Royer Oscillator Applied to Ultra Low Voltage（Energy harvesting）

[gagagazi]

Hello, I'm learning about the Royer oscillator. Through a paper on IEEE, I got the following circuit topology. This is an oscillator that can start oscillation at very low input voltage（ a non isolated push full boost converter）. In order to better understand the structure of the circuit, I try to reproduce the circuit in the paper, but unfortunately, the circuit can not run.what is wrong?I'm not sure if I can upload the paper so here are the attach file and the circuit diagram.
In the original paper, the author's input is as low as several mV. I'm curious about how to do it. If you are interested, I put the paper below.Thank you very much

[KE5FX]

Sounds like the circuit relies on a small asymmetry in real-world JFETs to start up, so you might try adding a large resistor from one of the gates to someplace else (drain?) to unbalance things slightly.

Also, Cgs might just be on the schematic as a parasitic element inherent to the FETs, not something you're supposed to actually add to the circuit. 

The extra Cgs probably doesn't matter at these frequencies, but from what I can tell from the paper, they intend for L1 and L2 to be equal ("center tapped"), which makes sense.  So that sounds even more likely to be an issue.

Can you post the LTSpice .asc file?

[basinstreetdesign]

Could it be that there is no noise in the circuit to provide some random signals for the transistors to amplify at start up?  It is my understanding that oscillators will rely on this to provide initial signal and signal unbalance.  It filters this for the preferred frequency and continues amplifying that until full-amplitude oscillation is obtained only limited by the power rails.

[Zero999]

Is 1µH right for L1? The other inductors are 100mH. I thought they were supposed to be the same values.

[RoGeorge]

If it doesn't start in the LTspice simulation, then please attach the .asc file, so others can run and debug the circuit.
If it doesn't start in practice, post a picture with your circuit.

For the IEEE paper, post the DOI number, or name its title/authors.

Another topology of a very low voltage oscillator is the Peltz oscillator, can run at about 0.7V and has a single coil.
https://www.eevblog.com/forum/projects/simple-sinusoidal-oscillators/msg4315813/#msg4315813

[Terry Bites]

 Make sure that DC operating point is disabled, its setting up equilibrium conditions, not very useful in an oscillator!
Try adding some initial charge to C2 or C3.

[mawyatt]

Usually some form of "perturbation" will start an oscillator in simulations. Changing Vcc to a step voltage at time zero + should start any oscillator, altho achieving steady state may take some long simulations times. There are other techniques for achieving steady state is a shorter simulation time and still starting the oscillator.

Best,

[jonpaul]

not a Royer push pull CT, saturation transformer.

FETS used will give poor efficacy

Academic papers rarely give practical engineering designs

Jon