery pack, but the only thing it is doing is damage the cells !
This is not me saying this, but Jack Rickard from EVTV : he has been converting cars to Electric Drive for 4 years now and shown by practice (real world charging & driving) that this is the best way to protect these expensive cells

It seems so simple and inexpensive that after reading this some people will still not believe it, but this is the simple reality

My EV experience (since november 2009) has proven the same thing to myself too

Typical LiFePO4 Cell Discharge Curve


- Bottom Balance each cell to 2.75V (Drain them with a resistor until they reach this voltage = empty)

- Connect them in series to form a string of the desired voltage

- Under Charge the string to 3.50V * number of cells with a LiFePO4 Charger
NB: A LiFePO4 Charger charges at Constant Current, then close to the pack voltage, switches to Constant Voltage and reduces Current slowly until it reaches the set pack voltage, and eventually Shuts Down Completely)

- When discharging the battery pack, set a lower limit to 3.1V * number of cells, start alerting that you are reaching the end of the pack and reduce maximum current drawn on the controller; When the pack voltage reaches 3V * number of cells, do not go below, stop the discharge at this point by disabling the controller

That's It !!

- Bottom balancing each individual cell has to be done only once
- If you have to add some cells to the pack (to get more pack voltage), you will have to "marry in" these new cells by bottom balancing ALL the cells again (old and new ones).

This Method is based on facts: Real World EV Charging & Driving, detailed below:

-Back in 2008, Jack Rickard distroyed many cells trying to Top Balance them to perfection (keeping them at a very equal voltage); When he drove his GEM EV, as he was going towards the second half of his pack capacity, he suddenly lost several cells, because they had reached a low voltage limit of 2V, and even below, going into reversal, the stronger cells driving current through the weaker cells that had past 0V

- He explained in details, with wood sticks, how each cell has its very own capacity, and if you try to top balance each cell, you will in fact reduce even more the smaller capacity cell's capacity, and as you discharge you pack, they will be the first ones to hit a lower voltage and if you continues will be damaged or destroyed
=> His Blog post: November 13th, 2009:  "GET RID OF THOSE SHUNT BALANCING CIRCUITS !"
=> His explanation Video: EVTV News Friday Show November 13, 2009. Part 1 of 6. LiFePo4 Cell Battery Management. The truth about current shunt regulator battery management systems. Latest data indicates they do not protect cells, and can actually kill them:

Full length Video can be found at EVTV Video Archives by searching for " November 13, 2009 "

- Watch Jack Rickard explain the process has been using for years on half a dozen of converted cars, using the same tehcnic again here with the new build of Speedster Nippon (starts at 54:34)

- In early 2010, after 4 months of driving my Engineer 4kWh Prius PHEV Kit, I experienced several cells loss because of Cells Balancers, that were 24/7 drawing little current from the cells trying to have them all at he same exact voltage (it is also Top Balancing); Same as Jack Rickard, I lost one cell, then another, and did not understand why at this point... Replacing the damaged ones by brand new, I ended up drestroying some more and eventually almost burned my office underground parking while charging ! At this point I stopped everything, scared of the potential consequences of this unexplained problem ... and disconnected everything from my Prius for several months
I understood later, watching Jack's EVTV Show why this had happened and what caused it

- After that I started working on LiFePO4 Battery based  Energy Storage for Solar at home, and have been experiencing only good things, using the cells (the remaining good ones) of my Prius PHEV Kit, for a year and a half now; So there was no weak cell, but only BMS (8S Cell Balancer) causing some cells to become weak !
Jack 's Method works !

- A lot of EV converters applied his Method, and do not have any Battery problem ever since; Many of them were at EVCCON last september

- CALB has released a new serie of great LiFePO4 cells, the CA serie, replacing (?) the SE serie
If you get these amazing cells (Jack Rickard's words after Charging them at 3C and dischanging at 12C !),
and damage or destroy them because of a bad method or BMS / shunt balancers, you will be very sad ...


CALB CA Serie (New & Very good)

* I will had more details to this post on this subject (To Be Continued) *

August 29th, 2013


To prove you that this Method works, here are videos of regular voltage checks performed on my DIY ESS Battery Pack, Bottom Balanced and installed in January 2013:

March 2013

May 2013

July 2013

October 2013
=> Conclusion: It works: All the cells are still perfectly in balance after 6 months ...


  1. Hi !

    Are you buttom-balancing your cells individually or the hole pack connected in parallel ?


  2. Hi Peter,

    The bottom balancing is performed on each cell, then they are assembled in a series string

  3. In the Typical LiFePo4 Cell Discharge Curve figure, is it that the single LiFePo4 cell has capacity of 195.1Ah? Or is it the capacity of a pack? Thanks. -Katrina

    1. Hi Katrina, Actually these new CALB CA180FI have a Real Capacity much higher: from 190Ah to 195Ah ; amazing isn't it ? :-)

    2. Wow, that's a lot of capacity.. one more question: so I tried and it really worked for one string ! Does this method work when I have multiple strings in a pack (each string consisting of same number of cells) ? -Katrina

    3. Hi Katrina,
      Yes it does, you can as many strings in parallel as you want

    4. Thank you, Christophe. Your method is amazing! I'm a few days away from completing my pack! lol

    5. You're welcome, I am glad it helps :-)

  4. lifepo4 battery management: Suppose you used power mosfets to switch a battery into the series circuit if in discharge and the voltage across the battery>2.85 volts, or to be bypassed if less, and similar in charging at the 3.55 volt level?

    1. Michael,
      Nothing should be done at a cell level, otherwise you start creating inbalances and problems; Always put/remove energy in/from the WHOLE pack of cells; Of course after each cell has been initially bottom balanced like the other one son the same series string.


  5. Hi Christophe, can I ask will this method work on smaller cells for an E-Bike?

    1. Hi Chris,

      Yes à y size of cell, it works as long as it is LiFePO4

  6. Hi Christophe,
    I have Winston 100Ah cells. Can I use same voltage levels - lowest 2.75 and highest 3.5V, for these cells?
    Thank You.

    1. Hi,
      Same thing for all LiFePO4 cells, but 2,75V is for initial bottom balance, after that, set lower limit to 3V or 2.9V not lower, and upper to 3.55V. Merry Christmas !


  7. Hi.
    Very interesting post.!
    I have had the same problem With some cells slowly decading, and have also come up With the same solution, exept the initial draining of all cells.
    This is to find and detect the "smallest" cell, as the battery pack will have no more Ah than the cell with the least capasity.
    BUT, I think you are missing one important step.(?)
    After draining the cells, then ALL consequent charging should stop as soon as the first cell reach 3.55V. Just measuring the total pack will not do it., and the more cells in series, the more important this is.
    No two cells will have exactly the same capasity. So when the charging starts, the voltage on each cell will vary very slightly according to the difference in capasity.
    When the smallest cell is full, reaching 3.55 (Or 3.6 should also be fine), the total pack is NOT 3.55 times number of cells. Continued charging will then overcharge the smallest cell EACH time the pack is fully charged, soon killing the cell.
    The same is equaly true under discharge.
    SO, the only way I can see you can care for these expensive batteries, is to constantly monitor EACH cell, both when charging and when discharging.
    Thank you for the important tip of the start drain. I will redesign my charger according to this.
    Best regards svein

    1. Hi Svein,
      That is right and this is why you have to Bottom Balance each cell, and Under Charge the pack to 3.5V x number of cells
      This method works perfectly: after a year of charge & discharge everyday my cells are still in balance. Jack RICKARD now has a 100 cell VW Thing running, so the number of cells does not matter really


    2. Hi again.
      Agreed! If you undercharge to 3.5 you should have ample room for the smallest cell.
      But then again, your not using the full capasity of the battery, probably not more than 80%.
      I'm working on a 12V 1000ah emergency supply, and leaving 200Ah is a bit much.
      (Thats a lot of dollars, i could use on my wife....)
      But there are so much disinformation out there! Look at EV :
      They recomend top charge of 3.8 Volt and initial 4.0 charging. The opposite of what at least we think...
      SO, I will still go ahead to charge to the first cell reach to 3.6V, and disconnect when the first cell Reach 2.8V.
      Of coarse only after I have first used Your genius initial draining!
      (if not I as usually have totaly misunderstood everything..)
      Best regards svein

    3. I did not invent this but followed Jack RICKARD's work over 5 years of conversions using LiFePO4. I did exactly like him, almost, cutting the charge at 3.5V to prevent any problem but anyways like he says "there is nothing much after that"
      I suggest you watch every episode of EVtv on YouTube and learn like I did :-)
      Follow people with experience not only trying to sell these batteries and giving exotic charge figures

    4. Does the initial charge to 4V have to be made on new batteries before your bottom balancing, or not?

      Best regards

    5. Herbert,
      No the initial charge is done by the manufacturer
      And never charge more than 3.65V for LiFePO4

  8. Sorry but I dont believe this is true. I have 28 headway cells that I labourously discharged to 2.7 volts each and re-assembled...charged untill one reached 3.6 volts..(they were not all the they rose in voltage from 2.75 volts they remained the same untill some reached 3.5 volts..some took off and others stay around 3.4+ after using the pack at about half capacity, recharged and they were constant...HOWEVER, after using the cells and discharging to about 75%, they began to widen and differentiate in the cell voltage..this only took 5-6 going to try this again but I think the same thing is going to happen...

    1. Sorry to hear that but I heard about problems with Headway cells (from Jack RICKARD)
      I use CALB only
      I applied the same technique to 18650 and leaf cells (different voltage limits off course) and no problem

    2. Can you point me to a link in whcih he talks about Headway cells?...I have tried this technique again this time, with 2 BMS modules connected. Now they have remained constant at the bottom of the SOC.I charge ( and monitor each cell) untill one hits 3.7 volts and at this point usually all are within 3.45-3.55 volts...when they are discharged the remain very close to the same as they approach 3.2 which point I stop the discharge. I have 28 headway 15 ah cells and charge to 99.5 volts. Not sure why it appears to be working this time...perhaps the BMS is working as I charge or discharge to keep them balanced..

  9. This comment has been removed by a blog administrator.

  10. I have an 8 cell pack of the newer Calb (light grey) cells. Bottom balanced using your exact method and a powerlab 8. Cells were all with pin .01v. Then I charged to 28v using the same voltage for both absorption and float. The pack sits around 28.1v. Testing each cell, I found one cell at 3.69 and another at 3.84v. I could probably live with the one that hit 3.69 just by dropping my charging voltages to 27.8v or so. Any thoughts here? I'm guessing I have a couple of runt cells in my pack.

    1. Bottom balancing using the device will drop the voltage down to 2.75V maybe but you have to do it to the same cell several times and let is sit there for 24h and do it again, until it really is at 2.75V
      There is no absorption or float in LiFePO4, and I do charge up to 28V, then shutdown the charger, so the pack voltage then drops down to 26.6V usually, each cell remaining then, at rest, at 3.32V
      Maybe your cells were used and have pretty significant difference in capacity/ MAybe they were used using a 'BMS' that damages them, I don't know what could causes this high cell voltage, I don' thave any of this

  11. Hi Christophe,

    What voltage limits do you recommend for Leaf cells?

    1. Hi Jack,
      Check out my mini ESS built using AESC LEAF Cells :
      You will know the voltage limits used


Please Share this Post if you Liked it !

Please Share this Post if you Liked it ! Thanks !
Related Posts Plugin for WordPress, Blogger...