Hello,
Crystal frequency 17kHz, no output what's wrong with this circuit?

Thank you

 

hi 777,
Welcome to AAC.
Please post your LTS asc file.
E

 

Please find the attached LTS asc file

 

Hi 777,
You need to set Vhigh=5 in the Inverter model, also the 'u' symbol is missing off the 5.366 d-15u.
Doing further tests on your circuit.
E

 

hi 777,
The circuit will not run with the basic inverter delays etc. so tried with 74HCT04 as a test.
The 5.66 e-15 is not recognized by LTS, so used a 5.66p
What is the purpose 0F capacitor across the 100k ?
Also, LTS reported an open Out pin, so added a 1meg.
There appears to be a design problem.
E

 

hi 777.
If you want to experiment with LTS digital components, check out this video, Fresz is 1st class.
E

 

Hey Eric,
Thank you so much for the help!

 

The following simulation demonstrates how to model a crystal and how to determine both the series and parallel resonant frequencies. You can then take these parameters and make them into a subcircuit.


Green trace is the circuit on the left., and the blue trace is the circuit on the right.

 

It may not be obvious from the above simulation, but the FREQ, Q, C and RS values define the values of the motional components, AND the motional components define the FREQ, Q, C and RS values.

 

You have created the wrong scheme. You are trying to use a digital element in an amplifier mode, which is wrong! Use an inverter on CMOS transistors.

 

changed the inverter - TC4429

 

The above output is asymmetric with 50% duty cycle. How can I avoid that?

 

The above output is asymmetric with 50% duty cycle. How can I avoid that?

You could run the output through a D-Flip-Flop, configured as a divide by 2 circuit, to get a 50% dutycycle square wave at half the frequency of the crystal.

 

Why is the 0.33pF capacitance of C2 your output capacitor "almost less than nothing"?

 

See

 

Bordodynov
Thank you so much!
What is this .opt reltol=1u for?

 

hi 777.
Check this link for LTSpice parameters.
Relative Tolerance: The relative error allowed between the compressed data and the uncompressed data.
E
https://ltwiki.org/index.php?title=Control_Panel

 

See
View attachment 292899

Hi Bordodynov,
The output of this spice results are analyzed by using matlab threshold analysis method. The jitter starts out quite large but narrows after that. I did simulation upto 3seconds, steady state arrives after the initial jitter. How can I avoid this initial jitter?
Thank you for your time!

 

I don't think you can get rid of it completely. Is there a small amount of time where the startup transient would be acceptable? If not, then I fear your quest will be fruitless.

 

Low frequency quartz oscillators are slow to start. In a wristwatch, for example, start-up is allowed up to 5 seconds.