Lead dress tips?

[daz]

can anyone either give me a quick lesson in what to consider when running leads from the board to tubes or point me to a layman's tut on the subject?

[bulldogguitars]

Rg has an explanation of lead dress at www.geofex.com.

[tubeswell]

Hi Daz

Don't know if this helps, (don't wanna tell you how to suck eggs) but when I did my 5F2A a couple of months ago, I used shielded cable from the input jacks to the grid of V1A/Voltage Amplifier, and again from the Vol wiper to the grid of V1B/Driver. I grounded the cable shields at one end only (at the input socket end and the vol pot end respectively). I kept all the other leads as short as possibile and as far away from other leads as practicable to prevet unwanted signal coupling from wires, and where necessary I had wires crossing each other at right angles. I used cable ties to kep some wires away from other wires

I had a split ground with all the pre-amp grounds, including pre-amp decoupling/filter cap going to a tagboard with one oead going to the (one) input socket with the ground. I had all the other grounds, including heater CT, PT shield, HT CT, Mains AC ground, Reservoir cap and other filter caps and output cathode ground going to another tag board with a single lead running from this to one of the PT bolts.

I had all pairs of AC wires (HT winding, Recto heater, 6V heater twitsed together to cancel EM field.

I had the heater winding going straight to the pilot lamp first , then to the output socket and thence to the pre-amp sockets. With the heater CT grounded I had each side of the 6.3V winding going to each side of the heater filaments (i.e. one wire going to pin 2 of the output socket and pins 4 and 5 of the preamp socket, and the other wire going from pin 7 of the output socket to pin 9 of the pre-amp socket. With the heaters wired this way I ran the heater wires tucked into the bottom rear of the chassis away from everything else. This enabled me to put the signal, plate and cathode wires straight down onto the socket from the main tag board. I understand you ca run a heater CT to the cathode pin to further eliminate hum, but with my amp wired this way, there's virtually no unwanted hum.

I keep a wet sponge in a tray nearby to my soldering iron to wipe the iron tip on after each bit of soldering to keep the tip clean. I tin the wire ends first in most cases, and when doing the joins, I hold the iron tip onto the join for a few seconds to heat it up real good before appying the solder, and keep the parts still while letting the solder set. I also tin other contacts like pot lugs and tag board lugs before soldering together. Don't let flux from your solder run into the pot when you're soldering the lugs - I find best way is to tin these is with pot shaft clamped into a vice or other wise securely held down. I also wouldn't apply the iron for too long on the pot lugs. Where possible make sure all parts are firmly held down while applying the soldering iron tip. I find old-fashioned desoldering braid works best for me to take off unwanted solder before resoldering something.

I use soldering lugs to attach wires to chassis bolts etc. I also use secure part mounting accessories with cable ties to mount big caps etc. Easier to get off if you have to change one, and it looks neater too (if you can say that about my stuff?)

[daz]

You pretty much do everything the way i do. My only question really is about lead dress from the board to the tubes. I've seen people talk about longer wires so they can keep them all away from each other, then others saying short as possible but only crossing each other at a 90 degree angle and no paralleling. I did the latter and it's pretty quiet. But i just wanted to be sure i'm doing it right when i build my second amp which will be my main amp hopefully. I guess my question really is, what should be my main concern when routing wires from the board to the sockets. I seemed to have done well on my first build, but my second will hopefully be built from my experience with my first and i want it to be right the first time w/o having to redo anything. So short as possible with no parallel wires? Or are there other considerations? I already know the things like shielded cables for grids with one end grounded. Common sense stuff like that. I think what i'm talking about is a bit more of that mojo stuff tho.

[tubeswell]

Hey whaddaya know? I'm on my second amp too, but I have a sneaking suspicion that it won't be the last one I build ;-)

Basically I think that the shorter the connecting wires in the signal path are, the better. My understanding is that the EM field that surrounds the wires radiates out from the wire core in a circumferential pattern sort of equidistant from the core along the length of the wire, and this simply means that longer wires just end up with more of this field radiating around in the chassis, so that the chances of other EM fields from other wires (or other components) coming into close proximity to any other given signal component are greater with longer wires, and thence unwanted coupling.

Other than that I don't know about any special mojo for those wires.

But I have heard that there is a 'technically correct' way (apparently) to ground your heater CT to the (common) cathode pin of the output tube(s) when you bias the Heater circuit that way. I understand that it involves doing some circuitry on the CT (like a voltage divider? or a shunting cap?? or something?) from what I recall, but I'm hazy on the details. Maybe someone else will chime in if they know that one. I did read soem stuff on it a few months back but I can't remeber what I did with it.

[PRNDL]

After my last amp repair (a Vibroverb with "problems" that had been to several techs over several years without addressing the issue of intermittent horrible hum), I've come to a greater appreciation of lead dress.

The chopstick test
This test involves moving the wires around with a wood stick (i.e. chopstick) while listening for changes in hum or noise. It works best when you turn the amp volume up.

In the Vibroverb, moving several wires to better locations (usually parallel to or farther away from other wires) lowered the noise, each contributing a little bit.

Moving the shielded input cable wire had a harmonica (or pan flute) effect, which turned out to be one of the amps main problems. In this case, Leo's layout had the wire routed through holes underneath the board - away from stray EM. It also locked the wire in place to stop movement permanently.

Some builders choose their amp wire with lead dress in mind, which means you want wire that stays where it is.

[daz]

The chopstick test
This test involves moving the wires around with a wood stick (i.e. chopstick) while listening for changes in hum or noise. It works best when you turn the amp volume up.

yes, i have tried that. And in fact it's what got me thinking harder about this because i heard differences in noise. Problem is, i used stranded 18 gauge and it's hard to move and impossible to stay, plus they're too short to move much anyways. I now have a big roll of 18 gauge solid i'll be using for #2.

But my main concern is this....keeping in mind that the shortest possible leads are what you want, how are you supposed to do this? In other words, if they are short you won't be able to move them enough if you get noise and lead dress is the reason. I made mine as short as possible, so when i tried to move them i barely could. So what to do? Make them a bit longer than necassary?

[PRNDL]

Problem is, i used stranded 18 gauge and it's hard to move and impossible to stay, plus they're too short to move much anyways. I now have a big roll of 18 gauge solid i'll be using for #2.

That's exactly what I meant about choosing the correct wire!

Also, cloth or braided wire is easier to work with. It's called "push-back" wire, because you cut the end and push back the cloth to expose the wire for soldering.

keeping in mind that the shortest possible leads are what you want

That's not an appropriate goal.

The goal is to keep the wires neat, which means keeping them parallel and apart, plus having as few cross over each other as possible.

Check out this vintage Fender chassis - a 63 Vibroverb. The wires curve and most lay flat on the chassis.

Although it's not the amp I fixed, you can see the red and yellow wires that go through and under the board and to the second tube. They are on the right side of the board to the left of the cathode R/C pair. One of those was the problem wire, even though it was shielded, which is supposed to be better than plain wire (in theory).

The amp I fixed was a Silverface to Blackface conversion. Some say those can be a real problem if you don't follow the original layout exactly, since it was optimized by Leo. Minor circuit changes (mods) can also result in problems because the wires aren't as neat and orderly, and more will cross over each other.

Notice the two solder blobs in the third photo for the transformer and board grounds. Some would say that it would be faster and easier to make one blob for both, and there's no difference electrically. Consider, however, that it is there for a specific reason, and most likely requested by Leo because there were problems in some builds that had only one blob.

[daz]

Thanks. Re-doing it now.