Vibration mounting a socket with a tube shield

[tubedood]

Hi all!

I have seen a few postings about folks trying to reduce microphonics in their amps by using little rubber mounting bushings to insulate the preamp tube socket against vibration. If that stage happens to have a tube shield on it does this null the effect of the shields ability to help with any noise suppression and make it just a heat sink?

Should a "floating" preamp socket that has a tube shield on it have a jumper wire to keep the shield connected to the chassis?

Have a great day,
Dale

[Enzo]

Yes, that was often done.

[Leo_Gnardo]

Should a "floating" preamp socket that has a tube shield on it have a jumper wire to keep the shield connected to the chassis?

Ideally yes. Some tweed Fenders had a strand of thin braided wire soldered at one end to the tube base, the other end to chassis. And a Gibson GA45T I just fixed mounted their shielded sockets with hollow rivets, ran a skinny ground wire thru the rivet, mighty clever!

[g1]

Solder wick works nice as a flexible braided wire.

[Tom Phillips]

Here is a photo of the Fender method that Leo_Gnardo described.
In the 1959 example shown the pre-amp tubes used the old style snap-on tube shields.
The '59 5E3 also had shock isolated octal power tubes and rectifier sockets with ground wires connecting the metal socket part to the chassis (Edit: For the two 6V6 power tubes).

[tubedood]

Thanks guys... You all are great!

Dale

[trobbins]

If you are using rubber bushings to alleviate microphonics, then aim to keep any wiring to that socket very flexible, otherwise stiff wiring will defeat the purpose of the rubber bushings.

Hopefully your preamp socket also has a centre spigot, which along with the socket metal surround and removeable shield, are all connected to the local star ground point for that stage to assist in shielding.

Adding a shield can assist with reducing microphonic output, as it increases the mass of the socket and valve, which then lowers the self-resonant 'sprung' frequency of that mass and the compliant rubber bushings, and hence increases attenuation of higher mechanical disturbance frequencies that come through the chassis and get to the valve's internal structure.

[tubedood]

Thanks trobbins for the tips... This site is great!
As a secondary question what are peoples thoughts on the tube shields for a Fender tweed style circuit? It seems a lot of those models sported them on the preamp tubes... Do they provide a lot of benefit? If so is that added benefit in the interference shielding or as a heatshield?

My little amp is based on the tweed Princeton so to keep the look authentic I will use the tube shield.
Dale

[Enzo]

Heat schmeat, tube shields protect the tube from electrical noise.

[mhuss]

Heat schmeat, tube shields protect the tube from electrical noise.

I agree, but doesn't the anode/plate inside serve much the same purpose (in terms of shielding the internal grid)?

[Helmholtz]

I agree, but doesn't the anode/plate inside serve much the same purpose (in terms of shielding the internal grid)?

You are right. The grid hardly sees the shield. Otherwise Miller capacitance would be noticeably increased.
I think the main benefit is mechanical: they protect and secure tubes. In rare cases it may be beneficial that they prevent coupling between the plates of different tubes. Steel shields also shield against magnetic interference.

[Enzo]

You can demonstrate the shielding by pulling them off. Not every tube benefits, just as a microphonic tube sucks in V1 but is fine in the PI.

[Helmholtz]

You can demonstrate the shielding by pulling them off. Not every tube benefits, just as a microphonic tube sucks in V1 but is fine in the PI.

I never noticed much of a difference. But the shield shields the tube's plate(s). And the plate is the input of the following stage. So some effects seems possible.

[nickb]

I never noticed much of a difference. But the shield shields the tube's plate(s). And the plate is the input of the following stage. So some effects seems possible.

The effect is real but it does depend on the tube, the signal level and the circuit. I just had exactly this issue on a Mesa Lonestar special when in high gain mode.

My thoughts are the connections to the electrodes (i.e from the base to the electrodes) are all directly susceptible to interference. While the plate does partially screen the grid it is not fully enclosed. Further, the circuit impedance at the anode is often around 35K ohms and so not a direct path the ground. Thus noise can thus couple directly to the plate.

[Helmholtz]

My thoughts are the connections to the electrodes Ii.e from the base to the electrodes) are all directly susceptible to interference.

I would think that this partial shielding is provided by the metal flange of the tube socket alone.

Further, the circuit impedance at the anode is often around 35K ohms and so not a direct path the ground. Thus noise can thus couple directly to the plate.

This is what I meant, when I said that the plate is the input of the following stage (except cathode followers).


Anyway, it seems that most of us agree that shields should be used.