OT wire routing under board

[Gaz]

When building in the typical Fender/Marshall turret board style, it ends up that the OT's primaries and secondaries exit somewhere under the board. I always try my best to get them as close the output tubes as possible, but currently I have one where the wires are gonna exit right under the phase inverter circuitry and maybe part of the preceding FX loop recovery stage. Am I okay here? Do I have to worry about any potential crosstalk, or just hum?

I wanted to know your guys' rule of thumb when the OT has to go under the board. Do you just try to keep away from the first preamp stages, or the PI too?

This is an area of the layout I haven't experiemented or thought too much about.

Thanks.

[Stevem]

Hum will not be a issue, but oscilation might!
Power it up and see what happens and then post back.
You want to keep the OT wires away from the input to the PI, the signal level out of the PI plates is so high that radiating signal from the OTs secondary wires seems to never be able to impress itself back to the primary side.
What is most important is the length of the plate wires off of the PI tube.
Up to the point where the plate wire gets to the coupling cap is considered the plate circuit, after that cap becomes the grid circuit of a output tube, and if that run of wire or whatever is too long ( some 4 inches or so) than problems can crop up depending on the rest of the layout!
So like I said, try it and see, if it sounds good and open with no strange noise your ok, and if you have a O-scope you can prove that out.

[Leo_Gnardo]

Hum will not be a issue, but oscilation might!

Had a case like this with a Fargen. Made a better radio transmitter than amp. I made a shield out of copper foil sandwiched between layers of fish paper, stuck that between the bottom of the board and the OT wires, ran a ground wire from foil to chassis and bingo, it became an amp!

For a fix that cheap and easy why not put in a shield anyway.

[M Fowler]

Can run the wires on outside too like Henry H. (Redplate showed me).

Also, put wires in a shielded outer jacket but I haven't had problems with this personally, good luck is probably why

[Gaz]

Hum will not be a issue, but oscilation might!
Power it up and see what happens and then post back.
You want to keep the OT wires away from the input to the PI, the signal level out of the PI plates is so high that radiating signal from the OTs secondary wires seems to never be able to impress itself back to the primary side.
What is most important is the length of the plate wires off of the PI tube.
Up to the point where the plate wire gets to the coupling cap is considered the plate circuit, after that cap becomes the grid circuit of a output tube, and if that run of wire or whatever is too long ( some 4 inches or so) than problems can crop up depending on the rest of the layout!
So like I said, try it and see, if it sounds good and open with no strange noise your ok, and if you have a O-scope you can prove that out.

Steve, that's all great info. So you are saying that it's best to keep output tube grid wires short and the PI plate wires longer if I have the choice? I was also wondering how I might observe check for unwanted coupling with the scope?

Hum will not be a issue, but oscilation might!

Had a case like this with a Fargen. Made a better radio transmitter than amp. I made a shield out of copper foil sandwiched between layers of fish paper, stuck that between the bottom of the board and the OT wires, ran a ground wire from foil to chassis and bingo, it became an amp!

For a fix that cheap and easy why not put in a shield anyway.

What type of circuit was the Fargen? Do you remember how far under the pre the OT was? I was also wondering if you could determine if it was the primaries or secondaries that were the culprit, and generally, which are more likely to cause the issue?

Can run the wires on outside too like Henry H. (Redplate showed me).

Also, put wires in a shielded outer jacket but I haven't had problems with this personally, good luck is probably why

hat's a really good simple idea about running the wires above the chassis. That's a big ass shield right there! My chassis is aluminum, so I'm guessing that it works signal coupling, but not any shielding against magnetic coupling, correct?

Thanks, all!

[Stevem]

In 50s era a Gibson amps they when as far as to have shielding on the OTs primary wires !
a amp is acting up with hardy oscillation taking place you do not even need a scope, just hook a voltmeter set for ac across the speaker, plug in the guitar and close down its volume control then crank everything on the amp up, if you read less than 750 MV on the meter your likely good to go!
But yes, all grid wires at least in regards to the first two gain stages need to be kept as short as you can, even if that means making plate wires longer!

[Gaz]

Thanks, Steve. I guess I a little paranoid because in my last build there was some crosstalk that I did not realize until later on. I felt kinda dumb because I thought I had laid everything out just fine. The amp did sound okay, but the treble improved at low master volume setting when I got rid of the crosstalk.

Also, would anyone mind helping me find some shielded tubing at Mouser? I've having some trouble...

Thanks!

[rp]

Hum will not be a issue, but oscilation might!

I have the same question regarding an upcoming (and slow) 5G9 build. Usually I stick w/ PTP to avoid wiring complications I don't fully comprehend but a vibrato PTP looked like a PITA so I'm doing a board.

I laid it out compact and JTM style, figuring if Marshall got away w/ it so will I (hahaha I know where copying others mistakes will get me). The board will have the bias circuit on the far left then the typical power/PI/preamp moving to the right. In Marshalls the wires run more towards the bias side than right under the PI. The OT wires in mine will be right under the PI caps.

Anyway, you all think it's best to shield the OT wires, both sides, right from the start, or just go with it and hope for the best? It'll be a bear to go back in later and shield it. I have some braided shielding snake I can slide over the OT leads. It's a used OT so the wires, especially the secondaries were kind of short, I could also lengthen them and move them farther to the left under the bias.

Am I bound for trouble? My other concern here is if shielding the OT leads will rob treble and sparkle like it can when using shielded coax in the preamp stage?

[Leo_Gnardo]

Hum will not be a issue, but oscilation might!

Had a case like this with a Fargen. Made a better radio transmitter than amp. I made a shield out of copper foil sandwiched between layers of fish paper, stuck that between the bottom of the board and the OT wires, ran a ground wire from foil to chassis and bingo, it became an amp!

For a fix that cheap and easy why not put in a shield anyway.

What type of circuit was the Fargen? Do you remember how far under the pre the OT was? I was also wondering if you could determine if it was the primaries or secondaries that were the culprit, and generally, which are more likely to cause the issue?

Fargen combo s/n 7 was more or less a one-channel Fender copy with reverb. The dreaded 3M3 470k 220k junction was directly atop the OT's primary wires sprouting thru the chassis, I'd say a quarter-inch away. A major candidate for oscillation, and so it did, howling incessantly at ultrasonic frequencies. Capacitive coupling, my best guess.

That dry/reverb junction in Fenders and their copies is a very sensitive spot, and I've occasionally built it right onto the tube socket pin to keep noise pickup minimized.

The copper sandwich I mentioned, grounded and placed between the transformer leads and the bottom of the circuit board, oscilllations went to zero and the amp started to work right.

Shielding around OT leads doesn't necessarily have to be shielded-cable type. I've put up a "fence" of copper foil sometimes on those "folded" designs like Mesa. Sometimes a swatch of grounded copper between input jack area and output transformer leads will clear up ultrasonic noise.

[Stevem]

What also helps and many times makes mass produced amps even sound better is to pay attention through out the amp for the proper placement of all signal / bypass caps.
what you want is to have the last wrap of the foil insulator and its connecting lead that it is made with to be on the out bound, or down stream side of the install.
installing them in this way acts like a shield so the signal is far less likely to radiate out and couple to other parts of the circuits and wires.
On most caps this is easy to do as the out bound side will most times be to your right with the lettering that's on the cap facing you.
If in question and you have a cap meter you wrap a section of tin foil around the cap and hook the meter up to the foil and each lead one at a time, the lead with the smallest reading will be the out bound one.

[martin manning]

Another thing you could do is twist the primary wires together where they pass under the board. This will put a few pF across the primary wires, but running them along the chassis, or shielding them and grounding the shield will put a few pF between them and ground.

Steve, I think you have the outer foil direction backwards. To minimize noise the outer foil lead should be connected to the lowest impedance to ground. This is usually the upstream side wrt signal flow. Also, evidence and reports from people who have talked to manufacturers about it say the printing is applied randomly so you have to test each cap. Aiken has a good write up on this subject.

[Gaz]

Another thing you could do is twist the primary wires together where they pass under the board. This will put a few pF across the primary wires, but running them along the chassis, or shielding them and grounding the shield will put a few pF between them and ground.

Yes, I'm already doing this in general. I kinda wanna just use the shielded tubing for fun... if I can find it. Anyone know where to find some?

[Stevem]

Go to a well stocked Hobby shop and or hardware store as most have brass or aluminum tubing in usable sizes, in fact with a big enough iron you can mount them and ground them at the same time by soldering them to the chassis.
Just note that the bass is much harder to cut!

[rp]

Posts here got me worried, I lengthened and moved the OT secondaries away from the PI and routed them down btwn the power tubes, grid wires can then run perpendicular to it, that'll have to do. I'll skip the metal braid for now and see what happens.

I don't recall where I got the metal braid, I've been sitting on it a long time, meaning most likely AES or someone like Michael Percy. If anyone knows I'm curious too.

Metal tubing would look snazzy on the outside - steam punk element, I can see it being a pain on the inside given tight spaces and inflexibility. Anyone know what the procedure for something like this is, as it's supposed to be grounded at only one end yet metal braid/tube would contact the chassis along its entire run? Would you have to put nylon snake or shrink over it? Might get kinda bulky, and adhoc and klugey looking at that point. Maybe nylon P-clips w/ tubing as they would raise it up a tad, that could look slick and purposeful.

[martin manning]

I think less is more. Just twist the leads and see if that does it.