The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Design >> RF Design >> Leakage Current through control volatge in LC oscillator
https://designers-guide.org/forum/YaBB.pl?num=1358747754

Message started by Virvasav on Jan 20^th, 2013, 9:55pm

Title: Leakage Current through control volatge in LC oscillator
Post by Virvasav on Jan 20^th, 2013, 9:55pm

Hi,

Due to asymmetrical waveform in LC oscillator, I found a low frequency leakage current though control voltage. This leakage current causes reference spurs in PLL loop. Please help me to get rid of this problem.

Thanks,
Virvasav

Title: Re: Leakage Current through control volatge in LC oscillator
Post by raja.cedt on Jan 21^st, 2013, 12:02am

hello,
Could you provide some more details like control voltage waveform,due to asymmetry in VCO two voltages, you will have even harmonics on the control voltages which intern gives spurs at even multiples of carrier frequency (of course which is not a big problem due to lower BW of the loop). Coming to dc component (as you pointed out 'low frequency') which might cause reference spurs at reference frequency, but this can't happen with a good design. So check you schematic, is there any schematic entry error or some thing else....

Thanks,
raj.

Title: Re: Leakage Current through control volatge in LC oscillator
Post by Virvasav on Jan 21^st, 2013, 12:59am

Hi raj,

Actually in LC loop, due to asymmetrical waveform there is no proper charging and discharging of control voltage and thus a certain amount of charge sustains at tuning voltage which will cause some increase in voltage at the reference frequency.

Virvasav

Title: Re: Leakage Current through control volatge in LC oscillator
Post by raja.cedt on Jan 21^st, 2013, 1:13am

Hello,
How can you say "there is no proper charging and discharging of control voltage". Looks like some strange problem and can't be root cause if there is no waveforms. So please show me control voltage waveform and vco output's.

Thanks,
Raj.

Title: Re: Leakage Current through control volatge in LC oscillator
Post by loose-electron on Jan 22^nd, 2013, 11:09am

schematics, simulation results and data?

mind reading does not work well...

Title: Re: Leakage Current through control volatge in LC oscillator
Post by raja.cedt on Jan 22^nd, 2013, 11:37am

Dear looselectron,

what's your opinion on this ripple on the control voltage, i think i wont be a problem as i posted earlier, but i don't know how to observe in simulation just because while designing VCO i would rather connect a voltage source at the control voltage so no ripple at the same time in the PLL locking simulations again due to filtering nature. So i am surprised who he observed this effect.

Thanks,
Raj.

Title: Re: Leakage Current through control volatge in LC oscillator
Post by loose-electron on Jan 24^th, 2013, 3:30pm

raja.cedt wrote on Jan 22^nd, 2013, 11:37am:

Raj - I think more information is needed before you can say anything. A lot of PLL simulations are faulty due to errors in math resolution and similar.

I usually will do the PLL part by part (like you describe with the VCO control voltage fixed and open loop) and then when all of thats good do closed loop simulations at the behavioral level (Matlab-Simulink or Verilog AMS)

Jerry

Title: Re: Leakage Current through control volatge in LC oscillator
Post by Virvasav on Jan 30^th, 2013, 2:08am

Hi all,

Sorry for the late response.
Actually I am not facing any problem in the VCO level but while using this vco in top level it creates the problem.
For your assistance I am attaching the a test circuit diagram and plot of vtune. The thing to look here is that, I have fixed the control voltage at 0.5 V but while low pass filtering it I found that control voltage settles at the 0.63 V, it means some low frequency (or dc) current flow into the vco.
I think now you got my problem. Thanks.

Virvasav

Title: Re: Leakage Current through control volatge in LC oscillator
Post by loose-electron on Jan 30^th, 2013, 2:38pm

Is this leakage current seen in simulation?
Or in silicon?

If it is simulation, then you should
be able to find the path (KCL the currents)

You have a large number of reverse biased diodes
in your design that could be leaking. A diode does not have zero current when reverse biased.

See:

http://electronicdesign.com/analog/academic-simplifications-produce-meaningless-equations

Jerry