BJT Design Question

[chickenyoyo6]

Howdy. I'm an electrical engineering student currently taking an analog class. For my final design project I'm designing a stomp box. My plan is to create a multistage bjt amplfier, with a common collector for the input so I have a high input impedance, and a common emitter for the gain. I was wondering if anyone knows what impedance I should expect for the load. More importantly, what amount of gain should I be designing the amplifier for? I really have no idea what voltage I can have going into my amp. I was told to expect an output from the guitar of 1v?
Any help is appreciated, thanks!

[PRR]

Welcome!

> I was told to expect an output from the guitar of 1v?

The stage amplifier only needs 20mV to make full output.

A HARD-strummed guitar can touch 1V peak, though 200mV is more reasonable.

So we need 20mV, we have over 200mV...... WHY do you need a booster in between?

(That's a seriously deep question you won't answer this semester.)

Guitar interface can be approximated as 50K source 500K load. A wide-open guitar may be under 10K most of the audio band. Many g-amps are 1Meg (||100pFd) input. For consistency you should observe the usual 10X mis-match. But heavy loading is the "sound" of some pedals.

Also note 30pFd per foot of cable. A 30-foot cord can be a LOT of load at the top of the audio band, even the guitar band (compute it).

> what amount of gain

If you don't know, why are you already planning a multi-stage design? (OK, if that is the assignment, the boss is always right and for now the prof/TA is the boss.) I won't detail your homework, but you should be able to show that if "Gain required between 50K source and 500K load is 2", then a single high-hFE device can do the job.

Finally.... plagiarize plagiarize plagiarize!! Only call it "research". (Hint: "LPB".) You are not the first to enter this jungle. Know what has worked before. It may shorten your path. It can also (in industry) help avoid recent patents. (Audio patents are a very debatable thing. However if your plagiarism research finds a high-price highly touted "new approach!" the promoter is super proud of, it may be best to take another path.)

[antonis]

Nothing to add but a little "theory" help..

My plan is to create a multistage bjt amplfier, with a common collector for the input so I have a high input impedance, and a common emitter for the gain.

This is a SINGLE stage amp, because the input follower doesn't amplify anything relative to signal's voltage..
(actually, it dominates a little the amplitude of incoming signal..)

On the other hand, you may call it a multi-purpose amp..
(taking in mind that the emitter follower is actually an current amplifier..)

So, you have Voltage and Current amplification and, according to your semester class, you may name it POWER(*) Amplifier..
(which will be ready to drive a speaker with a little help of a matcing impedance device..)

(*) Waiting for comments about the unconventional amplification sequence order..

[chickenyoyo6]

Really appreciate the input guys. I understand how I could create the circuit using a single CE circuit. I understand that the Emitter follower is unnecessary for the circuit, but I want to still include it to make the project look a little more "complex" since our last project was a basic common emitter amp. Since I have never worked with stomp boxes before, how does the physical amount of gain affect the sound? I was planning to design this circuit so that when the input is amplified, it is still under the max voltage swing so I won't get clipping. But isn't this just making the signal "louder"? Is clipping desirable? Should my design keep it in the active region of the bjt? Also, saying my input from my guitar is 200mv, and my guitars amp only needs 20mv, I should design it to have a gain < 1 right? I think I'm going to design my CE amp with 2 emitter resistors, 1 will be bypassed with a capacitor. This will keep my max swing voltage high, and my gain lower then.

[ElectricDruid]

Is clipping desirable?

Aaaah....How long have you got?!?!?

Yes, sometimes it's desirable. The history of Fuzz/Distortion/Overdrive is the history of deliberate clipping.

Without *some* distortion, then the answer to another question is also "yes" - gain is just making the signal louder.

T.

[Gus]

You could a search of this web site
There are sims screenshots showing interactions with the guitar,cable to effect
Seach for
buffer
Booster
buffer to boost
Emitter follower (EF)
BJT gain stage
Lots of stuff over the years
etc.

[robthequiet]

Ditto all of the above and probably all of the below. If I read your question correctly, you are building a device that will go between a guitar and an amplifier. My 2 cents is about the apparent volume between device _on_ and device _off_. In other words, if I stomp on the box to make it on, do I want a huge difference in volume, and when I stomp off, do I want the same apparent volume or less? Maybe some adjustability. My 2 cemts -- good luck!

[PRR]

> Is clipping desirable?

Have you ever heard UN-clipped electric guitar?

Yes, occasionally; usually for contrast against a clipped passage.

Main reason is that voice can SCREAM. Most musical instruments will raise overtones when pushed hard: flute, sax, brass, violin, piano... Often the total sound Power does not rise much, but they clearly get "louder". This is dynamic emphasis, a key musical trick/tool.

The naked solid-body electric guitar has very little "scream", and only at arm-strain strumming level. By adjusting amplification, the guitarist can find "scream" and at a convenient strumming level.

You go back to the 1940s, amps were lame, clip was considered rude. Even so, you find guys like Texas Playboys "fuzzing" parts of their solos.

Link Wray was not just souping amps, he was slashing cones to get nasty sounds. Not widely popular, but fairly influential among musicians.

1960s, gain gain gain. Hendrix. (No) Satisfaction. American Woman. Even Dylan and Donovan distorted. At the lowest levels, you can play two notes and hear a third note making a "chord". At high levels you can bounce complex (or simple) passages just in-and-out of overload for very exciting sound, or just hit the listener like a ton of bricks. In time, over-over-OVER!-drive amps with 0.5mV sensitivity appeared for headbanger music.

Various gains are "useful". This is a feature not a problem. More knobs = more money (and profit). Make gain adjustable-- bonus points for design evaluation.

[petemoore]

  listening to Clips and studying the associated schematics provides a really great resource for understanding what does what.
  Nothing academic applies exactly to what you have or will have, there are too many variables.
  A booster might open new territories to a distortion [compare Tonebender to FF, the TB is a Fuzz Face with boost stage added.
  Altering the input to any pedal will change it's character.
  For hands on...a boost allows the next stage to amplify [~cleanly with sufficient headroom, or distorted if not] a stronger signal, and that signal is re-multiplied in strength many times in successive stages of amplification before it gets to the speaker. Boosts invariably distort a little, and may cause subsequent stages and/or the speaker to heavily distort. 
  For tone shaping, a 7 band EQ, it takes care of boost and shows a graphic of your tone settings preferences. The graphic settings can be used as a guide to build a simpler tone circuit with fewer parts / lower noise, which replicates the EQ.
  Runoff groove has a library of various effect schematics, associated in-depth texts and soundclips.
  Plenty of youtube demo's etc. and online schematic availability.
   
 

[antonis]

Also, saying my input from my guitar is 200mv, and my guitars amp only needs 20mv, I should design it to have a gain < 1 right?

Theoretically right, but..

1. Let's assume that your amp is working with +20V -20V and is biased at 0V.. If you want to have an max undistorted output of 20V for an input of 20mVp you must have a gain of 1000..
If this gain comes from a single stage BJT amp(*) you'll have a great amount of distortion (linear and/or non linear)..
(*) It can't be done in this case because max gain in an CE configuration is limited to 20 times Vcc  - put the blame on BE diode's internal resistance..

2. In case of gain < 1 you practically have an attenuator which is easily obtained by a humble voltage divider..

IMHO, your project would be fine from the side of "compexity" if you design a single stage CE amp with negative feedback and emitter degenaration with emitter resistance bypassed with a capacitor and a series pot wired as variable resistor for gain setting - taking in mind of course of some "elementary" acceptances and rules of thumb like +9V single supply and 1/10 in-out impedance matching..

[induction]

Since this is for a grade, bear in mind (as noted above) that most guitar pedals of this kind do indeed intentionally generate distortion. If your teacher or grader is not a guitar player, he or she should be informed that distortion is part of the goal. Many or even most very successful guitar pedals would get a failing grade if they were submitted as final projects in an EE class, as guitar pedals and 'proper' audio design don't often play by the same rules.

If you decide to borrow design ideas from commercial pedals, be aware that schematics from Devi Ever and Death By Audio (among others) have been known to make EE professors' heads asplode.