I am trying to make a multistage transistor to amplify a sine wave from 10Hz to 2MHz. I am using the 2N3904 transistor to do so. When I run the circuit in simulation it works fine.
I created single stage first and tested it on DSO. It worked fine and gave 3X gain. But as I built the second stage it failed poorly. Doesn't show any output as such. I tried to lower the R8 and R9 but a multiple of 10, even then it doesn't respond. What could be the probable cause.
I think that C2, C4 etc. are reversed. The collector voltage is always higher than the emitter and base voltage.
What gain are you trying to get? If the first stage works and gives you a gain of 3, then presumably 9? Frankly that is not a sensible way to design an amplifier, very wasteful on components, probably would distort badly, no real way to set the gain. What you need is a "DC feedback pair" - Google will help, you could try this link:
http://www.geocities.ws/leon_heller/dc_amp.html.
There's a better configuration with an extra transistor called the dc feedback triple but I can't find a link to it. Is there a reason why you aren't just using an IC? If the circuit isn't working as you planned, most likely a wiring error or as DrD suggests the electrolytics are the wrong way round and upsetting the biasing.
Supply voltage?
This design only works within a limited supply voltage range.
Best supply voltage is where collector voltage is about 60% of supply voltage.
I guess, form the high base voltage devider ratio, that it's designed for 12-15volt.
Google "self-biasing transistor circuit".
Leo..
Where are the mystery lines from Q2, Q3 base going?
Thanks for pointing me to the right direction, will definitely go through the dc feedback pair study. Actually I am trying to design an amplifier for function generator application. It will be needing a fine control and another coarse control. I was hoping to control the gain by replacing the collector resistor with a potentiometer.
They are connected to the oscilloscope.
My supply voltage is +24V DC as I needed to swing the output upto 20V peak to peak. I will go through the suggested circuit.
Main reason I am using the transistor configuration is to control the output impedance as I do not have any knowledge about controlling the output impedance of an op-amp.
What value would output impedance need to be?
Usually you want it low...
Should not be a problem using opamps.
You can buy 4 in one IC at less than 1 Euro.
If you use the last as a buffer opamp you will have almost 0 output impedance...
Then you have pretty little margin and would need somewhere a gain control.
Otherwise clipping will occur. Also, how would you keep the middle in the middle?
Well - lets examine the design of one stage.
For a 24V supply the base voltage is 2.5V which gives 1.8V across R4 and 6.8V across R3
the UNLOADED stage gain is 5.6 / 1.5 = 3.7
Since R5 is 20 * R4 the components R5, C3 will have very little effect.
Now lets connect the second stage. The input resistance is R6//R7//(B Re) or about 5k
That will drastically reduce the effective gain from the first stage.
I'd suggest you do the calculations on one stage to see if the amplifier is capable of providing such a large output. Why do you need that? And what is the impedance it will be driving?
For your type of application, for a function generator, an op-amp is tailor-made. You would however need to choose one with a sufficient gain-bandwidth product to get your 2MHz upper cutoff. It will provide you with effectively zero output impedance, if you want a definite output impedance you add a resistor in series with the appropriate value (e.g. 50 ohms).
But I'm puzzled by the need for an output amplifier as most FG circuits are designed to generate the required levels directly. What waveforms are you generating? A common method is to have an integrator at the output to generate a triangular waveform from a square wave input, then the triangular output is fed back to a comparator to generate the square wave. If you want sine then a simple diode shaping circuit can do a reasonable job from the triangle unless you need to do audio amplifier measurements. I'm sure if you look at op amp datasheets you should be able to find circuits for FGs. There is (or was) a dedicated chip that would do the job, the Intersil 8038. Google is your friend.
This topic was automatically closed 180 days after the last reply. New replies are no longer allowed.