Single input plexi - switching vs standard input jack

[pjd3]

Hello, just curious here.

For my latest stereo plexi build, each amp/channel with have only a single input.

Is there really an advantage to using a switched input jack as opposed to a standard 2-conductor input jack? I see that with a standard input jack you would still have 1Mohm resistor from ground to the two 68K grid resistors.

I think it just came to me. You still want the right cut off frequency determined by the 68K resistors coupled with the Miller capacitance. So the 1M input resistor would raise that cutoff freq way too high to be effective.

Does that sound right?

I think I should just go with the switched Jack.

Thank you,
PJD3

[LOUDthud]

The main issue with the 68Ks is the noise. Without the second input, use a grid stopper of 10K. If high frequencies are a problem, add a small cap to ground at first tube.

If you need to use both channels at once, try this

Both_Input.GIF

[xtian]

I much prefer a single jack. I always jumper four-holers, anyway.

[sluckey]

Is there really an advantage to using a switched input jack as opposed to a standard 2-conductor input jack?

Yes. The shunt allows you to ground the input when nothing is plugged in. This will eliminate noise that may be picked up by the non-switched jack.

[LOUDthud]

For my latest stereo plexi build, each amp/channel with have only a single input.

I missed the part about it being stereo. Is it stereo all the way through, power amp and all ?

[pjd3]

Thanks everyone, although this will be a stereo amp, it will really be two discreet amps in one cabinet with even having separate power transformers.

I'm kind of doing two Sluckey 6V6 plexi's in one cabinet. In his redrawn Mark Huss schematic it is one input jack (switched) fanning out to the Normal and Bright volumes through 68K resistors. It appears that with no plug the grids would be seeing the 68K resistors out through the 1M ohm resistor to ground. Although the grid in that case would be seeing ground it would be over 1M ohm. Whether that would be too high a grid resistance that could cause any instability I"m not certain although it struck me that the whole idea of those grid resistors is to filter out frequencies over about 15Khz or so to avoid oscillations or RF pickup. With the switched input jack the grids would be seeing the intended grid stop values or darn close to it. When there are two inputs, a low and high, I can see that the switched input jacks are essential for the intended operaton of the inputting.

I'm just going to go with a switched single inp jack which will likely just be the prudent and optimal choice. I had a couple of Switchcraft standard 1/4" jacks so, that had me question the prospect of using them.

Best,
PJD3

[sluckey]

It appears that with no plug the grids would be seeing the 68K resistors out through the 1M ohm resistor to ground. Although the grid in that case would be seeing ground it would be over 1M ohm

Look again. With no plug the 1M is shorted. So, each grid sees only 68K to ground.

[pjd3]

Yes, I did see that, sorry for not being clear.

What I meant was that having a standard non-switched input jack would include the 1M ohm resistor in the grid path to ground. Did have my Wheaties today.

Thanks Sluckey, no standard input jack for my amp(s).

I was just trying to be cheap and see if I could use my standards. Nope. Aint doing it!

Thank you very much,
PJD3

[GlideOn]

That 1M resistor is NOT a part of the signal path, but rather an integral component for the Grid of a 12AX7 to have a load connected at all times, whether your guitar is plugged in 250k -1M load via pots) or is it shorting/jumpering the 1M resistor to ground path when unplugged, there is a load connected at all times.

IMHO, this is why raw schematics can be a bad way to solely interpret the function in an amp and they are usually poor when it comes to representing mechanical function vs electrical.

You either need a shorting jack or a relay with resistance. Most amps just use shorting jack because...well, it's simple and cost effective. Ain't broke don't fix it.

An if you want less signal or high content cut, go lower in resistance with the 68ks or use single 10k to 33k resistor, metal film or oxide too as they produce 10% of the thermal noise as carbon ones do which are immensely useful at the very beginning of the amplifying stage.

[Roe]

As a plexi geek I never ever jump channels. This is just an attempt to tame overly bright lead amps.

But the two channels are useful for other purposes. You can raise the volume of the unused channel to change the tone, eq and gain, since it affects the voltage divider and bright cap of the channel mixer. If the bright channel is too bright, just turn the normal channel to 7-8 without plugging in to it. The effect of the 500pf bright cap across the mixer resistor is then minimized, giving less mids and more gain, esp. in the lows.

Also, it is possible to use different pedals for the different inputs and to switch between these. That said, most of us hardly ever use the dark, Clapton channel

[pjd3]

Thats interesting Roe, I never thought of that.

I do find it useful to mix in a lttle normal channel volume for some added body and fullness, and from time to time when using very bright/harsh overdrives its very useful for compensating for that - turning the bright volume way down, and the normal much higher than usual.

But, I gather that might work (increasing unplugged normal volume ) only on Plexis with separate normal and bright volumes? Will have to look at that.

Thanks and best,

PJD3

[cdemike]

Agree with Roe in terms of limited utility of channel jumpering. IMO the only reason to jumper is if you want to reduce output impedance/get more gain from the first gain stage by paralleling V1. Even then, it'd probably make more sense to redesign it to take better advantage of the lower output impedance (e.g. eliminating the mixer resistors, lowering the grid stoppers from the jack to V1's grids, increasing the input impedance back to 1M at the jack, etc.), which would also allow improvement to signal-to-noise ratio pending component value and material selection. Otherwise, I think it makes a lot more sense to just adjust coupling cap and the first stage's cathode bypass cap's values to taste or make them adjustable on the amp and use the triode for something else.

[pjd3]

Thank you for giving me some good things to think about.

I'll be putting these approaches to good use!

Best,

PJD3

[GlideOn]

There's a Merlin article or two about using as low as input resistor as possible to generate the least amount of "hiss" or noise as everything is amplified from there.

Keep in mind though that many, many designs rely on some resistance from the grid in order to keep everything tight and in check the rest of way of circuit.

You may find that 10k or lower generates too much bass, too much flubby signal and simply rasing it to 20k or 33k (many modern Friedmans, Bogners, use this) is a fair compromise between noise floor and fidelity.

Also, the other X factor is that playing in a mix, any high powered PA or Bass amp will likely hiss louder than any guitar amp, so getting it acceptably quiet should be the main goal. There is no such thing as physically dead-silent.

[Helmholtz]

An if you want less signal or high content cut, go lower in resistance with the 68ks or use single 10k to 33k resistor, metal film or oxide too as they produce 10% of the thermal noise as carbon ones do which are immensely useful at the very beginning of the amplifying stage.

As there's no DC current flowing in the input grid stopper there will only be thermal noise.
Thermal or Johnson noise solely depends on resistor value and temperature, meaning that all resistor types will perform the same independent of resistor technology.
Type differences will only show with current (distribution) noise, e.g. with plate resistors.