More Grounding Misconceptions

[Mookakian]

Ok, so i wired it up, good news is it works fine, later today ill throw on the flouros and a few neons near the rig and see what happens

 

[Lyle Caldwell]

Tim,

Remember it won't be impervious to noise, just quieter than standard. And as your guitar isn't shielded, you will only have a piece of the puzzle in place - it will still be susceptible to neon, flourescent bulbs, etc.

 

[Mookakian]

Yep, your right, i cant really hear any difference here. But it would be a miracle if i could remember how much noise got in before. I guess radio waves would be where the real advantage of this comes in, and im a country dude, minimal RF around.

 

[Lyle Caldwell]

One of the frustrating things about "doing everything right" is that you usually won't know that the extra effort was worth squat.

You'll have a fully shielded guitar with separated signal negative and shield, and well, it will just sound like a guitar. It might be a little quieter when you're touching the strings, but not much. Just a guitar...

...until you share a stage with another guy with a similar guitar and his guitar is noisy as hell or starts picking up AM radio signals.

That's when it's nice to have "just a guitar" coming through your amp.

Like so many other things in having a great instrument and rig, it's a lot of little things that add up to good tone. No one thing by itself is a revelation.

 

[Mookakian]

Couldnt agree more, thats why i get so frustrated when ppl say "i cant hear anything, must be BS"

Silly Mohekans

 

[Quill]

Lyle, this makes great sense - I think everyone wonders about this. Great posts, thank you very much, I'm going to see if I can make this work for me, too. Really appreciate your contribution here.

 

[teddis]

hi Lyle,
Thanks for the thread. Just stumbled upon it and I am just about to redo my controls.
Two questions questions if I may:

1)Cavity shielding, is better to paint or put copper ? Would you do all compartments on an LP axcess, Control cavity, pick up recesses, trem bay, and toggle housing ?
2) Grounding a Floyd Rose. I see a hole in the Trem Claw, looks like I can solder a wire there easy. Where would I run it to ? or is this not the best ground point for a Floyd Rose ?

thank you

 

[Lyle Caldwell]

As far as cavity shielding goes, copper foil is best for an instrument with vintage value as it can be peeled off later with no change to the instrument.

It also gives a more perfect shield, but two good coats of shielding paint gives a more than adequate shield.

For an LP Axcess I'd recommend the paint as it won't hurt the resale and it is easier to do a good job with it.

You can get small cans of the paint for about $12 on Ebay - enough to do two guitars.

You should shield every cavity where the signal is present, so this means the pickup cavities, the control cavity, and the switch cavity. You do not have to do the trem springs cutout.

Once each cavity is shielded you must tie the cavities together. I use #4 wood screws and #4 lug washers with small gauge wire running from cavity to cavity. Shielded paint which is not tied to ground does nothing. So get all the cavities tied together and then have this system tied to the signal neg (ground) at only one place.

Note: you only have to do the screws/lugs/wires from each pickup cavity to the control cavity. The pickup selector switch will tie the switch cavity to the system shield and also shield the wires going to/from the switch.

For every distance traveled through an unshielded area (the rout from switch to controls, the rout from pickup cavities to controls) you must use shielded wire otherwise you undo all the shielding.

Run a wire from the trem claw to the lug in the control cavity - this ties the strings to the system shield.

Just to be clear, inside the control cavity will be a small screw going into the side of the cavity wall. This screw will hold a small lug washer against the shielded paint. There will be wires going from this lug to matching lugs/screws in each pickup cavity, a wire from this lug to the trem claw, and one wire from this lug to the signal negative, typically at lug 3 of one of the volume pots.

Each pot case will be tied to system shield through the shielding paint.

 

[teddis]

Thank you, I will follow this when the gerar arrives

 

[Mouse]

There is no grounding misconceptions, once you build lots of amps you begin to understand that there is no better way to doing grounds (star, bus..whatever), but there is things that works fine for given amp or instrument and that's whatever takes to get noise to reasonably acceptive level. good soldering technique certainly helps, a mess of cables too, conductors that are tidy and running at parallel will induce more hum.

 

[Mookakian]

Thats one hell of an effort shielding the guitar lyle, and a great how to write up. Im definately going to try this out on the good old fake epi zak when i get time, if i get no problems, ill do my other guitars for peace of mind. Good man

 

[FourT6and2]

A mess of cables too, conductors that are tidy and running at parallel will induce more hum.

Really? Every one of them? Interesting. I guess I should rewire my Metro build to be a rats nest, then? Maybe the noise that isn't there will go away even more! http://fc07.deviantart.net/fs71/f/2010/317/8/f/metro_build_33_by_haftelm-d32srow.jpg

 

[Mouse]

Really? Every one of them? Interesting. I guess I should rewire my Metro build to be a rats nest, then? Maybe the noise that isn't there will go away even more! http://fc07.deviantart.net/fs71/f/2010/317/8/f/metro_build_33_by_haftelm-d32srow.jpg

Great build, congrats. Well thats well built, the heaters wire is not parallel but twisted of tightly (it's (arguably) not really necessary cause it's AC ), and conductors which seems to run parallel are well off each other (problems with hum and oscillation beetween two or more conductors is less with distance).
Now, lets imagine situation that your cables have no distance and they are neat and there is no much distance on the turret board etc. so every conductor and passive element like resistor or cap is spreading it's Electromagnetic field on the each other which is pretty normal once the current goes through conductor. Would you then rather have conductor and elements that are running not parallel or rattle nest or some kind point to point in all directions rather than parallel like older Matchless amps did?

 

[Lyle Caldwell]

AC produces a lot more "noise" than DC. While the output tubes and PI are balanced and relatively immune to heater noise, the preamp tubes definitely need that tightly twisted pair unless the heater supply is well-filtered DC.

I agree about the parallel wiring, though his amp looks to have enough distance between them. My own solution for my amps is to use really heavy duty 4 layer PCBs with shielding between heavy traces and to use shielded wire for every off-board connection.

 

[FourT6and2]

Great build, congrats. Well thats well built, the heaters wire is not parallel but twisted of tightly (it's (arguably) not really necessary cause it's AC ), and conductors which seems to run parallel are well off each other (problems with hum and oscillation beetween two or more conductors is less with distance).
Now, lets imagine situation that your cables have no distance and they are neat and there is no much distance on the turret board etc. so every conductor and passive element like resistor or cap is spreading it's Electromagnetic field on the each other which is pretty normal once the current goes through conductor. Would you then rather have conductor and elements that are running not parallel or rattle nest or some kind point to point in all directions rather than parallel like older Matchless amps did?

The reason to twist the heaters is to keep the wires parallel. Look at amps like Soldano, Diezel, older Mesas and etc. Their heaters (AC, not DC) are straight, parallel lengths of bus wire. And there's no noise, especially since those amps are high-gain. Twisted pair wiring is done to ensure that the wires are parallel at all times. That's what cancels hum. There are multiple ways of getting to the same point. Twisted pairs, parallel bus wire, elevated heaters, DC heaters, etc. The idea, though, is to keep them parallel...

Think of a double helix (DNA) each point is parallel with the point directly opposite. That's what twisting does!

 

[Mouse]

AC produces a lot more "noise" than DC. While the output tubes and PI are balanced and relatively immune to heater noise, the preamp tubes definitely need that tightly twisted pair unless the heater supply is well-filtered DC.

The reason to twist the heaters is to keep the wires parallel. Twisted pair wiring is done to ensure that the wires are parallel at all times. That's what cancels hum. There are multiple ways of getting to the same point. Twisted pairs, parallel bus wire, elevated heaters, DC heaters, etc. The idea, though, is to keep them parallel...

Gentleman, Direct Current is different than Alternative Current (phase angle) and you by twisting the wires with DC supply reduce the hum just by doing that, making wires not parallel or crossing each other all the time.
Now tell me what are you doing by twisting the wires with AC supply?

Basics...AC the movement of electric charge periodically reverses direction. In direct current (DC), the flow of electric charge is only in one direction.

 

[Lyle Caldwell]

AC must be twisted, not run in parallel.

In a pair of wires carrying AC, each wire carries the same current in opposite directions, and they create magnetic fields with opposite polarities. If the wires are tightly and symmetrically twisted together, these fields will cancel out.

If in parallel there is no cancellation and they can inject hum into the signal path.

This is especially crucial on the preamp tubes as the field from an AC heater supply is much larger than the signal coming into the amp, so it is very easy (almost unavoidable) that the incoming signal picks up hum from the nearby AC wire (signal comes into pin 2 of V1, very close to pins 4,5, and 9 of V1). Tightly twisting the heater wires greatly reduces this hum. Using a DC lift lowers it a further 30dB or so. Using a full DC heater supply completely eliminates the hum.

In the phase inverter and power tubes any hum common to both sides of the tube will be canceled out, so the tight pair is not critical there, but it is good practice to twist it anyway (also makes for a neat and self-reinforced wire dress).

 

[Mouse]

AC must be twisted, not run in parallel.

In a pair of wires carrying AC, each wire carries the same current in opposite directions, and they create magnetic fields with opposite polarities. If the wires are tightly and symmetrically twisted together, these fields will cancel out.

If you use flat pair wire (such as cheap speaker wire) it will cancel hum just as effectively as twisted. You can see this in some of the original 18W amps made by Marshall's Ken Bran. Just make sure to wire all the heaters in the same phase. lot's of people on the 18watt.com builds their higain amps with straight wire method with great results.

Tightly twisting the heater wires greatly reduces this hum. Using a DC lift lowers it a further 30dB or so.

Yes, simply connecting the PT heater winding's centre tap to the power tube cathodes will provide enough DC elevation voltage for the preamp tubes. That's because you are switching off the stray current between filament and cathode besides raising the heater voltage above cathode v which is now again 6,3v but now is floating on the top of the DC reference voltage.

 

[FourT6and2]

AC must be twisted, not run in parallel.

If the wires are tightly and symmetrically twisted together, these fields will cancel out.

If in parallel there is no cancellation and they can inject hum into the signal path.

Countless amp companies run their AC heaters as parallel runs of wire without any noise or back EMF. A few examples you can look at are Soldano, Diezel, Mesa Boogie, Wizard and Engl.

Twisted pairs = parallel. Same thing. The twisting does not make the wires cross each other over and over. It keeps them PARALLEL at any single given point. Geometry, people! That's how perfect spirals, helixes and twisting works. Believe what you want, I guess. I know it's counter-intuitive, but it works.

Here's a good explanation The Valve Wizard

I'm not going to argue about it any more, I don't care enough to waste my time.

 

[Lyle Caldwell]

I don't care to argue either. But parallel is not enough - they have to be parallel in extreme proximity. You cannot have heaters in parallel with much space between them and have any cancellation. In a PCB, parallel is fine. For wired connections twisting the wires ensures as little an air gap as possible and offers mechanical strength.