I decided to design a Colpitts oscillator with an output frequency of 1,2 MHz followed by a buffer stage. No other specs for the moment.

I would prefer to keep unchanged the coil (I got it salvaged from old radio gear) and the transistors while using a +9V PSU.

Once having it working I would like to be able to adjust (vary) the frequency and increase/decrease the output voltage. How much? No idea.

I am most concerned about the value of C4 and the DC coupling to the buffer but I would be not surprised if I am told that the whole thing is just a piece of garbage.

Prior to starting soldering I would appreciate any comments / suggestions.

 

It would be helpful if you could post a screen cap of the schematic in addition to the .asc file; not all of us use LTSpice.

 

Here we go. Forgot to do it. Thanks for the reminder.

 

None of the many circuits in Google Images shows your circuit. Instead they show the junction of the two capacitors grounded and the collector at one end of the inductor and the base at the other end.

 

In fact, AGag, I was "inspired" by a working circuit proposed by a known member in ETO forum. The oscillator is basically the same. The capacitive divider, I found it in several sites.

 

Hmm. That is not the circuit I have ever used for a Colpitts oscillator.

Looking it up, I see that it is a common collector version. I have always used the common base version, like this:

https://wiki.analog.com/university/courses/electronics/comms-lab-colpitts-osc

I find it very easy to get this circuit to work.

Bob

 

It is a Colpitts. The direct coupling of the output transistor is novel congratulations -you save at least a resistor doing that.

 

Hmm. That is not the circuit I have ever used for a Colpitts oscillator.

Looking it up, I see that it is a common collector version. I have always used the common base version, like this:

https://wiki.analog.com/university/courses/electronics/comms-lab-colpitts-osc

I find it very easy to get this circuit to work.

Bob

With "this" you mean mine or Analog's?

 

With "this" you mean mine or Analog's?

The one you posted is the common collector version of a Colpitts oscillator. The one I posted is the common base version of same.

Bob

 

Which is easier according to you?

 

See

 

See

Hola Bordodynov. Thanks por taking the pain.

What are those two diodes in antiparallel for?

BTW, would you mind posting the .asc file? Gracias.

 

Diodes for amplitude limitation.
Here's the diagram, but you'll have to download my library from my web page.
http://bordodynov.ltwiki.org/

 

Diodes for amplitude limitation.
Here's the diagram, but you'll have to download my library from my web page.
http://bordodynov.ltwiki.org/

Thanks for the .asc file. On to test and see what comes out.

A couple of months ago I downloaded / installed your files but today had to do it again. (¿¿??) Is it that after every update of LTSpice I should do that again? Puzzling. Gracias again.

 

My collection of models is used by thousands of people. This is the first time I've met your problem. The truth is, I am using Windows 7 and 10.

 

If you install the library in the LTSpice default directory, it will disappear when LTSpice is updated.

Change your preferences to use a different directory. Then after an update, replace the standard filesin your custom directory.

 

My collection of models is used by thousands of people. This is the first time I've met your problem. The truth is, I am using Windows 7 and 10.

OK Alex. Win 10 here.

I will keep an eye on this. I was myself, really perplexed when ZZZ disappeared.

Be well.

 

If you install the library in the LTSpice default directory, it will disappear when LTSpice is updated.

Change your preferences to use a different directory. Then after an update, replace the standard files in your custom directory.

That means that LTSpice obliterates the content or just adds new sym and sub?

 

That means that LTSpice obliterates the content or just adds new sum and sub?

Obliterates the content.

 

Getting excellent results from BJT Colpitts is not easy. The expectation <4% THD can be compared to Colpitts other than BJT.
An op amp circuit have been known to do that with less messing with extensive parasitic enhancements.
I think you found a troublesome puzzle, probably an embarrassment to some who think it's just a simple fix.
If only I could order that special part, If only that test equipment would tell me the right parameters.
If our expectations not based on the real cause for Colpitts BJT instability a leading cause of Colpitts syndrome.

Ltspice obliterious, startup seems off. I would like to see more instruments to uncover those discrepancies. The waveform is very accurate.

An extensive time consuming study was done. It also is a tedious effort. They all have little errors here and there.
Analysis better than most for 2N3904 Colpitts In table 3.2 using his circuit an adjustable inductor 15uH should get a simulation close.
https://www.arpapress.com/Volumes/Vol16Issue1/IJRRAS_16_1_04.pdf