Observations about wire as a result of debugging

[KGW]

A little background (there is a question at the end):

I built a TWEX clone last winter using ApexJr's solid core/teflon coated wire. I really liked working with this wire due due how well it stays put. So, I decided to rewire my DIY Marshall 1987 (50W). This amp had EH EL34 tubes. It seemed to worked fine, but after a few months, I started noticing occasional popping sounds. I decided to try some SED EL34 / 6CA7 (WINGED C LOGO) tubes instead. This resulted in parasitic oscillation. I have mostly fixed the amp now. I did the following (in order - each step was an improvement):

1. Cleaned up the tone stack wiring so that all wires run close together and lay against the chassis
2. Rerouted the wire from the impedence selector to the presence pot so it is off the chassis (floating).
3. Same for the wire from the presence pot to the PI.
4. Twisted the primary OT wires (not including the centre tap)
5. Changed the wire going to the power tubes to twisted stranded wire.

#5 is the interesting one. I noticed that moving these wires affected the oscillation, so I unsoldered them and used some jumpers (the kind with alligator clips and stranded wire) in their place. Voila! the oscillation was gone. If I put either of the original solid core wires back, the oscillation would return. Eventually, I tried stranded wire and no more oscillation.

So, it would seem that in this case, stranded wire is better than solid. I have my ideas why, but I thought I would ask the experts here what they thought. That is my question.

BTW, I used some Kimber TCSS wire about 6 years ago in a phono preamp. This stuff is 19AWG stranded bare copper with teflon insulation. The strands are fairly thick so the wire stays put. I'm thinking of trying this next.

thanks

[VacuumVoodoo]

It's not the stranded vs solid wire but how you run them. Connection from PI to power tube control grids - which is what I assume #5 wires are - is one of the most sensitive connections in any amp. Run them too close to any other high signal level wire and you're asking for oscillation. Mounting grid stopper resistors anywhere else than directly on pin 5 of power tube sockets is also begging for oscillation.
One additional way to keep an amps tendency to oscillate under control is to dress OT fly out leads in a shielding sleeve and ground the shield.

Like this:

[IMG:556:262]http://i26.photobucket.com/albums/c149/ ... ceened.jpg[/img]

[Bob-I]

One additional way to keep an amps tendency to oscillate under control is to dress OT fly out leads in a shielding sleeve and ground the shield.

Like this:

[IMG:556:262]http://i26.photobucket.com/albums/c149/ ... ceened.jpg[/img]

Excellent idea. Do you ground the shield?

[gearhead]

"built a TWEX clone last winter using ApexJr's solid core/teflon coated wire"

FWIW, he also sells teflon/twisted strand silver plated wire.

[KGW]

Thanks for the answers so far.

I also like the shields OT wires.

Anyway, the solid vs stranded wire from the PI to the grids DID make a difference. It could be that the ApexJr wire had a really thin coat of teflon coating. My TWEX clone was wired with this wire and doesn't have oscillation problems. The Kimber wire that I mentioned has much thicker teflon. It is also almost $1.00 per foot.

I have heard/been told from many that stranded wire is better for signal wire because of the multiple magnetic fields from each strand as opposed to a single solid wire. Sonically, I have found solid wire to sound better than stranded for HIFI. Guitar amps are different though.

Any more thoughts?

[KGW]

"built a TWEX clone last winter using ApexJr's solid core/teflon coated wire"

FWIW, he also sells teflon/twisted strand silver plated wire.

I knew that but thanks. I have used Welborne Lab's teflon/stranded as well.

[tomrasdf]

1. Cleaned up the tone stack wiring so that all wires run close together and lay against the chassis
2. Rerouted the wire from the impedence selector to the presence pot so it is off the chassis (floating).
3. Same for the wire from the presence pot to the PI.
4. Twisted the primary OT wires (not including the centre tap)
5. Changed the wire going to the power tubes to twisted stranded wire.

In regards to 1 and 5: I know Gerald Weber is always preaching short grid wires. He suggests mounting the tone caps right on the pots and the PI coupling caps right to the power tube grid stoppers. The later takes some finageling with the bias setup, but I've done both on a BF bassman as well as a dmbl inspired build with favorable results. Gerald claims that the long grid wires going to the power tubes on plexis are to blame for the oscillations you get with the both the presence and treble on full.

[KGW]

1. Cleaned up the tone stack wiring so that all wires run close together and lay against the chassis
2. Rerouted the wire from the impedence selector to the presence pot so it is off the chassis (floating).
3. Same for the wire from the presence pot to the PI.
4. Twisted the primary OT wires (not including the centre tap)
5. Changed the wire going to the power tubes to twisted stranded wire.

In regards to 1 and 5: I know Gerald Weber is always preaching short grid wires. He suggests mounting the tone caps right on the pots and the PI coupling caps right to the power tube grid stoppers. The later takes some finageling with the bias setup, but I've done both on a BF bassman as well as a dmbl inspired build with favorable results. Gerald claims that the long grid wires going to the power tubes on plexis are to blame for the oscillations you get with the both the presence and treble on full.

My current layout is more like a TWEX than a plexi, but it would be better to get the wires from the PI to the power tube grids a bit shorter