How much MORE gain do you need for an on-board boost?

[Mark Hammer]

Sitting through focus groups today, and while they're reading the material to comment on, I was sketching out something I wanted to try making this weekend.

Let us say one was making an overdrive of some kind, and you wanted to incorporate an adjustable stompswitchable boost to the basic overdrive sound for solos and such.  What, for you, would constitute a sensible amount of boost to what is already a boosted signal?

Imagine you had an overdriven signal that was already hotter than bypass level (subject to attenuation by an output level control), and you could set an on-demand clean boost of that overdriven signal.  Would a simple 3x boost be about as much as you'd really need?  10x ?  20x ?  How much "more" does one need for a solo?  I'm thinking 10x is probably enough, but when it comes to hearing and higher volume levels, a non-gigging musician may have different standards.

[Bill Mountain]

When working with tube amps and you want to make a dent I would go with 20dB.  You could maybe adjust it with a trim or external pot.  I have an input stage on a bass overdrive that has a switch for 0dB or "about" 18dB.  Now...I'll usually run out of headroom before I get all 20dB but it'll allow you to get as much gain as possible and the resulting clipped boost stage will add some girth to the signal.

Now...some people use a Distortion + as a boost and that is quiter than many guitars by themselves!

[R.G.]

Bill correctly identifies amp headroom as an issue.  When the amp is already putting out all the audio power it can, feeding it a hotter signal only makes the signal more distorted, not louder. And an amp input can only accept so big a signal before the input overloads. A 12AX7 as usually biased can only accept something like 1.5V peaks before going into grid-positive clipping no matter what its plate is doing or what the other stages can amplify to.

And you're correct that this is a bigger issue in the practice room than it is on stage. On stage, you may be near the amp's input overload already.

Then there's the quirk of the human ear's amplitude response. To get to twice as loud needs (about) 10x the power, or 20db more signal voltage if there is nothing already clipping and keeping you from getting that +20db amplified.

It strikes me that it might be interesting to cut in more amplifiers, in line with my modular amplifier/speaker theory. Leave the incoming signal the same, but un-mute another half-dozen or so amplifiers. 

[midwayfair]

It strikes me that it might be interesting to cut in more amplifiers, in line with my modular amplifier/speaker theory. Leave the incoming signal the same, but un-mute another half-dozen or so amplifiers. 

Honestly, give that people work so hard to reach the sweet spot where their amp stops getting louder and starts distorting but then they want to be able to boost for solos, this is probably the best boost solution anyone could ever ask for. Two amps in a stack = more sound.

[R.G.]

I'm fond of the idea of putting a 50-60W LM3886 power amp (and the power supply) in a cab with a speaker, then making as many of these as you need. Small gigs need one. Clubs, maybe two. Shea stadium, take a dozen or two. 

[Paul Marossy]

It strikes me that it might be interesting to cut in more amplifiers, in line with my modular amplifier/speaker theory. Leave the incoming signal the same, but un-mute another half-dozen or so amplifiers.  

Honestly, give that people work so hard to reach the sweet spot where their amp stops getting louder and starts distorting but then they want to be able to boost for solos, this is probably the best boost solution anyone could ever ask for. Two amps in a stack = more sound.

Isn't this one of the principles behind a lot of "two channel" amps? I suppose not necessarily as one channel might simply create more distortion than the "clean" channel does but not necessarily much of anything in terms of dB and they both often share the same tone stack. I know, it depends on the amp design. I'm thinking of my Fender Hot Rod DeVille right now, but I have another amp that does have two channels with their own tonestacks. You can have quite a difference in sound levels depending on how you set the controls.

[R.G.]

It is, and it isn't. Human hearing happens in both the ear and brain. The brain takes the raw data from the ear and massages it - a lot. The ear will hear different mixes of signals with more/less harmonics, intermodulation, etc., as louder or quieter, even with the same peak change in actually sound pressure changes. There is a tendency to think of things that are more distorted as louder, even if they're not really.

And yes, on amps with channel switching, one or the other channel can be louder than the other, so switching can do what Mark is suggesting. But there's a limit at what the amp can put out; it has a power output limit. Channel switching can't push you over that if the power supply and power amp won't do it.

As long as the power amp still has some room to get louder, you can do exactly what Mark says: make the input bigger. But if the amp is already near it's power output limit, bigger input won't make it louder. And if it's near its input overload edge - which is different - a bigger input makes it more distorted, but not actually much louder, after you subtract out the psychoacoustics.

[Gurner]

If you want a clean boost too, then the limiting factor is not just the rail  of the follow on amp, but the rails of the boost cct itself....else your boost will start colouring the sound (with clipping)

For a typical 9V supply, there's not a lot of headroom to work with. A typical humbucker puts out something like 1V to 1.5V Ii.e. when the string is plucked hard). Most 9V stompboc ccts can't get to full rail swing (due to the components they typically use...TL072s etc)....you're likely looking at 7v swing max....so you can only really expect to squeeze 4x voltage gain out of a 9V circuit before you start getting clipping going down in the initial plucked note.

I reckon if you want a good boost, then don't look towards the front of the signal chain )stompbox)....look towards the end of the chain (a better amp with decent switching flexibility)

[Gurner]

double post - deleted.

[Paul Marossy]

And yes, on amps with channel switching, one or the other channel can be louder than the other, so switching can do what Mark is suggesting. But there's a limit at what the amp can put out; it has a power output limit. Channel switching can't push you over that if the power supply and power amp won't do it.

Here's what I have always done: I don't usually ever have the volume on my amp past 4. I can go all the way up to 10, so in my mind, the power amp section can still put out a lot more volume. So if I hit the preamp with a pedal with a "high output", the amp seems a lot louder. Is it that the power amp rises to the challenge or the preamp can't handle it and it's all happening in my mind?

I guess I ought to get a dB meter and measure it. This has me a little curious now.

[Mark Hammer]

If you want a clean boost too, then the limiting factor is not just the rail  of the follow on amp, but the rails of the boost cct itself....else your boost will start colouring the sound (with clipping)

For a typical 9V supply, there's not a lot of headroom to work with. A typical humbucker puts out something like 1V to 1.5V Ii.e. when the string is plucked hard). Most 9V stompboc ccts can't get to full rail swing (due to the components they typically use...TL072s etc)....you're likely looking at 7v swing max....so you can only really expect to squeeze 4x voltage gain out of a 9V circuit before you start getting clipping going down in the initial plucked note.

I reckon if you want a good boost, then don't look towards the front of the signal chain )stompbox)....look towards the end of the chain (a better amp with decent switching flexibility)

Thanks to all for the comments so far.  Very helpful.
In my own case, the intended circuit is a simple 2-op-amp device with the clipping from a 2+1 GE diode complement to ground between the stages.  Since the range of the signal feeding the boost stage is clamped at a maximum of +250mv/-500mv (more or less), in theory one could achieve something close to 20db of headroom/boost in the second stage if the Vref was "adjusted" for the 2nd stage to be at the 33/67 point rather than the 50/50 point.  Of course, realistically, the signal never really remains at the ceiling created by the diodes for very long, so the amout of headroom varies depending on where in the note's lifespan you happen to be.

[fpaul]

After 40 years of playing I'm finally learning to use the vol on my guitar to do this.

[Paul Marossy]

After 40 years of playing I'm finally learning to use the vol on my guitar to do this.

I use a volume pedal and guitar volume as well. I do everything but use the other channel on my amp.

[Ronan]

When using dirt pedals, or a cranked-up amp, I found the guitar volume wouldn't control the amp volume, it just varied the amount of dirt. Getting a boost to cut through for lead can be difficult. Years ago I used to use an arrangement with an LDR mounted on the master volume of the amp, footswitchable, which gave a volume cut. When the LDR was switched off, it became a "boost". It wasn't easy to set up, and required a dedicated footswitch box and an extra lead to the amp, but it worked.

After about 15 years of not playing, I am building/gathering pedals and amps again and rethinking my old setup. I put a boost switch into a Tonemender build using an LDR in the feedback network of the IC. Since Mark was asking "how much MORE gain do you need?" I got curious as to how much gain I was adding in the boost section of that pedal. I just measured the boost setting I last used (live) and it is set to around x 1.4 voltage gain. I am surprised it is that little, I was expecting at least x 2. The box is the last in line before the amp, which is set reasonably clean.

[fpaul]

it just varied the amount of dirt.

I agree with that and that is usually what I'm trying to do.  If you want to boost with the same tone you probably would need something else.  Your solution may not be simple, but I believe I would try it before I spent the time and money to build another half dozen amps.

[Ben N]

Mark said nothing about his boost being clean. As RG and others have noted, boosts generally aren't clean, because they slam into the headroom limitations of the amp right quick. But for most people that isn't a bad thing at all, and it doesn't necessarily mean that there's no volume increase available, either--depends on how you set your amp. If the amp is already at saturation, then yeah, slamming the input will only get you a modest volume increase at best. But if you run the amp clean--and note that in Mark's application, there is overdrive going before the boost kicks in, so presumably that is intended to provide the base distortion level, not the amp--there is enough headroom on tap that boosting will give you some amp saturation and clipping (how much comes from what stages in your amp is another question), but also a significant increase in volume. But yes, it is tricky to get all the levels right so the clean sounds good, and the boost actually boosts. A bit of tone shaping on the boost goes a long way in extending the perceivable volume increase from the boost.

[Mark Hammer]

Sensible comments.

I'll augment them by differentiatng between a clean boost and a clean outcome.  In that sense, we're talking about a boost which does not intentionally add any coloration of its own, but which may produce coloration in subsequent stages (and we should not assume that the amplifier is necessarily the next stage).

As others have noted, if we are talking about a traditional/canonical 9v-powered pedal, there are finite limits to how much gain can be applied, no matter how it is distributed throughout the circuit, and how much headroom there is, when applying gain to a signal of the general amplitude of a guitar pickup.  So, even where the design of the added boost stage attempts only modest additional gain, and makes no deliberate steps towards coloration, that's entirely separate from whether there is any more voltage swing left in the circuit by the time the input signal reaches that last output stage.  The circuit, removed from its context, will look as clean as hotel towels on paper, but may be asking the power supply to do the impossible, and produce coloration as a result.  And, as noted at the outset, that hotter signal, even though constrained by the power supply of the pedal, will still elicit coloration from the amp because of its limitations.

In a sense, an overdrive pedal with a switchable boost stage turns into triple clipping, because one has a) the overdrive itself, b) the headroom limitations imposed on the boost stage, c) the headroom limitations imposed on the amplifier.  And we're not even touching on how gain is distributed within the amplifier and where the clipping occurs there.  So "triple clipping" may well underestimate the number of loci where coloration is introduced.

Now that I'm thinking about it, I wonder if this might not be a good point to begin learning more about charge pumps.  And possibly using 9v for the overdrive stage, but a higher supply voltage for the boost stage.  Of course, if the circuit was originally intended to be just a 4558 chip. this split supply thing means I'll need to consider two single op-amps.  Certainly feasible, and permitted by my parts bins, but layout gets more complicated since there is not only the two 8-pin DIPs but the charge-pump circuit as well..

[DougH]

My Peavey Windsor has a boost function. Basically it provides an overall volume boost but more importantly it adds some EQ to push the mids more. It seems pretty effective. By re-EQing the signal you can really hear it stand out in a mix more.

[Mark Hammer]

True dat.

On the other hand, what you do to the boost will depend on what helps, and what helps will depend on where you introduce that help.  For instance, in my case, as something in a pedal, I am assuming I'm going to elicit more harmonic content from the amp, not simply raise the level (as may be true in the Windsor example).  So, I'm thinking that the boost function may want to shave off a little top end at the same time so that the lower order harmonics stick out more.

[R.G.]

Good idea, Mark.

I was just listening to some Animals and a few lesser-known groups from the 60s/UK and I was struck by how mid-range-y the guitar sound was. Not today's boom-shhhh hockey-rink sound or fully-produced studio sound. Some of the live recordings are even more so.

A little tone shaping on the boost would be a good step into getting a bit 'more'.