RC Oscillator Circuit Adjusting Frequency

[freakywarlock]

Hi,

I am working on RC OScillator Circuit, but i had problems with adjusting frequency. I can adjust frequency but it effects amplitude too (amplitude is changing while adjusting frequency), how can i separate these two variables ? I mean, when I am adjusting frequency, I want to amplitude remains same, and also for amplitude too. By the way I am using this circuit : (Ignore the Op-amp Circuit, it is converting sinusoidal wave into square wave.)

 

[LvW]

In your circuit (phase shift oscillator) a frequency change will also influence the oscillation condition (Barkhausens rule). That is the main reason for the observed effect. More than that, the oscillation will disappear for a larger frequency shift as performed by you. From this, two qestions result:
1.) Why didn`t you select another oscillator topology which allows a frequency change (preferrably using one componenet only) without touching the oscillation condition ?
2.) If you need a squarewave signal, why didn`t you - from the beginning - use a squarewave generator?

 

[freakywarlock]

Thanks for enlightenment. About your questions :

1) Actually, this is our laboratory project, and we have to use BJT in the circuit, that is why I chose this topology.(If there is another oscillator topology with using bjt let me know it, thanks.)

2) Laboratory project wants both sinusoidal wave and square wave, so after creation of sinusoidal wave, it is easier to convert it to square wave.

 

[LvW]

OK - I see.
In this case, some questions:
1.) Are you reqired to tune the frequency of oscillation?
2.) Over which range?
3.) How do you intend to tune the frequency (which elements are varied)?

 

[freakywarlock]

1) Yes. Also we need to tune square wave's frequency and sinusoidal signal's frequency separately for extra points but that is another topic.
2) About 500 Hz- 1500 Hz. But 1000 Hz - 1500 Hz is also acceptable.
3) I tried to vary the resistor R4 in figure, it gives a little bit change (around 250 Hz) but also there is a change in amplitude again. But now change in amplitude it is not too much, around 150 mV.

Thank you again.

 

[LvW]

As I have mentioned in my first answer - changing one of the elements within the frequency-determining network influences the phase and amplitude characteristic of that network and requires another gain of the transistor amplifier.
If the gain is not changed accordingly, there is only a relatively small tuning range (which depends on the amount of excess gain that is available).
Theoretically, you must tune all 4 resistors or all 4 capacitors of that network at the same time. However, this seems not practicle.
Therefore: Did you consider also other transistor based oscillator topologies (Colpitt, Hartley, Clapp9 ?

 

[tech99]

As I have mentioned in my first answer - changing one of the elements within the frequency-determining network influences the phase and amplitude characteristic of that network and requires another gain of the transistor amplifier.
If the gain is not changed accordingly, there is only a relatively small tuning range (which depends on the amount of excess gain that is available).
Theoretically, you must tune all 4 resistors or all 4 capacitors of that network at the same time. However, this seems not practicle.
Therefore: Did you consider also other transistor based oscillator topologies (Colpitt, Hartley, Clapp9 ?

I think the Wien bridge circuit will work with a BJT and can tune over wide range using a dual gang potentiometer.

 

[LvW]

I think the Wien bridge circuit will work with a BJT and can tune over wide range using a dual gang potentiometer.

Yes - in principle, I agree.
However, the OP is required to use transistors (no opamps).
Hence, it is not easy to decouple the WIEN network from the transistor amplifier.
As another problem: The amplifier must have a positive gain of slightly larger than "3"!
Therefore, perhaps a long-tailed pair (diff. amplifier) is the best solution?

 

[freakywarlock]

It seems I need to consider about the other topologies, I do not have any idea about where to insert Wien bridge into circuit.

 

[LvW]

In principle, it is simple: You need a closewd loop of (a) the WIEN network and (b) an amplifier with a gain of "3" (or slightlxy larger).
But the problerm with transistor stages is: Both parts must be decoupled. This is not a problem for opamp based amplifiers.

 

[NascentOxygen]

It seems I need to consider about the other topologies, I do not have any idea about where to insert Wien bridge into circuit.

When searching for circuit ideas, remember...

There is a 2 transistor Wien Bridge oscillator here: http://www.nutsvolts.com/magazine/article/bipolar_transistor_cookbook_part_5
though you'll probably have to build it in order to establish its frequency range.