Winding a high current choke – How, theory and practice?

Discussion in 'The Projects Forum' started by Dyslexicbloke, May 9, 2011.

  1. Dyslexicbloke

    Thread Starter Active Member

    Sep 4, 2010
    420
    19
    Hi folks, I need some help with making a choke

    I am using an inverter based power current source, a welder, to charge a large battery bank.

    With no additional parts I was getting about 800mV ripple at 75Khz when the current is between 30 and 70 amps.

    In an attempt to reduce this I wound a choke using appliance wire on a torroid core I had which reduced the ripple to about 300mV.with a current of 50A or so.
    I couldn’t run much more current than that because it gets alarmingly hot due to the wire size I had available.

    The core is 75mm OD / 55mm ID and 25mm deep, it has 20 turns of 6mmSq wire on it.
    It appears to be a spiral of thin flat material covered in a protective plastic coating
    (Its an old 24v transformer core)

    How can I work out what value this choke is?
    How much bigger will it need to be to reduce the ripple to 50mV?
    Will the core and windings contribute to heating, or is it simply the wire size causing the problem?
    Since I want to block 750mV at 75A will I be dissipating circa 60W or is some of that energy pushed back into the circuit?
    Is a torroid the best way to go and would a ferrite one be any better requiring fewer turns?

    Looking forward to being educated, thanks,
    Alistair
     
  2. Bernard

    AAC Fanatic!

    Aug 7, 2008
    4,170
    395
    I come up with # 2 wire- rated at 90 A, so core material is probably the problem, powdered iron or ferite might be better. But why the concern about a little ripple in battery charging?
     
  3. ifixit

    Distinguished Member

    Nov 20, 2008
    638
    108
    Hi Dyslexicbloke,
    1. Where do you see the ripple, across the battery?
    2. Why do you need to reduce the ripple? Isn't it harmless in this application?
    3. If that core was for a 50 or 60 Hertz application then it will not be suitable for 75KHz. Use an air core design instead.
    4. Is the 75KHz ripple a sinewave, more or less?
    5. From what you said, there is 800mV of ripple on the battery with 70A of charge current. Therefore the battery impedance would appear to be approximately: 0.8V/70A = 11.4 milliohms.
    6. In order to keep the wire temperature below 75°C you need to use AWG6, or 3 of AWG10 in parallel. AWG10 is approx. what you have now.
    7. To attenuate the ripple by 100, the choke impedance should be 1.14Ω at 75KHz. Try 2 or 3uH.
    8. Connect 3 lengths of your 6mm^2 wire in parallel. Wind on a 1 inch dia. form. Rough estimate.
    Good Luck,
    Ifixit
     
  4. Dyslexicbloke

    Thread Starter Active Member

    Sep 4, 2010
    420
    19
    Thanks Ifixit,
    The ripple is measured at the battery terminals and aparantly AGM cells dont like it, it reduces life, charging voltages have to be carfully controlled.

    Sine - Not realy more square with rounded corners.

    I hadnt considdered an air cored choke, I will have a play with that, perhaps look up some dimentions on the web.

    Alistair
     
  5. ifixit

    Distinguished Member

    Nov 20, 2008
    638
    108
    Hi Dyslexicbloke,

    I only suggest air core because it won't saturate at high currents. There may be other materials that might work as well, but I don't think you need them.

    Maybe you don't need to worry about ripple current at such a high frequency. The ripple current concern was more for 60 or 50 Hertz frequencies where the cell had time to charge/discharge at the ripple rate. Even at 50Hz a 0.5% p-p ripple was acceptable and caused minimal temperature rise.

    Google: AGM cell "ripple voltage"

    Let me know what you think.

    Ifixit
     
  6. Bernard

    AAC Fanatic!

    Aug 7, 2008
    4,170
    395
    With 6 mm wire + some insulation, est. .235 in. Choosing 9 turns, r = 1.25 in, l= 2.115 in, then L=3.9 μH. Resistance about 14 mΩ.
     
  7. #12

    Expert

    Nov 30, 2010
    16,261
    6,770
    Just confirming, cores can get hot because energy is dissipated in them. They get magnetized, saturated, and stuff like that. At tiny little microhenries like this job, air core is foolproof. Air doesn't saturate (magnetically).
     
  8. Dyslexicbloke

    Thread Starter Active Member

    Sep 4, 2010
    420
    19
    Great info, THANKS folks.
    I have to admit I still need to work out the most practial approach for my spaciffic task but now I feel that I have a good chance of getting it to work because I have some idea where to start.

    Thanks again
    Alistair
     
Loading...