Hoffman Amplifiers Tube Amplifier Forum
Amp Stuff => Tube Amp Building - Tweaks - Repairs => Topic started by: joesatch on March 31, 2022, 08:29:42 am
-
I have found on the interwebs an individual that uses a shielded cap instead of a resistor on the input grid. He claims it reduces noise by 70%. I normally dont link other sites but this is the only instance of this strategy i have come across. Anyone have experience with this?
https://www.marshallforum.com/threads/shielded-wire-questions-2204-build.91685/post-1755515 (https://www.marshallforum.com/threads/shielded-wire-questions-2204-build.91685/post-1755515)
-
In general - anytime percent is used to describe a difference in volume, I’m very sceptical.
-
More to it than just a cap. The grid leak resistor is also in that shielding tube. I suspect if there was much merit to this that it would be more widely used.
-
Most pedals and other electronics have an input cap just to block DC voltage from screwing up the bias. For AC signals the cap is irrelevant.
Shielding the input jack and wire can reduce noise, and removing the grid stopper removes the Johnson noise caused by that resistor. The grid stopper has other important effects, though (https://valvewizard.co.uk/gridstopper.html).
-
Certainly, in a 2204, the 68k grid stopper adds a lot of noise.
But I suggest to be sceptical of stuff written by the individual behind the soundguruman / ampmadscientist / crumb etc identities.
Those caps look to be pointless?
Merlin shows a valid way to reduce grid stopper noise, whilst retaining its benefits http://www.valvewizard.co.uk/gridstopper.html
-
... a shielded cap instead of a resistor on the input grid. He claims it reduces noise ...
More to it than just a cap. The grid leak resistor is also in that shielding tube. ...
My Standel-like build used an input coupling cap (because the original amp did).
Much Guru-smoke in that thread. I notice that by removing the series-resistor, we are not getting the RF roll-off that we normally obtain from the Series-R (of the grid-stopper) interacting with Input-C (of the tube's Miller Capacitance). This doesn't normally get down into the audio-range, but Guy says he doesn't want to mess with treble (okay, as long as you never get strong RF).
So he has a series cap, which is really just a coupling cap. Total Noise is related to bandwidth, so trimming some bass bandwidth should reduce some noise. Except 0.011µF (his two 0.022µF in series) working against 1MΩ is -3dB at ~16Hz. Much more fat could been trimmed.
The only "special thing" about putting all this mess in a copper tube is that he's spent all this time using shielded wire from Input Jack to V1 grid, and the capacitor has un-shielded leads. It's hard to be sure, but the photo-fragment appears to show the Input Jack is also in its own copper enclosure. Not sure that's truly called for if the open side of the chassis is pressed against screen in the cabinet, but...
This individual stressing you must have the caps' outer-foil in the correct direction inside the copper tube seems to be evidence this is bogus. The outer-foil focus is only about giving partial-screening to the stuff inside the cap (though it does ignore that both leads are typically un-shielded).
-
In addiction to the lower value resistor + cap suggested by Merlin linked by pdf64
I remember that Mesa Boogie in place of the input resistor used a ferrite bead for the same purpose
BTW when I've read shielded cap I've think to this
(real shielded capacitor)
(https://el34world.com/Forum/index.php?action=dlattach;topic=11559.0;attach=23364;image)
or, at least, to this
(through capacitor)
(https://el34world.com/Forum/index.php?action=dlattach;topic=11559.0;attach=23370;image)
EDIT: Here is one of those Mesa that uses Ferrite Bead instead of resistor
(https://el34world.com/Forum/index.php?action=dlattach;topic=15099.0;attach=35134;image)
Franco
-
Used a LOT of those filters, every wire that entered the MRI scan room went through one! someone had to hook up the wires to 'em :BangHead: