Transistor Audio Filter Design

[electrodoc]

Hi all, this is a forum linked from another thread i started for a Low Pass Filter which would give a visual display of the Bass beat from a disco system. Here's the link to that...

www.electronics-lab.com/forum/index.php?board=13;action=display;threadid=332

[electrodoc]

Hopefully this new theory thread will help in my own and other users quest's to understand How these filters are worked out..

A few questions to sart the ball rolling would be...

1)..I can see that R2 + the volume control (RV2) sets the gain for the primary amplifier but what about R3 and R6, is R3 for stability and are R6 and R7 needed?

2)..How do I work out the gain that I want and then work out the resistances required?

3)..Is the gain affected by any bandpass filters and vice versa?

[russlk]

The gain of T1 is nominally RV1/R6, but there is negative feedback thru R2, so the computation is complex. The purpose of R6 is to stabelize the DC current and increase the input resistance of T1. The input resistance is rb+beta*R6. Probably around 4700 ohms. The input capacitor (100nF) is working against the 4700 ohms, which gives an high pass cutoff of 340 Hz. You need a larger input cap to get good bass. Voice frequencies are about 300 Hz and up, so you need low pass starting at 100 hz or lower to have any attenuation at 300 hz.

[audioguru]

Electrodoc,
Transistor amplifier and simple filter? Geez, I have been using opamps and state-variable filters for 30 years!

Russ is right, the voltage gain of T1 is roughly RV1/R6, reduced by negative feedback through R2. But its gain is also reduced by the low input resistance of the T2 stage, which is in parallel with RV1.

That is a poor design where its gain goes up when you turn the volume down. It also causes a simple capacitor filter across RV1 to not work properly.
If all resistor values of the T1 stage were lower and most resistor values of T2 were higher, then the volume control and simple filter would work much better.
But then the lower input resistance of the T1 stage would load-down the microphone (reducing its output), and and the higher output resistance of the T2 stage would be loaded-down by the low input resistance of the T3 stage (reducing the voltage gain of the T2 stage).
So what are we going to do? Add another couple of stages? Not yet. Now we are simply going to turn-down the volume control so that the low input resistance of the T2 stage does not load-down RV1 as much.

Design of the T1 stage:
1) Choose an input resistance of the T1 stage to be about 5 times the output resistance of the microphone. The microphone has the drain of a FET as its output, which has a very high resistance, so the output resistance of the microphone is just R1.
2) Choose the output resistance of T1 to be 1/5 the input resistance of the following stage. T1's output is its collector, which has a very high resistance, so the output resistance of the T1 stage is simply RV1.
3) For battery operation, choose the current-drain of the T1 stage to be low, therefore the value of RV1 must be high.
4) T1 is a preamp, so its voltage gain must be high. So the value of R6 must be low.

Since this circuit doesn't have enough stages, then compromises must be made:
1) The input resistance of the T1 stage cannot be high enough since its gain must be high, but the value of R6 must be low. Its input resistance is almost Beta of T1 (about 200) times R6. As a compromise we can allow the input resistance of the T1 stage to be equal to R1, which is 10K. R1 cannot be decreased since that would reduce the microphone's output level.
With a chosen input resistance of the T1 stage of 10K, then the value of R6 is calculated to be about 50 ohms. Use 47 ohms.
2) RV1 must have a high value so that the voltage gain of the T1 stage is high, and that its drain-current is low. Choose a value of 220K for RV1. (to be continued)

[Kevin Weddle]

Keep in mind that the capacitors have a specific impedance and could be replaced for resistors at the frequency you want to keep. They look like they attenuate the lower frequencies. Look at the high gain of T3. It's amazing that this is typically done with IC's when we know how unstable it is to not have emitter bias.

To answer your question about the gain and the capacitors. The gain and reduction of the signal tends to act against each other, but one will overcome the other with a net result. I would say that the capacitors contribute to voltage reduction while playing a smaller part in gain.

[audioguru]

Kevin,
T3 in this circuit is not an amplifier and does not need an emitter resistor because it is simply a high gain switch. It is on, or off.
The capacitors have nothing to do with gain, they couple stages that have different DC voltage levels and simply attenuate the lower frequencies.

[Kevin Weddle]

Any collector resistor contributes to gain. Most of the time, the capacitor will contribute to voltage reduction instead of gain because the parallel elements make the capacitor independent of gain. When you go to the series circuit, you can't make anything independent so you are left with voltage reduction.