Shielded wiring... where to use?

[Snufkinoob]

I've read up on a little regarding shielded/screened wire and the general consensus is that you only need a little, specifically at the Input connections, but I've seen a few boutique fuzz pedals and even board diagrams depicting more copious use that cover all the Out points, except Board Out to Vol Pot. (Here for example: http://2.bp.blogspot.com/-L4b9kcunKEs/T13pi-Rcn8I/AAAAAAAABEc/TFdrwq4Tx2E/s1600/WEM+Rush+Pepbox+PNP+GE.png)

For my fuzz builds, It's probably a little excessive as the runs aren't long, but everything is close together, and if I can eliminate any potential noise then great.

So I'm thinking shielding from:

Input Jack to Foot switch (GND at Jack)
Foot switch to Board In (GND at Foot switch)

But I'm not sure about the Out connections:

1)Vol pot to Foot switch
2)Foot switch to Out jack

I'm assuming in that example above, that since the Vol pot is essentially a variable resistor, and so part of the circuit, it doesn't need shielding, but what about those two Out points? Overkill, negligible, useful, essential? One more so than the other?

[R.G.]

Every electronic part has both a major characteristic, or several of them, and a long list of side effects, special cases and imperfections. Even wire.

The short answer to your question is that it's very, very hard to give a succinct answer to you. Shielding of all kinds is needed when there are interfering signals around and when the circuit details are such that your circui will pick up the interference and have poorer results because of it. That doesn't help you any, because what happened in your head as you read that sentence is that you were wondering "what interfering signals and how would I know?" and "WHAT circuit details?"

The simple answer just pushes the questions back into more questions about more complicated stuff.

The reality is that wire has not just (low) resistance, but also inductance and capacitance. If there's another signal nearby (including the same signal the wire is carrying, perhaps amplified in voltage or current and phase shifted) the wire can pick up that signal by  virtue of its side-effect inductance and/or capacitance. This may be incredibly, crucially important or not matter at all depending on what the rest of the circuit is that the wire's connected to.

IF the wire picks up interference from other wires or the rest of the circuits, and IF your circuit is susceptible to it, and IF a conductive shield on the wire excludes the interference, it will be great. If the interference isn't capacitively coupled, or if the circuit isn't susceptible, the shield is a waste of time and money.

And now we're getting nearer to the internet wisdom you've seen. High impedance inputs are more susceptible to capacitive coupling, and wires that are packed densely next to each other are even more susceptible.

There are some very few pedals that need shielded wire in certain places. But these are few and far between. My take on this is to ignore shielded wire until you can prove that you need it.

[GibsonGM]

Bada-bing ^

I find it *may* help on inputs, where you have a low level signal picking up environmental noise before amplification. Esp. in actual AMPS rather than most pedals.  If and when I do it, it's more for peace of mind than any need I could prove.

Once amplified above the noise floor, I see no real need to have shielded signal wires.   Unless you've done something to create that magical condition R.G. is describing, where you've made an antenna....that is a design issue, not something you 'cure' with shielding.  I think that analyzing what is in your circuitry (transformers? Other EMF sources?) and laying designs out properly are a much better way to spend some time!     2 cents...

[Snufkinoob]

Thanks for the replies.

What made me consider using it was that I used to have a Roger Mayer Voodoo Axe fuzz pedal, which had no noise issues until I went to sell it at a local shop, and when it was plugged in, there was a tremendous amount of noise no matter what guitar, amp, power supply, battery etc was used, so we put it down to the shop environment, small space packed with gear, fluorescent lighting nearby etc.

So now I'm building fuzz pedals myself, I want to try and eliminate that particular potential side effect, and shielded wire is all I can think of as a possible solution. I've made one up without any shielding, and will put some on the next build, take them in and see if there's a difference between the two in that environment to see if I need to continue with it or not.

But beyond that (I don't know if that specific issue could be resolved with shielded wire, it's just a guess) I'm interested in the actual usefulness of doing it anyway. Seems like it's one of those extra little flourishes with fuzz pedals solely to mimic the original builds and/or add to the aesthetics of the circuit.

[merlinb]

What made me consider using it was that I used to have a Roger Mayer Voodoo Axe fuzz pedal, which had no noise issues until I went to sell it at a local shop, and when it was plugged in, there was a tremendous amount of noise no matter what guitar, amp, power supply, battery etc was used, so we put it down to the shop environment, small space packed with gear, fluorescent lighting nearby etc.

If it is already in a grounded metal box then shielded wire is very unlikely to make any further difference to external noise pickup. It's shielding within shielding!

[GibsonGM]

^^ Yes.       

There WASN'T any shielded wire in the original builds, Surf!     Maybe just cloth-covered stuff that looked like it.    Like Merlin says, the metal box is grounded, and acts like a Faraday cage (ok that's extreme, but same principle).  If the circuit is designed and built properly, it should not make a LOT of noise.  Yes, some high gain stuff can hiss, and if you aren't careful it can feed back, esp at high gain settings,  etc etc --  however, it really seems like something else was going on with your fuzz. 

Perhaps the connection TO the box was corroded, dirty and the like?  That kind of thing WILL contribute to noise, in a big way.  Sometimes they make that connection via a pot, and it ages and conduction stinks, leading to lots of noise.  If a nearby flourescent/transformer is bothering your effect, it might be getting into the chain another way, too - maybe thru the cord, or what have you.   You should have to get RIGHT UP on a transformer or something to have its field cut into your stuff enough to hear it.    Pickups (single coils) are another story - not shielded, they are antennae...and dirty pots can make intermittent connection, sounding a lot like ground problems...

I wouldn't worry about shielding INSIDE your effects, myself...but wire placement, yes.

[Snufkinoob]

^^ Yes.       

There WASN'T any shielded wire in the original builds, Surf!     Maybe just cloth-covered stuff that looked like it.   

Some might not have, but some definitely did...

"All the early Tone Benders, right up too the introduction of the MKIII, also relied heavily on the use of shielded cabling throughout."
http://www.stompboxes.co.uk/MKI_5.html
http://www.stompboxes.co.uk/Wifey.html

Although tone benders have that weird mounting method, so in general, I'm sure you're right, most stuff might not have had it, but some certainly did. (Doesn't look like Fuzz Faces used it though)

I'm certainly not arguing the point that shielded/screened wire is useless, there's plenty of info out there that echoes the above comments about it being fairly excessive, and if that's based on first hand experience then great, But there are instances like this one: http://hownotomakeapedal.blogspot.co.uk/2011/02/fuxz-faxtory.html where he uses a screened wire on the input to eliminate radio interference with success. Obviously, there are always going to be loads of variables, even if everyone where building the exact same circuit, and this example may be the exception to the rule, but some people do seems to get results with it. On the other hand, it seems like for the most part it's aesthetic for "completeness".

Always interesting to hear first hand experiences with it though.

[GibsonGM]

Yeah, not that NONE used it, but Mayer didn't as far as i know.  I didn't think folks needed to open the unit to see the shielded wire inside, and get a warm feeling from that, lol!      You CAN certainly use it if you want!  Don't think it will harm a thing, as long as you only ground one end.   I find it cumbersome, and don't like having to plan out where my grounds are going to connect.    I'm not on a soapbox about this, by any means...hey - it's YOUR build.    IMO, If someone 'gets results' with it, it means their wiring layouts are lacking, all other things being equal....

In a tube amp, where you have much higher signal voltages, properly shielding your wiring is certainly appropriate.  Especially from input to first stage, where you may be near a later stage and get some very nasty coupling.   It does have uses, don't get me wrong...only that if you take 99 of 100 of our little 9V builds and shield IN and OUT, I am willing to bet you'll hear *very* little difference when the circuits are in a metal box that is grounded.   If you start running outs and ins next to each other, parallel, you CAN couple and get bad results, which is where the shielding "trick" probably arose.     Good layout prevents this....just sayin'...the old chopstick tells no lies. 

[Processaurus]

RG is right, noise in analog circuits is a broad subject, but some practical guidelines I follow for when to use shielded cable or not, is for high impedance connections, such as input wires  to the circuit, or inputs to opamps, when there's A) lots of gain, like distortion or compression effects, B) a long way to go with the wire, like further than ~6", or C) there's a source of noise in the box that can be picked up capacitively, like an LFO circuit, or microprocessor, things with a clock.  I'll use it preemptively if there's any combination of the three.

Good layout can alleviate the need for shielded cable in many cases.  Good layout can keep wires short (like from input jacks to bypass footswitch to board, and short wires from pots and switches to the board), and can physically separate sources of noise, like LFO, digital clock, AC power input, from high impedance inputs or offboard wiring.

One alternative, when noise problems present themselves, is if there is a section of the circuit generating the noise, is to contain it with a little shield, like a grounded box made out of a piece of tin can, that can cover the offensive circuit part.  I remember seeing this idea in the Boss tuner pedal, they put a little tin hood around the input jack to shield it from the processor that was close by.