Why is it so important to choose RC first, then calculate IC, and finally IB? Why I can not at first choose IC, then determine RC and then determine IB? Both ways seems to work to me.

Hi deral,
I think your problem is that you have not considered the final application of the circuit, it is by knowing the purpose of the amplifier that you choose the actual circuit design.

Can you post a short design specification of a theoretical circuit that we can discuss, and hopefully help you resolve this problem.

E

 

I think your problem is that you have not considered the final application of the circuit, it is by knowing the purpose of the amplifier that you choose the actual circuit design.

Can you post a short design specification of a theoretical circuit that we can discuss, and hopefully help you resolve this problem.

Thank you for your interest. I'm designing an amplifier for educational purposes only. There is no practical application for this circuit; the main goal was to apply the theory I’ve learned. The most important takeaway for me is that I now see I must be careful when using transconductance to calculate resistances (see my first message, where I tried to use transconductance to calculate \( R_C \)).

The only specification I have is given in the first message of this thread.

Background: Almost a year ago, I took an exam on "Electrical Circuits." This exam covered the basics of analog electronics, especially amplifiers. I prepared mainly from a theoretical perspective, focusing on calculations for given values, with almost no practical experience in this area. Now, I finally have some time to combine that theory with some practical experience.

 

See
View attachment 330910

Thank you! Could you please elaborate on what I should be seeing? I only see that the transistor is biased differently than in message #12. However, the end result is the same as what I achieved in message #12: amplification of the input signal by a factor of approximately -20.

 

There is no practical application for this circuit; the main goal was to apply the theory I’ve learned

Hi deral,
To apply the theory you have learnt, you should consider the final application specification.

Set yourself simple design speciation for the amp.

For basic examples,
The Input impedance?
The output impedance?
Frequency response.?
Gain?
etc.

These requirements of the circuit will determine the component values which have to be used.
Which will give you a more practical approach to your studies.
E

 

Thank you! Could you please elaborate on what I should be seeing? I only see that the transistor is biased differently than in message #12. However, the end result is the same as what I achieved in message #12: amplification of the input signal by a factor of approximately -20.

The load is 2 kOhm. A smaller collector resistor should be selected. I have chosen 1 kOhm. It is desirable to set half power on the collector. To do this, select the resistor in the base to get this. Then by selecting the resistor in the emitter you get the necessary gain to get an amplitude of 25mV*20=500mV at the output. Then I calculated the distortion for this. They turned out to be acceptable for such a simple circuit.

 

you are driving 2k load... maximum power transfer occurs when impedances match. for better control of the load you want output impedance to be lower ... because high impedance output cannot properly drive load with lower impedance. but ... do not go extremely low like in first post. you saw results yourself. it is all about finding the right balance...
so if load is 2k, 1k is a good initial choice for Rc. now that you have matched output to load, you can calculate Ic and from that Ib. this is what everyone was telling you... you start from known things (specs/requirements) and work your way back.

doing this the wrong way means you would assume the Ib (no criteria known), then determine Ic. and it may work, probably for part of range but even there it is very likely to experience clipping/distortion. so you go back and guess again and recalculate everything... again and again. and after several attempts you may find the reasonable solution, but doing things the proper way ensures that you get good result first time.

 

you are driving 2k load... maximum power transfer occurs when impedances match. for better control of the load you want output impedance to be lower ... because high impedance output cannot properly drive load with lower impedance. but ... do not extremely low like in first post. you saw results yourself. it is all about finding the balance...
so if load is 2k, 1k is a good initial choice for Rc. now that you have matched output to load, you can calculate Ic and from that Ib. this is what everyone was telling you... you start from known things (specs/requirements) and work your way back.

doing this the wrong way means you would assume the Ib (no criteria known), then determine Ic. and it may work, probably for part of range but even there it is very likely to experience clipping/distortion. so you go back and guess again and recalculate everything... again and again. and after several attempts you may find the reasonable solution, but doing things the proper way ensures that you get good result first time.

Do you care about efficiency? If you take and make the collector load 2k ohms, you will reduce the current draw. At 500 mV output amplitude, you will have high distortion! To reduce distortion at a given output amplitude, you need to reduce the value of the collector resistor. To save power, it is advisable to use a more complex circuit with complementary transistors at the output. This will radically reduce power losses. At the same time it will reduce the output resistance. 1 kHz is not the frequency when impedance matching is required.

 

you are driving 2k load... maximum power transfer occurs when impedances match. for better control of the load you want output impedance to be lower ... because high impedance output cannot properly drive load with lower impedance.

Ok, impedance matching is another topic I hadn't considered. The output impedance of my common-emitter amplifier is determined by the collector resistor \(R_C\). I chosed \(R_C\) to be 2 kΩ. Mr. Bordodynov, along with others, suggests that it should be lower. You also mentioned that "1k is a good initial choice for \(R_C\)." Is there a rule of thumb for selecting \(R_C\) when the output impedance of the load is known? For example, if my possible output impedance is in the range [X, Y] Ω, where X < Y (and of course both X and Y are greater than 0), what value of \(R_C\) should I choose in this case?

 

As a rule of thumb, if you do not want the load to have any affect on the driver, make the driver impedance 10 times lower than the load.

For example, if the load is 2kΩ, make Rc 200 Ω or lower.

 

As a rule of thumb, if you do not want the load to have any affect on the driver, make the driver impedance 10 times lower than the load.

For example, if the load is 2kΩ, make Rc 200 Ω or lower.

Good. I will replace three steps at the beginning in the message #12 by the following:

1. Determine the load impedance range. Take the minimal value \( R_{L_{min}} \) from this range.
2. Select \( R_C \) to be e.g. 10 times lower than \( R_{L_{min}}\)
3. Determine the collector current \( I_C = \frac{U_b}{2 \cdot R_C} \)

The explanation is the better impedance matching: To better control the load, I want the amplifier's output impedance to be lower than the minimum possible resistance of the load. Is this correct?

Do the other steps in message #12 still apply?

 

Wondering why nobody has yet suggested voltage divider polarization.

 

1 kHz is not the frequency when impedance matching is required.

it does not need to match but driver/load coupling should always be considered, regardless of frequency. that includes DC.

 

No, the beta does not tell you what collector current to use, it tells you what base current to use to get the desired collector current.

The way I determine the collector current is by setting the collector voltage to 1/2 the supply voltage. So:

Ic = Vs / 2Rc

Where you have a Emitter resistor Re (R4 in this case), the collector current is controlled by Re, the transistor \( \beta \) does not have much effect. You can vary Rc, but Ic will not change.
Ve/Re sets the collector current, Ve is determined by the potential divider of Rb and Re with the 0.7 Volts correction.