Why is there distortion in my Clapp Oscillator?

[BlownUpCapacitor]

Hi, I am trying to make a Clapp oscillator for my project, but every time I made one, there have been no oscillations until now, where I get at least SOME KIND of oscillation.

I have looked at my connections and see nothing wrong. According to LT Spice, there should be none of this "pulsed" rubbish and should work perfectly.

I have worked on this little shitter for a few days and only until 1 in the morning today did I get it to "work". But I need a relatively pure sine wave, or rather some kind of stable & reliable oscillation.

I am using a clapp oscillator because I plan to turn it into a VCO using a few varicaps I have. But I first need to get this thing to work without the varicap.

You can see 3 images. The LT spice operation and schematic, my build on the copper clad, and the scope waveform.

Frankly, I am way too tired for this am desperately need sleep because of the busy week I had.

Good night, I'll see y'all in the (later) morning. Perhaps even the afternoon in PST time.

[jonpaul]

squegging, poor bias

j

[Zenith]

As jonpaul said.

https://www.e-magnetica.pl/doku.php/Squegging

Circuit simulators such as LT Spice are very useful, but they won't tell you everything, such as the effects of parasitic inductance and capacitance.

Try a couple of different transistor types.

There comes a time with a problem where you do best to set it aside for a day or two, and come back to it fresh.

[KE5FX]

Kind of a long shot, but C6 may not be big enough to act as a good emitter bypass at the frequencies involved.

You could try a simpler topology, maybe.  I recently encountered a circuit that rivals the simplest, quickiest/dirtiest Clapp oscillators in ham radio literature, but it works well in something that probably cost north of $100K:



Notice that they ground the base rather than the emitter in this approach, using a 0.1 uF cap even though the oscillator runs near 100 MHz.  They don't worry much about the RF side, but they do want a stable operating point in this particular application, at least in the short term.

Also, I'd probe this circuit by clipping your probe's ground lead to the probe tip to make a loop, and holding it near the coil.  Your probe is adding some extra loading here, and there will be some parasitic effects.  The LC junction in a series resonant circuit is a high-impedance, high-voltage node.  (Which is actually why I had to repair the circuit above -- the voltage was high enough to burn a hole in the insulation of the coax that acted as a 3 pF series capacitor.)

[T3sl4co1l]

Squegging results when the time constant and impedance of the bias network, and the feedback magnitude, are such that rectification occurs, pushing the amplifier into such deep class C that loop gain, averaged over the cycle, drops below unity, and it rings down.  The network then recovers, slowly, and the process restarts.

It's a blocking oscillator, "with extra steps".  That is, the overall behavior is that of a relaxation oscillator, where it just so happens the regenerative (negative resistance) phase occurs with high-frequency oscillation, but the overall effect is the same.

That there are two different time constants (base and emitter) is peculiar.  This isn't necessarily a problem: they act more or less in parallel, so it's a pole-zero in the response (assuming a linearized operating point, which we're only occasionally crossing in actual operation anyway), not a double-pole (that would cost phase margin and suggest oscillation).  They should be equal (i.e., C1 * (R2 || R3) = R1 * C6, but probably C6 has to be larger because r_e is a variable term in its response as well).

Feedback amplitude must be reduced to just what is required; with no loss in the circuit (no load -- remember, SPICE components are ideal down to numerical error*), power only has one place to go, the base (well, and whatever the scope probe is taking).

*But do check the properties of everything: LTSpice has a bad habit of hiding default-value parasitic parameters behind dialogs.

Why SPICE hasn't captured the squegging, dunno, but there may be further differences between your actual circuit.  Note that supply impedance matters, too (bypass it just in case).  Ground-return and series component inductances may matter (at least, probably not a whole lot for frequencies a 2N3904 can reach).

Huh. There are also two degenerate nodes in your simulation.  RSHUNT or LT's modified solver (see settings) are probably enough to deal with that, but in general it is a bad idea to connect capacitors in series in SPICE.  (These are C1, C4/C5, and C2/C3, C5 via L2.)  The usual error is "singular matrix" (the matrix version of divide-by-zero), because the DC voltage on those nodes has no solution, it can be literally anything; tweaks, as above, can determine it, but leaving the matrix near-singular often costs simulation speed.

Tim

[BlownUpCapacitor]

Thank you all for your tips.

I managed to get my oscillator to work by using a larger emitter bypass cap and a transformer for the inductor. This way the oscillator is more isolated to whatever it is connected to.

I also adjusted some capacitor values. Once it worked, I replaced one the the caps in the tank circuit with a varicap. My oscillator now works with the vari cap and does perform as a VCO.

Thank you all!

[BlownUpCapacitor]

I have another question: The amplitude output changes based on the frequency output/VCO control voltage input, and this is undesirable. Does anyone know how to fix this?

[KE5FX]

Could be variations in the varactor Q with applied voltage.  Output is weaker at higher frequencies, I assume?  You're using two varactors back to back to avoid forward bias, right?  Try a parallel combination of two series and two parallel varactors, maybe.

Otherwise, the output coupling transformer could be a suspect, or inadequate bypassing.  Did you add a supply bypass as Tim suggested?  If not, do that.

[T3sl4co1l]

Consider the effect of varying Zo = sqrt(L/C) of the resonant tank.  What effect does this have on the oscillator's behavior?  How is the oscillator limiting, is it voltage limited, current, power?  What effect does all this have on amplitude?

The Q factor of the inductor also varies a bit with frequency, though usually not very much.  You also need to avoid rectification in the diodes themselves (peak signal level less than bias).

Tim

[BlownUpCapacitor]

Output is weaker at higher frequencies, I assume?  You're using two varactors back to back to avoid forward bias, right?

Output is stronger at higher frequencies. Maybe this has to do with the type of transformer I'm using? I'm using an old IF transformer after I've cut off the resonance capacitor. Maybe the ferrite works better at higher frequencies?

And yes, I am using two varicaps, with the "kathodes" connected to eachother, and that's where I inject the control voltage through a 33k ohm resistor. I don't know the exact values of it, but I'm pretty sure it's a 30pF-580pF varicap. I find all of my parts out of scrap boards. The RF components such as the varicap, and transformer came out of an old electronic radio tuner. The varicap was taken from the AM tuning part.

Otherwise, the output coupling transformer could be a suspect, or inadequate bypassing.  Did you add a supply bypass as Tim suggested?  If not, do that.

I find that if I increase the value of the emitter by-pass cap, squegging is a bigger problem. I checked in LT-spice as well and the results are the same. I decrease the value of that cap from 47nF to 3.3nF. That seems to be sweet spot.

I also added a 2.2nF supply bypass cap using some of the old RIFA caps from my Tek 2230 after I replaced them. I know they're good at removing EMI yes?

Consider the effect of varying Zo = sqrt(L/C) of the resonant tank.  What effect does this have on the oscillator's behavior?  How is the oscillator limiting, is it voltage limited, current, power?  What effect does all this have on amplitude?

The Q factor of the inductor also varies a bit with frequency, though usually not very much.  You also need to avoid rectification in the diodes themselves (peak signal level less than bias).

Tim

Being honest, I have very limited knowledge in RF. I know what I need to cram as much RF knowledge into my brain for my school project, but I am currently lacking at this point.

I am not sure how to vary the impedance of the LC tank circuit.

I have no load on the transformer output except the scope probe, which is 13pF I think of capacitance, and at 7MHz, (or was it 4MHz?), 1.7k ohms of Cx.

I am also confused about the diode rectification? I would assume that the diode characteristics of varicaps would be negligible with kathode to kathode right? Or am I wrong?

[DimitriP]

I am also confused about the diode rectification? I would assume that the diode characteristics of varicaps would be negligible with kathode to kathode right? Or am I wrong?

I guess it depends on how "creatively" you added the diodes
An updated schematic would help.

[BlownUpCapacitor]

An updated schematic would help.

Yes of course!

Here ya go!

[T3sl4co1l]

L2 is floating at DC.

Tim

[BlownUpCapacitor]

L2 is floating at DC.

Unsure by what you mean, for me the oscillator works just fine in LT-spice, except the wave is not sinusoidal.

Maybe it's because of the different varactor I used? I'm 96.7% sure the varicap I have is the KV1236Z.

[fourfathom]

L2 is floating at DC.

Having L2 floating isn't necessarily bad, but i does complicate the biasing of the varicaps.  I prefer to have the inductor grounded when practical, just to make the circuit perhaps less sensitive to stray capacitance / coupling.  But in theory it's OK.

You need have the anode of D1 at ground potential-- it's currently floating, so who know how your tuning voltage is biasing it?  Often in a circuit of this type having the L2 inductor at DC ground gives you a way to keep that diode properly biased, but in this case the various fixed capacitors make that a challenge.  You might be able to provide a DC ground reference by connecting a large-value resistor from D1 anode to ground; this also depends on the AC signal levels at that point.
One more thing:  That control voltage needs to be connected to the varicaps by a high impedance (a large resistor or a choke).  Otherwise you are putting a load at that point and almost certainly disturbing the frequency and amplitude of oscillation.

[RFDx]

Unsure by what you mean, for me the oscillator works just fine in LT-spice, except the wave is not sinusoidal.
Maybe it's because of the different varactor I used? I'm 96.7% sure the varicap I have is the KV1236Z.

Anode of D2 has no DC path (e.g. 100k resistor or similar) to ground. There is no indication in the schematic where the output voltage is taken from.
An example with AM varactor diodes in series with the inductor is attached. The transistor is of course also acting as a limiter, the rather high output voltage is of constant value. The frequency coverage is 4.6MHz-5.3MHz with a control voltage of 2...12V.
If you are looking for a cleaner, constant output voltage with low harmonics you have to consider a more practical oscillator topology with an additional, symmetrical (soft) limiter stage in the loop and not rely on the main oscillator transistor doing that.

[BlownUpCapacitor]

Ah yes, sorry, I had mentioned earlier in a previous reply that I had been using a 33k ohm resistor for the control. Initially it was 100k, but I wasn't getting nearly enough range out of it, so I opted for the 33k ohm, which was the exact value the tuner system in my scrap tuner board was using to control the varactor.

I was hoping to get a 10MHz range out of it, but appears not. Perhaps using a higher frequency of oscillation is better, but I haven't been able to get it to oscillate higher than 8MHz. I know there are J-FET based designs, and they look promising. I'm justing using a Chinese 2N3904 as of now. But I have a K117 (audio) and a K168, which are both Japanese manufactured and from the original tuner. Maybe I could use the K168 as it lists Local Oscillator as one of its applications.

I'll try adding pull down resistor to D1's anode. D2 is already anode grounded.

[MarkT]

Change the inductor to 8.2µH or 6.8µH perhaps?