Reducing battery charging ripple, Winding a high current choke?

Discussion in 'General Electronics Chat' started by Dyslexicbloke, May 7, 2011.

  1. Dyslexicbloke

    Thread Starter Active Member

    Sep 4, 2010
    Hi folks,
    I am using an inverter-based welder as a battery charger. I have control of the output so can control the current and thus the charging voltage of my AGM cells but I am concerned about ac ripple on the pack which is at about 17KHz

    The 400AH pack needs to be current limited until it reaches sufficient terminal voltage to manage its own float charge after that I am controlling the output of the welder to maintain an acceptable float voltage.
    My question to twofold ….

    How much ripple is acceptable?
    Without any modification the ripple at 55A was about 800mV, I added a choke that I made from an 80mm torriod core with 20 turns of 6mm tri-rated on it. This reduced the ripple to 300mV but also got quite warm in the process.

    Will doubling the sqr area of the wire and doubling the number of turns, which would be stacked and therefore further from the core, give me half the ripple and reduce the heat disipated.

    I will happily buy some heavy enamel wire but don’t want to spend money on something that isn’t going to work..

    Any and all input will be gratefully accepted.
  2. marshallf3

    Well-Known Member

    Jul 26, 2010
    Personally I wouldn't worry about it for two reasons:

    1) Batteries themselves serve as huge filter caps effectively reducing the ripple.

    2) It's actually good to have a little ripple in a battery charger - at least a lead-acid celled one - since the ripple current helps to dislodge the hydrogen bubbles that build up on the plates and insulate that area. That's the reason that when I'm charging a regular battery - especially a motorcycle battery you can actually see the accumulation of bubble in - I tend to jostle the battery a bit every half hour or so to help obtain a more even charge distribution over the plates.
  3. Dyslexicbloke

    Thread Starter Active Member

    Sep 4, 2010
    Yep that all makes sense and the big cap analogy is definatly what I am seeing but I have AGM's ... so no bubbles and the measured ripple is with the bats connected and charging.

    It was 800mV, I added a choke I wound on on old torriod core and got it down to 300mV
    Both readings were taken at a chargeing current of 55A

    The ripple I have now is just within the minimum and maximum recomended float voltage range which puts it well below the absolute charging voltage which is effecivly the gasing voltage.
    Obviously if you gas an AGM you will kill it due to water loss that cant be replenished.

    My gut tells me that a controlled ripple might actually be benificial, disloging sulphates and the like, but with a big pack that I cant aford to replace, its a huge chance to take if I am wrong.

    Somwhat unhelpfully the manufaturers simply recomend their chargers ... Go Exide!
    Like I can afford one of those for my second hand cells which the origional owners changed to keep their insurers happy.

    I wish I knew more about battery chemestry ....
    Last edited: May 8, 2011
  4. Kermit2

    AAC Fanatic!

    Feb 5, 2010
    Gel cells DO gas if over charged, but the process only creates 'bubbles' in the gel and thereby removes that portion of the plates from any charge or discharge activity. Effectively reduces the capacity of the battery in other words. It might not be noticeable at first but over time the effect builds with the increased formation of more bubbles. Since the plate area is slowly being reduced by this process, the remainder of the active plate area will see an increase in current flow, which will cause further increase in bubble formation. It becomes worse as more and more plate area is removed from activity and the remaining plate area is subjected to ever increased current flow density due to this effect.