# Designing a Colpitts Oscillator

Discussion in 'Homework Help' started by Mysteryname, Oct 6, 2011.

1. ### Mysteryname Thread Starter New Member

Oct 19, 2010
4
0
Hello,
as a project we have been given, we are to design a common base colpitts oscillator with the following parameters

Vcc = 12v
output
frequency 1khz
3vp-p

I've grabbed some value out of the air with the oscillator, calculated the resonant frequency to be at 1khz but the simulation does not go as expected.

The resistance over the inductor is really low and allows the DC to pass, how do I make it oscillate?

here is a image of my work so far.

File size:
25.9 KB
Views:
220
2. ### t_n_k AAC Fanatic!

Mar 6, 2009
5,448
782
While the output waveform won't be an ideal sinusoid, the one shown in the attachment is probably more reliable. Some of your values are not ideal.

• ###### 1KHz colpitts.png
File size:
3.2 KB
Views:
301
Mysteryname likes this.
3. ### Ron H AAC Fanatic!

Apr 14, 2005
7,050
657
Your power supply is shorted out by the two inductors.

Mysteryname likes this.
4. ### Mysteryname Thread Starter New Member

Oct 19, 2010
4
0
Cheers for that, I didn't think about moving the feedback network. You mentioned that the values that I had were not ideal. what are the general rules for designing a oscillator of any type?

Thanks, While I knew that the circuit was a effective DC short. I didn't know how to fix that, I looked at the example in the textbook and used that as a reference.

~Regards
Mystery.

Last edited: Oct 6, 2011
5. ### t_n_k AAC Fanatic!

Mar 6, 2009
5,448
782
With respect to the grounded base version then Wikipedia has some useful notes.

http://en.wikipedia.org/wiki/Colpitts_oscillator

As to the design of LC oscillators in general, then are many web sites that cover that topic in detail. Google LC Oscillators and have a look for yourself to see which explanation best suits your requirements.

BTW: Where did you find the applet from which your schematic comes - the one shown in your first post?

Last edited: Oct 6, 2011
Mysteryname likes this.

Dec 26, 2010
2,147
300
Clearly your first mission is to obtain a viable circuit structure. Several variations are possible on the Colpitts theme, without forming a DC chain of coils across the supply. Here are a few general ideas about what you might do after that, but note that the last posters advice to study a good detailed text on the subject is a good one. I suspect however that you may be in need of getting a better "feel" for how circuit values relate. Here I think the keyword is proportion.

Having selected a workable circuit, a general rule would be be that the reactances of the capacitors and inductors need to be correctly related to the impedances presented by the transistor, biasing components, and any external load.
Remember, XL = 2πfL, XC = 1/(2πfC) (In this case, f should be the intended oscillation frequency

In your circuit, for example, the 10pF grounding the base circuit seems very small for a 1kHz oscillator. Typically the reactance of this capacitor is relatively small compared to the impedance looking into the base Actually, your circuit could almost be OK for base grounding, but only because you have used a very low base bias chain resistance. What is the impedance of 10pF at 1kHz -a bit big, eh? Conversely, the 100μH inductor in the collector circuit is going to have a tiny reactance (how much?).

For the tuned circuit, a few special considerations apply. One of these is that capacitors and more particularly coils have parasitic effects that limit their range of operating frequencies. In practical terms, this tends to limit the range of easily obtainable inductance to work well at a given frequency. Too much inductance, and parasitic capacitance may be a limiting factor forcing the resonant frequency down. With too small an inductance, the inductive reactance may not be big enough compared to the effective series resistance, so the Q may be poor.

The size of the tuning components again needs to be considered related to transistor impedances so that you obtain a reasonable Q. Finally, the capacitor tapping needs to be appropriate so that there is enough loop gain for the oscillator to start, but best not be too generous or the waveform may be poor.

Mysteryname likes this.
7. ### Mysteryname Thread Starter New Member

Oct 19, 2010
4
0
Thanks for the information,

I sourced the program from the this website http://www.falstad.com/ via my lecturer.