grid stopper question - 300B TubelabSE

[discomonkey]

Hi All,

This is my second TSE 300B. The first one has PCB mounted tube sockets and this one, I am thinking of moving them off the PCB. My question is in regards to the grid stoppers.

For the 5842, do i move the grid stoppers off the PCB and add them directly to the socket pin? Or should I keep the 4.7Kohm resistor on the board and add a 100ohm resistor at the pin? What should I do to the resistor on the grid of the 300B (if anything)?

Thanks in advance,

Mike

[rknize]

Keep the grid stoppers near the grids. Especially for the 5842.

[discomonkey]

Thanks Russ,

I appreciate your help on this build. I'll ask you another question in here, rather than start a new thread: What do you think of wiring up my caps in parallel rather than using single capacitors. There are 2 reasons I would do this:
1) Allows me to use boutique capacitors which I only have in lower values and
2) reduces the overall ESR

Thoughts?

Thanks

[tomchr]

Originally Posted by discomonkey

For the 5842, do i move the grid stoppers off the PCB and add them directly to the socket pin?

What matters is the inductance (length of wire) between the PCB and the grid. Where the resistor is placed along this wire should not matter. The current flowing in the wire "doesn't know" where the inductance or resistance is...

I know the general recommendation is to have the grid stopper as close to the grid pin as possible, but from a circuit analysis it doesn't really make any difference. Unless I'm missing something (in which case, please point it out).

Originally Posted by discomonkey

What do you think of wiring up my caps in parallel rather than using single capacitors.

I think that's a fine idea for the reasons mentioned. Plus:
3) Effective ESL becomes smaller as well.

The only draw-back is the total physical size of many caps in parallel versus one big cap.

~Tom

[discomonkey]

Thanks Tom,

I appreciate your comments. The overall length of the wires will be less than 2" and I've read of people having success with such lengths - so here's hoping I do as well.

[rknize]

I am no expert, but I believe it has something to do with the grid leak resistor. The inductance of the wire on the driven side of the grid stopper is in series with the output impedance of the previous stage and in parallel with the grid leak resistor. The inductance on the grid side of the stopper is in series with the capacitance of the grid, creating a potential oscillator. The way to slug it out is to increase R and decrease L as much as possible.

Hopefully I got that right.

[tomchr]

Originally Posted by rknize

I am no expert, but I believe it has something to do with the grid leak resistor. The inductance of the wire on the driven side of the grid stopper is in series with the output impedance of the previous stage and in parallel with the grid leak resistor. The inductance on the grid side of the stopper is in series with the capacitance of the grid, creating a potential oscillator. The way to slug it out is to increase R and decrease L as much as possible.

That's my hunch as well, though, I don't have any theory to back it up. I'd imagine that if you keep the inductance down in the loop formed by the grid stopper, grid leak resistors back through ground, you'd be OK.

Still... Say 10 pF tube input cap, 50 nH series inductance (2" = 50 mm, 1 nH/mm) --> 225 MHz resonance. Unless you're working with tubes that are made for operation at RF, you're not likely to get into trouble. Unfortunately, the 12A*7-series has gain at RF...

~Tom