Talk:Vackář oscillator

Original circuit

Let's include the circuit from the original paper. Reason is, when I put the diagrammed circuit (created by user:Rainglasz), into an LTspice simulator to look for oscillation, I found it was too damp to sustain oscillation with the parameters given. The problem was, the JFET operates at lower voltages than do tubes, so the value for C₀ needed adjustment. But, it does oscillate in the simulator if C₀ > 6680pF and given a pulse to start it. So I'm guessing that the relation Cv Cg /(Cv + Cg) = Ca - C₀ will give about optimal behaviour. -- Waveguy (talk) 03:52, 14 March 2010 (UTC)

Okay, reading Vackář's original paper, I notice in his figure 5 that there is an additional resistor in parallel to Cg; and that he does specify C₀:Ca = Cv:Cg for optimal behaviour. -- Waveguy (talk) 17:38, 14 March 2010 (UTC)

so this yields correct results with the missing resistor -- Waveguy (talk) 22:07, 14 March 2010 (UTC)

The Resistor in Fig. 5 between gate and cathode is just to provide a DC path; it is normally about 1MegOhms. Otherwise, the gate could be driven to negative voltages by static charges, and inhibit operation. It would also be advisable for a MOSFet. In real practical circuits, I would use it also for JFETs, but only about 100kOhms and not smaller. Rainglasz (talk) 13:25, 18 March 2010 (UTC)

Startup of oscillators in LTSpice (and probably any simulator) can be boring, because a circuit will properly start outside the simulator, where it keeps to be stable and not oscillating. When testing the circuti in LTspice, I use the startup option in the transient menu, which starts power from 0V and thus provides enough noise to start the oszillation. Alternatively, in particular with power converters, I use a very small initial current in the inductor (i.e. .ic I(L)=1n), that is sufficient to start the circuit. I prefer this over your solution to insert continously a driving signal at the drain, because in general the control statements for the simulator are not shown in the schematic (this is not an electronics workbench). Rainglasz (talk) 13:25, 18 March 2010 (UTC)

But you are perfectly right that I should have used values closer to the paper. But I have to check this more thoroughly, so give me some days for that, so that I finally can prepare a new schematic. If you like to create SVG drawings from LTSpice, see my user page here. Rainglasz (talk) 13:25, 18 March 2010 (UTC)

The problem with LT spice is, it doesn't work properly. Dave — Preceding unsigned comment added by 88.146.180.186 (talk) 23:24, 8 February 2012 (UTC)

Dave, if you have a better solution with a better simulator show it to us.

It doesn't look the thread was updated since March of 2010. I think Rainglasz wanted to revert to the original schematic. I assume the one in the article shows the original values. Also, it's not clear from the above discussion if a 1MΩ resistor should be used and where it should go. I see R and R₁ and none of the two is a 1MΩ resistor. I also don't understand what is V2. If it works with R1 to provide current, the current is 100nA and not 1nA like mentioned above.

ICE77 (talk) 02:53, 4 August 2015 (UTC)

Comments and questions

1. How is Vackář pronounced?

2. The paragraphs under "Invention" should be in the past tense (not half in the present tense and half in the past tense).

3. "Vackář reports that VFO circuit being used by the Czechoslovak Post Office since 1946."

This sentence needs revision.

4. After reading the "Invention" section, it's not clear if Vackář or Radioslava invented the oscillator. It seems Vackář credited a company called Radioslava for developing the circuit in 1945. If that is the case, why is that this circuit is not called Radioslava oscillator?

5. Is there an expression for predicting the frequency based on L and C values?

ICE77 (talk) 02:55, 4 August 2015 (UTC)

3, 4. The 1945 Radioslava VFO is commonly referred to as the Vackář oscillator. Vackář described the circuit to the world, and it became known by his name. There is a slight possibility that Vackář worked for Radioslava. Vackář offered significant design improvements over the Radioslava design, but those extended-range designs that are properly his are seldom used.

5. Yes. read the paper. Glrx (talk) 18:23, 2 September 2015 (UTC)