Do any amps use this idea, and what's your take?

[iknowjohnny]

This is just an idea i had. Sometimes certain things such as gain staging or treble attenuation might work better in one area than another. I was thinking what effect it would have to decrease the gain at the input by using a small ground resistor, then adding gain later on. I'm talking high gain preamp by the way. But the way that i calmed the high end was with large grid resistors which also curbed some blocking at high gain knob settings and oscillations.

So i started thinking about this and it seems to make sense that if you are having to resort to large grid resistors to calm issues that are gain induced, why not remove them altogether or at least make them small (say 47k) and kill the gain at the first grid? I know you'll tell me use voltage dividers, but i've tried every imaginable scenario and the was the preamp is now is the best it gets, and VD's have always killed the tone in this thing. So what is wrong with my theory? You all know i'm a retard when it comes to electronic circuits, so i may well be missing the obvious. Or has anyone used this idea? I don't think i've ever seen a ground resistor at the input less than 1M, but even if there is certainly nothing down to like 100k which its i think what it would take to remove say 30% of the gain. I'm sure there are tonal implications too.

[selloutrr]

A mod for fender amps refered to as the "DRIVE MOD" is to remove the 1meg resistor on the input and make it a variable 3~5K resistor (potentiometer) You can also remove it all to together or leave the 1meg and add a pot.

You can also do the same with the 68K resistor using a 100K pot for "voicing".

I'm sure I'm following exactly what you are thinking but I think what you are after is the Drive and/or Voicing mod.

[iknowjohnny]

What i'm thinking is a swap....swapping gain in 2 different areas. In other words, i want the same amount my preamp now has. But i'm thinking of taking the gain down in one area and adding gain in another to make up for it. The reasoning being that with all things, treble attenuation, etc etc, the amp's voice and feel, dynamics, everything can change when switch WHERE certain things take place. As an example, i used to have the amp set up with 2 gain knobs at different stages and if i turned gain 1 down and gain 2 up the tone would be quite different than if I turned gain 1 up and 2 down even with the same amount of gain happening. I've always been interested in how things are affected when you choose a different area in the circuit to try things.

On that drive mod tho....how would you get any gain with a 3-5k pot in place of the 1M? You'd be pretty much killing the entire signal !

[selloutrr]

your approach to gain and makeup gain is used in studio gear. compressors how hard you hit the signal then makeup gain.
or you can run the signal through the unit and use mostly makeup gain and little input so you hardly compress but get the tone of the circuit.

Take a look into the 1176LN and LA-2a circuit.

here is a link the drive mod If it's not working look on page 6.

http://www.amazon.com/Great-Tube-AMPS-G ... 0933224907

[iknowjohnny]

I see....you meant to say 3-5 *meg*, not *k*. But i was talking about the opposite....a SMALLER ground resistor than the 1 M thats there. Somewhere down around 100-220k probably. If i recall i once tried that and at around 100-180k, somewhere in there, the tone went bye bye. But when i tried that i wasn't attempting to throw away gain and make it up later like i'm talking about now. I was just trying to lose gain period.

So yes, it IS like in a studio and is called gain staging which is exactly what i'm talking about. because i think it's one key in a amplifier that can make or break the tone.

[Cliff Schecht]

I don't like putting gain later in an amplifier, perhaps just because it's against my engineering training. Friis noise formula predicts that the first stage of a multi-stage amplifier will contribute the most noise to a system and each subsequent stage gets its noise power divided down by the previous stages gain (or stages if you are past two). So the idea is to squeeze as much out of the first stage as possible to get the best possible signal to noise ratio. While this formula is typically applied to RF systems in terms of power, it is adaptable to audio and essentially the same ideology holds true. I tend to follow this rule in my tube amplifier designs more out of habit than anything else but I haven't really played around with seeing if more gain later really does add more noise. But I can say that with other designs I've done (discrete analog stuff) this idea definitely holds true!

http://en.wikipedia.org/wiki/Friis_formulas_for_noise

It's funny how there are some design paradigms that many people seem to neglect on a whole, probably more out of ignorance and lack of experience with general analog electronics than anything else. Not to poke fun at the OP as this isn't something I would expect him to know, and even many of the top manufacturers and boutique builders seem to not understand this idea. It's funny how we throw conventional logic out the window in order to achieve the best sound possible.

[iknowjohnny]

The thing is, i have had this amp in more variations than you would ever imagine. This is my 3rd amp of this design, and i wore out the others from soldering and resoldering parts in them if that gives you any indication. I mention this because in all those different variations it's always been very quiet. So noise isn't an issue for me for whatever reason. I guess the ground scheme i used is just good, i dunno. But i DO understand how gain staging can result in noise, i'm just not worried about it. Especially considering i can always go back. I was more interested in what thoughts people would have on this idea as far as if it's been done, if they'd tried it, and if they had any thoughts on it that i might have missed. You added your thoughts on the latter in the form of noise issues, and thanks for that. But it's not likely going to be an issue since at this point it never has.

I guess what i should do is put a 1M pot in place of the input ground reference resistor and after removing the grid stoppers on stages 2 and 3, turn the pot down till theres no perceivable blocking and the right amount of gain and see how it sounds.

[Firestorm]

Replacing the 1MR with a 1M pot is fine, but attenuating the input signal (however you do it) has essentially the same effect as turning down the guitar volume pot which is in parallel with the 1M. In fact, if you look at the input circuit on three-input Vibrolux amps, you'll see that the "high" input has no grid-to-ground resistor at all: it uses the guitar volume pot to find its ground reference.

To tweak gain, you might want to play with the plate loads on the preamp tubes. Some modern high-end designs (I think I've seen it on Komets or something of that ilk) even put a cermet pot in series with the plate resistors, so you they can dial in the sound in final assembly. If the preamp stage has an unbypassed cathode resistor, you can play with that too.

[iknowjohnny]

I know, and i have done all that. Split load plates included. But thats not what i'm trying to do. And the input resistor being in parallel with the resistor, yes, i know that also. But in fact the reason o thought of doing it this way was to act like i'm using a very low output guitar, even lower than the lowest pickups. But I don't believe it affects the tone of the pickups the same as turning the guitar down, especially considering i have a treble bleed on the pot. Plus thats not an option anyways because then i have to range left on the pot to clean up.

I think i just haven't been able to explain what my intention and motivation is, so i'll just leave it there. Thanks anyways.

[Ian444]

I believe Steve Ahola was trying exactly what you are suggesting johnny, the smaller resistance to ground to cut gain, at the input of a Blues Express amp. He called it a trim control. Bottom of page 2 and page 3.

http://www.blueguitar.org/new/articles/ ... uesxpl.pdf

I have considered putting a JFET buffer (gain of 1) right at the amp input, followed by a pot, and use that pot as the regular "volume" or "gain" control of a typical amp. Then use the other vol or gain pot as the "master". But what would I know, just started playing around with guitar ccts recently. Its a lot harder than hifi tube amps, you change one cathode resistor in the preamp and the tone changes, WTF?

[sliberty]

In one of my Express builds, I did a number of thing to reduce gain as I wanted a more managemable version of the Express. One of the things I did was to replace the 1M resistor on the input jack with a 1M pot. As mentioned above, turning this pot down will allow more signal to bleed to ground, which is similar to turning down your guitar vol. But, since you can STILL turn down your guitar vol, it really does add value. I call it input sensativity, and you can set it to whatever value makes you happy for max level. Then, use your guitar pot for reducing level (and gain) further during appropriate times.

It did what i wanted for that amp, but truthfully, I haven't used that trick since in any other amp. I think, for me, it would be better to use a pot during development of the amp, and settle on a favored value, which can then be installed in the amp permanently. But thats just me.