Home Product Directory Topics Note Pad MTT-S Video Coverage Back Issues RF Blogs Military Electronics Subscribe News Online News Design Features Web Seminars PartFinder Whitepapers Microwave Legends Newsletter EuMW 2007 WebConnect RF Design  RSS


PART SEARCH :
GlobalSpec - The Engineering Search Engine


Related Resources

  
Reprints   Printer-Friendly    Email this Article    RSS        Font Size     What's This?

[Components]
Planar Resonators Arm Tunable Oscillators
Voltage-controlled oscillators (VCOs) based on novel self-injection-locked compact coupled planar resonators (CCPRs) feature high Qs and low phase noise at microwave frequencies.

Ajay Kumar Poddar, Ulrich L. Rohde  |  ED Online ID #19385 |  July 2008

Last month, Part 1 of this article introduced novel self-injection- locked compact-coupled-planar-resonator (CCPR) oscillators. Part 2 concludes this article with more details on CCPR technology and some product examples.

Edward5 proposed a novel, compact, high-Q multilayer integrable printed helical resonator that offers an optimum ratio of loaded quality factor to unloaded quality factor (QL/Q0 ) for minimum phase noise for a given VCO topology. Figure 5 shows an integrable planar helical resonator coupled to coplanar waveguide (CPW) for VCO applications.6 But such high-Q helical resonators are limited in tuning range for given phase-noise, size, and cost requirements.5,6

A recent publication11 described the design of an extended resonance oscillator (ERO) in which the resonator group delay is maximized for low phase noise. From ref. 6, the oscillator’s loaded Q factor, QL, is

where
φ(ω) = the phase of the oscillator’s open loop transfer function at a steady state and
τd = the group delay.

Figure 6 shows a typical ERO circuit using an N-way power divider and combiner where the condition for coherent power combining can be obtained by making phase difference between successive device output ports (θdn) equal to the phase delay between the corresponding device input ports (θgn).

From ref. 7, QL is proportional to the absolute value of the group delay; therefore, the main design objective for the ERO is to maximize group delay by incorporating (N >2) multiple devices. From ref. 11, the group delay τd of the N-device ERO depicted in Fig. 3 can be described by

where
Ii = the input current,
I0 = the output current, and
V0N = the voltage at the output of the Nth device.

From ref. 11, the figure of merit (FOM), F, can be given by

where
τd = the group delay and
L = the insertion loss.

The relative noise contribution of the N-device ERO circuit with respect to a two-device ERO can be given by11:

where
τDn = the group delay of the N-device ERO and
LN = the insertion loss of the N-device ERO.

From ref. 11, an eight-device ERO will yield about a 13-dB improvement in phase noise in comparison to a twodevice ERO, but there is a limitation in the number of devices for a given tuning range, noise factor, and power dissipation. The typical ERO circuit shown in Fig. 6 is limited to narrow/fixed frequency applications, sensitive to temperature variations, and requires larger real estate and power, therefore, not a promising alternative to DROs.

The new approach presented here simplifies the limitation of the ERO by incorporating a stub-based tuning mechanism to maximize the group delay for a given operating mode. The present work describes a novel topology that improves the Q factor in compact size, and also suited for MMIC process. Figures 7 and 8 show typical stub-tuned planar-coupled resonators (STPCRs). They use open and shorted stubs depending upon the injection strength for a given mode, operating frequency, and tuning range. Figure 7 shows open stubs of lengths l1 and l2 (l1,2 = λ0/4±Δl), which form the self-coupling mechanism (without using a coupling capacitor). The two unequal planar open stubs exhibit resonant frequencies below and above f0, in which the lengths of the resonators are symmetrically offset by the amount ±Δl (Δl<<λ0). This approach provides a selfcoupling mechanism without a lumped capacitor as a coupling element. The input admittance Yi(ω) for this configuration is given by Eqs. 12-17.

where
Y0 = the characteristic admittance,
Z0 = the characteristic impedance,
vp = the phase velocity,
φ = the phase shift,
γ(ω) = the propagation constant,
Gi(ω) = input conductance, and
Bi(ω) = input susceptance.

From ref. 15, Rp can be found by Eq. 18.

From refs. 14 and 16, Cp and Lp can be given by Eqs. 19 and 20 (Fig. 4 ):

Continue to page 2


<-- prev. page     [1] 2 3     next page -->




Reprints   Printer-Friendly    Email this Article    RSS        Font Size     What's This?




Reader Comments

superb approach, DRO is not affordable due to narrow/fixed frequency characteristics for satellite com applications we are looking for 11.8-12 GHz planar VCO for SATCOM applications. could not find this model in the catalog, can author give us the availability of this part? regards, Nijhaman

Jacob Nijhaman -September 04, 2008   (Article Rating: )

Ulrich, i like this manuscript, well written and self-explanatory. Thanks, - Putin Jamdagni

Putin Jamdagni, Ph.D. -August 18, 2008   (Article Rating: )

informative article but need prerequisite knowledge of advance mathematics...

Arun Judeza -August 16, 2008   (Article Rating: )

Hi Ajay, Good work,scientific flavor, and state-of-the-art approach !!! Rgds, Jeena

Dr. S. Jeena, Senior Scientist, Delhi

Subramanyam Jeena -August 15, 2008   (Article Rating: )

what kind of bandwidth we expect in IC version versus disrete circuits, which are described in this paper.

Putin -August 09, 2008   (Article Rating: )

good article but very difficult to correlate mathematical derivation pertaining to noise reduction approach as cited in the second part. like to read extension of this paper if any in future.

Uzerian Mirza -August 09, 2008   (Article Rating: )

Readable but not easily digestable. I am impressed with the mathematical part of this paper that makes sense while correlating conventional approach to new one.

Juliot -August 06, 2008   (Article Rating: )

We have immidiate requirement of 11.2-11.8 GHz VCO in 0.3x0.3 inches size. We have limitation in supply voltage and current (Vcc=3V, Ic=20mA) but flexible for tuning voltage. Can this technlogy provides solution for lower supply voltage? The phase noise performance is -95 dBc/Hz @ 100 kHz offset. David Solanki RF Component, Tokyo

David Solanki -August 06, 2008

Ulrich, Outstanding paper. What is maximum operating frequency of this technology? What we understand that physical dimension of the planar resonator may dictate the over all Q factor and performance. BiCMOS process is the technlogy to go for this kind of configuration for a given cross-coupling in integrated circuit.

George -August 06, 2008   (Article Rating: )

paper makes sense but i am not sure about the thermal stability. can authors explain about the thermal drift (ppm ???), and how one can control drift in IC. PLL based design can control the thermal driftb(any idea???).

Rebiz Joyce -August 06, 2008

What is chip dimension, We are looking for 0.25x0.25 inches size for our production units. Quantity can be several thousands???

T. Subramanyan -August 03, 2008   (Article Rating: )

This work attracts me, although it's bound to be big turn-around till we get chip in hand for commercialization and replacement of existing DROs market. John, RFMD

John -August 03, 2008   (Article Rating: )

This work attracks me, although it's bound to be big turn-around till we get chip in hand for commercialization and replacement of existing DROs market. John, RFMD

John -August 03, 2008   (Article Rating: )

Dear Rohde, et.al, Can we use your circuit layout for integrated circuit for my educational research work? If yes, then advance thanks. If not then what I'm supposed to do for getting permission. This is purely academic work and will be properly referenced with the patent number as shown in your layout. Dr. T. Rukediza, Brazil

Rukediza -August 03, 2008   (Article Rating: )

Ulrich, let me concur first that this is superb paper on planar resonator based signal sources. As you described about the novel MCSTPR approach, example tunable DCO/DXO VCOs using a SiGe heterojunction- bipolar-transistor (HBT) active device (a model BFP 740 from Infineon) were fabricated on Rogers substrate material with a dielectric constant of 3.38 and thickness of 30 mils (microstripline); can we understant that the phase noise performance and other characteristics will remain unchanged in other technologies such as GaAs or BiCMOS. Please comment, may be other readers might have similiar queries on this matter. Thanks, Prof. J. Bahal

Prof. J. Bahal -August 03, 2008   (Article Rating: )

Authors may like to extend this article in part-III, and describe in detail about the technological pitfalls while selecting processing (GaAs,SiGe, BiCMOS,and other possible choice). This part is just Wow !!! First time reading, it was boring but in second time it was interesting, and while reading third time, I am now fascinated with this article.

Kim -August 02, 2008   (Article Rating: )

New dimension and new approach...both part of the article generates immense interest while reading...

Liyao -August 01, 2008   (Article Rating: )

Can we get samples for my PhD work, we are interested to build point to point radio using 0.3x0.3 inches VCO (11.8-12.2 GHz). Thanks, Rajeev Agrawal Caltech,USA

Rajeev Agrawal -July 30, 2008   (Article Rating: )

We seldom find state of the art work in journal, but this is solid piece of work that has significant vaues in electronic industries. Author's work will be referenced in my article (IEEE Journal of solid state), and I convey my personal thanks to Ulrich.

George Steinbach -July 26, 2008   (Article Rating: )

state of the art article but real breakthrough... M. Sundarajan Prof., EED, Delhi

M. Sundarajan -July 24, 2008   (Article Rating: )

read both parts but still not clear cut answer is given about the technology transition from discrete to integrated. authors may like to address this issues in next part for clear understanding. i read author's oscillator book, found some useful mathematical derivation, which is helping is understanding this article.

Tom -July 23, 2008   (Article Rating: )

This article is simply great! this is the real piece of work in printed coupled resonator VCO that can be seamlessly transformed into integrated oscillators. I read both part of the papers and previous papers on the subject by the authors and there is much more to know about the concept of mode coupling. It is well organized, to the point, and fresh for new RF engineers who are working in this field.

Umrendra Srivastava -July 22, 2008   (Article Rating: )

Ulrich, Keep it up, solid piece of design work... Rgds

David -July 21, 2008   (Article Rating: )

Can authors highlights drawback of this approach. This is a solid piece of design work that contains high level of mathematics with clear correlations what they are talking about. We seldom find this kind of paper that has mathematical derivation in support of CAD simulations and experimental results. As describe in part-I and part-II, we understand design approach but fail to digest the concept of Q factor in IC, which is bound to degrade, resulting poor phase noise. Would it be possible to you to describe the concept of compensating/nullifying the Q degradation phenomena in lossy IC substrate. FYI, this techniques as a whole would change the design philosophy of integrated TX-RX circuit; authors are highly appreciated for their innovative and novel concept utilized in this research work. Dr. Roshan Giakwad Analog Devices, Bangalore

Dr. Roshan Giakwad -July 20, 2008   (Article Rating: )

Can authors highlights drawback of this approach. This is a solid piece of design work that contains high level of mathematics with clear correlations what they are talking about. We seldom find this kind of paper that has mathematical derivation in support of CAD simulations and experimental results. As describe in part-I and part-II, we understand design approach but fail to digest the concept of Q factor in IC, which is bound to degrade, resulting poor phase noise. Would it be possible to you to describe the concept of compensating/nullifying the Q degradation phenomena in lossy IC substrate. FYI, this techniques as a whole would change the design philosophy of integrated TX-RX circuit; authors are highly appreciated for their innovative and novel concept utilized in this research work. Dr. Roshan Giakwad Analog Devices, Bangalore

Anonymous -July 20, 2008   (Article Rating: )

Hi Ulrich, I enjoyed reading your paper in which part-I introduces the mathematical basis for designing integrable oscillator (inexpensive alternative of DRO, and subsequent verification of the design via measured data. The second part adds practical circuits using MCSTPR theory to classical terminology, expands Leeson noise model to dynamic phase noise model, and includes beta opt and mopt concept for oscillator analysis. This piece of work can ca help readers, provided some insights for transformations from discrete to the integrated circuits while maintaining comparable performances. My PhD students are building some of the example circuits for academic purpose with a proper reference (not for commercial use), will update you the outcomes of your discrete example circuit in IC configurations. Regards, Prof. T. Jacob

T. Jacob -July 20, 2008   (Article Rating: )

New approach

Anonymous -July 19, 2008

New approach

Anonymous -July 19, 2008

This part (II) demonstrates the feasibility of printed resonator based regenerative circuits that can partially substitute the DROs (if we need more tuning range).

Riya Joseph -July 17, 2008   (Article Rating: )

I enjoyed reading both parts (part 1 and 2) of the article, would like to express my thanks to authors who are pioneer in VCO's field. It is but natural to expect breakthrough in technology but and it seems that gradually printed structures are winning other territories in faster pace than expected in reality. The only risk what we anticipate is the thermal stability and real challenge witnessed by design communities to compensate the frequency drift due to changes in temperature and associated cross-couplings and substrate noise in integrated VCOs.

Jacob -July 17, 2008   (Article Rating: )

POST YOUR COMMENTS HERE
Name:

Email:
Rate this article:

 less useful more useful 
1
2
3
4
5

Your Comments: