# Feedback oscillator

1. Jun 25, 2014

### rar0308

A circuit is here.
I built this circuit. People say a frequency is determined only by tank circuit and C2 is nothing to do with the frequency.
But when i change C2 value, the output frequency changes.
for example when C2 is 6pF 150Mhz, 12pF 120Mhz 18pF 108 Mhz.
why is this happening?
i'm struggling to fit the frequencty to FM range i.e. 90-100Mhz
I used C1=30pF and inductor 1uH, i get 150Mhz
which is totally disagreeing with the formula i.e. a reciprocal of (2 pi Root LC) = 29Mhz.
when i used 0.1uH it was still 150Mhz. It seems the output frequency is not solely determined by the tank circuit.
and I'm using low q factor variable capacitor. is this affecting this also?

edit : what is the official name of this circuit? I want to know the name to find information about this. people say it's not Hartley nor Colpitts.

24pf 98Mhz 30pF 92Mhz. Vpp is 500mV in Oscilloscope.

I connected 1000uF i.e. 10^9pF paralell to L. The frequency dropped from 105Mhz to 82Mhz.

A thing I built on PCB.

also i used different transistor KSP2222A instead of 2N2222

Last edited: Jun 25, 2014
2. Jun 25, 2014

### jim hardy

At RF frequencies layout becomes important.
Probably some of the resident Ham operators can see at a glance what's up.

my guess would be that your 1000 uf capacitor is an inductor at radio frequencies. That's why you usually see big electrolytics paralleled with a 0.1 ceramic.

It could be as simple as too long wires between battery and the transistor,
or your 10uf needs a small ceramic bypass. But i'm not a RF expert by any means.

http://230nsc1.phy-astr.gsu.edu/hbase/electronic/npncb.html

http://www.cde.com/catalogs/AEappGUIDE.pdf

3. Jun 25, 2014

### rar0308

I thought i should have uploaded a photo of a thing i built but usb port malfunction deterred me.
By increasing C2 value I managed to jam FM 95.9Mhz from my radio without knowing why it works.
i suspect low peak and wide bandwidth due to low q factor of variable capacitor C1 lead to make Barkhausen criterion inappropriate to this configuration.

Last edited: Jun 25, 2014
4. Jun 28, 2014

### Staff: Mentor

I'd say Jim is spot on with his diagnosis (as usual). At VHF you can't use convenient low-frequency lay out construction techniques. Discard the breadboard, you should build most of this in mid-air like a rat's nest and supported by component leads, with component leads clipped very short. No ceramic caps in the RF circuit, they are hopeless here. The variable capacitor should be small, with air dielectric. Look at winding the inductor yourself, using an air-cored former and stout copper wire with turns spaced apart like you see in VHF equipment. The battery and electrolytic should be close to the transistor, with short leads.

So that your experiments do not disrupt radio signals on that frequency in your locality, house this in a shielded case or metal box, and earth it.

Good luck!

5. Jun 29, 2014

### Averagesupernova

It is not correct to say that C2 has nothing to do with the frequency of the oscillators operating frequency. Any time we hang something right on the tank it will affect the frequency. The emitter is a very low impedance point so naturally tying something to it is like tying it to ground. Xc of C3 at 100 Mhz is less than .25 ohm, so this impedance is lowered through the beta of the transistor even farther.

6. Jun 29, 2014

### jim hardy

First few pictures here show how RF layout is kept compact.
http://circuit-zone.com/?cat=fm_transmitters

Any loop of wire has inductance so the approach is to minimize wire lengths. Especially around that tuned circuit.

You can build "in the air" as N O suggested over a copper clad board; the clad makes a good ground plane.

And i think maybe N O meant to say:
You have a common base amplifier with positive feedback via C2 to make it oscillate.
C2 sets the amount of feedback, which you want to go all into the emitter. Stray capacitance in parallel with R3 steals from your feedback so be careful there with layout. C3 should have short leads to keep base to ground impedance real low at RF.

I couldn't tell if your inductor is air core. If not, make sure it's core material is a variety good at 100mhz. Spec sheet for the inductor should say.

Good for you for building it. You learn a LOT more that way

Ever heard of the "Consumptive Usher" ?
Another day..

old jim

7. Jun 29, 2014

### MrSparkle

as I understood it, the inductor was just 6 turns of wire (which I didn't see in the circuit).

8. Jun 29, 2014

### jim hardy

I'd guess it's the green object on the rear of the board.
Can't tell what it is
power chokes made for switching inductors probably aren't suitable
the Bourns RLB0608 1 uh for example self resonates at ~105 mhz
http://www.bourns.com/pdfs/rlb.pdf

i think you want a genuine RF choke

but as i said i'm not a good RF guy.

9. Jun 29, 2014

### Staff: Mentor

It goes without saying that electrolytics are unsuitable at RF.

I really did mean not to use disc ceramic capacitors at VHF. They are their own little tuned circuit.

Air dielectric variable caps with ceramic bead insulators are another thing entirely.

10. Jun 29, 2014

### jim hardy

Thanks - i learn something every day...

http://www.vishay.com/docs/22019/cergenin.pdf

page 13 of the pdf:

as i said, i'm no RF expert.

11. Jun 29, 2014

### davenn

You had really better clarify what you mean

ALL the HF, VHF, UHF and microwave radio gear I have ever seen and worked on are full of ceramic caps. From power supply stages right through to RF output stages

ALL components have self resonance values that need to be avoided. Values and physical sizes of the particular components concerned need to be taken into account

Dave

Last edited: Jun 29, 2014
12. Jun 29, 2014

### AlephZero

There are two "grades" of ceramic caps, imaginatively labeled Class I and Class II by standards bodies like IEC.

Class I are high stability low loss components intended for high frequency signal paths. Class II are mainly used for power supply smoothing, etc.

I suppose NascentOxygen was using "disc ceramics" as a non-standard term for Class II.

See Page 1 of Jim Hardy's pdf link for some differences in the two specs.

Last edited: Jun 29, 2014
13. Jun 30, 2014

### Staff: Mentor

I had in mind the ubiquitous dun colored disc bearing one number, and seen everywhere. What you buy from Tandy.