Dirt simple amplifier demystification requested!

[Ratbones]

Ok, so this isn't really stompbox related, but I wonder if you guys can help me to understand a bit about this very simple amplifier circuit.

Some setup: So what I'm planning on doing is miking a gascan repurposed as a drum...
I want to be able to feed the signal into an fx processor to get dub sounding echoes and reverb.
I'm not expecting clean pretty sounds. In fact, I want it to be kinda crusty and nasty... to a degree...

Now, I know this is probably about as simple as amplifier circuits come, but I'm squealing-naked-baby new to amplification, so I wanted to start with baby steps.
I've constructed this little circuit, and as can be expected, the frequency range is very limited.

What modifications would allow this to produce a greater range of sound? Particularly bass?
 


To start, I've got 1uF caps on input and output, and Rc is 1k; Rb is 100k.  Transistor is a 2n3904.
I've also got 2N5088's on hand. I dunno if more gain would be more desireable...

If this circuit is just plain not suited to this, that's cool, I'm excited to explore other options.
I would also greatly appreciate any other circuits that you may recommend for my application.  Thanks for any direction you can give me!

[John Lyons]

More importantly at this point is which mic/element you are going to use for the gas can.
Then we/you can tailor the amplification for the sound pick up.

Are you looking for a mainly boomy sound or more like a tom/floor tom with some attack and resonant low
or just BOOM (low end)?

[merlinb]

What modifications would allow this to produce a greater range of sound? Particularly bass?

Bigger input / output caps. At least 100uF say.

[CynicalMan]

Short version: Your input impedance is too low. You should add emitter resistance.

Long version: The input impedance at the base of a common emitter amplifier is equal to hfe*(r'e+Re) where r'e is the emitter-base ac resistance and Re is the value of the emitter resistor, if there is one.
zin=hfe*(r'e+Re)
r'e=25mV/Ie (emitter current)
For collector feedback bias,
Ie=(Vcc-Vbe)/(Rb/hFE+Rc)

Vcc is supply voltage, so 9V, Vbe is the emitter diode drop, for silicon it's around 0.7V, Rb is 100k, and Rc is 1k.
hfe in a 2N3904 can be from around 100 to 400. For now I'll go with 100.

So,
Ie = (9-0.7)/(100k/100+1k)
    = 0.00415A
r'e = 0.025/0.00415
     = 6.02 ohms
zin = 100*(6.02+0)
     = 602 ohms

Making hfe and hFE 400, zin still comes out to only 1506 ohms.
So, you circuit's input impedance was around 600 to 1500 ohms, how much that affects things depends on your mic, but with a guitar pup, anything less than around 500k will roll off too many highs for normal purposes. Also, that means that the input filter had a corner frequency of 1062Hz to 2654Hz, way too high to let through much bass.

The easy way to fix this is by adding in an emitter resistor. This increases the input impedance by hfe*Re. It lowers gain and changes bias, but you could just add in a 1k resistor and see how it sounds. Oh, also increase the input cap.

[petemoore]

  Hook a speaker to it, try an alnico...lol.
  Use a Peizo and get better highs maybe, and a Jfet current booster.
  Mic...if there's a mic to try.
  See how that works then if voltage amp is needed [to boost signal input].
  Clean or dirty but the dirt doesn't 'pile' like guitar, it comes in 'clumps' like the input peaks. Perhaps a set?s of diodes across the speaker [or it's voltage boosted output] as limiter...this kind of thing may be very demanding on a speaker asked to have less than peak levels sound audible among other instruments, count on the peaks being brutal.

[Ratbones]

Thanks for the replies guys; I really appreciate the input and the dissection.

More importantly at this point is which mic/element you are going to use for the gas can.
Then we/you can tailor the amplification for the sound pick up.

Are you looking for a mainly boomy sound or more like a tom/floor tom with some attack and resonant low
or just BOOM (low end)?

The microphone i'm using is actually a four inch (pretty much the exact diameter of the opening in the gascan) 8 ohm salvaged speaker...
I'm not sure if using a speaker as a mic necessitates any special alterations... Any ideas?

Ideally, what I'd like is a boomy, bassy sound.

Short version: Your input impedance is too low. You should add emitter resistance.
The easy way to fix this is by adding in an emitter resistor. This increases the input impedance by hfe*Re. It lowers gain and changes bias, but you could just add in a 1k resistor and see how it sounds. Oh, also increase the input cap.

Thanks so much for your explanation.  If I've got this correct, then if I swap out my 2N3904 (which I measured to have a gain of around 110) for the
2N5088, which measured 450 something, and then add in an emitter resistor that would bump zin up to around 500k it should produce a greater range of frequencies...

something like:
500k = 450*(6.02+x) if the emitter resistor was 1k ohm, zin would equal out to just over 450k, right?

From here, I'll throw in a larger input cap and experiment with that emitter resistor to produce a range of input impendances to see what gives me the best output.

and thanks petemoore and merlinb for your help too.


I'll update you guys with my findings later tonight if I get a chance to experiment.
Thanks a ton

[CynicalMan]

something like:
500k = 450*(6.02+x) if the emitter resistor was 1k ohm, zin would equal out to just over 450k, right?

Not quite, because changing hfe and adding an emitter resister both change Ie, but it would be very close to 450k.

[Ratbones]

CynicalMan et al,

So i swapped in a 220uF input cap, and a 100uF output cap, switched in the higher gain 2N5088, and added in a 1k resistor between emitter and ground...

Huge difference.  The range of expressed frequencies has been expanded a ton.
There's still some experimentation to do to increase the bass response further, but it's definitely getting there! 

I have a few questions though... First, I understand basically that a higher input cap allows for more bass to be passed, but what does a larger output cap do? What's the method to block the higher frequencies? Lastly, is there an ideal input impedance to shoot for?  Should I be trying to match the 500k input impedance of a guitar?

Thanks for all your help guys.

[Ratbones]

I got this circuit to perform at a level fit for tiny perfboard commitment, and I'm pleased to say it actually sounds pretty cool! So thanks to everyone for your input!

CynicalMan, your math help turned out to be very educational, as I decided to switch transistors up, and used your example as a template to figure out exactly what resistor I needed for the emitter. First time in a long time doing enjoyably applicable math!

So thanks!