Input voltage (from guitar) for design?

[nashvillebill]

Might seem like a basic question, but here goes: When designing V1, what value for input signal (coming from a typical guitar) is generally used? I've run LTspice using 1 volt p-p but is that low/high/or about right?

I guess to really do it right, the guitar pickup ought to be broken out into an equivalent inductor/resistor/current source, but that's more than I can wrap my tired mind around...

[martin manning]

I believe 100-200 mV is typical. I've never tried to measure it, but I understand a hot humbucker can produce over 1V.

[MWJB]

An input stage in a guitar amp will amplify the signal by anywhere between x30 and x60 with a 12AX7 tube, lower end may utilise an unbypassed cathode, higher end (& most larger guitar amps) will have a bypassed cathode. By the time the signal hits the power tubes, it might have been amplified by anything from 2000 times for a simple SE amp to x1000000+ for a big, high gain amp...so input voltage is not a huge factor in the big scheme of things...the typical amp input can handle signals of several whole volts.

The guitar source signal doesn't really need too much consideration, player might want to amplify an accordian/keyboard/microphone/mp3...?

A large part of an amp's character/voice comes from V1, so that is probably the thing to be thinking about. Running higher plate voltages at V1 tends towards a cleaner, brighter sound, as will larger cathode resistor values. Lower voltages will be browner/distort earlier, as will lower cathode resistor values.

[nashvillebill]

Thanks! That's getting more like what I am trying to figure out...how much gain to shoot for, not only on the first gain stage, but on successive gain stages. 3 gain stages at 30x each would be 27000x overall, then that would be attenuated to provide correct grid signal amplitude on the power tubes.

Thanks, looks like I'm going to be asking Santa for some more books for Christmas...maybe Merlin's books and Kevin O'connor's Principles of Power.

[MWJB]

", then that would be attenuated to provide correct grid signal amplitude on the power tubes.", well yes, the amp's volume & tone controls would take care of that. Also check out the schems at Fender.com, they have ac voltages as well as dc.

30x30 seems on the low side unless the amp is a small single ended design, if push-pull the type of PI will make a big difference, x30 is at the high end for a LTPI, at the low end for a cathodyne with preceeding gain stage, my point being, you have to consider the big picture, the amp as a whole rather than modular, unrelated stages.

[martin manning]

In reality the signal will be attenuated between stages, not at the PI output. In fact the PI output is likely the only place where it is not attenuated at all so that the ~50V P-P required to drive the power tubes is available. The first gain stage might increase the signal 40x, but then it will be reduced, essentially back to its original level, by a tone stack and volume control. This might happen a couple of times, more or less, through more gain stages with the net voltage gain at the PI input being around 40 (you only need ~5V P-P there for a typical 12AX7 LTP). The PI output might be 400x the input voltage, with the signal at the power tube plates being ~2500x.

[MWJB]

...indeed, but there can be as much as 10,000 times input voltage at the PI output & over 50,000 times the input voltage at the power tube plates.

[GregS]

You can measure your guitar's output by connecting it to a jack with your DMM leads clamped to the terminals and taking note of the AC voltage when you play.

You can calculate/approximate what the signal levels should be throughout your design on paper and be in the ballpark. However it helps to measure what the signal levels actually are with an oscilloscope and signal generator once you've built your design. That way you can decide if you need more gain or attenuation at any particular place to ultimately get the results you want. This was my experience with a design I recently did and the test equipment certainly helped me to structure the gain exactly as desired throughout the stages. Also you can determine where distortion is occurring first as you increase volume etc.

Hope this helps,

Greg

[GregS]

One more thing. Kevin O'Connors book TUT5 Tone Capture has information on gain structure throughout amplificaton stages starting at the guitar that might be useful to you.

Greg

[jimboyogi]

I once measured the output of a Jazzmaster with stock single coils, using an oscilloscope. Vol and Tone on full.

Chords - peak at ~ 2V p-p, sustained at ~ 0.8V p-p

Single notes - peak at ~ 1.2 V p-p, sustained at ~ 0.4v p-p

A standard input stage (12AX7) is biased to at least 1.5V, giving a headroom of ~ 3V p-p, which according to the above measurements is big enough for single coils, but may struggle with transient peaks from a higher output pup eg. a humbucker.

If you put a pedal in front of V1 with any sort of gain then you could quite quickly get some distortion from too much input voltage to V1a.

I usually bias V1a to ~ 1.8V to give a bit more input headroom.

[nashvillebill]

I suspected that starting with "a blank piece of paper", so to speak, would wind up with too many unknowns. So ultimately I'd need to compare classic amp designs and decide what amplifier scheme to go with: for example, classic Fender clean with only a couple of gain stages like an old Bassman , or super high-gain like the Peavey Triumph....then work out details from there. I have looked at soem of the gain stages on that Peavey triumph with the sig gen and scope, it seemed like there were lots of stages that had no gain at all.

I originally had thoughts of simplifying (and making smaller/less powerful) something like my ancient Ampeg V4, but that's a pretty complicated design.

[GregS]

Although some stages may not add to the gain they build up harmonic content which makes the sound fuller so they do serve some purpose.

Greg