@MarcelvdG If the input coupling capacitor shown in the circuits given by @rectifryer is by chance left out, the low output impedance and any DC offset of a typical signal generator will completely destroy the DC bias, operating point and AC gain provided by R1, R2, Rl and Re.
Same sort of comment applies to the output.
If the resistor values were correctly calculated and the voltage gain was quite small compared to hfe then the distortion would be small and can be calculated. Typically a design voltage gain of say 3 would work really well if the transistor had a hfe greater than say 300.
Same sort of comment applies to the output.
If the resistor values were correctly calculated and the voltage gain was quite small compared to hfe then the distortion would be small and can be calculated. Typically a design voltage gain of say 3 would work really well if the transistor had a hfe greater than say 300.
Last edited:
…….
Adjust the base-ground 22k resistor to get better DC point according to transistor Beta. Here it would be 47k. Omit 22uF emitter resistor bypass if you do not need full voltage gain. With 22uF, input voltage will be only in mV.
Adjust the base-ground 22k resistor to get better DC point according to transistor Beta. Here it would be 47k. Omit 22uF emitter resistor bypass if you do not need full voltage gain. With 22uF, input voltage will be only in mV.
Last edited:
@PMA that circuit is a great heads up for @rectifryer but don't forget that for 1Khz and the 22uF Emitter capacitor it will have no negative feedback and its massive open loop voltage gain really will drift around with temperature and choice of transistor etc.
Yes but we are not talking about highend here. The circuit works with Beta from 100 to 500 without problems. Some gain drift does not matter for first attempts. DC bias holds regardless Beta change. The ac gain is about 100x with 22uF bypass capacitor. One needs to put the 4k7 in parallel with h22 and model current source when doing assumptions.
The DC bias point for this circuit does not matter too much as long as it is stable because this circuit is providing voltage gain rather than power gain (very high R5).
Trouble with capacitors and Emitter capacitors is they introduce phase shifts and phase shifts are a loop stability problem in multi-stage amps.
Try running your simulation again without the 22uF Emitter capacitor and you will find that voltage gain is very stable as you adjust the value for Beta (hfe).
I think this "Voltage Divider Circuit Bias" setup is capable of excellent, stable, very low distortion and good dynamic range if its voltage gain (Av) is kept low. The values of R1, R2, Rl and Re really need to be calculated for the required voltage gain (Av) according to equations typically like those given by @rayma 's link.
Trouble with capacitors and Emitter capacitors is they introduce phase shifts and phase shifts are a loop stability problem in multi-stage amps.
Try running your simulation again without the 22uF Emitter capacitor and you will find that voltage gain is very stable as you adjust the value for Beta (hfe).
I think this "Voltage Divider Circuit Bias" setup is capable of excellent, stable, very low distortion and good dynamic range if its voltage gain (Av) is kept low. The values of R1, R2, Rl and Re really need to be calculated for the required voltage gain (Av) according to equations typically like those given by @rayma 's link.
See how awful the variation of open loop current gain with temperature is for the 2N3904 (and most other NPN transistors).
But you do have a good point !! You can of course still achieve a large voltage gain without using the capacitor across Re - simply by making Re quite a low value. As you can see I'm not a fan of electrolytic capacitors in audio amps. Its best to avoid them like the plague 👍
Last edited:
See post #3.
If R3 were much smaller than hfe/gm in the circuit of post #1, it would be a voltage-driven common-emitter stage with about 1 % of second-harmonic distortion per millivolt of peak input voltage due to the exponential voltage-to-current relation of a bipolar transistor. As R3 is actually higher than that, it is somewhere in between voltage and current drive. That reduces distortion, but it won't be great.
That is exactly why I already told that the circuit will run with hFE from <100; 500> without any issues. I think it is assumed to be tested by OP as a 1st trial on the breadboard, not as a part of complex amplifier structure with GNFB. I do not feel like arguing about this anymore, these are all very basics and I am not here to learn, I only wanted to suggest to OP something that is definitely working. We were investigating such circuits 55 years ago, right? At least I was.
Accept that you are not here to learn but possibly @rectifryer is wanting to learn and that's why I have tried to explain carefully according to established theory, with respect I find your posts unhelpful from the point of view of someone trying to learn and relate his circuit simulation to a practical constructed circuit. To be honest I don't make much sense of what you are trying to say. The links posted give the common emitter bias circuit and typical design equations. The Emitter resistor capacitor is confusing because it makes the design equations fairly pointless when only voltage gain is needed.
I think you could understand that the Emitter resistor Re creates negative feedback and by so doing stabilises voltage gain and reduces distortion, it is very like OpAmp feedback to it's negative input.
I think you could understand that the Emitter resistor Re creates negative feedback and by so doing stabilises voltage gain and reduces distortion, it is very like OpAmp feedback to it's negative input.
Last edited:
Temperature change 0, 25, 100°C makes almost no difference in gain. Beta change 100 - 500x makes almost no difference in ac gain as well. Gain is not equal to hFE, sorry. 4k7 collector resistor defines the gain. With emitter resistor bypass capacitor, Gain = (Beta/rbe)*Rc. Emitter bypass capacitor to be enlarged 10x to 220uF, and we are almost perfect.
Distortion
Distortion
Last edited:
I don't like electrolytic capacitors in audio preamp designs and they are not used inside OpAmps packages obviously.
There is a need for the equations to calculate R1 R2 Rl and Re for the required voltage gain Av.
I'm sticking to what I said about Re providing negative feedback, putting an electrolytic capacitor across it simply takes this negative feedback away,
As you I could work all this out but I don't want to escalate this when it's really for @rectifryer to come to his own understanding of how to bias transistors.
If you are a true scholar here you will appreciate what I said about the magical grounded Base configuration as being little used but very cunning. Have a good day 👍
There is a need for the equations to calculate R1 R2 Rl and Re for the required voltage gain Av.
I'm sticking to what I said about Re providing negative feedback, putting an electrolytic capacitor across it simply takes this negative feedback away,
As you I could work all this out but I don't want to escalate this when it's really for @rectifryer to come to his own understanding of how to bias transistors.
If you are a true scholar here you will appreciate what I said about the magical grounded Base configuration as being little used but very cunning. Have a good day 👍
Last edited:
what has silicon band gap got to do with the forward bias voltage across Vbe, this is an entirely different physical property of Silicon semiconductor Physics??
In a bandgap reference circuit, a voltage that is proportional to the absolute temperature is added to a junction forward voltage to cancel out the first-order temperature dependence. The sum voltage is then the bandgap voltage of the semiconductor plus some extra term. It's normally 1.23 V to 1.25 V for silicon.
Conversely, when you take 1.24 V and subtract the forward voltage of a silicon diode, the difference will be nearly proportional to the absolute temperature, precisely what you need to cancel the temperature-dependence of the transconductance of a bipolar transistor.
Conversely, when you take 1.24 V and subtract the forward voltage of a silicon diode, the difference will be nearly proportional to the absolute temperature, precisely what you need to cancel the temperature-dependence of the transconductance of a bipolar transistor.
I think the problem is not using formulas and arbitrarily plugging in values in a simulator. When going to DC coupling, The method that should be applied is an input resistor replaces the coupling capacitor at the same value as the target input impedance.
The values chosen aren't completely arbitrary. I chose around 20-50mA collector current based off the rise time performance at that collector level. That could be in error, but I worked back from there until I got the voltage gain I needed.
I am trying to do a low distortion class A amp that scales a voltage from 50 mVpp to 3Vpp. The output has to be in phase with the input, so I've added another 2n3904.
My main technical hangup is deriving the internal transistor resistance, sometimes referred to as Ree. Wouldn't it just be the Voltage drop divided by the base current? Vforward/ib?
I am trying to do a low distortion class A amp that scales a voltage from 50 mVpp to 3Vpp. The output has to be in phase with the input, so I've added another 2n3904.
My main technical hangup is deriving the internal transistor resistance, sometimes referred to as Ree. Wouldn't it just be the Voltage drop divided by the base current? Vforward/ib?
Last edited:
Then do not use the circuit that you have suggested. It would be a miserable class A amplifying stage.
That is what I was afraid of and why I asked for help. I am not married to the design, only constrained to the transistor.
Why not go back to your simulation as in your post #6 . Delete R5 and increase the value of R3 and R4 (50mA is way too high for Ic). Graph both the input waveform and output waveform so you can compare them as regards gain, phase shift and distortion. Try different resistor values so you get a feel for what's happening. If you feel confident try to apply the design equations that relate R1, R2, R3, R4, voltage gain (Av), Ic at operating point etc. Your worst case is probably high frequency (20 KHz) so rerun the simulation at different frequencies. This transistor should be fine for the full audio frequency range.
I think that an Ic bias current of about 1mA would be fine for this application. Experiment with adding an Emitter capacitor but try not to rely on it. Electrolytic capacitors in audio amp circuits are bad practice.
Try adding an identical second stage in series and so on.
I think that an Ic bias current of about 1mA would be fine for this application. Experiment with adding an Emitter capacitor but try not to rely on it. Electrolytic capacitors in audio amp circuits are bad practice.
Try adding an identical second stage in series and so on.
Last edited: