Newbie Problem:Class-A Amp

Zoomby · Joined: Aug 28, 2006 Posts: 18 Location: Germany

Hi,

can someone explain why this simple amplifier "doesn't work"? The problem is the bandwidth. It acts like a lowpass with cutoff around 500Hz. I tested it between a guitar and a guitar amp.

Here's the circuit:
Posted Image, might have been reduced in size. Click Image to view fullscreen.

Posted Image, might have been reduced in size. Click Image to view fullscreen.

Bye
Chris

[BlueHell: made your image [ http://img138.imageshack.us/img138/8495/amp1oy3.jpg ] appear in line for easier reference]

blue hell · Posted: Sat Sep 30, 2006 10:30 am Post subject:

The input impedance of the circuit can be roughly estimated to be 47 kOhm (assuming a hFE of 100 for the transistor) and this possibly is too low for a guitar element. In the following link it is suggested that it should be around 1 Meg. Ohm.

You could try to change the emittor resistance to a higher value (like 2k2 or something, or try to replace the transistor with one having a higher hFE value, for instance by using a BC107C (if it exists, otherwise you'll have to look through some datasheets to find another one - both changes may result in clipping though, can't really quickly judge that).

And the promised link is : http://www.geofex.com/Article_Folders/impednc.htm

blue hell · Posted: Sat Sep 30, 2006 10:39 am Post subject:

For possible transitor alternatives see : http://www.datasheets.org.uk/search.php?q=BC107C&sType=part - a BC107C does seem to exist with a minimum hFE of 450 which might just be high enough.

Zoomby · Joined: Aug 28, 2006 Posts: 18 Location: Germany

Thanks for the reply Blue Hell!
I'll take a look a the link you posted!

Zoomby · Joined: Aug 28, 2006 Posts: 18 Location: Germany

You said the input impedance is 47K. 100 (hFE) * emitter resistor, this is obvious. But why exactly. Rolling Eyes

Can you explain it easily to a noob?

blue hell · Posted: Sat Sep 30, 2006 12:01 pm Post subject:

Couldn't find a nice website to explain it, but I can try myself ...

The hFE tells you what collector current will flow into the transistor for a base current of 1 (whatever units, but lests say micro Amperes). So with a hFE of 100 there will be 100 times as much collector current than there is base current.

In the circuit above the current flowing into the collector will flow out of the device at the emitter where it has to go through the emitter resistor (470 Ohm) into the circuit's ground.

This means that the voltage at the emitter will get higher (Ohm's law). But as for a bipolar NPN tranasistor the base voltage is always about 0.6 Volts higher than the emitter voltage (relatively independent of the base current) the base voltage will go up with the same amount.

The amout being ( hFE + 1) * 1 * 470 [ + 1 is for the base current that also flows into the emitter, * 1 is for the input current that we assumed to be 1]

This means that the base voltage will rise hFE times as much as expeted from the emitter resistance alone (in a case where no collector current would flow), whcih means the "experienced" input impedance has to be hFE (+1, strictly) times higher than the emittor resistance.

This assumes that the resistance between base and emitter of the transistor is near zero, which can be done safely as long as there actually is a base current.

Macaba · Joined: Jul 13, 2005 Posts: 160 Location: UK

The circuit makes no sense to me. I'm used to circuits with a potential divider on the input to bias the transistor to halfway between the rails, as shown in the first diagram on this page:
http://www.st-andrews.ac.uk/~jcgl/Scots_Guide/experiment/lab/expt5/page1.html

Zoomby · Joined: Aug 28, 2006 Posts: 18 Location: Germany

Thanks for the great explanation Blue Hell! Smile

blue hell · Posted: Sat Sep 30, 2006 4:33 pm Post subject:

Uncle Krunkus · Posted: Sun Oct 01, 2006 5:52 am Post subject:

I know the DC resistance of a guitar pickup is in the 2-10K kind of range. I assume that it's impedance would be a lot higher than that, yeah?
The input impedance of this preamp needs to be at least 10 times the output impedance of the guitar, yeah? (I don't know these things for certain, so correct me if I'm wrong Jan) So if it's not, that would stuff with the bandwidth wouldn't it?