Designer’s Guide Community

vivkr Community Fellow Offline Posts: 780	Frequency divider for sinewave input Jul 26^th, 2005, 4:57am Hi, I would like to know how one can divide a sinewave frequency by say, 2. How would one realize a frequency multiplier for sinewaves alone? Thanks Vivek
Back to top	IP Logged

rf-design

Senior Member

Offline

Reiner Franke

Posts: 165
Germany

Re: Frequency divider for sinewave input
Reply #1 - Jul 26^th, 2005, 1:41pm

Answer to the second question first:

Take math

sin(2*pi*f*t)^2=0.5*(1-cos(2*2*pi*f*t))

a MOS is a quite reasonable squarer. If you AC couple the result of the squarer you could get up to -30dBc clean 2. harmonic.

The first question is more difficult because you missing to state the exact intention, or spec. If you are not interested in simply divide by 2 but to generate quite clean divided sine the following will help.

If you analyse in detail the operation of a bipolar divider you will find that the input sinewave switch the M+S latches from latched to transparent mode. Because of the capacitive load at the outputs each latch work similar to a leaky integrator. You can also oberserve that there is a small frequency range where the required input amplitude for divide operation is very small. Within this frequency range the M+S latches operate more as an IQ oscillator locked to input at double the selfoscillating frequency. If you modify these latches further for more linear operating you get an locking IQ oscillator with quite reasonable harmonic purity.

I am not shure that I am too detailed?

Back to top

IP Logged

vivkr

Community Fellow

Offline

Posts: 780

Re: Frequency divider for sinewave input
Reply #2 - Jul 27^th, 2005, 7:12am

Hi rf-design,

I dont quite understand your explanation. Anyway, to rephrase what I mean by the first question (dividing the frequency of a sinewave by 2), I was thinking in terms of an analog PLL designed as a frequency multiplier. The input frequency is multiplied up by a certain factor xN. So, the VCO frequency must be divided by this number and compared to the reference frequency.

I just wanted to know if such circuits exist or not and how practical would they be. It is a purely academic question.

Thanks
Vivek

Back to top

IP Logged

rf-design

Senior Member

Offline

Reiner Franke

Posts: 165
Germany

Re: Frequency divider for sinewave input
Reply #3 - Jul 27^th, 2005, 7:37am

Sorry,

I did not understand you first because I thought within this forum PLL is a common topic.

Yes PLL's are doing this function. There are one of the function found in nearly every communication device. They couple in a regulated manner two different frequency sources. There are PLL's where the ratio between both frequency sources is an integer and other where it is a fraction.

Back to top

IP Logged

vivkr

Community Fellow

Offline

Posts: 780

Re: Frequency divider for sinewave input
Reply #4 - Jul 28^th, 2005, 12:22am

Hi,

I think there is a bit of confusion here. I am aware that one can use PLLs to perform clock multiplication very easily. But usually, this uses square wave clocks.

The question I had was about frequency multiplication of a pure sinewave to get another nearly pure sinewave using a continuous-time variant of the normal PLL. Such a PLL would probably use a Gilbert cell-based phase detector but would need to divide the Nx times faster sinewave to get another sinewave that matches the reference input. Unlike the conventional PLL, this would not involve sampling the phase error at discrete time intervals but would just work like a feedback loop which is always closed.

As I said, the question is of academic interest only.

Anyway, thanks for the feedback.

Vivek

Back to top

IP Logged