[SOLVED] Low power colpitts oscillator trouble

LvW said:

For an oscillator, I expect that the loop phase at the gain crossing point (0 dB) also is zero (360 deg).
.

Click to expand...

Yes - FvM is right. my comment applies to the IDEAL oscillation condition only.
Hence, it is better and more realistic to say: At the frequency where the loop phase crosses the 0 deg line (360 deg) the loop gain (magnitude) must be larger than unity.

I wrote:
"When a third capacitor is added in series with the inductor, then it becomes a Clapp oscillator."

FvM said:

No. A clapp oscillator has a completely different topology, non-inverting amplifier versus inveting amplifier with colpitts oscillator.

Click to expand...

Here are several sources in line with my statement.

From:
http://en.wikipedia.org/wiki/Clapp_oscillator

"The Clapp oscillator is a Colpitts oscillator that has an additional capacitor placed in series with the inductor."

From:
**broken link removed**

"The Clapp oscillator is distinguished from the Colpitts oscillator by the addition of C0 in series with L."

From:


"The Clapp oscillator is simply a Colpitts oscillator with an extra capacitor in series with the coil."

--------------------------------------------

The above could be called a generalization, of course. It doesn't have to mean that any and every Colpitts type can be turned into a Clapp type

From collecting Colpitts and Clapp oscillator schematics, I find there are many variations. These are different depending on the placement of the LCC tank. (How it is fed and at what moments in the cycle, what components it connects to, whether the supply is in the loop, etc.)

So you could be correct, in the sense that not every Colpitts variation may be adaptable to being turned into a Clapp type.

Example, a transistor which needs DC bias current. If that current comes through the coil, then it will not work if we put in a series capacitor. It will block DC current.

I see that my comment about different topology doesn't hit the point.

It's right that the clapp oscillator belongs to the family of colpitts oscillators. A colpitts oscillator can be turned into a clapp oscillator by adding a series capacitor to the inductor. In a first order, the series connection of all three capacitors determines the oscillation frequency. The posible C and L values are of course constrained by the oscillation condition for given amplifier input- and output impedance and gain.

The difference in topology is the difference between common source and common drain amplifier circuit. But for an unloaded oscillator, it's only an apparent difference. You can flip the common terminal around without changing oscillator operation. Also the third common gate configuration can be chosen, but the voltage differences and currents inside the oscillator are kept.

Good news! (kind of)

We got the oscillator working with the original values by ALD (1mH inductor), so the circuit does atleast work.



Now the question is just, why can't the inductance be lowered by F=1/(2Π√LC). It could very well be the impedance change, thats probably the problem. We're going to do some tests next week. But we're glad for now we got atleast something working :grin:

The thing is, we have to fit the inductor onto a contact lens PCB so we can't work with 1mH.

The gain is above 0DB at the phase crossing point.


Thanks guys

While seeing the photograph of your test-set I am afraid that the situation will be different after removing all the connection lines.

LvW said:

While seeing the photograph of your test-set I am afraid that the situation will be different after removing all the connection lines.

Click to expand...

The setup is a little messy because it was the end of the day and we had to go. It will be tested more extensively tomorrow. The top oscilloscope was measuring inaccurate, we fixed it after the picture.

You have two options:
- design the oscillator with a resonator impedance similar to the original circuit
- redesign the amplifier to work with a low impedance resonator at realistic Q values. Requires at least considerably higher transistor current or possibly a different transistor.