The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Design >> Analog Design >> Crystal oscillator powered by LDO
https://designers-guide.org/forum/YaBB.pl?num=1400150333

Message started by Hrishikesh on May 15^th, 2014, 3:38am

Title: Crystal oscillator powered by LDO
Post by Hrishikesh on May 15^th, 2014, 3:38am

Hi,

I am working on a Pierce type parallel resonant crystal oscillator. I am able to see the oscillations when I apply an ideal power supply. But when the supply is from an LDO, the oscillation buildup is too slow for the oscillations to be recognizable. The supply level, ramp rate are all similar. I don't see any difference in the amplifier negative resistance or the current output from the supply. Anybody has any idea of the possible cause?

Title: Re: Crystal oscillator powered by LDO
Post by ic_engr on May 15^th, 2014, 6:58am

What is the Frequency of oscillation.
Are you aware that you need to add an initial condition ic=(some current) in the Lm (inductor) model of the crystal, else your crystal will not start.

If it is MHz frequency try making ic=5uA. If it is kHz frequency make ic=500nA and give it a try.

Regards

ic_engr

Title: Re: Crystal oscillator powered by LDO
Post by Hrishikesh on May 15^th, 2014, 9:47pm

Thanks for the reply. The frequency of oscillation is 8 MHz. I was not aware of setting an initial current. But I am able to start the oscillations when I simply ramp up the ideal supply from 0. So is this something relevant only when I use an LDO instead of an ideal supply?

Title: Re: Crystal oscillator powered by LDO
Post by ic_engr on May 16^th, 2014, 10:08am

The LRC model of the crystal in simulation is only ideal. There are no inherent noise in the crystal model in simulation to start and let the oscillations grow.

Therefore, the simulation needs an initial condition of some current in Lm.

There are other ways to start, but usually the simplistic approach is the use of initial condition.

Does your LDO have a decoupling capacitor that is large. It could be the ramp up is slightly longer such that the transient on LDO is unable to start the oscillation, contrary to the ideal supply.

Your other test is: use ideal supply as DC (don't ramp-up set DC voltage only by using VDC) then use initial condition current ic=xxx check what is the minimum you need to let the oscillations grow in simulation. May be 1uA or 5uA I don't know.
Remember if you give too large, then the amplitudes will clip therefore give lower values as explained before. If you can succeed in starting with some initial condition for a true ideal DC supply (with no ramp), then the same condition should be used for the case of LDO and your crystal oscillator should start with no problems.

ic_engr

Title: Re: Crystal oscillator powered by LDO
Post by loose-electron on May 16^th, 2014, 1:49pm

ic-engr has good advice, also look on the site here (search time) because I answered a bunch of questions on this a while back.

jerry

Title: Re: Crystal oscillator powered by LDO
Post by Hrishikesh on May 18^th, 2014, 10:39pm

Thank you very much for all the inputs!