shielding sensitive wires?????

[geoffro]

Hi, sorry to be painful, but i am trying to improve my pedal building and thought that proper shielding must make a difference,
but what is proper shielding,
which wires in a typical pedal should be shielded,
what wire should be used,
how is it done,
now i know i have seen pictures either on this site of rg's or someones
but do you think i can find them now.
not on your nelly
i have searched the forum here and came up with not much.
a little help would go a long way
thanks heaps
cheers
geoff

[Seljer]

use coaxial cable
the wires between the input/output jacks, footswitch and circuit board
hook up the sheild part of the cable to ground

thats as much as I know

[geoffro]

thanks for that,
i have bought wire that i hope will do the job,
it is decribed in the jaycar book as single screened audio cable.
cheers
geoff

[soulsonic]

I like to use RG-174 coax cable. It's usually well shielded and it's a good skinny diameter for use in tight spaces.

[railhead]

I like to use RG-174 coax cable. It's usually well shielded and it's a good skinny diameter for use in tight spaces.

You use that for all the wiring in your peds?

[Ed G.]

Shielding makes a difference in high-gain applications or I guess in applications where a component is bleeding something like a clock signal into the audio path.
The input and output wires usually make the most difference.
In your basic effects it's usually not necessary.

[QSQCaito]

Hey! Let's say something important. If you use shielded cable, that you ground the shield part, ground only one extreme, not both, to avoid ground loops if i', not wrong

[geoffro]

thats the sort of ifo i need
so if i shield the input from the jack to the board and the output from the board to the output jack as a minimum i need to conect one end of the shielding to the ground on the output jack yes??
the shielding ends that go to the board are just cut off??
cheers
geoff

[Pushtone]

the shielding ends that go to the board are just cut off??
cheers
geoff

There are a couple of places that you can connect the ONE end of the shield, but basicaly, Yes.

The idea is that any RF or EMI picked up by the shield will drain to ground, preferably to the output jack ground and not the board ground trace.

Did you see this thread?
http://www.diystompboxes.com/smfforum/index.php?topic=53352.0

In the fist post John Lyons links to his shielding tutorial.
Very good nuts and bolts stuff on how to actually accomplish it and a little on why.
Check out his tutorial in that thread.

[geoffro]

ah thats it, thankyou very much pushtone
that is exactely what i was looking for.
funny that it didn't come up when i did a search before.
thanks again
cheers
geoff

[Processaurus]

Shielded wire is enough of a pain that I usually avoid it unless there are problems, or there is a long run of wire (>4") that goes from a high impedance source to a sensitive, high impedance input, like opamps.  You see, the amount of energy noise that gets picked up by a wire compared to its signal has is lessened if more current is being sunk into the signal from a low impedance source.  Imagine noise can put .01 mA (no idea if that's a reasonable example, BTW) on a wire, and the signal is from a weak source, that also puts out .01mA.  The noise will have as much energy, and as loud as the signal.  Like two tiny streams meeting, one's dirty, ones clean. Now say we can lower the impedance of the thing making the signal, and source more current into the signal, 1mA.  The noise now only has 1/100 of the energy the signal has, like a dirty little stream meeting a big clean river.

Guitar pickups are high impedance (unable to put much current into the signal), electronically buffered things are low impedance, so a really good test to see if something needs its input wire shielded is to try it, in its box, connected directly to the guitar, or with a buffered pedal (like a boss pedal) in between, and if there is an improvement with the buffer, that means things can be helped with a shielded input cable.  If not, then the input isn't the problem and shielding it won't help.

Another good technique with problem pedals is have it on, plugged into an amp, and move wires around with something that isn't conductive, lots of times changing how a wire is routed is as good as shielding it.  Running wires next to the box, which is grounded, is like half shielding it, as is twisting a grounded wire around it.

Big EH pedals in particular, that aren't particularly noisy, get away with murder, like running super long input wires right over clock sources  down to the switch.  I've never seen shielded cable in an EH pedal, come to think of it.

[soulsonic]

I like to use it for the input and output wires whenever practical to at least cut the chances for radio station interference and to help avoid output-to-input EMF feedback issues. Usually, the only time I don't use shielded wire is when they are VERY short; like, less than an inch. Within the circuit itself, I usually don't bother with it, unless a specific issue comes up where it needs to be used on a certain problem wire.

In an amp, I'll use it mostly just on the input, but sometimes if a signal wire is very long - like going from the board to a line out jack halfway across the chassis, then I'll shield that too. I learned early on that there's no good to be gained from shielding EVERY wire.

[GibsonGM]

I've found that 99% of my "hiss" is from opamps/trannies in high gain applications.  No amt. of wire shielding has changed that, it's thermal noise and noise generated in biasing resistors and the like.   I only go to shielding when as Proc. says, there is a problem, such as a FF picking up holy rollers from Indiana ;o) 
Clock circuits are notorious offenders, and I will twist signal wire with a grounded 1 end wire if passing near them, and this works very well!  I'd like to try out some of that RG-174, tho, sounds useful...