Twist plate & cathode wires to increase stability?

[JazzGuitarGimp]

It seems that by twisting the plate and cathode wires on a 12AX7 gain stage between the socket and board, you would add a bit of capacitance from plate to cathode, thus adding a bit of LNFB and improving stability. Too much capatance would cause a loss of high frequency response, I imagine. Just curious if anyone has tried this, and if so, was it sucessful?

[JazzGuitarGimp]

This is what I have in mind. It's the blue / orange pair. I'm going to give this a try and I will report back if the results are useful.

[John_P_WI]

Yes, the low impedance cathode snd plate wire can also be twisted with the grid wire. The three wires together in effect act as shielded cable providing some immunity to noise and high frequency oscillations.

[Stevem]

Mesa was forced to do pretty much the same thing in there early Boggie amps to stop oscillations due to the tight layout!
They took another wire off the plate, insulated the end and wrapped it around that stages grid wire 4 or 6 times.
Note that this works but the phase canceling taking place is part of what gave those amps there Mid heavy lead tone.
I would by far just use sheilded grid wire and make sure my plate/ coupling caps are installed the right way to not radiate out!

[brewdude]

Is t Dumble touching on this concept by running the plate cathode wire in close proximity? -- Fine tuning the frequency response of the amp--especially in overdrive.

[martin manning]

That's the understanding. A couple of inches of parallel wires is a few pF at most, but there is also the capacitance to ground from running the leads along the chassis surface.

[Malcolm Irving]

Whenever you can identify a 'go' and 'return' current in an amp, I think it can be a good idea to keep them close together or twist them. This does add a few pF of capacitance between the conductors, but has the benefit of greatly reducing the pick up of interference. Equally, it also reduces interference radiated from that circuit (important for big signals and power supplies).

If there are three (or more) conductors, where the total current adds to zero (taking all +ves and -ves into account), the same idea applies.

[JazzGuitarGimp]

Excellent! I appreciate all of you taking the time to share your knowledge.

Many thanks,
Lou

[pdf64]

Capacitance links everything to everything else; trying to minimise it may only go so far. It can be beneficial to tweak the lead dress to slightly increase the capacitances that support a little local NFB, in order to counteract PFB between stages.
So to implement that, I like to put a twist or two on the grid and plate wires of each voltage gain stage.