Hf filter for ballast

Discussion in 'Electronic Design' started by JohnS, Oct 17, 2006.

  1. JohnS

    JohnS Guest

    Hi Guys

    Anyone point me in the right direction. I am playing with a 300 watt ballast
    for an amalgam UV lamp and the screened cable from ballast to lamp is
    radiating excessively in the region 30-40MHz. I have tried the usual ferrite
    beads at the ballast O/P but have not had much success. Can anyone suggest a
    reasonable low pass filter cutting off at about 20MHz that can handle lamp
    currents of 3A, or a good site with some design data to give say 12db
    attenuation. I am more used to LC filters in lf range with equal impedances.

    Thanks

    JohnS
    JohnS, Oct 17, 2006
    #1
    1. Advertising

  2. JohnS

    Joerg Guest

    Hello John,

    >
    > Anyone point me in the right direction. I am playing with a 300 watt ballast
    > for an amalgam UV lamp and the screened cable from ballast to lamp is
    > radiating excessively in the region 30-40MHz. I have tried the usual ferrite
    > beads at the ballast O/P but have not had much success. Can anyone suggest a
    > reasonable low pass filter cutting off at about 20MHz that can handle lamp
    > currents of 3A, or a good site with some design data to give say 12db
    > attenuation. I am more used to LC filters in lf range with equal impedances.
    >


    Here is a free (ad-supported) program to calculate LC filters in the HF
    range:
    http://www.aade.com/filter.htm

    Have you considered common mode measures? For example, looping the cable
    as a whole a few times through a big #43 ferrite toroid at each end?

    --
    Regards, Joerg

    http://www.analogconsultants.com
    Joerg, Oct 17, 2006
    #2
    1. Advertising

  3. JohnS

    JohnS Guest

    Hi Joerg

    Yes, I tried individual ferrites and common mode, it seemed to change the
    points at which the radiation peaked but did not suppress over the range.
    Just about also to try a foil and braid screened cable, as teh original was
    just braid, but it seems that hf ballast manufacturer's in general are not
    paying much respect to EMC on the output of their ballasts. It is annoying
    bacause I was paying for test house time.
    Thanks for the web site suggestion

    JohnS

    "Joerg" <> wrote in message
    news:BL9Zg.18723$...
    > Hello John,
    >
    >>
    >> Anyone point me in the right direction. I am playing with a 300 watt
    >> ballast for an amalgam UV lamp and the screened cable from ballast to
    >> lamp is radiating excessively in the region 30-40MHz. I have tried the
    >> usual ferrite beads at the ballast O/P but have not had much success. Can
    >> anyone suggest a reasonable low pass filter cutting off at about 20MHz
    >> that can handle lamp currents of 3A, or a good site with some design data
    >> to give say 12db attenuation. I am more used to LC filters in lf range
    >> with equal impedances.
    >>

    >
    > Here is a free (ad-supported) program to calculate LC filters in the HF
    > range:
    > http://www.aade.com/filter.htm
    >
    > Have you considered common mode measures? For example, looping the cable
    > as a whole a few times through a big #43 ferrite toroid at each end?
    >
    > --
    > Regards, Joerg
    >
    > http://www.analogconsultants.com
    JohnS, Oct 17, 2006
    #3
  4. JohnS

    Joerg Guest

    Hello John,

    >
    > Yes, I tried individual ferrites and common mode, it seemed to change the
    > points at which the radiation peaked but did not suppress over the range.
    > Just about also to try a foil and braid screened cable, as teh original was
    > just braid, but it seems that hf ballast manufacturer's in general are not
    > paying much respect to EMC on the output of their ballasts. It is annoying
    > bacause I was paying for test house time.



    That's what happened to most (new) clients of mine, blowing through the
    first EMC lab visit and fail. But not all is lost because then you have
    the baseline plots. You can rent an antenna set and possibly an analyzer
    if there is none in the lab and experiment on your own until you reach
    enough margin. There is never a guarantee but this greatly increase the
    confidence level for the next round at the EMC lab.

    Shielding will help but only if there is a tight shield connection at
    the generator, ballast and preferably at least the frame of the load.

    For the ferrites make sure it's indeed #43 material or similar. #77
    looks deceptively similar. Unfortunately what I find a lot during visits
    is that all the ferrites have been placed in bins and after a few trial
    and error sessions the cores of same sizes are often mixed. The only way
    to re-separate is an inductance meter, some wire and lots of time.

    --
    Regards, Joerg

    http://www.analogconsultants.com
    Joerg, Oct 17, 2006
    #4
  5. JohnS

    Jake Guest

    On Tue, 17 Oct 2006 21:19:27 GMT, Joerg
    <> wrote:

    >
    >For the ferrites make sure it's indeed #43 material or similar. #77
    >looks deceptively similar. Unfortunately what I find a lot during visits
    >is that all the ferrites have been placed in bins and after a few trial
    >and error sessions the cores of same sizes are often mixed. The only way
    >to re-separate is an inductance meter, some wire and lots of time.


    For what it is worth.

    Buy some cans of spray enamel and paint them after sorting. Do the same
    with new stock.

    Small beads and cores can be strug on wire and dipped in a thinned mix from
    a can of paint.

    You can even use this for threaded cores.

    Jake.
    Jake, Oct 18, 2006
    #5
  6. JohnS

    Joerg Guest

    Hello Jake,

    >
    >>For the ferrites make sure it's indeed #43 material or similar. #77
    >>looks deceptively similar. Unfortunately what I find a lot during visits
    >>is that all the ferrites have been placed in bins and after a few trial
    >>and error sessions the cores of same sizes are often mixed. The only way
    >>to re-separate is an inductance meter, some wire and lots of time.

    >
    > For what it is worth.
    >
    > Buy some cans of spray enamel and paint them after sorting. Do the same
    > with new stock.
    >
    > Small beads and cores can be strug on wire and dipped in a thinned mix from
    > a can of paint.
    >
    > You can even use this for threaded cores.
    >


    I usually write the material code onto them with a silver or gold
    indelible ink pen. Then I don't have to remember colors. But those pens
    dry out quite quickly.

    Nail polish works as well but my wife wouldn't appreciate that ;-)

    --
    Regards, Joerg

    http://www.analogconsultants.com
    Joerg, Oct 18, 2006
    #6
    1. Advertising

Want to reply to this thread or ask your own question?

It takes just 2 minutes to sign up (and it's free!). Just click the sign up button to choose a username and then you can ask your own questions on the forum.
Similar Threads
  1. eeh

    T LP filter and RC LP filter

    eeh, Apr 1, 2005, in forum: Electronic Basics
    Replies:
    2
    Views:
    286
  2. Tom Becker

    PLL input filter vs. loop filter

    Tom Becker, Apr 9, 2006, in forum: Electronic Design
    Replies:
    44
    Views:
    1,710
    Tom Becker
    Apr 19, 2006
  3. Pedro

    Using a fluro ballast as a Filter choke?

    Pedro, Nov 1, 2004, in forum: Electronic Repair
    Replies:
    1
    Views:
    770
    jhill
    Nov 3, 2004
  4. KeeperOfLogic

    Volume (decibel) Low-pass filter (NOT frequency filter)

    KeeperOfLogic, Oct 21, 2006, in forum: Electronic Basics
    Replies:
    9
    Views:
    974
    Bob Masta
    Oct 22, 2006
  5. mercer7
    Replies:
    10
    Views:
    1,563
    Homer J Simpson
    Nov 27, 2006
Loading...

Share This Page