Another thread about grounding

reluctant-builder

Hero Member
Messages
816
I hate soldering to the backs of pots. All of my guitars have come this way from the factory, and so if I even want to undo it, I've got to apply some head to the pot housing. The solder in my Swede must be from Hell; it doesn't melt. It's not my iron; it works fine in every other application and my tech, who uses a more powerful iron than my 40W, also had trouble getting the factory solder to melt. The pots sure get hot, though. Oh, joy.

Cagey is big proponent of Solder Lugs. I don't blame him; they seem like a great idea. I've got four that he sent me, lying in wait to use on my Jazzmaster build. But I don't want to buy more of them unless I've got things enough to order that would justify the shipping cost ... so I'm looking for an alternative grounding method to Cagey's alternative grounding method.

No backs of pots. No solder lugs.

So, what, then?

tfarny mentioned "Star Grounding" and the concept sounds interesting. I'd be all set to do it, but I'm kept at bay by that every wire in my Swede seems to have a hot portion and a bare, ground portion. The wires are, in fact, two wires, one with white insulation, one with no insulation, both wrapped in a gray outer layer.

This means that every connection in my guitar has a hot and a ground.

If were to remove all the wire and try to run new 24 AWG wire in its stead, which is just one spool of copper wire with plastic insulation -- no bare wire companion -- how would I effectively ground the guitar, with the absence of these bare wires?

I've attached what is probably absolute nonsense, but I am trying to illustrated how I'd want to join all things that would normally ground to the back of pots and have them run to the pup toggle, the tone filter toggle and to the jack sleeve.

Edit: I modified the diagram again because, after disassembling the wiring, I could take out the tone filter switch and see now that there is a jumper wire between the two hot outer terminals.
 

Attachments

  • bs-wiring-diagram.gif
    bs-wiring-diagram.gif
    24.3 KB · Views: 334
That braided outer wire that surrounds the inner signal wire is a shield. It gets grounded so if any interference is encountered, it gets conducted to ground and the signal wire never sees it. You can't run a separate wire and call it a shield. The signal line has to be inside the shield in order to be shielded. The shield also acts as the reference point for the signal, and the return line that completes the circuit.

 
I don't like grounding to the pots either.  It seems everything that wire is supposed to attach to is made to do so with an eyelet of some sort.  The back of the pot doesn't yet it's entirely accepted as such.  To make matters worse things don't like to stick there and excessive heat is bad for the plastic guts.  Go figure.  The solder lugs are a great thing, but they are not the only thing.  A 5/16" crimp lug is basically the same thing and can be found at most hardware and automotive stores.  Most Wal-Marts sell them 24/7.  I've used them before.  With smaller guage wire, I strip of twice as much insulation, twist it, fold it over on itself, then crimp the hell out of it.  For extra piece of mind, after crimping hold the soldering iron tip to the crimped area until you see the finish change from the heat, then place the solder where the wire goes into the lug.  It sucks the solder right in like you were soldering with a torch on copper pipe.

You can also make your own solder lug by drilling a smaller hole in a fender washer.

Here's something I made for one of my builds out of an electrical box cover.  I cut it, bent it, and drilled holes for ground wires from the pots, pickups, jack, and bridge.

P1140645.jpg


P1140641.jpg


P1140640.jpg

 
Super Turbo Jack Ace Deluxe Custom said:

Ooh, that's nifty.

I saw someone's Gibby with a metal plate in the cavity, to which the pots were affixed with the grounds soldered to the plate. Silly bugger had it removed and went with the back-of-pots method from a custom kit.
 
Cagey said:
That braided outer wire that surrounds the inner signal wire is a shield. It gets grounded so if any interference is encountered, it gets conducted to ground and the signal wire never sees it. You can't run a separate wire and call it a shield. The signal line has to be inside the shield in order to be shielded. The shield also acts as the reference point for the signal, and the return line that completes the circuit.

It's not actually braided wire in my Hag. It's just regular, silver-colored wire. There are two lengths, one is covered in white rubber, the other is bare, but both are wrapped in gray rubber.

I just don't see the point in grounding pot to pot to pot to pot to switch to switch to jack. That's how the Swede was wired before I took all the components out. What a mess of needless extra wire and gobs of needless solder.

Why not run individual ground wires from each individual component and have them converge at one point from which runs the wire to the jack sleeve that sends the signal to ground?
 
Rather than a braid, there may by a foil wrap shield under that outer jacket. Running the bare wire along with the signal line automatically connects the foil, so it acts like a shield. It's not uncommon. You see pickups wired with that stuff all the time. You get four conductors, a bare wire (sometimes called a "drain"), and they're all wrapped in foil with an insulation jacket over that.

726px-Shielded_wire_4F.jpg

Of course, there can be more or less conductors, that's just an example.

If you run separate signal and ground wires, you're going to end up with a very noisy guitar, even if it has humbuckers. You can ground things in a star, a triangle, a plate, to lugs, whatever. What's important is that everything gets grounded and no signal wires run around unshielded. A certain amount of exposed signal line is unavoidable, but keep it as short as possible.

 
Cagey said:
If you run separate signal and ground wires, you're going to end up with a very noisy guitar, even if it has humbuckers. You can ground things in a star, a triangle, a plate, to lugs, whatever. What's important is that everything gets grounded and no signal wires run around unshielded. A certain amount of exposed signal line is unavoidable, but keep it as short as possible.

OK. So, single conductor wire is therefore pretty useless? That bums me out, since I bought 75' of it (it only cost $7.50 bucks, which might be exorbitantly expensive at $0.10 a foot, but still).

I take this all to mean that I need to have two-conductor wire. My pickups have only the two wires, hot and ground, but I didn't gather that I'd absolutely need every hot wire to run concurrently with a ground wire all wrapped together.

Any recommendations on the best place to buy two-conductor wire? Braided or otherwise?

Edit: Could you explain WHY running the separate ground would make the guitar noisy? I totally get you're saying THAT it will happen, but I'm curious to the actual cause of the noise and why the scheme of running ground wires in separate courses from the hot wires is unreliable.
 
reluctant-builder said:
OK. So, single conductor wire is therefore pretty useless? That bums me out, since I bought 75" of it (it only cost $7.50 bucks, which might be exorbitantly expensive at $0.10 a foot, but still).

I wouldn't call it "useless". It's still wire. It just may take you longer to use it up. You will likely find uses for it. For one thing, sometimes short interconnections will defy the practical use of shielded wire, and for another guitars aren't the only things in the world that need wire. At $7.50, pretend it was a 6-pack you already drank and pissed away <grin>

reluctant-builder said:
I take this all to mean that I need to have two-conductor wire. My pickups have only the two wires, hot and ground, but I didn't gather that I'd absolutely need every hot wire to run concurrently with a ground wire all wrapped together.

It's best.

reluctant-builder said:
Any recommendations on the best place to buy two-conductor wire? Braided or otherwise?

The usual suspects for guitar supplies all sell it. Warmoth, Luthier's Mercantile, Stewart MacDonald, et al. It's also available in a variety of configurations from any electronics supply house, but they're unlikely to sell you short (less than 100') lengths. They all get a pretty penny for the stuff, but you'll find if you try to buy a standard reel of it, you'll have to sell a kidney to pay for it. StewMac might be the best bet. Theirs comes in 25' lengths. Warmoth's is less expensive, but there's a lot less of it.

reluctant-builder said:
Edit: Could you explain WHY running the separate ground would make the guitar noisy? I totally get you're saying THAT it will happen, but I'm curious to the actual cause of the noise and why the scheme of running ground wires in separate courses from the hot wires is unreliable.

Well, it depends. First, running them separate doesn't make the guitar noisy, it prevents it from being quiet. I know - semantics - but sometimes it helps to think about things differently.

Second, it's actually perfectly fine to run them separately if you're using regular single coil pickups, because there's not much you can effectively do to quiet things down. You'll notice on old Strats and Teles they don't shield anything. All the wires are run separate because it doesn't matter. But, if you're running humbuckers, the only place EMI isn't cancelled out is in the wiring. So, you have to shield it.

Finally, separate wiring isn't "unreliable", it's just susceptible to induction. So, let's talk about that a bit.

Any time you have an electrical current flow, a magnetic field builds up around it. The inverse is also true - any time you have a magnetic field passing through a conductor, you induce an electrical current flow. When it comes to both guitar signal and noise, this basic little law is where it all comes from.

We are surrounded by magnetic fields from a wide variety of sources, not the least of which are the power lines that enable our entire society. So, any conductors are going to have some amount of current induced into them. In most cases, it's so small an amount that it can usually be considered non-existent. But, when you have an amplifier that's trying to make extremely tiny signals big enough to do things with, you'll find it has a super-ultra-mega sensitive input. Ever touch the tip of a cord that's plugged into an amp? That noise you hear is what your body is picking up from around you. It's barely there, but the amp is just making it a LOT bigger. Your guitar's wiring is picking up the same thing. So, you have to shield it.

What shielding does is provide a conductor for all that surrounding magnetism to generate some current in. Tie the shield to ground, and that current gets drained away. Anything inside the shield won't see it, so no current gets generated for the amp to see. There are about a million other details, but that's it in a nutshell.

It's key that the signal line be inside the shield, though. If it's outside, the shield still drags current to ground, but it doesn't shield the signal line from it. Think of it like this: if you want to stay warm, dry and useful, you don't wanna be submerged in the ocean next to a submarine, you wanna be inside it. Otherwise, the submarine will still be warm and dry inside, but you'll be useless. The sub is your shield against the ocean.
 
Thanks, Cagey, for the detailed reply. The analogy about the submarine and the ocean does make sense on its own, and I can sort of relate it to the ocean of magnetism through which the current must flow, in my guitar. But I still don't understand why this:

bare-white.jpg


Is more desirable than this:

red-black.jpg


The wire in the first photo doesn't have any aluminum shielding like the image of four conductor wire you posted. I stripped back that outer insulation to check. It's just gray rubber around a bare silvery wire and a white-rubber-coated silvery wire.

The second photo is supposed to indicate the "hot" and "ground" running independently. I gather you might want to shake either head or fist at me and cite the ocean, again. But I don't see any discernible difference between the two except that the wires in the first example are closer together and one of them does not have individual rubber coating ... and I still don't quite understand how only those differences mitigate the ocean from seeping in, since there is no shield that I can see.

Finally, I've attached a better representation of my "star" ... it actually looks like a star now, instead of a game of Qix.
 

Attachments

  • bs-wiring-diagram.gif
    bs-wiring-diagram.gif
    25 KB · Views: 8,750
Shielded cable is manufactured a number of different ways. The shield can be a wire leader inside a foil wrap, it may be a braided wire sheath, it may just be a bundle of thin wire wrapping, etc. The important part is that it encloses the internal lines it's wrapped around. Depending on how the shield is made, it will be more or less effective against different frequencies. Think of it as a filter or a sieve. For instance, some braid is tighter (closer knit) than others, all of which are usually tighter than a loose wire wrapping, none of which are as impervious as foil, but all of which are stronger.

The first piece of cable you show above appears to use the loose wire wrap method of shielding. If you were able to peel back the outer insulation jacket covering it without disturbing anything, you'd be able to see just a bunch of thin bare wires wound around the inner conductor. When you strip back the outer jacket, you see something like this...

cabletalk_01.jpg

Once the outer jacket is stripped back, the exposed loose wires are usually twisted into a singe strand to facilitate connection to [whatever - usually ground].

Because that inner conductor (wire) is totally enclosed by the shield, if the shield picks up any signal, it drains it to ground. The inner wire is unaffected, and any signal on it doesn't get degraded or have anything (noise) added to it.

The second set of wires you show above are separate from each other. The hot wire is not protected from outside interference. If you find yourself in even a weak magnetic field, that wire is going to get corrupted by currents induced by the magnetic field. That interference ends up on the amplifier's input, where it gets amplified to the point you can cuss at hear it.
 
Ah! Brilliant. Thanks, Cagey. I totally understand now. What I was thinking of as just a bare wire running concurrent to the insulated wire is not, in fact, concurrent but enveloping.

It makes sense now why I saw these "bare wires" soldered to the grounding terminals at every hot juncture. When I visualized them running just in double courses, I didn't fathom that the bare wire actually encompassed the white ... becoming the submarine of your analogy.

Thanks again. This elucidation is most helpful.
 
You're welcome. Picture's worth a thousand words, isn't it? <grin>
 
So, any qualms you may have about "vintage" wiring aside, what do you think of this mock-up of a potential wiring scheme?

I've tried to depict braided, shielded wire with ground wires soldered to the braided shield at the typical grounding points, while still avoiding soldering to the backs of pots.

Edit: Whoops! Had the tone ground coming from the wrong lug. Fixed.
 

Attachments

  • 50s-wiring-braided.gif
    50s-wiring-braided.gif
    29.7 KB · Views: 3,128
I'm assuming that lower pole on the upper switch is for an identical pickup circuit? Other than that, it looks like it'll work.
 
Yes, indeed, it's for the identical pup circuit.

It looks, to me, like I've got some grounding loops in there ... I've read the mash about how they're not a big deal, though many still obsess over them. I can't recall individual opinions of ground loops, however. What's yours?

Edit: I think I eliminated the loops ... the ones that I saw, anyway. Image attached.
 

Attachments

  • 50s-wiring-braided.gif
    50s-wiring-braided.gif
    29.6 KB · Views: 297
There are as many ground loops possible in a typical electric guitar as there are aircraft carriers in a puppy's tummy.

In order to even have a ground loop, you need two different grounds that are at two different potentials. Since you're connecting the guitar to another device through a two-conductor cord and nothing else, and one of those conductors has to be ground and the other "hot" for it to even work at all, where are we going to get this other ground to loop to?

I don't know how this ground-looping guitar wiring myth ever got started, but it's certainly one of the sillier ones out there.
 
On my Soloist, I used Belden (can't remember the number right now, but it's the same cable that Duncan and Dimarzio use for their pickups) shielded cable from the volume pot to the output jack. The distance from the Volume pot to the selector switch is so short that I just used standard 22 AWG copper wire and twisted them around each other. Everything works great and looks great.
 
Street Avenger said:
The distance from the Volume pot to the selector switch is so short that I just used standard 22 AWG copper wire and twisted them around each other. Everything works great and looks great.

I was just looking through a blog on building a Strat (which Cagey posted in a different thread) and saw just that idea, braiding two lengths of single-conductor wire.

DSC_0053-4.jpg
<-- Bizarre ... there must be a permissions issue with linking directly to image paths on his server, or load issues ... this image is only working intermittently for me.

I'm still not sure if I want to reuse the original wire from my guitar, which would allow the only improvement to be better solder joints; use the single-conductor wire I bought from Radio Shack and braid it together like you mention and is shown in the above photo; or buy the admittedly very cool looking and purportedly quite easy to work with braided outer shield wire from Stew-Mac.

The former two options cost me nothing additional, in terms of materials, while the latter puts me out the $26.58 + $8.30 or so in shipping. I do want to get some DPDT On/Off/On toggles, though, and three or more at Stew-Mac are $6-something. It's too bad their CTS pots are so much more expensive than at GPR, else I'd throw some pots into the order, too.

Before I do anything, though, I need to buy some contact cleaner. My pots crackle when turned and the shafts are covered in corrosion. Want to see if they're salvageable before buying replacement pots.

It irks me, though, to pay the same or more for shipping than the cost of any item. Does anyone know where I can pick up that "Blow Off" stuff live and in person? Also, from where can I get the best in-person price on a good size spool of 60/40 rosin-core solder? The sad, little Home Depot in Chelsea had only a pathetic little spool for 10 bucks. It was kind of maddening.

Is it worth spending $18 on a seemingly all-purpose Klein Tools wire stripper/cutter / bolt/screw cutter? Or would some cheapo stripper / cutter do the trick, and to hell with shortening screws?

Thanks.
 
Twisted wires are not shielded, and the twisting trick only works with differential transmissions, which you don't have in a guitar. Everything is relative (unbalanced). Plus, it's usually for long distances and/or higher frequencies. For instance, telephone lines are always twisted, because they run long distances, and ethernet pairs are because they carry high-frequency signals. Don't fight it or try to cheat; you need shielded cable or a high tolerance for amplified noise.

Don't waste your money on contact cleaner. It's very nearly useless except for unreliable and very temporary fixes. If the pots are noisy, they're shot. Replace them. Anything else is turd-polishing, and you'll be opening your guitar up frequently for a fresh buffing, then after a short time you'll need to replace the pots anyway. Do it now.

Solder is surprisingly expensive, even online. But, you can do better at Parts Express.

re: good strippers vs. garbage - never buy cheap tools. You'll be sorry you did. You may wreck something or be forced to do a half-assed job or just generally have a rough time trying to accomplish whatever it is you set out to do. Then you'll have to buy the good tool anyway, making the cost that much higher. It's always faster/easier/cheaper to do it right the first time.

Sorry for all the bad news, but better to learn it now than after you've spent time and money that you have to repeat later on the same job.
 
Cagey said:
Sorry for all the bad news, but better to learn it now than after you've spent time and money that you have to repeat later on the same job.

No, that's quite all right. I very much appreciate being told "like it is." My gut told me to just fire on the new pots and buy the Klein stripper, but I still like to confer a bit instead of making a pure gut decision. The only time I go right with my gut is at the rink, since there's no time to doubt it.

Thanks again for the forthright rebuttals.

 
Back
Top