Lf oscillators

Hi Harry
Yes. I believe it could as I have home made direct probes on my scope that are not too good for RF work. It does look ok on other oscillators though.

Could that be shunting caused by the impedence/reactance of the 'scope? Just a thought....

Thanks Harry
I see what you say. I will watch it a little closer on the scope the next time I change the capacitors. This is what I noticed so far that I am still unsure about. While adding a cap across the coil (which is 100uh air wound) The frequency increases and while adding to any cap in the circuit it drops. I didn't do this with the scope connected so maybe this is doing something to the wave form.
\The idea I have is to later be able to sample this from the DC power on the coax line that feeds this circuit so all that I think I need here is a good solid oscillator that shifts as the coil does. I should be able to monitor if anything is near the coil in this way.

I have a lovelly little frequency counter that was sold by a CB shop in England. It is powered by 12v and covers up to 500MHz. It is a really great machine, but if you know the construction of it then you learn that you cannot always trust these things as far as you can throw them.

At 1.8MHz I had a TTL crystal oscillator and it sounded nice and pure on the 160m receiver. The counter however told a different story - it drifted from about 1MHz to 3MHz. About 10 years later I watched the same oscillator on my Spectrum analyser and I can see every odd harmonic just a few dB lower then the previous and every even harmonic just a little stronger than the previous. At 110MHz they all meet in amplitude.

The counter probably works very well at low frequencies if the prescaler is bypassed, but with the prescaler it cannot be trusted below about 20MHz. If an oscillator uses too much feedback then the waveform distorts, and it can even be seen on the scope as double peaks, which the counter can count seperately.

From all this I learned:

1 - Reduce oscillator feedback until oscillation stops, then increase it a little until it starts reliably when the osc is switched OFF/ON.
2 - Never try to count a waveform that is not a pure(ish) sine or square.

I have had a lot of success using a wavemeter to get a good estimate of frequency, but you can couple the wavemeter to a counter and use it as a filter to get rid of the crap and see a nice clean count. I sometimes use my GDO as a wavemeter.

Now watching this thread I see that this lesson is being learned once again. Try using a tuner and diode level meter in front of the counter at the frequency you are interest in. If you make a coil then make two.

Finally, one other lesson I learned was to make a multivibrator using 10nF ceramic caps. I then touched a cigarette on the caps and watched the frequency rise more than tenfold. You can calibrate this amd make a decent wide-range thermometer using this technique, but caps should be chosen with care if you want a stable frequency.

Ok, just a few observations. Sorry I have not been on the board for a while. I now have a new job, new salary and looking for a new house. In a few weeks time we will be looking at properties in Nerja :-) - yes the holiday season is arriving and we will be having another adventure - booked flights but not too sure where we will be after the first week outside Malaga.

BR Harry - SM0VPO

I finally got a chance to play with the oscillators again.
Ok I found my oscillator was wired wrong causing to oscillate and a frequency way above what I calculated. Now I still have what looks like a pulse train instead of a sign wave. doesn't that mean I have way too much feed back?
It's now running at 440k and I calculated 455k I think that part is great.
I have .004 across the emitter resistor and a .004 base to emitter and a .0022 from coil to base as per Harry's oscillator circuit.

Hi DragonForce
Yes I have build those out of ceramic filters and the seem to run just slightly lower that 455k but always look good.
I was mainly playing with a circuit I found in one of my book that show a 9" coil with only 5 turns runing at 455Khz. I could believe that was possible so I build it. It did run at 402Khz but I did a lot of strange things and the wave form was not a good sign wave on the scope. You talked about harmonics am that might be what was going on there. The circuit I coppies it from was for a search coil on a metal detector.

Hi Jalex

Not sure what you're trying to do... but

I've had success reaching 455KHz with 200 turns on a ferrite cored former with 220 pF in series. I don't have a scope so have no idea what the waveform looked like, but my frequency counter had no problems reading it.

One thing to mention here is harmonics - mine seems to lock onto the strongest signal - it's not necessarily the fundimantal.

Capacitors used in tuned circuits need to be NP0 ceramic or high quality polystyrene/mica. Use a crappy cap from the spares box and it can cause all sorts of trouble - stability among them.

Be aware also that semiconductor junctions can exhibit capacitance when reversed biassed. All sorts of "hidden" capacitance like this can cause a lot of head scratching.

To be honest, when I built my alignment RF generator I used one of those little ceramic resonators on 455KHz - it works very well - until you get an IF frequency of something other than 455KHz to align.

Hi Ivan
I am learning all the things you talked about above. I use a victor VC3165 2.5 gig counter. It has two bands and all below 30 mhz is on the low band. I am a little unsure how to couple it to the circuits though. It came with a shielded cable with just two alligator clips. I am grounding one side and feeding though a .1 capacitor to the other. I tried winding a little pick-up coil like Harry said but it didn't seem to work that well either.
The oscillator is like the one on Harry's useful circuits page. I added a .001 capacitor from + rail to the base and the frequency is now 336 Khz. To me this could mean that I have too much signal on the base. I used a .01 uf cap from the coil to the base and a 270k bios resistor and emitter at 1k..
I also wound another coil 9 inches that read 60 uh the frequency raised to 402 khz. The original coil is 102 uh.
What can cause an oscillator to speed up when I add more capacitance as I have seen it do that.
I have never tried to make an oscillator run on a exact frequency before and didn't know how hard that could be I usually just work around what ever they come out when I get them in the ball-park.
Another thing I can't figure out is why I can't seem to change the frequency very much by playing around with the capacitors.
What I am trying to do here is make an oscillator with a coil that is very sensitive to things around it and send the signal back up the supply line.

Hi Jalex,
please keep in mind, that the LC product affects frequency and the L/C ratio affects impedance of the tuned circuit. Changing the frequency in a really wide range by capacitor only, one may get out of the oscillating conditions easily. And the amount of feedback is important, too.

BR from Ivan

If you are not getting the results you are expecting, ask yourself how reliable your method of measuring is. I dont know how you are measuring the frequency, but if you are using a frequency counter it may not be triggering properly if the waveform is ugly.

Hi
I am just playing with oscillators. I have a air coil 5" with 30 turns. With the caps I chose I get 55Khz I was shooting for 455Khz.
when I increase the capacitance across the coil I get 156 Khz I thought that would lower it. The wave form looks more like a pulse train rather than a sine wave.
Whats going on here? Does this mean I have too much feed back?