1. Home
  2. Electronics Design
  3. Colpitts Oscillator, less "universal"

Colpitts Oscillator, less "universal"

150K 120K +-----/\\/\\----+-----/\\/\\--+--->

| | | | | |/------+ --- +--------+---| 2N2369A - | | |\\ | C1 --- V | 330p --- | | | | Cld --- +-----+ 22p --- | | | C2 --- \\ o 250p --- / 4.3K Xtal | \\ o | | | | | | | | +--------+-----+ | --- -

With C2 at 330pF, only the 200Khz tuning-fork crystal oscillates, and only with a supply voltage near 2.5V; At 3V to 20V supply, 2Mhz-10Mhx crystals work and the 12Mhz does not. With C2 at 250pF, no tuning-fork crystal will oscillate, and those from 2Mhz to 12Mhz does. However, the higher frequency crystls oscillate at a lower amplitude with a given supply voltage. This circuit gave me a surprize; it is clear at 2Mhz, that it operates as a class C circuit from about 5V up. Put a 50 ohm resistor in the collector and pick off the signal with virtually zero disturbance of the oscillator.

Reply to
Robert Baer
Loading thread data ...

Isn't this really a Clapp oscillator?

A cute circuit I came up with some time back does this:

!/ ---!!-+---! ! !\\e --- ! --- ! ! ! +-----+ ! ! ! \\! --- !--- --- e/! ! ! +-----+ ! ! --- \\ --- / ! \\ GND ! GND

It makes a bigger negative resistance effect be seen by the tuned circuit.

[....]

h

I've built basically your circuit with a tuned circuit in the collector and taken a harmonic off. Mistuning of the harmonic circuit has very little effect on the crystal.

Reply to
MooseFET

Listed as Colpitts; except for the base biasing values, all else was "borrowed" from the web. Yes, picking off of the collector with a low-Z tap of a tuned circuit gets a very decent signal; fundamental or harmonic. And since this is a grounded collector design, keeping that pickoff low-Z will ensure almost zero effect on the basic oscillator. Not much sensitivity to the supply voltage, either.

Reply to
Robert Baer

I have done some fiddling with C1 and C2 and have been unable to get the other tuning-fork crystals to work. Any suggestions?

Reply to
Robert Baer

Crystal properties vary widely depending on cut, frequency, and mode, which is why there are so many different circuits.

Those tuning fork crystals have big ESRs, like 30-50k @ 32KHz, dropping as frequency rises.

I'd think a JFET-based Pierce, having a high-Z input, could be the ticket for the high-ESR types. There's one in the National FET Applications that went something like this:

Vdd -+- | | .-. | | R1 | | XTAL '-' _ C1 | | | || | .--|| ||---||--+ | |_| || | | | | |-+ | | Q1 +----------->|-+ | | | | .-. === | | R2 GND | | 10M '-' | === GND (created by AACircuit v1.28.4 beta 13/12/04

formatting link

I once made a very nasty, ugly little 32,768 Hz tuning fork oscillator like this (it breaks just about all the rules--for curiosity only!)

+1.5v -+- | | .-. .----. | | R2 | | | | 1M(?) .-. | '-' R1 | | | | 10M | | '----+ '-' _ | | | | X1| (R1-2 values are guesses; I don't +--|| ||--+ remember what I actually used) | |_| | | | | |/ Q1 '-------| MSPA18 |>

| | === GND (created by AACircuit v1.28.4 beta 13/12/04

formatting link

Neither of those shows proper crystal loading--that's crystal-specific, and up to you.

To adapt your existing circuit use a much higher gain transistor and increase impedances--the emitter load resistor, and, especially, the bias resistors (they're loading the crystal something fierce right now)

--and you might get it to oscillate.

Cheers, James Arthur

Reply to
James Arthur

Thanks; will try those ideas.

Reply to
Robert Baer

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.