Connect with us

RF ground in an apartment.

Discussion in 'Electronic Design' started by Steve, May 21, 2007.

Scroll to continue with content
  1. Steve

    Steve Guest

    I live in an apartment, on the second floor. I am finally finishing
    up a tesla coil, but I don't have an rf ground. Would placing a
    ground plane of wire mesh work? How big should it be? Would say,
    4x4ft suffice? The coil is appx 3ft tall, 7.5kV NST, 17.5in tall
    secondary, really not too big.

    Also, I want to build a faraday cage for the coil. It isn't
    necessarily a large coil, but I don't want to cause any interference
    for anyone. Is aluminum screen acceptable for a cage?

    Also, I have a standard box power filter, mfg: corcom, model: 20VK6.
    To keep rfi from entering back into the power line, do I hook the
    filter up normally, with line going to line and load going to load, or
    backward, with line to load and load to line?

    Thanks for all help,
    Steve
     
  2. Wimpie

    Wimpie Guest

    Hello Steve,

    When you do it right, your Tesla coil resonates like a quarter wave
    resonator. That one needs a ground (the same way as a mono pole
    antenna needs a ground or counterpoise).

    The second reason for a ground is to reduce RFI. Unless you make a
    faraday cage, you cannot avoid RFI completely. I don't know whether
    you are making a pulsed or CW Tesla Coil, but the power involved is
    that high that some interference may occur.

    You can enhance the effect of the ground (and reducing probability of
    RFI) by adding many wires to your ground surface and spread them out
    (radially) in your apartment (in fact you extend the ground with the
    wires). You can even run them upwards (recommended). When you setup
    the coil in a corner, you can use the walls to hold the vertical part
    of the wires. In fact you make a faraday cage without a roof.

    Regarding the filter, the enemy will be so called common mode noise.
    The best thing is to place any filtering for signal and power lines at
    that point where you have maximum metal or mesh surface (so the mesh
    and radial wires are in between the coil and cabling). So all
    decoupling will be done with respect to that point.

    I would install the Telsa setup (coil with ground plane) of the
    ground and decouple all cabling (power supply, control, etc) with
    respect to the ground mesh. All cables (inclusive the safety ground)
    run beneath the ground plane (with wire extension).

    When it is not possible to build the Tesla coil off the ground, I
    would extend the mesh to at least one side and let it go upwards (for
    about 6 ft). The lower corner I would use as a reference for filtering
    (AC decoupling capacitors), entrance of cables and safety ground
    connection (PE wire).

    To further reduce RFI I would provide large ferrite cores to all
    cables (control, power, safety ground) that leave the Tesla coil setup
    (do not put ferrites around the radial ground extension wires.

    I hope this will help you a bit. When you feel something is not clear
    (sorry, I have no image), you may contact me.

    Best regards,

    Wim
    PA3DJS
     
  3. Steve

    Steve Guest

    Sorry for not specifying the type of coil. It is a standard spark gap
    coil, with the basic components: hv transformer, capacitor, spark gap,
    primary, secondary, topload.

    I think I understand what you mean. What size capacitors should I
    use? Based on my calculations, it should oscillate somewhere in the
    500kHz region. Just to make sure, decoupling capacitors go in series
    with the circuit, correct)? I had planned on my control wire to just
    be a switch between mains and the neon sign transformer.

    Also, my transformer states the center tap is connected to transformer
    case and ground stud. I have read to connect RF ground to the center
    of the spark gap as well as the center tap of the transformer. Should
    I do this, and just leave AC ground off the transformer, and will this
    cause any shock danger, being that the RF ground isn't connected to
    earth and could be floating?

    Thanks for the help,
    Steve
     
  4. Wimpie

    Wimpie Guest

    High Steve,

    I sent a reply directly to you, I hope it did reach you, I had some
    problems sending it.

    Best regards,

    Wim
    PA3DJS
     
  5. default

    default Guest

    You really should have a good earth ground. Is that out of the
    question? House wiring is more like an antenna then RF ground - it
    snakes around forever before going to ground at the entry point power
    panel (meter). I sunk a 10 foot copper pipe outside the wall where my
    TC sat with a 5 foot heavy gauge wire to it.

    Most TC's of the size you are talking about would work at a lower
    frequency like 100 KHZ, but that is all a matter of how many turns of
    wire etc.. - the faraday shield/cage would also increase the capacity
    and lower the frequency.

    Put in a safety gap at the input to the transformer. And a couple of
    ferrite chokes with Teflon or HV insulation on the wire for the chokes
    in each leg of the HV - assuming the NST is center tap grounded.

    A safety gap prevents reflected RF from getting into the transformer
    and back into the power lines. There will be SWR problems until the
    coil is in tune . . . AND of course a good power line filter.

    Consider liability - you aren't just talking about TVI, you could set
    fire to the apartment, computers on the same power line or in close
    proximity to the TC may die - my TC's could eat DRAM chips and reset
    the CMOS settings on my computer 6 feet away from a 1 KVA TC.
    (unplugging the computer AND modem AND all peripherals from the power
    line would keep it safe) Other sensitive equipment in the vicinity may
    be ruined (or not) my speakers took some direct hits to the wood cases
    and the mosfet amp didn't suffer. You probably don't want neighbor's
    with pacemakers around.

    The pupman forum is an excellent source for all kinds of TC/HV
    questions. It is operated as a mailing list - you can use your email
    account and regular reader or use a web-based email account
    (recommended). You have to join the list to post but can read
    anything in the archives without joining.

    You download the list mailing via a regular news reader or web based
    email. Read some of the archives before posting so you have a feel
    for how things are done.

    http://www.pupman.com/ If these folks don't know it, no one does.
     
  6. Steve

    Steve Guest

    Thanks to everyone for the replies. I will have to look at the forum.

    I can't put in a good earth gound because I'm on the third floor, and
    the apt won't let me run any wires outside etc... so all I have is the
    mains ground, which I'm sure is probably shared with a few units.

    So, that's why I was trying to find out if just the Faraday cage and
    ground plane would be sufficient. If it's iffy, and I could hurt
    someone elses equipment, I'll just wait until I get a house. I'm not
    out to make anyone elses life difficult.

    Thanks,
    Steve
     
  7. Chris Jones

    Chris Jones Guest

    I would think about finding somewhere other than an apartment to run the
    coil, as it is quite likely to upset the neighbours, and damage their
    appliances etc. Do you know anyone who lives on a farm by any chance? If
    not, the Faraday cage is a good start. If the coil is inside the cage, and
    the ground of the coil is connected to the bottom of the cage, then the
    thing you need to take care of is any wires that go in or out of the
    Faraday cage. At the point where any wire goes through the cage, it needs
    to be connected RF-wise to the cage. It doesn't matter whether the cage is
    connected to the real ground two stories below, but the cage should be
    connected to the ground of your power cable, and the live and neutral wires
    should go into the cage by way of a filter, the ground casing of the filter
    being connected directly to the cage. You will need a very good filter.
    The most important thing about the Faraday cage is that any joins in the
    mesh need to be really well connected, and all the way along, not just in
    one place. It is unlikely that you can get a really good connection
    between pieces of aluminium mesh if it is anodised or painted - you would
    have to get that coating off and it would be difficult. If you can get
    hold of some bare metal mesh then that would be ideal, maybe chicken wire
    (galvanised iron wire with about half inch holes) might be OK if you could
    solder the joins.

    Chris
     
  8. default

    default Guest

    If you go for the faraday cage, try to find wire screen that is
    hot-dipped after it is woven - like chicken fence or "hardware cloth."
    Hot-dipping in zinc insures a connection at the joints between wires.

    A clean wire brush in an electric drill will buff the zinc so it will
    take solder - then you just need something like a large soldering iron
    and some sheet metal to make mounting points for the cage filters.

    A few of the pupman list subscribers are in the business of building
    and displaying large TC's they build cages for them to use inside
    shopping malls and museums with temporary facilities and probably no
    earth grounds

    Richard Hull (one list subscriber) had some of his coils shown in
    National Geographic years ago (it may be July 93) the article was
    about lightening and they sensationalized it but the pictures were
    worth the price. Richard has moved on to "Farnsworth Fusers" but may
    still put in a word from time to time. He was working in the class of
    devices that used "pole pigs" (pole mounted distribution transformers)
    reverse connected to produce the excitation sparks for his coils.
     
  9. Steve

    Steve Guest

    I think you're right. I may just wait until I have a house. I don't
    want to risk damaging someone elses equipment, that wouldn't be very
    responsible of me.

    So, if the cage itself is connected to power ground, which in my case
    is AC ground, as long as all wires going in and out are filtered
    properly, there should not be any RF present on the ground wire?

    Just to clarify, how do I properly decouple the wires going through
    the cage? I'm used to seeing feedthrough capacitors in RF equipment,
    something along this principle? I'm having some trouble figuring out
    which ground is acceptable to connect where, but I will do my
    research, and likely put this project on hold until I have a more
    acceptable location.

    Steve
     
  10. Even a small unit like a Violet Ray may toast all of your electronics - and
    your neighbors' as well. You need a rock solid filter for the AC line
    alone - and good insurance.
     
  11. default

    default Guest

    That makes sense.
    Yes feed through filters are good for this - cap and inductors in one
    device. The NEC power cord filters one sees on computers and such may
    work well. Switching power supplies are up in the TC coil range these
    days, so those types of filters may work.

    A device used in auto radios of old (50's) was something called a
    "spark plate" (had nothing to do with sparks). It was a piece of mica
    (high dielectric constant, low loss) on the outside of the chassis
    often with a rivet or bolt carrying the battery connection to the
    outside world, and a square piece of copper. Only a few pico farads
    but very effective for high frequency noise. The chassis formed one
    side of the capacitor and the copper the other (somewhat harder to
    make safe with mains voltage).

    I assume it got the name "spark plate" from auto mechanics that
    accidentally grounded the hot copper piece exposed on the outside of
    the chassis.

    I've used that idea to decouple gunn diodes used in radar jammers.
    Solder a nut to the thin pcb material and it decouples the power and
    provides an isolated mount for the diode. (gunn diodes won't
    oscillate if the voltage lead has inductance).
     
  12. Steve

    Steve Guest

    Interesting, I"ve never even heard of them.

    So, if I decide to change gears completely, lose the Faraday cage and
    ground plane altogether and put in a proper RF ground, I assume the
    same filtering still applies to the mains wiring to keep RF out of the
    115V supply, it just won't be connected to a cage? I will probably
    put this project on hold until I have the proper facilities, at which
    point I'll do the proper research and figure out how to do it right.

    Thanks to everyone for their help, I think this will wait until a
    later day.

    Steve
     
  13. Wimpie

    Wimpie Guest

    [main part deleted]
    Hi Steve,

    You need an RF-ground, because you are using a quarter wave resonator.
    When you don't make a construction that serves as RF-ground, all
    return current will flow through the power cable (so the power cable
    and everything connected to it serves as RF-ground). This is
    undesirable.

    The reason for the faraday cage or large ground plane is to force the
    E-field lines into your artificially created ground, so the induced
    current can return to the base of the coil via a known safe path. In
    other words the cage or ground plane is the RF-ground. For safety
    reasons you connect your RF-ground to PE. When you did de construction
    correctly, there will be little RF current in the wire that is between
    RF-ground and PE.

    Off course, when you have a large terrain available, you can use
    mother earth as an RF-ground (in same way as medium wave broadcast
    transmitters). www.pupman.com is good site to start with.

    BTW, did my mail arrive?

    Best regards,

    Wim
    PA3DJS
     
  14. Steve

    Steve Guest

    Yes, I received your mail. Thanks for the picture, it was very clear.
    I tried sending you a reply using two different email accounts, both
    undeliverable. Thanks for all your help,
    Steve
     
  15. Chris Jones

    Chris Jones Guest

    You would still need the cage with any scheme where the coil is operated in
    your apartment, both for radiated interference and to stop an arc from just
    directly hitting your light fittings or wires in you walls - that would
    cause some conducted interference!

    Chris
     
  16. Steve

    Steve Guest

    Good point. Just to clarify, by proper facilities, I meant moving
    into a house where it's just my wife and I, but I may still do the
    cage so I have no worries. I am a paranoid person by nature.

    Steve
     
  17. Also use a damn good filter on the AC lines - like an IsoBar.
     
  18. default

    default Guest

    Good, you'll live longer. Not paranoid - safe.

    "there is no jesting with sharp edged tools"

    Your main danger is the primary circuit - it is easy to mistake a gap
    that has stopped firing for one that is turned off, check, double
    check, and put the plug in your pocket while making adjustments with
    one hand only. I good bright rotating beacon wouldn't be a bad idea.

    Folks that use distribution transformers will add a time delay and
    horn sounding before the mains voltage can be connected. Switches
    have be operated from a safe distance, and in sequence.

    Make sure others don't try to work the equipment when you aren't
    there.

    Make sure the tank cap has a way to discharge and know how long it
    takes to reach a safe voltage. (normally only a second or two by
    discharging through the NST) - but all it takes is a loose connection
    to keep it charged, so use a shorting bar when you get ready to get
    both hands in it.

    Use ventilation - these things produce ozone like you won't believe.
    Situate it so that an open window can exhaust the ozone.

    Put a lock on it if you have children, or children will be present.

    Warning signs aren't a bad idea
     
  19. Chuck

    Chuck Guest

    Hi Wim,

    It seems if you use a counterpoise or
    whatever you want to call an RF ground
    plane on the 3rd floor, you will reduce
    lossy current flow through the floor,
    etc., and maybe reduce current flow
    through other wiring.

    But the effect of doing that is to
    increase the efficiency with which the
    resonator radiates. That will probably
    decrease its Q through loading. But is
    radiation what you're trying to achieve?
    I would think that the last thing one
    wants to do is efficiently radiate RF
    energy into everything in the far-field
    as well as the near-field! One can have
    a resonant LC circuit without
    significant radiation.

    Does that make sense?

    I concur with suggestions to use filters
    on the AC wiring. And the Faraday cage
    may provide benefits as well.

    Chuck
    NT3G
     
  20. Wimpie

    Wimpie Guest

    [upper part deleted]
    Hi Chuck,

    As all (most) coilers want to get the sparks at the top of the coil
    (and not at the bottom), you need an electrically quarter wave circuit
    (or LC series circuit where the top capacity + coil capacity forms the
    capacitor). Another option may be a center fed (dipole) structure.
    The problem with these is that air-breakthrough is not that
    symmetrical, so when the air on the left side breaks first, you get
    strong asymmetry that may cause a breakthrough from center to primary
    feeding coil.

    The quarter wave monopole has low impedance (in particular when it is
    a short one [3ft at 500 kHz). All quarter wave circuits (whether
    shorted or not) require a ground or something to pull out the drive
    current for the coil.

    When you make it large enough (or even extend the wire mesh on the
    walls), you provide a return path for the displacement current (and in
    the end a return path for streamers).

    With respect to radiation.
    You are right, some energy will leave as radiation, but it will be a
    very small fraction of the input power. As le/lambda = (about)
    0.0017, the radiation efficiency will be very low (compare it with a
    0.13m monopole antenna at 80m). Over an infinite ground plane the
    radiation resistance will be about 1m Ohm. Assuming about 30 ohms of
    AC resistance of the coil, the radiation efficiency will be 0.004%.
    When someone adds a large top capacitor, the radiation resistance will
    increase, but with factor 4 maximum.

    As soon as air-breakdown occurs, all excess power will be converted
    into heat. I think, the Q-factor is not limited by radiation losses,
    but by conduction losses and at higher power levels Q-factor will
    reduce due to air-breakdown. The same is valid for small loop
    antenna's, Q is limited by heat losses, to much input power causes the
    tuning capacitor to arc.

    You can reduce the far field radiation (and near field to the adjacent
    apartment) by bending the ground plane upwards (on the walls of the
    apartment). When you extend it higher then the top of the coil, you
    made a coaxial quarter wave resonator (with low near field and far
    field radiation).

    So in my opinion providing a defined "return path" for the bottom feed
    current of the coil will result in maximum power in the ionized air.
    When you also want to minimize near field and far field radiation, you
    should construct a faraday cage (or similar structure with wires).

    Best regards,

    Wim
    PA3DJS
     
Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Electronics Point Logo
Continue to site
Quote of the day

-