In a class I once taught:
1) Keep inside fields inside (the Faraday cage).
2) Keep outside fields outside (the Faraday cage).
The result is that fields generated inside the enclosure stay inside and
can't get out and interact with outside fields or radiate and, of course,
vise-versa. Sounds pretty simple, but in practice, it takes some thought
for each individual application. The outside of the chassis supports
external fields and currents, if things are properly done. So, involvement
of the 'green wire' should have no effect on what goes on inside the
appliance or radiation of energy inside the appliance.
3) The 'green wire' or 'ground' wire is just that and only that. It's for
safety and only safety. The final RFI/EMC solution should be independent
of the
'green wire' or conductors connecting individual pieces of equipment. I
often hear from amateur operators on the air, 'well, do you have you
station 'grounded'? I prefer the term 'earthed' instead of 'grounded'. My
reply: If connecting an earthed conductor changes or modifies 'RF in the
Shack', part of your antenna is missing. Think about it. Besides, an
'earth wire' some 33-feet long on 7.0 MHz is an open - a shorted (earth
end) 1/4-wavelength results in an very high impedance resulting in no
connection at 7.0 MHz to earth. However, its good for safety, and properly
installed, can help with lightning (remote strikes only) remediation.
4) Remember AC power currents have a return be it a conductor or earth.
RF fields have only a 'return' embodied in 'free space'. For example, the
Crab Nebula emits copious amounts of broadband RF energy into space, but
there is no 'ground' involved. The ISS has no problem communicating with
earth, but the RF fields responsible for that communications do not require
a 'ground'. Electromagnetic energy, RF et.al., require only the "magic" so
well condensed in the four Maxwell equations to radiate through space. No
'ground' (GGGggrrrrrrrrr......) is required!
Dave - WØLEV
On Sat, Mar 2, 2019 at 3:11 AM Sean Waite <waisean@gmail.com> wrote:
> That does raise a question. If the chassis of the computer is bonded to the
> electrical ground as well as the station ground bus, does that increase
> your chances of ground loops or potential increased noise from dirt on your
> house power circuit? I have my PC bonded via bolt through one of the van
> vents on the chassis. If it's doing much I can't tell, there's so much else
> going on at my house that my noise floor is pretty high anyway.
>
> 73,
> Sean WA1TE
>
> On Fri, Mar 1, 2019 at 8:27 PM Roger (K8RI) <k8ri@rogerhalstead.com>
> wrote:
>
> > Bonding to a computer?
> > There have been many a tale of RFI from computers and no few of station
> > RF getting into the computer.
> >
> > A GOOD computer will not generate RFI!
> >
> > HOWEVER there are many big name brands that do. Price and brand name are
> > no guarantee you'll get a good computer.
> > Bonding for safety is a good idea. You can't always keep lightning out,
> > but if all lines in and out rise even to thousands of volts at the same
> > time it's unlikely any damage will occur.
> >
> > Computers are a commodity and if a dime can be saved here and there,
> > it'll soon add up to hundreds of thousands of dollars...or more
> > If I were to undergo a project like this, I'd pick the band, or bands,
> > with the most interference and use that as a base , or starting line,
> > but don't neglect the other bands and after adding a device to the
> > bonding, check for any change from, or to the computers.
> >
> > NOTE: The green wire in your electrical wiring and in the power cord
> > should be bonded to the power supply enclosure. They aren't always.
> >
> > Consumer grade computers/PCs tend to have plastic cases (IE: No
> > shielding) and inexpensive power supplies ($10 - $15 range) in stamped
> > Aluminum cases with little or no RFI suppression. If accessible I'd
> > ground directly to the Al sub chassis. The Serial and USB ports do have
> > a ground although tiny. I'd not depend on a ground through #22 or
> > smaller wire.
> >
> > Custom built and gaming computers tend to have metal cases such as
> > small, mid, and full size tower cases, typically made of steel. At 180#,
> > I could stand on any of mine. They are strong, but they are also large
> > and heavy and not inexpensive. The power supplies may run as much as
> > 1200 Watts. I use 850 W supplies that even have power factor correction
> > built in and generate no RFI. BUT computers often connect to phone
> > lines and the station. Add to that the AC line and the network you have
> > at least three additional routes for lightning into the station. Good
> > power supplies run from around a $100 to well over $300
> >
> > I haven't read the entire thread so I apologize if I'm repeating things
> > already posted
> >
> > I doubt many build their own computers, but now days it has become quite
> > simple, at least for the mechanically inclined. They are modular and
> > relatively simple to plug together, But RTFM as there are things like
> > the motherboard, CPU, and memory that must be compatible. The underside
> > of the CPU is a mass of tiny, fragile, pins and no place for heavy
> > hands. Install with care. There is a tiny mark on the CPU and a
> > corresponding mark on the socket for alignment. It should drop right in.
> > If any force is needed, you have a problem.
> >
> > 73, Roger (K8RI)
> >
> > BTW, many (certainly not all) new computers now include a serial port.
> > Often there is a small socket on the motherboard. In that case you have
> > to purchase a small adapter plate that has the serial port and a plug to
> > fit the motherboard
> >
> >
> >
> >
> > On 2/28/2019 2:34 PM, N4ZR wrote:
> > > I want to try to manage RFI in my shack, among other things, by
> > > bonding all chassis together, including my shack computer, but the
> > > question has come up - where can I find chassis ground on a PC without
> > > serial or parallel ports? Is there a design standard, such as
> > > connecting the power supply chassis to ground buses on the
> > > motherboard, that makes this easy or does it have to be figured out PC
> > > by PC?
> > >
> >
> > --
> > Roger (K8RI)
> >
> >
> > _______________________________________________
> > RFI mailing list
> > RFI@contesting.com
> > http://lists.contesting.com/mailman/listinfo/rfi
> >
> --
>
> Sent from my Motorola DynaTAC 8000X
> _______________________________________________
> RFI mailing list
> RFI@contesting.com
> http://lists.contesting.com/mailman/listinfo/rfi
>
--
*Dave - WØLEV*
*Just Let Darwin Work*
*Just Think*
_______________________________________________
RFI mailing list
RFI@contesting.com
http://lists.contesting.com/mailman/listinfo/rfi
|