Gain stage vs. Booster stage - What's really happening to that signal?

[Earthscum]

I have been looking and reading diligently trying to find this info out... basically, what is the difference between a gain stage and a booster stage?

Basically, why can a single transistor boost deliver so much louder of a clean signal than a gain stage? I understand stuff from a mechanical point of view, ie A happens, which causes B to react between C and D, which work together to produce product E. What happens to the signal as it passes through these stages, what happens to the signal? Like, for example, comparison of why a larger sounding signal can pass through a booster without clipping, but in gain stage the signal gets pounded off the rails. I understand changing component values and am starting to get my head around biasing (just need to sit for a night and run numbers and get a comparison chart so I can visualize it better), but you hardly ever get a decent explanation in reading of WHAT is really happening to the signal throughout the entire chain. What happens to the signal that comes through the feedback loop? What exactly does that signal do when it merges with the clean input signal? Does it even go through the feedback loop?

Basically, does someone have the in-depth knowledge to explain, from input to output, what is happening to the signal in gain vs booster comparison? If there's another besides boost and gain, throw that in as well. The more knowledge, the better.

[earthtonesaudio]

Maybe you should include some pics or links  to what you're talking about.  In my mind boosters and gain stages are exactly the same thing. 

Perhaps you're talking about a single transistor "clean" boosts vs. op-amp distortion/overdrive circuits?

[Ben N]

If I understand you correctly, you are using the term "gain stage" to mean a distorter, right? OK, first you should understand that although many dirt boxes label what their distortion control "Gain," gain does not equal distortion, at least not always. Gain just means the amount a signal is being amplified, in electrical terms (usually voltage). It just happens that we guitarist achieve distortion by exceeding the "normal" limits of an amplifier that is designed to quickly saturate, so the more gain there is, the more distortion there will be. When the signal is amplified to the point where the device saturates, the peaks of the signal are "clipped"--and the more you try to amplify the signal beyond that point, the more of the signal will be clipped.

How this clipping happens can vary. Tube preamps typically use cascaded stages with theoretical gains of 100x. So that first stage will not generally produce much distortion, unless you have a booster or hot pickups. It just makes the signal much bigger. However, the second stage, assuming it gets the full brunt of that boosted signal, is trying to amplify by the same amount a signal that is already much bigger than the input. That effort is doomed to failure, because the magnitude of the signal that the tube stage can pass is limited. When it tries to exceed those limits, the signal is clipped. Some dirt boxes work the same way, cascading gain stages to achieve saturation. Others create artificial limits by using diode clippers that pretty much do the same thing: An amplifier amplifies the signal, but when it exceeds the threshold of the diode clippers it gets clipped.

But if the amplifier is not designed to saturate quickly, it is said to have more headroom, and so an increase in gain will not cause an increase in distortion, just more volume.

Now, in response to your question: when clipping occurs, it generally tends to reshape the signal towards being a square wave. When it gets to square, there is no place else to go--it is at the absolute limits of the circuit to amplify it any further. So clipping doesn't jujst change the wave form, it also caps the amount of volume that you can get, and compresses the dynamics of the signal. That's why with many amps as you turn the volume up past where it gets good and crunchy, you get a tonal/waveform change, but it stops getting louder.

An example: if you look at the schematics for an MXR MicroAmp (a clean booster) and an MXR Distortion + (a crunchy overdrive), you will see that there is not a whole lot of difference other than a pair of diodes to ground at the output of the opamp. In fact, IIRC Tonepad uses the same PCB for both, and lots of DIYers build them as a single project that switches between distortion and clean boost. But the big knock on the Dist+ is that is does not produce enough signal to overdrive your amp's input stage, whereas the MicroAmp can produce a pretty hefty boost. Now you should be able to see why this makes sense (with a bit of oversimplification to help make the point): both may have roughly the same gain from the single opamp amplifier stage, but in the Dist+ the clippers clamp down on that signal, imposing an upper limit on what will get by. That's why some similar overdrives, like the Voodoo Labs OD, provide a second stage after the clippers just to provide more volume.

[Mark Hammer]

Not that Ben is wrong, but, stated more simply....

Gain = Boost

When a circuit needs the signal to be at a certain level in order to acomplish some task, there will be some gain supplied by the circuit at one or more points.  For example, the Big Muff has four gain stages.  The first one brings the input signal up to a level that will permit the second stage to clip.  The second stage also provides gain in order to produce the clipping.

Because the very act of clipping places a ceiling on the signal, in order to clip the signal a second time, the BMP also introduces gain in the third stage.  And finally, because the tone control works be selectively bleeding signal to ground, the final stage provides gain/boost to compensate for what is lost and bring the signal up to a desired level.

A "booster" pedal is simply a gain stage (usually gain that can be adjusted, as opposed to a fixed gain), sitting on its own, that has no specific purpose, although most users of "boosters" will use them to deliberately overdrive something else.

Grass is grass, and if it's in front of your house you call it a "lawn".  If it's in back of your house, you call it a "yard".  But it's still just grass.  Similarly, in electronic terms, boost=gain=boost. 

[aziltz]

Grass is grass, and if it's in front of your house you call it a "lawn".  If it's in back of your house, you call it a "yard".  But it's still just grass.  Similarly, in electronic terms, boost=gain=boost. 

you have a way with words my friend.

[Ben N]

OK, Mark, I know I'm in trouble when YOU bust me for wordiness! 

[Mark Hammer]

 

[B Tremblay]

Grass is grass, and if it's in front of your house you call it a "lawn".  If it's in back of your house, you call it a "yard".

Sounds like something Oscar would say on Corner Gas!

[Earthscum]

Ok, I'm getting it, although what Ben and Mark both said is pretty much what I DO know.

I'm seriously baffled by my first project, which was a single transistor booster (I think from AMZ). That thing was clean, even when cranked all the way up, but it didn't seem to do much other than make my signal louder. I tried clippers in different positions, but all it would do is attenuate the signal and wouldn't clip. From what I could see in the circuits, and like what was said, the clippers seemed to be the only difference. I just couldn't get it to distort in any way shape or form. That's where I thought there's something minute that I was missing. When I built my GF's Bazz Fuss, I couldn't keep it from distorting... even with the clippers out, it was still a dirty boost.

So, I think my question comes down to biasing. So, if I'm running , say for instance, a 100k resistor from drain to battery positive in one circuit (and other resistors, just for argument's sake, matched to get 4.5V at source... correct me if I'm off, of course. maybe base?), and then in a different circuit I'm running 470K, the Transistor with the lower (current?) is going to have less headroom, or would the trans with the higher current have more headroom? Which one is more likely to slam the rails first?

Another thing concerning me is the purpose of the feedback path. I don't think I've ever got an explanation of why it's there, beyond "Just follow the schematic". Does the signal itself travel that path? What's it's purpose? All I can assume is it increases gain somehow, but I don't understand the method, I guess. That's why I was wondering if someone could do a walkthrough as the signal would travel. Why doesn't the signal pass through the junction and to ground when there's no resistor from source to ground?

Explain it like you are showing me how to break it... I firmly believe that if you don't know how to break something to the best of your ability, you don't really know how to fix it to the best of your ability. 

[earthtonesaudio]

Pictures or links to the specific circuits you're referring to would help a lot.

[Ben N]

I don't think I can answer all of your questions (what a relief, right?), but I'll give it a shot.

As for your booster that would not clip: that would depend on (a) the gain of the booster, (b) what kind of diodes you were using and (c) where they were mounted. Since it, whatever it was, was intended as a clean boost, I guess it was working right!

Same for the Bazz Fuss--that is a circuit that cannot do clean, so yours is working right. So far you've got a booster that boosts and a fuzz that fuzzes, so the problem is??? BTW, I'm not sure what you are referring to as "the clippers " in the BF. The basic design has a single diode between the collector and the base, but, as far as I understand it, that does not function exactly as a conventional clipper. But there are a lot of variations on the BF, so...

You are also confusing power supply with biasing. Biasing relates to a reference voltage that is applied to the input of the device. It ain't about current, and it is determined by the voltage divider that splits the power supply into two values, one of which becomes the main supply and the other of which is the bias supply. AFAIK the bias supply uses very little current. However, you are right that the Bazz Fuss is all about (mis-)biasing.

Headroom is AFAIK (at least in part) a function of supply voltage. I don't know if current matters, but I believe that a series resistor only affects current, not voltage--for that you need a voltage divider.

Finally, (negative) feedback: it's purpose generally is to linearize the output of non-linear amplifiers, like transistors and tubes. It does not increase gain, it reduces gain, because it mixes a portion of the device output out of phase into the input, leading to cancellation of part of the signal. For opamps, it is how you set gain. So, for example, the maximum theoretical gain from a non-inverting opamp is "open loop gain," i.e. no feedback from output to (-) input. Minimum (unity) gain is when there is a straight wire, no resistance, from output to (-) input.

[Earthscum]

Pictures or links to the specific circuits you're referring to would help a lot.

Basically just use the MXR circuits like Ben was talking about and you'll have A basic circuit comparison.

If nobody else posts a link before I get off work, I'll dig them up. Right now I gotta go, already late.

Thanks to everyone so far. I'll think about the info I have right now and see if i can pose a question that might get me to that point of what I am missing. I know it's gotta be something simple and stupid that I'm not getting. I guess I just want to be able to get inside the transistor and SEE what is happening, like when I was a kid... I was so confused about how you could put meat in the top of the grinder and it came out the bottom as hamburger. Then I tore the grinder apart and actually made it work by hand and then I understood the mechanism. In Jr. High, I got bored and made a model of a supercharger using a busted blow dryer and a soda can just to teach myself how it works and why it does what it does. I can't tear apart a transistor and look inside the same way because signals aren't visible like meat, lol. I'll see if I can figure out where there's a blank spot that I'm not getting and then I'll probably end up figuring it out.

[Mark Hammer]

Grass is grass, and if it's in front of your house you call it a "lawn".  If it's in back of your house, you call it a "yard".

Sounds like something Oscar would say on Corner Gas!

Him?  He's a *&^*&^.

[GibsonGM]

You should look up transistor amplification on Google, Earthscum.  Helped me a lot, to understand WHERE the signal lies on the device's characteristic curve (bias level). 

Look at it this way:  the active device (transistor, opamp) receives the signal, and it is connected to your power source.  The bias supply allows the device to amplify BOTH sides of your sine wave...without a bias, it will only work on 1 side and be terrible sounding.  It makes the device "think" the signal is at say 4.5V when it is at zero volts, so to speak.   As mentioned above, resistors are selected to give a desired level of GAIN (=boost=gain).  Feeding back some signal, out of phase (merely the opposite polarity of what was input) will mix with the input signal and bring it down, 'controlling it' to stay at the selected gain level.

Internally, the active device acts like a valve (why tubes are called valves in the UK)....allowing a very small voltage (or current) to control a larger voltage (or current).  Sort of like Spirograph, where you trace a small drawing and the armature reproduces it, only larger!  But I am almost 40, you might not remember that toy, lol.  The concept of what's going on in the transistor etc is pretty simple (altho the physics is not so). 

If you operate a device out of the linear portion of its curve, it will be unable to deliver an undistorted (faithful reproduction of input) output.  This can be done via opamp, transistor, FET, MOSFET.....diodes, altho they're not active devices.   You can either mis-bias the device, overdrive the input, or turn down the power to the device to cause this non-linearity.  SO, clean gain just = a device operating with plenty of power to amplify as much as you asked it to, which is properly biased into the center of its characteristic curve.   "Dirty gain" = a device which is driven into some clipping through overdriving the input, biasing it out of linearity, or limiting its power supply...this is why as you lower the voltage supply for tube preamps, say, you get more distortion....this is LESS HEADROOM.  Whereas, in a hi-fi PA amp, you will find tons of headroom to give flawless sound reproduction (what guitarists would call lifeless, ha ha).

Hope this helps a bit!  Check out "characteristic curves"; maybe this site will help, too:

http://www.angelfire.com/electronic/funwithtubes/index.html

[Mark Hammer]

Grass is grass, and if it's in front of your house you call it a "lawn".  If it's in back of your house, you call it a "yard".

Sounds like something Oscar would say on Corner Gas!

Him?  He's a *&^*&^.

Hmm, I thought I typed "*&^*&^", but somehow it got respelled.  Note: For those unfamiliar, the character in question is a cranky old man who calls everyone a jack-ass (there, a hyphen gets you through it safely).  In fact, the book we got on the show has a synopsis of each episode and shows the number of times the term is used in each episode!  Meanwhile, I had no idea there was any kind of filter that would respell such terms here.  learn something new every day I suppose.

[frank_p]

*&^*&^ :   Now that is real bilingual Canadian slang !  ç`»^Éâ»ï` !
"Gain" a few characters and "boost" your vocabulary.

Fûck !  Woops  ...  I learned something new today also ! Now we know we can type anything we want, Mark, with that "where's that damn key" type keyboard  .

[Earthscum]

GibsonGm:

Thanks a BUNCH! That is the thing I was missing. I wasn't getting the purpose of the feedback loop. Now I understand the difference between when it's connected from source or drain, I think.

See if I get this right:

When the loop is from drain (inverting signal), it is used to attenuate the signal rather than shunting part of the signal to ground beforehand (and thus creating signal filtering complications). When it is coming from the source, it will boost the incoming signal?

So, does adding in the inverted signal add any distortion in it's own right that may complicate or compromise a desired distortion characteristic? I understand filtering in the feedback path now, I believe (and this will help a BUNCH if I got it right)... basically, if I cut lowend in the feedback path, the amplified signal will be a percieved "bass boost", correct? And this is, I assume, how a MFB filter works?

BTW, I'm 33... I grew up with spirographs! (at least until we got MTV, haha!) It's funny that you even metioned that as an analogy... spirograph is what helped me understand the relationships between gears and their ratios.

[Ben N]

I'm reading Mike's reply and what you got out of it, and for the life of me I can't see what you see.

Feedback from the source of a single transistor amplifier is in phase with the input, or positive feedback, and is not used in amplifiers because it causes the device to go into oscillation. Unless what you want is an oscillator. http://hyperphysics.phy-astr.gsu.edu/Hbase/electronic/posfe.html

[Earthscum]

Well, now I'm utterly confused then. And, btw, I understand voltage biasing, I'm just not understanding where the signal is going in the path and what it's doing from input to output (in this case, a single transistor amplifier). I understand how the signal travels along the DC path, but I am not getting where these paths lay, which direction things are going, etc.

Ok, I'll start here. The signal comes in and encounters a junction. We'll assume this is the initial signal, and nothing has come through ahead of it (and I say that because I don't know how much signal lag there is before the signal comes back ito the path). So, the signal hits the base of the transistor, and then goes where, and affects what? I'm gonna go read up some more, or again... I'm sure I'll be re-reading the same stuff over and over again like I always do until I give up and move on to a different hobby.

Ben, bttw... I'm not quite getting your explanations... I think you are assuming I know electronics on an engineering level, and nothing else. Assume I know the basics of electronics (heck, I even know what an SCR is! And I've used them before... just that it was way back in 6th grade). I am an artist, musician, and basically am extremely mechanically inclined. I don't do the "These are the rules, deal with it." without finding out exactly how and why these rules came about, for example why you don't use less than a grade 5 bolt on any suspension component, which all boils down to the molecular structuring of the steel and it's properties caused by the structuring. Like I said before, I can't be satisfied just knowing that I put meat in and it comes out as hamburger, lol. I want to know why the holes were drilled at certain angles, and why the feeder screw has a certain pitch, and how it's sealed to keep meat from squishing out the wrong places. Dig? I'm really not trying to be difficult... although I know I am. Sorry for that.

[frank_p]

Please help us:
- What circuit did you use
- What active device was in there: a FET, an opamp (MXR microamp) or an ordinary BJT transistor ?
- Where was the feedback loop and what was in it.

What do you read again and again, what does it say ?

But you're talking also about drains and sources, so you might be on a FET circuit.

Maybe an animation could help ?
Negative feedback in opamps:
http://www.falstad.com/circuit/e-amp-invert.html
ex: Micro Amp shematic (filter in the negative feedback loop) (is that the one ?)
http://www.tonepad.com/getFile.asp?id=6

Common emitter BJT ?
http://www.falstad.com/circuit/e-ceamp.html
ex: Rangemaster.
http://www.geofex.com/Article_Folders/Rangemaster/atboost.pdf

Go see in the Wiliamson lab
http://williamson-labs.com/
Click in the left pane for transistors or opamp, then click for more animations.

If you forgot what was the circuit you were working on, help us know what circuit you would like to understand a bit more in this particular moment.
One thing is sure we must know what you want to know to help you.

Is it possible that you are thinking about the second transistor stage in the schematic that you had posted before ?
Are you talking about the 470k Ohm resistor ?
If so, read on the common emitter amplifier stage (or ask more questions on that particular point).

Are you thinking about all the stages of what you had posted at the same time ?



Take a step at a time to explain where you want to focus.