colpitts oscillator

Thread Starter


Joined Feb 11, 2007
It seems that my regen detector is almost the same as this one:

My Re is instead a 0.1uH inductor
My transistor is NPN, and the battery polarity is switched (to satify an NPN transistor)
A 0.15uH inductor is connected between Collector and the Cc/Rc junction.
I omitted T1
I omitted C1
and I have a capacitor between the Cc/Rc junction and the NPN base.

I have one question about this detector. Why is a capacitor C1 connected to the output of Cc instead of to the input of Cc?

Am I supposed to treat coupling capacitors as filters?

Thread Starter


Joined Feb 11, 2007
actually, my circuit is more like the left-most one in this image:

except the polarity is reversed, I'm using an NPN, and the emitter is next to the Q1 label. and the 100nF capacitor connected to the base, and C4 is grounded, and instead of a transformer, I use an inductor, and my antenna is connected to the NPN's base.

everything else is the same.


Joined May 16, 2005
It's not a detector, it's a Colpitts oscillator. Looks like something out of the NEETS modules. Your description of differences between your circuit and the one pictured tell me that they are two different animals. If you omit C1 and T1, you no longer have an oscillator, especially not a Colpitts. The tank is a defining structure of an oscillator. The capacitive divider is the defining structure of a Colpitts oscillator.

Coupling capacitors simply block DC. In a loose sense, you could say they were high-pass filters, but I've never heard them described as such. Treat them as DC blocking caps. That's all they are.

In the picture shown, Cc couples the amplifier portion of the oscillator to the tuned tank portion of the oscillator. C1 is part of the tank, not part of the amp.

Thread Starter


Joined Feb 11, 2007
an article that I found at:

indicated to me that I was on the right track so far. My second stage amplifer uses the fig. 3b model (see my last post above), and according to the article I should use fig. 3c model.

For some reason, I don't think figure 3c will work so well, but I guess I will have to try it out.


Joined Jul 17, 2007

Why don't you download one or more of the SPICE simulators I recommended to you yesterday, and start trying out some of these ideas of yours?

Start with a known-working example schematic, get the model working as it's shown, and then make modifications to it, and test your changes in software to see what your ideas of changes to the working examples do?

For some unknown reason, you seem to blatantly disregard good advice that's been given in good spirit by people with quite a bit more experience than yourself. For example, in your "My radio is ALMOST complete" thread, located here:
your schematic, help.gif located here:
shows no indication of implementing ANY recommendations that either Audioguru or myself had made in a prior thread about exactly the same type of audio amplifier, on which the two of us individually spent some time on.

You seem to like to make changes. You need to start using tools like I suggested that will show you how your changes will affect existing circuits. You are extremely fortunate to live in a day and age where such things are available.

Meanwhile, the suggested changes you apparently completely ignored will cause your output signal to be painfully distorted. If you model that portion of the circuit using a SPICE simulation, you will discover that I'm giving you factual information.

Additonally, you should have used the 4401/4403 or 3904/3906 transistors instead of the 2222/2307s, as the hFE is more favorable.

If you aren't willing to implement the suggestions offerred, what's the point in asking questions on here?


Joined Dec 20, 2007
How many threads for the same radio?
In the recent thread about this British super-regen receiver's audio transistor, I remembered his changes to another receiver and commented that his tank in the emitter destroys the gain and the missing parts at the emitter eliminates the quenching oscillator that makes automatic sensitivity adjustment.

Now he is talking about trying figure 3c but doesn't realize that the transistor is not a common-collector NPN, it is a common-emitter PNP. Now he says that the power amplifier is his second stage but his schematic in the other thread shows that it is his only stage.

This is a colpitts oscillator with the tank at the collector but it doen't have quenching oscillator parts at the emitter like in the two articles about a super-regen receiver.



Joined Jul 17, 2007
This thread was inadvertently closed for a bit, which caused the OP and myself to carry on via PMs. Just tidying up loose ends here ;)

SgtWookie said:
mik3ca said:
I wanted to reply to your post about my colpitts oscillator, but I couldn't because the thread was closed.
I didn't realize that the thread was closed. Perhaps a mod felt that I was being hostile. That was not my intent, however I was feeling frustrated.

As for why I don't use the tools? My computer may not be compatible with any spice tools. The only tools I might find accurate if they are accurate, are those tools that take input directly from the speaker and produce a waveform from it. Even those tools were slow for my system. If they made tools for DOS or a unix environment requiring little or no background processes previously loaded, then I would consider it.
SPICE is a circuit simulation tool, that allows one to simulate via software what a real circuit might do. The simulation is only as good as the data given to it. SPICE is an acronym for "Simulation Program with Integrated Circuit Emphasis", but there are a great many discrete components (such as resistors, transistors, inductors, capacitors and inductors) that are/can be modeled. SPICE programs don't require anything running in the background. You tell the program how accurate you want the results to be, by the time you allow the simulation to run, and the timestep increment. I know that you can run the Eagle Layout Editor (Free Version), and thus I know your computer is capable of running at least a couple of the SPICE emulators I mentioned.

I believe what you are referring to is some kind of "PC O-scope" that takes signals from the line input or mic input from a sound card. That is for taking measurements from real components/circuits; but the bandwidth is extremely limited.

Also, because I am in school and at work, I have no time to go all the way to Burlington to pick up the parts I need. The source (by circuit city) has a very limited selection of parts, and charges more money.
Nutech (in Hamilton) hardly has any parts, but charges arms and legs.
Then SPICE schematic capture/simulation should be of even greater interest to you, as you will have a rich selection of virtual components that you can simply select from a list to plug into your grand design, without spending a penny or leaving your computer room - as all of the SPICE packages I suggested to you can be downloaded and used for free.

I appreciate the fact that components can be hard to come by. and Digi-Key are a couple of great resources. Electronics Goldmine is another. You are not limited to your local mom & pop electronics stores; you can get what you need online for great prices if you are a savvy shopper.
I would like to accept suggestions, but when I am on one circuit, I would like to make very little changes to it. what is the point of throwing one working circuit out to make a completely different design (that I don't know if it will or will not work).
Both AudioGuru and myself spent hours simulating the results of your original circuit, and changes that were proposed for your circuit, including the results. AudioGuru was kind enough to post a simulation that showed the original distorted output, and the resulting relatively smooth waveform output from the proposed modification. As I obtained highly similar results from my simulation, I didn't post them.

I don't know what AudioGuru's background is, but he seems to be quite knowledgeable, and I don't feel that he would steer you in a wrong direction.

I've been involved in electronics since I was a pre-teen. That's 44 years. My initial replies to a given topic may be dubious; but I will quickly zero in on a solution to the problem at hand, which is my nature.

Get a couple of the SPICE tools I mentioned, and use them. Circuitmaker Student is the quickest to learn, but it's a dead end. LTSpice is the current "pick of the litter". I use both of them. You should, too. You will save a LOT of money on components.