The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Design >> RF Design >> Quadrature VCO OR Quadrature Divider ?
https://designers-guide.org/forum/YaBB.pl?num=1262240362

Message started by Mayank on Dec 30^th, 2009, 10:19pm

Title: Quadrature VCO OR Quadrature Divider ?
Post by Mayank on Dec 30^th, 2009, 10:19pm

HI all,

If i want Quadrature I-Q LO signals, There are two major ways :--

1. Single Phase VCO -- running at 2 * F_LO + a Quadrature Divider.
2. Quadrature VCO -- running at F_LO

Which one's better ??

thanx,
Mayank.

Title: Re: Quadrature VCO OR Quadrature Divider ?
Post by loose-electron on Dec 31^st, 2009, 2:17pm

you can generally get better duty cycle and lower phase noise out of the 2X VCO and digital (ECL arcitecture) type divider.

Title: Re: Quadrature VCO OR Quadrature Divider ?
Post by rfcooltools.com on Jan 2^nd, 2010, 11:17pm

I have always talked my self out of Quadrature VCO's on a IC when it comes to making an LO signal for Quadrature conversion here is why:
Open loop phase noise is worse since the VCO A will present Noise to VCO B at the most sensitive transitions of each others duty cycle.
Closed loop phase noise will be worse since ulimately the LO from only one VCO will operate the PLL phase detector. That VCO will get the normal loop suppression of noise, but since the other VCO tune voltage would be tied to the same point its noise will not be suppressed by the loop unless it is common to both and worse yet the VCO not operating the phase detector will get the noise of the one operating the detector added to it since it is uncorrelated. There may be a way using two PLL's, but Getting the Quadrature correct is not obvious.

EM Coupling: depending how close the coils are to each other the coupling of these coils and the phase angle the resulting transformer will present the equivalent of an inductor with a different value and potentially lower Q than intended and when combined with the PLL loop the satisfactory locked condition may be where the phase is not 90 degrees.

On chip inductors Q increases with frequency while the degradation of varactor Q is negligible at frequencies of say 3 to 4GHz when compared to other deQ'ing elements.

Higher frequency inductors are typically smaller in value and thus in physical size thus cheaper.

Almost always the divided frequency equivalent phase noise will be better both in current, area, and simplicity which is usually always a win win.

Title: Re: Quadrature VCO OR Quadrature Divider ?
Post by Mayank on Jan 4^th, 2010, 12:12am

thanx Jerry and rfcooltools,
your replies helped a lot. I also studied further on this topic & i also think Q-vco is not such a good idea as a Q-divider with a single-phase VCO.

thanx,
Mayank.

Title: Re: Quadrature VCO OR Quadrature Divider ?
Post by raja.cedt on Jan 4^th, 2010, 3:57am

hi,
apart from the disadvantages explained above, both vco's will not oscilate where Q will be maximum due to coupling Transistors current.

Thanks,
Rajasekhar.

Title: Re: Quadrature VCO OR Quadrature Divider ?
Post by Mayank on Jan 4^th, 2010, 5:37am

Hey raja,
Didnt understand your last post so much....Can you pls. explain in some further detail ??

--
Mayank.

Title: Re: Quadrature VCO OR Quadrature Divider ?
Post by pancho_hideboo on Jan 4^th, 2010, 6:18am

Mayank wrote on Dec 30^th, 2009, 10:19pm:

1. Single Phase VCO -- running at 2 * F_LO + a Quadrature Divider.
2. Quadrature VCO -- running at FLO

Regarding current consumption, "2" is superior than "1".
Regarding phase accuracy, "1" is superior than "2".

Also you should pick up other scheme such as VCO -- running at FLO with passive poly-phase filter and buffer as candidate.

However I don't know actual product which is realized by "2".