Simulation of a commercial crystal oscillator in ADS's SMT crystal library

[aleksidql]

Dear All,

I tried to test the performance of ECS-10.7-7.5 crystal oscillator, which i think is the same as the sx_tqz_CS10_TQ71_07_A_19930601 provided by System_lib in ADS. I tried to do the test according to the Fixed frequency oscillator DesignGuide, but the simulation broke down, how can i fix it?

Any help to test this oscillator will be appreciated. Thanks for your time!

 

[BradtheRad]

Did you start out with a sample circuit included with the program? Did it provide sensible results?

Alerts say capacitor values are large. (Apparently 4.9 Farad & .0214 F). When running a circuit at 10 MHz, reasonable values are in the pF or fF range.

By happenstance the Falstad simulator contains a 10 MHz crystal. Default values shown below:

 

[aleksidql]

Did you start out with a sample circuit included with the program? Did it provide sensible results?

Alerts say capacitor values are large. (Apparently 4.9 Farad & .0214 F). When running a circuit at 10 MHz, reasonable values are in the pF or fF range.

By happenstance the Falstad simulator contains a 10 MHz crystal. Default values shown below:

View attachment 177971

Hi, BradtheRad. Thanks for your time first!

I noticed that the capacitor values were large. But I guess it was because the unit of this commercial crystal had not been set, for exmaple pF. Instead, ADS takes the unit F. I don't know how to change the setting.(change the setting of schematic didn't work out, and I can't change the setting of libraries.....)
I was assigned to test this crystal's performance by ADS, I wonder if I could use the Falstad simulator to test it. Because the library porvided by manufacturer is needed.

 

[BigBoss]

Before using OscPort element, you should check the "necessary and sufficient oscillation condition" of the circuit.
In order to do this, you have to use OscTest element in s-parameter analysis palette. This is standalone simulation controller to predict the oscillation conditions. No need to place additional s-parameter simulation controller.
When you place this element adequately , it will make a small signal s-parameters analysis in a closed loop then you can see the oscillator has necessary and sufficient conditions or not.
If you plot S11 ( or S22) on a Polar Diagram, you will see S11 is close to 1+j*0 point or it encircled. If it encircles this point, it means you have RHP otherwise it will not oscillate ( Nyquist Criteria )

For more information, see help and other application notes on oscillator design.
Also, you have to check the negative resistance of the amplifier part and it must be at least 3 times greater than the real part of resonator's real part.

There are also additional techniques but no time to explain. Read something on.

 

[BradtheRad]

I wonder if I could use the Falstad simulator to test it. Because the library porvided by manufacturer is needed.

Edaboard thread below discusses crystal oscillators and has a few schematics. Post #6 has a link to a Falstad simulation with a model of a crystal (inductor, series capacitor, parallel capacitor, resistor). Click it to:

1) Open website falstad.com/circuit.
2) Load the schematic into the simulator.
3) Run it on your computer.
You can modify component values as desired although Falstad's does not use the Spice family library.

https://www.edaboard.com/threads/colpitts-clapp-overtone-crystal-oscillator.334817/