Audiodog wrote: ↑Sat May 08, 2021 2:06 pm
This has no bearing on the plate and grid lead dress. You have a huge signal on the plate and the grid is a super sensitive microphone. The grid will induct signal from the plate which is reverse phase. You get tone killing cancellation. Guaranteed.
martin manning wrote: ↑Thu May 06, 2021 12:54 pm
On my LCR meter, 6" parallel leads with insulation touching produces capacitance of about 1pF/inch, calibrated to 0pF with the leads parallel and 1" apart. If the leads are crossed at right angles, it's 0.2pF. As a feedback path in a typical 12AX7 stage, a couple of pF has, in theory, a negligible effect in the audio range, a couple tenths of a dB at 20kHz.
To check this inductive coupling/cancellation theory out I set up a 12AX7 gain stage on a breadboard. It's a typical 100k Ra, 1k5 Rk bypassed with 22u, and 22k Rg with 1M to ground. B+ is 280V and there is 180V on the plate, so 1mA of quiescent current. Applying 3V p-p signal at the grid, I measured the gain at 50, pretty much as expected. There is about 3.5 cm (1.4") of grid and plate wire length running parallel. Watching the input and output waveforms on a scope, and moving the green (grid) and blue (plate) wires together with a chopstick I observe no effect at either 1kHz or 10kHz.
Martin,
I would expect at some frequencies a induction effect, what about 100Hz and 250Hz? Ian reported better low end, that is where I would start searching.
alkuz1961 wrote: ↑Sat May 08, 2021 8:06 pm
But what about the heating wiring, about the influence of which everyone knows ?
Low voltages, but tends to penetrate the audio signal in the form of a hum if the wires are not twisted.
Inductive coupling is driven by current and the magnetic field it creates around a conductor. Twisting heater wires keeps them close together and parallel to each other, so that the field from one lead cancels the field from the other. In a 12AX7, the heater current is ~300x the plate current.
erwin_ve wrote: ↑Sat May 08, 2021 8:15 pmI would expect at some frequencies a induction effect, what about 100Hz and 250Hz? Ian reported better low end, that is where I would start searching.
erwin_ve wrote: ↑Sat May 08, 2021 8:15 pmI would expect at some frequencies a induction effect, what about 100Hz and 250Hz? Ian reported better low end, that is where I would start searching.
Nothing happening at 100 Hz either.
Is the parallel wiring before or after the grid stopper?
erwin_ve wrote: ↑Sat May 08, 2021 8:33 pmIs the parallel wiring before or after the grid stopper?
Before, same as in the D layout. Rg/Miller Capacitance won't filter frequencies that low anyway.
Interesting experiment thanks for doing this?
Any opinions on the effect on Harmonics or possibly the introduction of noise into the system with wires in proximity to chassis acting as a shield? (on plate)
I honestly cannot tell much difference yet I always fly the grid on V2a
Tony
" The psychics on my bench is the same as Dumble'"
Martin, thanks for setting up the experiment. I'm really surprised that it does not reveal anything. My amps all had a noticeable improvement after adjusting them.
Just out of curiosity, do you detect any changes when running the plate and cathode wires in parallel?
BTW? An easy way to prove the capacitance coupling theory on the grid is to take an alligator clip and clip one side to the insulation of the grid wire. Then you can probe around with the other end,attaching it to the insulation of the plate wire see if you get any degenerative feedback there..Plates that are out of phase will kill tone and have a wah like effect on the low end . Plates that are in phase will feedback and squeal..This should test out how noticeable it is on your amp. IMO every amp is different. Dumble seems to run both plates and grid far enough apart V2a to not cause any C.C IMO
Let us know what you find. Ian
Tony
Last edited by talbany on Sun May 09, 2021 12:58 am, edited 1 time in total.
" The psychics on my bench is the same as Dumble'"
talbany wrote: ↑Sun May 09, 2021 12:08 amAny opinions on the effect on Harmonics or possibly the introduction of noise into the system with wires in proximity to chassis acting as a shield? (on plate)
ijedouglas wrote: ↑Sun May 09, 2021 12:37 amJust out of curiosity, do you detect any changes when running the plate and cathode wires in parallel?
Running wires down on the chassis will introduce capacitance of about 1pF per inch to ground, which will only affect extremely high frequencies. I would expect similar results running plate and cathode wires close together on the chassis for a couple of inches, which will put a couple of pF between them, and also to ground on each. I can see that in a simulation, it's a loss of about 0.1 dB at 20kHz, but that's going to be too small to see on the scope.
Last edited by martin manning on Sun May 09, 2021 1:39 am, edited 1 time in total.
talbany wrote: ↑Sun May 09, 2021 12:08 amAny opinions on the effect on Harmonics or possibly the introduction of noise into the system with wires in proximity to chassis acting as a shield? (on plate)
ijedouglas wrote: ↑Sun May 09, 2021 12:37 amJust out of curiosity, do you detect any changes when running the plate and cathode wires in parallel?
Running wires down on the chassis will introduce capacitance of about 1pF per inch to ground, which will only affect extremely high frequencies. I would expect similar results running plate and cathode wires close together for a couple of inches, which will put a couple of pF between them, and also to ground on each. I can see that in a simulation, it's a loss of about 0.1 dB at 20kHz, but that's going to be too small to see on the scope.
So basically what you are saying is that the lead dress on V2 (V2a grid + V2b plate and cathodes) has a negligible effect?
I can’t find anything significant about the grid to plate lead proximity, at least not for one stage in isolation, and the snubber caps on those stages are huge compared to the plate to cathode lead capacitance... two orders of magnitude larger.
martin manning wrote: ↑Sun May 09, 2021 1:50 am
I can’t find anything significant about the grid to plate lead proximity, at least not for one stage in isolation, and the snubber caps on those stages are huge compared to the plate to cathode lead capacitance... two orders of magnitude larger.
Interesting. It definitely made a difference for me (maybe confirmation bias). Charlie W can lend his bat-ears next week and prove me bonkers
I'd love to know why Mr. Dumble looked to be pretty particular about this part of his lead dress.
I've had problems with the OD circuit oscillating at a high frequency and the cure for it has been moving V2a plate wire away from the grid wire. Usually the oscillation is happening at a frequency outside of hearing range.
