hi everybody
i have been 'trying' to experiment with transistor amps but cant seem to link one stage to another. i am using bc337-25's on a bread board with trim pots to adjust as needed. with a single transistor everythings fine, but when i try connecting the second stage it fails.
if anyone knows of any good articles/books that cover this specific subject, please let me know.
warm regards
gaz
i have been 'trying' to experiment with transistor amps but cant seem to link one stage to another. i am using bc337-25's on a bread board with trim pots to adjust as needed. with a single transistor everythings fine, but when i try connecting the second stage it fails.
if anyone knows of any good articles/books that cover this specific subject, please let me know.
warm regards
gaz
hi jan thanks for the reply
i have been using schematics from books that i have, none of which specified the bc337. i have googled bc337 amps, and could construct one but, then i'd only be copying a schematic from a screen and not really understand for myself how it works(if that makes sense)
i have been using schematics from books that i have, none of which specified the bc337. i have googled bc337 amps, and could construct one but, then i'd only be copying a schematic from a screen and not really understand for myself how it works(if that makes sense)
You asked a question about a circuit concept. You said you have a single stage, but wnat to add a stage.
If you show what you have and want to do, we can comment and help.
We can't help if we have no idea what you are talking about.
A BC337 is just one of about a million transistors available, and mentioning it is not helping.
Jan
If you show what you have and want to do, we can comment and help.
We can't help if we have no idea what you are talking about.
A BC337 is just one of about a million transistors available, and mentioning it is not helping.
Jan
hi jan, so currently the bread board is dismantled from frustration, but the last one i tried was this. i made it exactly as it was in the book, and after adjusting the pots i could get a signal but it was less than the input signal and was very distorted. (electronics sourcebook for engineers p103)
warm regards
gaz
warm regards
gaz
Gaz, if this is the first time you bread bord a little circuit it's perfectly normal that it doesn't work.
Take a break from it and retry. You'll learn along the way.
Perhaps the transistor pins were reversed, there was a bad connection somewhere, a wrong resistor value and so on.
Work slowly, double check every move and measure the voltages as you go.
I'm sure you'll succeed.
Hugo
Take a break from it and retry. You'll learn along the way.
Perhaps the transistor pins were reversed, there was a bad connection somewhere, a wrong resistor value and so on.
Work slowly, double check every move and measure the voltages as you go.
I'm sure you'll succeed.
Hugo
In "short": 
An input cap prevents the 1V bias voltage to be shorted to ground because the generator normaly has an output impedance of 50 Ohm.
Imagine a +/- 50 Ohm resistor from the input node to ground.
Also, you need an output cap to prevent the 8.5V of the output transistor to damage your speaker. They only like AC voltage, not DC.
The schematic doesn't show the caps because it focuses on the coupling of both transistors but you need caps in real life.
Hugo
An input cap prevents the 1V bias voltage to be shorted to ground because the generator normaly has an output impedance of 50 Ohm.
Imagine a +/- 50 Ohm resistor from the input node to ground.
Also, you need an output cap to prevent the 8.5V of the output transistor to damage your speaker. They only like AC voltage, not DC.
The schematic doesn't show the caps because it focuses on the coupling of both transistors but you need caps in real life.
Hugo
Last edited:
hi hugo
i have made a single transistor circuit that did not require input/output caps, 0-12v , no speaker, just the generator on input and scope on output, i got about 9V signal out and it was clean! did it work because i adjusted the bias pots to compensate ? if yes why cant i just do that all the time?
please suggest any reading material that you think is relevant to my question.
thank you for replying
gaz
i have made a single transistor circuit that did not require input/output caps, 0-12v , no speaker, just the generator on input and scope on output, i got about 9V signal out and it was clean! did it work because i adjusted the bias pots to compensate ? if yes why cant i just do that all the time?
please suggest any reading material that you think is relevant to my question.
thank you for replying
gaz
EDIT: I hadn't refreshed my screen in awhile so did not see posts 9-12 before writing this. sorry!
EDIT AGAIN: I agree with the coupling cap comments of course. They are included in the circuits below.
I think playing with these kinds of circuit is a great way to learn.
This circuit is designed for an enormous amount of gain (>10,000 from back-of-the-envelope), so even a 1mV input signal will be way too big! Here is a lower-gain version that keeps the same DC operating points but lowers the gain to something like 40 so will be easier to work with. Even then, input signals of about 50 mV are probably about as large as you can use before you see clipping.
If you are using the pots because you don't have the right fixed resistors, and just leaving them fixed at the design values then that would be fine. But I think it is a bad idea if you are using pots and randomly adjusting them after the circuit is built. My experience with directly coupled common-emitters is that they do not generally allow significant changes to resistor values without messing up the entire circuit (gains and bias points are very tightly coupled). Designing them takes careful thought, and randomly adjusting pots is unlikely to get you anywhere (or help you learn!).
By the way, when cascading common-emitter stages (especially if you only want modest gain from each stage) you will quickly run out of DC voltage. It is easier to design the circuit if you have both PNP and NPN transistors available. Here is one with a gain of about 12. I don't claim it is optimized in any way, but it should work.
Have fun!
jason
EDIT AGAIN: I agree with the coupling cap comments of course. They are included in the circuits below.
I think playing with these kinds of circuit is a great way to learn.
This circuit is designed for an enormous amount of gain (>10,000 from back-of-the-envelope), so even a 1mV input signal will be way too big! Here is a lower-gain version that keeps the same DC operating points but lowers the gain to something like 40 so will be easier to work with. Even then, input signals of about 50 mV are probably about as large as you can use before you see clipping.
If you are using the pots because you don't have the right fixed resistors, and just leaving them fixed at the design values then that would be fine. But I think it is a bad idea if you are using pots and randomly adjusting them after the circuit is built. My experience with directly coupled common-emitters is that they do not generally allow significant changes to resistor values without messing up the entire circuit (gains and bias points are very tightly coupled). Designing them takes careful thought, and randomly adjusting pots is unlikely to get you anywhere (or help you learn!).
By the way, when cascading common-emitter stages (especially if you only want modest gain from each stage) you will quickly run out of DC voltage. It is easier to design the circuit if you have both PNP and NPN transistors available. Here is one with a gain of about 12. I don't claim it is optimized in any way, but it should work.
Have fun!
jason
Last edited:
I assume you made the circuit as above with one BC337-25. Where were the bias pots?
This is what you should have seen with no coupling cap. Look at the DC values.
I have no idea how you got a clean 9V signal out of this circuit.
I don't know your book but I would think there's reading material in in that discuss your issue.
This is what you should have seen with no coupling cap. Look at the DC values.
I have no idea how you got a clean 9V signal out of this circuit.
I don't know your book but I would think there's reading material in in that discuss your issue.
hi jason
i do have fixed resistors but where i have been trying out different circuits, i have come to the conclusion that its easier to remove the pot, adjust it to correct value and then pop it back in, also being able to adjust it on the fly helps me see what affect it has on the circuit.
a quick question, on the schematics there is a 100k 'load' resistor, i havent used a load resistor yet in any of my circuits, they look like they form a high pass filter, is this what i would use to clean up some of the distortion i was getting in some of my earlier attempts ?
thanks for getting back to me
gazza
i do have fixed resistors but where i have been trying out different circuits, i have come to the conclusion that its easier to remove the pot, adjust it to correct value and then pop it back in, also being able to adjust it on the fly helps me see what affect it has on the circuit.
a quick question, on the schematics there is a 100k 'load' resistor, i havent used a load resistor yet in any of my circuits, they look like they form a high pass filter, is this what i would use to clean up some of the distortion i was getting in some of my earlier attempts ?
thanks for getting back to me
gazza
I just put the 'load' resistor so that the capacitor isn't floating - it is not strictly necessary but I always include them on capacitor-coupled outputs. Yes, it is a high-pass filter at about 1 Hz or so. No, it will not clean up distortion in general.
As Jason said, it better to stick to fixed ones that are carefully selected.
But we are past that point now.
It's a good idea to try Jason's second circuit.
You'll need both a NPN and a PNP.
In my previous post, discard the blue clipped square wave. I didn't refresh the sim.
It should be a stright line at 12VDC because by shorting the input to ground with the generator due to the lack of an input cap,
the transistor is not turned on an can thus not produce a output signal.
Keep us posted!
Hugo