Trying to make a simply NPN transistor audio amplifier circuit

[ShortCqt]

Hi All,

I'm trying to build a simple transistor audio amplifier circuit to augment a crystal set I'm in the process of building.

I've found a circuit diagram I'm following (attached below), and while it works, the output appears rather distorted.  The CRO output below shows the input signal, and the output signal (yes they have different scales so that they both fit on the screen).  I am expecting to see the output signal as a clean version of the input, just larger.  Is that wrong?

I've tried changing the value of input voltage (up and down), the bias resistor, as well as the filter caps, all with no beneficial result.

Can someone suggest where I'm going wrong, or if I'm perhaps just misinterpreting what I'm seeing on the CRO.

For reference, I'm using a 9014 npn transistor as that's what I have on hand, and have included a shot of my breadboard layout.  Hopefully you can see what is connected where   The blue and white wires from the left are the signal (from a tone generator on an iPad).  The blue breadboard strip is ground, red 5v.  The red wire leaving the board at the right is the amplified signal out.

I'd love some insight into where I'm going wrong.

Ta,
  Lance

[ataradov]

Where did you get this schematic? That's not how you bias BJT. You need a resistor from base to ground. R2 and this resistor will form a divider that will set a DC operating point to put transistor into a linear region.

[c4757p]

Where did you get this schematic? That's not how you bias BJT. You need a resistor from base to ground. R2 and this resistor will form a divider that will set a DC operating point to put transistor into a linear region.

Er, what? It's not how you bias a BJT well, but it'll work, and it won't cause this. In fact, your instructions won't work, without some emitter degeneration. You need feedback to hold the bias point. His circuit has that.

ShortCqt, looks like your circuit is oscillating. Try putting a capacitor of about 15pF across R2 ( 1/(2 pi 15pF 100k) = 100kHz = past the audio spectrum ) and see if that stops it.

That, or your output is really, really small and we're looking at zoomed-in noise. What are your actual scales?

[ShortCqt]

@ataradov

Google gave it to me

Having said that, I did try with a voltage divider in an earlier attempt, but was having grief.  Consider a voltage divider with R1 connected to V+, and R2 connected to ground, with both connected at junction 1 (J1).  No matter what I did, the volgate at J1 was always 0v.  Measuring the resistance, R1 was in the megaohms range.  Disconnect it and it was back to 10k (or whatever it was at the time).  I just couldn't figure out why a resistor would show a higher resistance when in a circuit.

Anyway, because I couldn't get it to work, I tried this circuit.

@c4757
I'll try adding the cap and see how I go.  As for the scales.... I really don't know, though I can tell you that it's definitely amplifying.  My scope is best described as finicky.  The same signal at the same scale on both channels displays vastly differently.  So in this case I set each channel scale while both were connected to the input signal so that they showed the same size.  Then set channel 2 to the output, and it was way larger.  So yes, it is certainly amplifying.

[ataradov]

No matter what I did, the volgate at J1 was always 0v.

That's not possible. If this is a divider, it should divide.

Measuring the resistance, R1 was in the megaohms range.

What do you mean "measuring"? This divider need to be calculated and proper resistors used.

I just couldn't figure out why a resistor would show a higher resistance when in a circuit.

Resistor in circuit can may measure as a smaller value, but never larger.

But c4757 is correct, with 2 resistor biasing, you will need an emitter resistor.

Anyway, because I couldn't get it to work, I tried this circuit.

I would actually try to make more conventional circuit work.

I'll try adding the cap and see how I go.  As for the scales.... I really don't know, though I can tell you that it's definitely amplifying.  My scope is best described as finicky.  The same signal at the same scale on both channels displays vastly differently.  So in this case I set each channel scale while both were connected to the input signal so that they showed the same size.  Then set channel 2 to the output, and it was way larger.  So yes, it is certainly amplifying.

Can you show both signals on the same setting. If one of the channels is faulty, then use only one channel (whichever you believe to be working).

[Seekonk]

That circuit should work fine.  R1 is to power an electret microphone.  Delete it.  As long as V collector is about 1/2 battery voltage, bias is ok.

[c4757p]

Anyway, because I couldn't get it to work, I tried this circuit.

I would actually try to make more conventional circuit work.

I wouldn't. This circuit is not broken. It doesn't perform particularly well, but it works. So when you have problems with it, don't give up and move on and hope they go away, that isn't how you learn. Figure out what the problems are and fix them. Then, once you get it working, you can play around with it and see whether it really performs well enough for you.

Clearly he's a hobbyist trying to learn, not an inexperienced new engineer trying to get a product out the door. Let him figure out what's going on.

[ataradov]

I'm not against learning, but learning through putting random resistors in random places will not get you very far. And common emitter amplifier is probably the best described circuit with reasonably easy calculations. It would make sense to go though the theory of operation and make sure that resistor values used are correct for given transistor and input voltages.

[Chalcogenide]

The circuit appears to be a crude transimpedance amplifier, given the feedback resistor between the collector and the base, and that is a problem as it is expecting a current input, not a low impedance input such as the headphone output of an iPad.
Try putting a series resistor between the input and the decoupling capacitance and see if something happens.
Also, the output dynamic of this circuit is very small, probably 0.3V or so before the transistor saturates. If you used a 10k input resistance one would expect a gain of roughly -10, so only up to 60mV pk-pk input will let the circuit stay in linear range.

[firewalker]

What are your requirements? What is the desired gain of the amplifier? The power supply voltage? You can always calculate your own C-E amplifier.

https://alexkaltsas.wordpress.com/2015/09/03/designing-a-transistor-amplifier-part-a/

Alexander.

[mikerj]

The circuit appears to be a crude transimpedance amplifier, given the feedback resistor between the collector and the base, and that is a problem as it is expecting a current input, not a low impedance input such as the headphone output of an iPad.

It's a voltage amplifier. The resistor is there is bias the transistor to a suitable operating point, and the negative feedback it provides helps to ensure the bias remains reasonably stable.  It's not the best way to bias a common emitter amplifier, but as other have said it does work.  A low impedance source feeding an high impedance input is not a problem (at least in audio circuits).

[Kalvin]

What are your requirements? What is the desired gain of the amplifier? The power supply voltage? You can always calculate your own C-E amplifier.

https://alexkaltsas.wordpress.com/2015/09/03/designing-a-transistor-amplifier-part-a/

Alexander.

+1

[Seekonk]

I was thinking last night that a LM431 might make a nice little three lead amplifier if you controlled the AC feedback.

[dom0]

Anyway, because I couldn't get it to work, I tried this circuit.

I would actually try to make more conventional circuit work.

I wouldn't. This circuit is not broken. It doesn't perform particularly well, but it works. So when you have problems with it, don't give up and move on and hope they go away, that isn't how you learn. Figure out what the problems are and fix them. Then, once you get it working, you can play around with it and see whether it really performs well enough for you.

Clearly he's a hobbyist trying to learn, not an inexperienced new engineer trying to get a product out the door. Let him figure out what's going on.

+1

[Chalcogenide]

It's a voltage amplifier. The resistor is there is bias the transistor to a suitable operating point, and the negative feedback it provides helps to ensure the bias remains reasonably stable.  It's not the best way to bias a common emitter amplifier, but as other have said it does work.  A low impedance source feeding an high impedance input is not a problem (at least in audio circuits).

If you apply an ideal voltage input you completely break the feedback, and you end up slamming the transistor in and out of saturation, unless the input is very very low. But even so, distortion will ge huge as the circuit would be a non-degenerated common emitter.
This circuit works fine with an electret microphone biased by the 10k resistor because the electret is essentially a current source, so the source impedance is given by the 10k bias resistor, and the gain would roughly be -(100k/10k). If you want the circuit to keep working -with the correct feedback - applying a voltage input, you must add a 10k resistor in series to the input.

[Seekonk]

Back in 1965 I was 14 and I got my first job repairing transistorized electronic organs. My brother Had to drive me to service calls because I didn't have a license.  I learned electronics by reading. I had no mentor.  I feel I had a great advantage learning because there was no internet.

[Audioguru]

I noticed that the breadboard circuit has no supply bypass capacitor and its input does not have a shielded audio cable. No wonder it oscillates at a high frequency.

[mikerj]

If you apply an ideal voltage input you completely break the feedback

Thanks, I see what you are saying now