Hi
I am having difficulty in making an oscillator. My schematic and results are below (please don't flame the diagrams - I know the unknown values dont help but the diagram is accurate and clear).

I would like to create a high peak to peak value - the wavelength is fine and I am not fussy about the saw tooth.

I was hoping someone could give me some advice as to what is going on here and how to correct it, nothing specific, just a rough guide ie larger capacitors, bigger coil or bigger resistors (or something along these lines).

 

How linear do you want the triangle wave to be? Would a sawtooth work?

Your oscillator is more of a RF oscillator (think sine wave). Add capacitance on the coil (there might be parasitic capacitance) and it becomes a resonant oscillator, a very common design at that. Thinking about it, the coil and feedback cap are a series resonant circuit.

Frequency and duty cycle are needed parameters for any project like this.

 

Lots of words.
Not enough numbers.

 

Another thing, measure the power supply to be sure the signal you are seeing isn't there too. Many times you will have RF from local AM radio stations riding on the DC lines. Made me crazy when I was calibration power supplies many 10s of orbits ago.

 

Here is an example of a Colpitts that has a more reasonable C1 to C2 ratio. Note the start-up time. Note the purity of the sine wave after it gets going;

 

I would like to create a high peak to peak value -

What is your end goal? Radio transmitter? The application dictates the design. Or as they say, form follows function.

 

For the low freq you are looking for the inductor will be large. You will also need to increase the capacitance into the 10's of microfarad range.

To get a larger rail to rail swing the transistor must conduct more. Reduce the size of the emitter resistor.

A quick and dirty simulation shows the values I came up with for getting your freq range close.

 

If you're looking to get a triangle or sawtooth wave with a 10mS period a Colpitts oscillator such as you show is not the best type.

 

Wow, so many responses - thank you!

If you're looking to get a triangle or sawtooth wave with a 10mS period a Colpitts oscillator such as you show is not the best type.

Like I said - I am not at all fussy about the shape of the wave, I just want it to oscillate. If it becomes a problem later - I will work on it then.

What is your end goal? Radio transmitter? The application dictates the design. Or as they say, form follows function.

I am working on a cheap and simple way to step up a small voltage to a large voltage, the oscillator being the first of three steps in the process

Another thing, measure the power supply to be sure the signal you are seeing isn't there too. Many times you will have RF from local AM radio stations riding on the DC lines. Made me crazy when I was calibration power supplies many 10s of orbits ago.

thanks for the heads up, I have checked the circuit out and residual RF energy is not the cause of this, it is all just component choice.

 

I am working on a cheap and simple way to step up a small voltage to a large voltage, the oscillator being the first of three steps in the process

Can you define 'large'? What other steps do you have in mind?
Does it have to be 100Hz?