Colpitts oscillator

[riccardo manfrin]

Hi,

I'm following instructions here (https://www.wikiwand.com/en/Colpitts_oscillator#/Practical_example) to build a discrete components above MHz oscillator.

I could not find a 150nH inductance so I used one I found on spare components, which has I think rather larger inductance.

The circuit appear to be working well and outputs around 10MHz, but in order to see the frequency on the oscilloscope, I had to probe between the two capacitors C1 and C2, that is, on the emitter pin rather than on the collector, as instead shown in the figure.

Probing the collector didn't show any oscillation at all.

Am I doing things completely wrong?

Thanks in advance.

Regards,
R

 

[Phil Hobbs]

On 9/25/19 6:48 AM, riccardo manfrin wrote:

Hi,

I'm following instructions here (https://www.wikiwand.com/en/Colpitts_oscillator#/Practical_example) to build a discrete components above MHz oscillator.

I could not find a 150nH inductance so I used one I found on spare components, which has I think rather larger inductance.

The circuit appear to be working well and outputs around 10MHz, but in order to see the frequency on the oscilloscope, I had to probe between the two capacitors C1 and C2, that is, on the emitter pin rather than on the collector, as instead shown in the figure.

Probing the collector didn't show any oscillation at all.

Am I doing things completely wrong?

Thanks in advance.

Regards,
R

Attach the ground clip to the probe tip. That'll give you a nice
inductive loop. Then put that near the inductor. You should have
enough signal to see easily.

Failing that, a small resistor in series with the collector (20 ohms
maybe) will let you probe.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC / Hobbs ElectroOptics
Optics, Electro-optics, Photonics, Analog Electronics
Briarcliff Manor NY 10510

http://electrooptical.net
http://hobbs-eo.com

 

[George Herold]

On Wednesday, September 25, 2019 at 6:48:37 AM UTC-4, riccardo manfrin wrote:

Hi,

I'm following instructions here (https://www.wikiwand.com/en/Colpitts_oscillator#/Practical_example) to build a discrete components above MHz oscillator.

I could not find a 150nH inductance so I used one I found on spare components, which has I think rather larger inductance.

The circuit appear to be working well and outputs around 10MHz, but in order to see the frequency on the oscilloscope, I had to probe between the two capacitors C1 and C2, that is, on the emitter pin rather than on the collector, as instead shown in the figure.

Probing the collector didn't show any oscillation at all.

Am I doing things completely wrong?

Thanks in advance.

Regards,
R

Are you using a x10 probe?
(lower capacitance)
If that's still too much C, I'll hang a 1-2 pF cap on the node
and attach to that with my 'scope probe. (kinda a x 100 probe)

I've also done Phil's trick of making a loop, but with a few turns
of magnet wire.

George H.

 

Measuring stuff like that unless you need real amplitude parameters is best done by picking up the radiation. That affects the circuit the least. Remember the law in science, you cannot measure anything without affecting it. If you think that through very carefully you will find it to be absolutely true but I know the usual question like telescopes, and I have the answer to that. But for now as my friend Raphael would say - fuck all that.

Depending on the gain your oscillator had, the probe capacitance could have killed it. If you use the loop method, you are going to need a very sensitive scope, I doubt 5mV/div would show you much. Capacitively might be better, just don't ground anything and get the probe near the collector. If you used the design on that page and there is not inductor the loop trick will not work well. When you use a coil then it is better.

The capacitance of the probe on the collector, remember even though it is loaded by the capacitive divider and no doubt that got it a lower collector resistor, you are still feeding the emitter. That makes the collector output "soft" in a manner of speaking. Like the focus voltage to a CRT, it usually has such low current behind it you just don't have to worry about it. Just your body will short it out and you might not even feel it.

That type of oscillator usually needs a buffer. If youo have use for two polarities of it then use an FET and take off the source and drain with equal load resistors on it, of whatever value you need. (then you pick the FET when you know all the parameters you require)

Or you can use an emitter follower. that is actually a misnomer because they call THAT stage an emitter follower but in it nothing follows the emitter.. It is actually a common collector configuration. The emitter follows the base and it presents much less of a load to the previous stage.

On common this and that, like your common base. Learning ? Learn this. ALL TRANSISTORS ARE COMMON EMITTER. The emitter is the terminal of the device that is common to the input and output. the difference in what they call common emitter, collector or base stages only means which way you pointed it in the circuits. The transistor hasn't changed. Which terminal is at signal ground ?

Common collector you get current gain but no voltage gain. Common base you get voltage gain but no current gain.
Common emitter you can actually have both but that does not make it right for every circuit.

If you are interested enough to get through this shit then you might just like this field. And understand I look at things in a very unconventional manner. However I submit this - I am successful at it.

Oscillators are not my strong point but I can see the circuit. the output impedance is higher than you would think at quick glance. Usually it is like the collector load resistor, and I guess it is. But that doesn't mean capacitance can't stop it. up in the MHz range it doesn't take much.

One thing you might like, I don't feel like finding the link right now so just go search for impedance nomograph. There will be a hit at some university. It is in HTML sorta, but I think Java or whatever. You can save it if you know how, then you don't have to get online to use it. But just bookmarking might be alright, it seems to be staying put.

If you have interest and drive you might like this field. But I have this to say, get your running shoes on because once you catch up you don't get a rest. Technology moves fast these days. You want to catch it you have to be fast.

 

[Cursitor Doom]

On Sat, 28 Sep 2019 06:57:13 -0700, George Herold wrote:

I've also done Phil's trick of making a loop, but with a few turns of
magnet wire.

Phil knows his stuphph.



--
This message may be freely reproduced without limit or charge only via
the Usenet protocol. Reproduction in whole or part through other
protocols, whether for profit or not, is conditional upon a charge of
GBP10.00 per reproduction. Publication in this manner via non-Usenet
protocols constitutes acceptance of this condition.

 

[riccardo manfrin]

Thanks for all your help guys.
I've managed to see the oscillation by looping the probe with ground. Than I tried again probing at the collector, this time with 10x instead of 1x, and I managed to see the 80mV 10MHz waveform.

At that point I wanted to linearly amplify with a simple single stage common emitter bjt, but had no success.

I think the biasing was correct.. I was in the linear zone, but I have the doubt if the input to output impedance ratio between the two stages (the osc and the linear amp) is killing the waveform input signal at the base of the bjt...

Can you give me some piece of advice?I'll be posting the biasing of the linear bjt amp when I'll be in front of the breadboard