Champ 5C1 Input Question

[rp]

I just can't get my head around inputs. I want to build this with just one input, the high one. I think these are set up as switching inputs but I'm not sure if the grid leak cap setup changes things.

So is it:

#1 - 75K series resistor -> cap.

#2 - 75K series resistor w/ 75K grid load to ground -> cap

...and I want #2 for the high gain input?

Or have I got it all wrong?

Anyone think of any good reason to keep that 1st input? e.g. sounds good with a harp, etc.

[martin manning]

Nope, #1 is the high, #2 is the low. If you plug into #2 you have a voltage divider that cuts the guitar signal in half, so it's a -6dB pad. One reason to keep it the way it is is that with the shorting jack on one input it will be quiet when no instrument is plugged in.

[rp]

with the shorting jack on one input it will be quiet when no instrument is plugged in.

It's a new build not a mod to an existing amp. If I wired it with only one input I'd use a shorting jack. Wouldn't that be fine? So, if I want only the high input it's:

guitar in hot -> 75K -> cap - > tube grid?

Are the inputs done this way w/o the typical 1 meg load resistor to ground because it's part of how the grid-leak bias works with the octal tube, or because it's just a primitive set up like the Harvard I built? I ask because I might want to try the more typical input set up seen in later Fenders.

[martin manning]

If I wired it with only one input I'd use a shorting jack. Wouldn't that be fine?

Yes that would do it.

So, if I want only the high input it's:
guitar in hot -> 75K -> cap - > tube grid?

Yes.

Are the inputs done this way w/o the typical 1 meg load resistor to ground because it's part of how the grid-leak bias works with the octal tube, or because it's just a primitive set up like the Harvard I built? I ask because I might want to try the more typical input set up seen in later Fenders.

Without the 1M to ground the input impedance will be 5M. With it included it is essentially 1M, which will change the way the guitar interacts with the amp. It would be easy to try it out both ways and see.

[tonewood]

"Are the inputs done this way w/o the typical 1 meg load resistor to ground because it's part of how the grid-leak bias works with the octal tube, or because it's just a primitive set up like the Harvard I built? I ask because I might want to try the more typical input set up seen in later Fenders."

Warning- I've never used a pentode on the input! Just triodes. But-

I think you need the >1M for the grid leak bias. I don't clearly understand the grid leak bias so I change them to the more common cathode bias and am happy with the results. Depending on your voltages, you can probably just remove the cap (maybe not remove the cap?), swap the 5M for a 1M and grid stopper & look at examples for a Plate,screen resistor and a cathode bias resistor and maybe a bypass cap for more gain.

Something like this may work:

http://www.diyguitarist.com/Schematics/OctalFatness.gif

BTW- what's primitive about the Harvard?
edited: added warning and added my cap question

[rp]

BTW- what's primitive about the Harvard?

Nothing about the amp, though compared to a modern 20 knob wonder I guess it's pretty darn primitive. I meant the inputs which were set up for mic and non guitar instruments like accordions for people to plug accordion, bass, and vocals all into a 12W watt amp and which people actually did. Pretty primitive amp, pretty primitive way to reinforce sound. From what I've seen these early suckers often left out the grid load R.

The 5C1 I'll build stock and then monkey with it if it doesn't sound outrageous on first flight. I'm looking forward to building it as a grid leak bias and hoping for the best. I've never played any original amp with octals.

[Cliff Schecht]

"Are the inputs done this way w/o the typical 1 meg load resistor to ground because it's part of how the grid-leak bias works with the octal tube, or because it's just a primitive set up like the Harvard I built? I ask because I might want to try the more typical input set up seen in later Fenders."

Warning- I've never used a pentode on the input! Just triodes. But-

I think you need the >1M for the grid leak bias. I don't clearly understand the grid leak bias so I change them to the more common cathode bias and am happy with the results. Depending on your voltages, you can probably just remove the cap (maybe not remove the cap?), swap the 5M for a 1M and grid stopper & look at examples for a Plate,screen resistor and a cathode bias resistor and maybe a bypass cap for more gain.

Something like this may work:

http://www.diyguitarist.com/Schematics/OctalFatness.gif

BTW- what's primitive about the Harvard?
edited: added warning and added my cap question

My little Silvertone 1331 has a grid-leaked 6SJ7 for its only stage and it's a rockin' little amp, especially for blues stuff. I don't usually do grid leak designs on modern builds but with vintage inspired stuff it works well (my 5C3 build is grid-leaked and I really dig its tone).

[martin manning]

The bigest drawback to grid leak bias is that the bias point will vary with each particular tube and also as the tube ages. You may have to experiment with the value of the grid leak resistor to get something that works and that you like. I can imagine that there could be desirable tonal differences, but cathode bias is much more predictable and stable.

[rp]

Isn't the series grid resistor, the 75K, also the bias resistor? So it's these I play with? I assume a lower value gives more gain just like cathode bias. What will I hear that indicates I want to try a smaller or bigger value, tone-wise or gain-wise? I should just do this by ear or can I tell anything form the voltage readings?

The bigest drawback to grid leak bias is that the bias point will vary with each particular tube and also as the tube ages.

That should make tube rolling extra challenging, more serendipity than usual needed and I only have 3 6SJ7s. So, if I don't like what I hear it won't necessarily be an indicator of a bad sounding tube, just a bad bias resistor match?

[martin manning]

Isn't the series grid resistor, the 75K, also the bias resistor? So it's these I play with? I assume a lower value gives more gain just like cathode bias. What will I hear that indicates I want to try a smaller or bigger value, tone-wise or gain-wise? I should just do this by ear or can I tell anything form the voltage readings?

The 75K's are doing two things. One is they serve as mix resistors to isolate the two inputs from each other, and the other is they function as grid stoppers which, in combination with the input cap and the Miller capacitance form a low-pass filter that shunts high (RF) frequencies to B+, which is AC ground.

The bias resistor is the 5M, and it works by resisting the flow of electrons on the grid to ground, keeping the grid more negative than the cathode (which is tied directly to ground). That is the one that you might have to play with. A good DVM should be able to measure the bias voltage across the grid leak resistor, which I think should be around 3V. You could also look at the voltage drop across the plate load resistor (250k) and calculate the current, which I think should be around 3mA. I have no idea what the operating voltages (plate and screen) are, though; these are the data sheet values. Since this is a pentode it's not as straight-forward as the triode case.

The bigest drawback to grid leak bias is that the bias point will vary with each particular tube and also as the tube ages.

That should make tube rolling extra challenging, more serendipity than usual needed and I only have 3 6SJ7s. So, if I don't like what I hear it won't necessarily be an indicator of a bad sounding tube, just a bad bias resistor match?

Yes. I guess you could roll a few grid leak resistor values through pretty easily, say 4 to 8M and see what happens. I'd just try the nominal 5M first, then once you know the voltages you can check the bias point with the data sheet. Here's a good one: http://www.r-type.org/pdfs/6sj7.pdf