Tony,
You clearly know what you're talking about and I can't fault your logic and
practical advice. I also suspect that most builders are concerned about the
right problem ie. radiated RF noise. However there is another problem which
you've left out, possibly to keep things simple, that is impedance.
Some consumers draw current interrupted at high frequencies (commutating DC
motors, transponders, switches, electronic ignition systems) and these
generate significant levels of RF in their supplies' return lines. The
RESISTANCE of the line back to ground may be low, but the IMPEDANCE of only
a few microhenries at umpteen MHz, becomes large - and it's easy to have
lots of microhenries in even a short line in the plane. This gives rise to
RF noise on the consumer's local ground, although the grounded point at the
busbar is quiet. When the consumer's local ground is also used as a
reference ground for, say, a microphone input, then this noise is introduced
on to the signal (we call this series mode noise). On most good quality
avionics, the designer will connect two types of capacitor across the supply
at the consumer's power input - one large electrolytic, whose size is
limited only by the surge current capabilities of the supply's fuse, and a
second, say 0.1 microfarad ceramic or similar, with good RF performance.
(this is because large electrolytic capacitors usually possess quite high
self inductances making them less effective for RF smoothing for the same
reasons I've tried to explain above). This normally does the trick. However
all avionics are not good quality and in some cases (eg. the old Narco
handhelds I know for certain), don't have this feature. There may be other
handhelds suffering from the same deficiency since they are all short of
space inside.
The solution is fortunately quite simple - connect a 50 microfarad capacitor
the right way round, in parallel with a 0.1 microfarad capacitor, to the
consumer's supply, right at its power terminals.
I had a similar interference problem on a two stroke Rotax powered
microlight some years ago - couldn't receive anything without the squelch
turned down, then it was unintelligible. I tried all sorts, including
antenna re-wiring and re-siting but to no avail. In the end I screened the
HT leads, screen grounded at BOTH ends, and enclosed the ignition coils in a
die cast box electrically bonded to the crankcase. That got the &$%&
*.
After that there was no difference between the squelch setting engine off
and full chat.
And there's another real problem with RF energy - it radiates from leads
coming out of the box, not just the box itself. I have a wonderful Lowrance
Airmap hand held GPS; wonderful that is except for it radiating interference
---From the antenna lead and antenna as well as the unit. I could screen the
lead of course, but I can't screen the antenna.
Isn't life a bitch sometimes.
Dave Simpson
----- Original Message -----
From: Tony S. Krzyzewski <tonyk@kaon.co.nz>
Subject: Re: Bundling of coax cables
>
>
> So a "negative" bus bar, with about a dozen connections, should be
> thrown and replaced with a single point connection for all the "circuit"
> return wires ,as well as a single point connection for the "grounding "
> wires? (Just convince me please that circuit negative wires are the same
> or different to "ground" wires?) Regards Bob
>
> A negative bus bar is a single point ground if all of your
> ground/negative wires terminate on the bar. If all your wiring
> terminates on this single bus bar then you have the best form of ground
> return other than terminating all of the wires at the battery.
>
> The key is to have all the circuit return wires terminate at as few
> (preferably one) point on the aircraft and to have minimal resistance
> (read big fat wire)to the current source return (battery). This single
> point can be a bus bar, a bolt or a plate with connectors. The point
> here is that the resistance between the individual return line
> termination points should be very, very low and a block of brass bus bar
> does this very nicely.
>
> Circuit negative wires are the same as ground wires which are the same
> as return wires.
>
> On a tin can aircraft the skin and structure effectively can act as a
> very large single point ground with the return current flowing over the
> entire skin. As it's a massive area electrically the current flow per sq
> cm is miniscule. Composite aircraft don't have the luxury of having this
> cheap return path so we have to build an individual return path for each
> and every circuit. Unfortunately electricity has the habit of finding
> the easiest path to ground and, if the returns form multiple paths to
> grounds and there is a variation in resistance of those return lines,
> then voltage differences can be induced between the ground points as the
> current flows between them. Bringing all return lines to a single point
> and ensuring that the single point has a low resistance to the aircraft
> true ground point removes the problem.
>
> I have built several large computer controlled model railway systems
> over the years which have amazing parallels to wiring the aircraft.
> Failure to bring the grounds back to a single point in those systems can
> induce so much electrical noise that the microelectronics simply fails
> to work.
>
> Tony
>
|