On Wed, 23 Mar 2005 07:13:45 -0800, Clay Curtiss W7CE wrote:
>Thanks for the link Jim. I'm in the process of digesting the info now.
>
>I assume that I need to wrap all three wires through the toroid (both 120V
>legs and neutral).
Yes
>I have a four wire connection from the generator to the
>cabin with a separate ground wire.
Good. The ground should probably NOT have a toroid on it.
>Since I'm connecting to the primary
>breaker panel, ground and neutral are connected directly together at the
>panel.
Do you have my Power/Grounding white paper? If not, grab it off my website.
There must be ONE, AND ONLY ONE, bond between neutral and ground, and that
should be where the "system" is established. According to NEC, that bond
should be at the generator. From an RFI point of view, it might be better at
the panel.
>Should the ground wire also be wrapped through the toroid, or should
>it be kept separate?
It should not be wrapped. From an RFI point of view, you want to "ground" the
RFI filter of the generator to the body of the generator, and tie the body of
the generator to the ground rod. But the ground rod is NOT the key to RFI, it
is simply there for safety. From an RFI point of view, you are tying the cold
side of the filter caps to the reference point of the source, which in turn
keeps the noise off the power conductors (by shorting it back to the source).
I don't have much practical experience choosing
>toroids so any advice is appreciated. I'm planning on 30A of current, but
>since the total current is balanced, I suspect that a large core is required
>only because of the wire size involved (3 or 4 10-gauge conductors).
That's essentially correct. To facilitate installing the choke (which should
be as close to the generator as possible, but after the line filter), you
could use a short length of smaller gauge wire (or even unjacketed wire like
twisted THHN) for the ferrite choke. Note also that the long run of the power
conductors ought to be twisted pairs. I would use one pair for +120 and N,
another for -120 and N. This gives you a double neutral, and any noise
radiated will be cancelled, at least to some extent.
When you are sizing these conductors, remember that current in electronic
equipment is drawn in pulses at the peak of the cycle. See my White Paper
about this.
>Another topic that is regularly debated is whether to have a separate ground
>rod at the generator and another at the service entrance to the cabin.
By all safety codes and good engineering practice you should, and they MUST
be bonded together (in this case, by the equipment ground conductor that you
are carrying between the generator and the distro).
>I'm
>not sure if that will have any affect on RFI or if it is purely a safety
>issue. It seems that two ground rods would increase the possiblity of
>ground loops which could increase RFI. Any thoughts?
Yes. In an ideal world, ALL grounds should be bonded together at a single
point as close to where things come into your shack (portable or fixed),
including antennas and rig.
The vast majority of so-called "RF in the shack" problems are really "pin 1
problems" -- that is, the mis-wiring of signal returns and cable shields
within equipment. See the applications notes on my website about this, and
also, AES48, an international standard on the connection of shields that I
helped write. You should be able to download a courtesy copy at no charge.
http://www.aes.org/publications/standards/
Jim
_______________________________________________
RFI mailing list
RFI@contesting.com
http://lists.contesting.com/mailman/listinfo/rfi
|