IEEE TRANSACTIONS ON MICROWAVE THEORY AND TECHNIQUES, VOL. 61, NO.

12, DECEMBER 2013

4353

Theory and Design of Octave Tunable Filters With Lumped Tuning Elements
Akash Anand, Student Member, IEEE, Joshua Small, Member, IEEE, Dimitrios Peroulis, Member, IEEE, and Xiaoguang Liu, Member, IEEE
AbstractThis paper presents octave-tunable resonators and lters with surface mounted lumped tuning elements. Detailed theoretical analysis and modeling in terms of tuning range and unloaded quality factor are presented in agreement with simulated and measured results. Based on the models, a systematic design method to maximize tuning ratio and optimize of the resonator is suggested. A resonator tuning from 0.5 to 1.1 GHz ranging from 90 to 214 is demonstrated using solid-state with varactors. A two-pole lter with a tuning range of 0.51.1 GHz 0.1% with a constant 3-dB fractional bandwidth (FBW) of 4 and insertion loss of 1.67 dB at 1.1 GHz is demonstrated along with a three-pole lter with a tuning range of 0.581.22 GHz with a constant 3-dB FBW of 4 0.2% and insertion loss of 2.05 dB at 1.22 GHz. The measured input third-order intermodulation is better than 17 dBm over the frequency range for the two-pole lter. Index TermsCombline lter, combline resonator, evanescent-mode (EVA) design, lter design, lters, full-wave simulation, measurement and modeling, modeling, tunable lters, tunable resonators, waveguide lters.

ECENTLY, there has been a growing interest in tunable RF/microwave lters. The driving parameters for these lters are low loss, wide tuning, low power consumption, small size, fast tuning, high power handling, and ease of fabrication at a low cost. Various demonstrated tunable lters excel in some parameters at the cost of sacricing other parameters. For example, planer microstrip lters with lumped tuning components are easy to fabricate, but the unloaded quality factor suffers due to the low of the planer resonators [1][3]. To achieve higher than planer structures, highly loaded 3-D evanescent-mode (EVA) resonators integrated with various types of tuning technologies have shown promising results [4][8]. For example, a tunable lter with a piezoelectric actuator showed unloaded quality factor of 700300 at 4.62.3 GHz [4]. Two EVA resonators with RF microelectromechanical systems
Manuscript received July 05, 2013; revised October 07, 2013; accepted October 16, 2013. Date of publication November 05, 2013; date of current version December 02, 2013. This work was supported by the National Defense Science and Engineering Graduate Program and the University of California at Davis under the Hellman Fellows Program. This paper is an expanded paper from the IEEE International Microwave Symposium, Seattle, WA, USA, June 27, 2013. A. Anand, J. Small, and X. Liu are with the Department of Electrical and Computer Engineering, University of California at Davis, Davis, CA 95616 USA (e-mail: akaanand@ucdavisedu.edu; jasmall@ucdavis.edu; lxgliu@ucdavis.edu). D. Peroulis is with the School of Electrical and Computer Engineering, Purdue University, West Lafayette, IN 47907 USA (e-mail: dperouli@purdue.edu). Color versions of one or more of the gures in this paper are available online at http://ieeexplore.ieee.org. Digital Object Identier 10.1109/TMTT.2013.2287674

I. INTRODUCTION

(RF-MEMS) tuners, one with a switched capacitor network and the other with a silicon diaphragm, achieved of 500300 at 5.584.07 GHz [5] and of 1000300 at 246 GHz, respectively [8]. While these technologies attain high , complexity arises in fabrication due to the precise assembly needed to either align the tuners with the EVA cavitys vertical gap (typically in micrometer) or insert the RF-MEMS switching network inside the cavity. To avoid complicated fabrication and yet maintain high , 3-D cavities are integrated with commercially available surface mount tuning components as an alternative medium between low planer structures and high 3-D cavities. In [9], packaged RF-MEMS switches mounted on a substrate integrated waveguide (SIW) are used to get of 13293 at 1.61.2 GHz. However, the tuning range is limited to a few states. RF-MEMS capacitor banks mounted on a combline resonator resulted in of 1300374 at 2.502.39 GHz with a limited tuning ratio (TR) of 1.05:1 [10], [11]. A surface ring gap resonator structure loaded with solid-state varactors reports a of 16040 with limited tuning range of 3.12.6 GHz [12], [13]. The authors of this paper demonstrated a continuous octave tuning substrate-integrated coaxial resonator with of 86206 and tuning range of 0.51.2 GHz using solid-state varactors and of up to 240 at 6.6 GHz using RF-MEMS varactors [14]. It is the intention of this paper to further investigate this surface ring gap combline cavity. Compared to previous works [12][14], this paper presents an in-depth theoretical analysis of the resonant frequency, TR, and . Effects of the surface gap capacitance and surface inductance are considered to show the compromise between TR and . A design method is suggested to maximize TR and optimize . This method demonstrates tunable resonators and lters with higher and tuning range than the state-of-the-art with similar technologies. A two-pole lter with tuning range of 0.51.1 GHz and measured insertion loss of 1.67 dB at 1.1 GHz is demonstrated. This two-pole lter maintains a constant 3-dB fractional bandwidth (FBW) of 4%. A three-pole lter with tuning range of 0.581.22 GHz with a constant 3-dB FBW of 4 0.2% and measured insertion loss of 2.05 dB at 1.22 GHz is also demonstrated. The measured input third-order inter-modulation is between 1730 dBm over the frequency range for the two-pole lter. II. SURFACE GAP COMBLINE RESONATOR Fig. 1(a) and (b) schematically compares the cross section of a vertical gap resonator with the cross section of the proposed surface ring gap resonator. In the surface ring gap design, the center post is extended to short the bottom and top of the cavity and a ring gap is created on the surface to isolate the center

0018-9480 2013 IEEE

4354

IEEE TRANSACTIONS ON MICROWAVE THEORY AND TECHNIQUES, VOL. 61, NO. 12, DECEMBER 2013

Fig. 1. (a) Vertical gap resonator and (b) surface ring gap resonator. Both have the same lumped equivalent circuit model.

post from the rest of the cavitys ceiling. To tune the resonant frequency, tuning elements are placed across the gap to vary the capacitance instead of having to change the physical gap. The gures show that the surface ring gap resonator still has the same lumped equivalent circuit model as the vertical gap resonator. The surface ring gap resonator in Fig. 1(b) resembles a shorted coaxial transmission line and is commonly referred as a combline resonator. This design has several advantages compared to the vertical gap. First, the structure is not limited to a particular tuning technology, but allows for various types of tuning components, such as RF-MEMS or solid-state varactors. Second, precise assembly is not needed since the tuning components are surface mounted. This makes high-volume-manufacturing possible and the resonant frequency and tuning range independent of fabrication and assembly tolerance. Third, this structure is easily implemented on a low-cost printed circuit board (PCB), which can be integrated with other RF components. However, as shown later, this type of tuning may not yield in the order of 1000 if solid-state varactors are used as the tuning components. Consequently, solid-state varactors are appropriate for applications that need in the order of 100500 without the added complexity of vertically aligned tuners. III. THEORETICAL ANALYSIS AND DESIGN A. Resonant Frequency Fig. 2(a) shows the dimensions of the resonator where is the height of the cavity, is the radius of the inner post, and is the radius of the outer conductor. In Fig. 2(b), the resonator is approximated with lumped elements where is the effective inductance of the cavity and capacitance in series with resistance models a varactor. The packaging and assembly, depending on the type of varactor (solid-state, RF-MEMS, etc.), can introduce other parasitics. To keep the model generic, package parasitics are excluded and the varactor is assumed to operate well below its self resonance so the impedance resembles a series RC circuit. The conductors of the cavity, such as the inner post, outer wall, and the top and bottom walls, have resistances of , , and , respectively. A simplied model for the resonator is shown in Fig. 2(c), where and are the equivalent capacitance and equivalent resistance of all varactors and is the sum of all other resistors. For an LC resonator, the angular resonant frequency is given as . Thus, from Fig. 2(c), and are needed to nd the resonant frequency of the resonator. The capacitance

Fig. 2. (a) Cavity resonator with dimensions. (b) Cross-sectional lumped model. (c) Simplied LC resonator.

is extracted from the equivalent impedance of the varactors , which is approximated as (1) (2) (3) is the number of varactors placed in parallel on the where ring gap. Since the resonator resembles a shorted coaxial transmission line, is extracted from the input impedance of a shorted transmission line (4) where (5) where is the characteristic impedance of the coaxial transmission line, is the dielectric constant of the material inside the cavity, and is the phase constant. The above equations assume very low cavity losses, which is typically the case for the dimensions considered in this work. It has been shown that near resonance, the reactance of is inductive [15] and the effective inductance is approximated as (6) where is the speed of light. Assuming low losses, resonant frequency can be solved numerically in (7)

ANAND et al.: THEORY AND DESIGN OF OCTAVE TUNABLE FILTERS WITH LUMPED TUNING ELEMENTS

4355

TABLE I CAVITY DIMENSIONS USED IN HFSS SIMULATION

Fig. 4. Theoretical capacitance is tuned,

versus frequency for three xed capacitances. As the increases with frequency.

Fig. 3. Simulated and theoretical resonant frequency versus for the resonator. The difference between theory and simulation (up to 9%) is addressed in Fig. 10. Fig. 5. Simulated and theoretical versus frequency for the resonator. The . conductors of the cavity are simulated as perfect conductors to extract

Equation (6) shows that depends on , but the change is is small (up to 3% for the cavity parameters and frequency range considered in this paper). Thus, neglecting the variation in , the change in or TR, is determined by the tuning factor, , of the varactor (8)

Substituting (7) into (9) and neglecting the variation in ratio as

gives

(10)

Ansys HFSS [16] is used to compare (7) with full-wave simulations with the cavity parameters given in Table I. The surface gap radius and surface gap width is explained in Section III-C. In Fig. 3, resonant frequencies from (7) and HFSS simulations are plotted versus . Resonant frequencies from theory is between 3%9% higher primarily because (7) does not include the effects of the gap capacitance and surface inductance (discussed later in Sections III-C and III-D). B. Quality Factor The resonators is dependent on the quality factor of the varactors and the quality factor of the cavity . Since is the resistance of the varactors, the series combination of and [see Fig. 2(c)] gives (9)

versus frequency for three different capacitances In Fig. 4, are plotted. For a xed capacitance, decreases as frequency increases. However, as the varactor is tuned to a lower capacitance, larger , (10) and (8) show that both and frequency will increase. Fig. 4 also shows versus frequency as is tuned from 2.67 to 0.63 pF: increases from 124 at 0.6 GHz to 248 at 1.2 GHz. In Fig. 5, (9) is compared to results from HFSS simulations. To extract from simulation, the conductors (post and walls of the cavity) are simulated as perfect conductors so that is approximately . The reason for higher in simulation is investigated in Section III-C. Since is the resistance of the cavity conductors, the series combination of with [see Fig. 2(c)] gives

(11)

4356

IEEE TRANSACTIONS ON MICROWAVE THEORY AND TECHNIQUES, VOL. 61, NO. 12, DECEMBER 2013

Fig. 6. Simulated and theoretical versus frequency for the resonator. In from HFSS, the resistance of the varactors are set to zero. order to extract

Fig. 7. Theoretical .

with respect to

. Optimal

is achieved at

which is essentially the quality factor of the cavity with lossless varactors. The sum of , , and in Fig. 2(b) is . From [17], (12) is the surface resistance. Surface current density in which on the top and bottom of the cavity varies radially from the outer conductor to the inner post. Thus, integrating in the radial direction gives (13) Summing (12) and (13) gives , (14) Substitution (6) and (14) into (11) gives 1 to 25 at four xed frequencies: 0.1, 0.5, 1, and 2 GHz. The optimized ratio is evident at all frequencies. Using (9) and (15) and the series LC resonator model in Fig. 2(c), of the resonator is approximated as (16) (15) is simulated in HFSS with cavity parameters The resonators from Table I. Fig. 8 shows that simulation results are larger than theory because was larger in simulation previously (Fig. 5). In this case, is much larger than due to the relatively large (Figs. 6 and 5) so . In Fig. 8, is initially limited by at lower frequencies (larger ), but starts to exceed at higher frequencies (smaller ). The limitations of is analyzed in more detail in Section III-C. C. Surface Ring Gap Capacitance In Fig. 3, the theoretical resonant frequency based on the lumped model in Fig. 2(c) was up to 9% higher than the simulated results. Part of this difference is due to the surface ring gap capacitance . As shown in Fig. 9(a), this capacitance is dependent on the ring radius , gap width , depth of the gap , and the dielectric constant . Since is in parallel with , from (7) is modied as (17)
Fig. 8. Simulated and theoretical is much larger than , versus frequency for the resonator. Since .

HFSS is used to compare (15) with simulation with cavity parameters from Table I. In order to get only, in simulation is set to zero and is varied from 2.67 to 0.63 pF. Fig. 6 shows a plot of theoretical from (15) compared to simulated . The simulated is about 3%4% lower than the theoretical one. Fig. 6 also shows that radiation from the ring gap has a negligible effect by comparing simulation results from a resonator with a shielding cap covering the ring gaps and varactors on the surface. Note that the ring gap width is about 100200 times smaller than the wavelength range considered in simulation. Part of the 3%4% loss in simulation comes from the resistance associated with the ring gap, which is considered in Section III-C and Fig. 12. The of a coaxial resonator is optimized when the ratio is 3.6 [18]. This optimum ratio is again veried by plotting (15) in Fig. 7. The outer radius and height of a copper air cavity are xed to mm and mm, while is varied from

ANAND et al.: THEORY AND DESIGN OF OCTAVE TUNABLE FILTERS WITH LUMPED TUNING ELEMENTS

4357

in (23) and how to minimize the variation is considered in Section III-D. Since is in parallel with , the total electromagnetic energy will be distributed between and . Previously, if the conductors were lossless, then . However, if is large enough, then a signicant portion of the electromagnetic energy will be stored in . Since the of is expected to be higher than (resistance associated with is small compared to ), should exceed . This is, however, accomplished at the cost of reduced tuning range. The equivalent impedance, in parallel with , is given by (19) where (20) and is the resistance associated with shown in Fig. 9(b) and is initially neglected . of this impedance then becomes (21)
Fig. 9. Capacitance exists in parallel with the tuning components (varacis dependent on the radius of the ring , width of the gap , depth tors). (a) due to the of the gap , and the dielectric constant . (b) There is resistance wall plates (dark color region) around the surface ring gap.

and

from (16) is modied as (22)

Fig. 10. Simulated and theoretical resonant frequency versus varactor capaciis included. The difference in frequency is reduced tance when the effects of to within 1%.

where is given by (23). The value of simulated at two different values ( cavity structure from

can be extracted from and ) for a xed

(18) where is the resonant frequency for and is the resonant frequency for . is extracted for simulated data from Fig. 3 to be pF. Compared to Fig. 3, when pF is included in (17), Fig. 10 shows the difference in resonant frequency is within 1% of simulation. The extraction of here assumes that is constant as or varies. The variation

is small and negligible, then (22) reduces to (16). Note that if A cavity with the parameters from Table I is simulated, expect with , so that the effects of are more signicant compared to . Fig. 11(a) plots (22) versus frequency when varies from 1 to 11 mm to show the effects of on and TR. As becomes larger, TR decreases and increases. Also plotted in the gure is theoretical at pF (maximum ). When is larger, maximum for each curve is not limited by maximum theoretical (dotted line) and far exceeds , but as decreases, maximum for each curve approaches maximum . In the case of ideal varactors or when , such as when no varactors are mounted in a static surface ring gap resonator [19], if . Fig. 9(b) shows that depends on the two thin layers of metal wall (dark regions labeled as ring gap resistance in gure) that border the surface ring gap. These two thin walls are simulated in HFSS as a perfect electric conductors (PECs) or , while the rest of the cavity surface is still copper and with parameters from Table I. Fig. 12 compares of this PEC walls simulation with the previous case in Fig. 6 when all of the cavity surface was copper (no PEC or ). As expected, Fig. 12 shows that with is higher than with . Before, in Fig. 6, was 3%4% lower than theoretical (15), but with , simulation is within 1% of theoretical . In Fig. 13, resistance of the varactors is included in HFSS simulation to compare when and . Since dominates both and , Fig. 13 shows that effects of is negligible for the parameters in Table I. Thus, the previous assumption of made in Fig. 11 is

4358

IEEE TRANSACTIONS ON MICROWAVE THEORY AND TECHNIQUES, VOL. 61, NO. 12, DECEMBER 2013

Fig. 13. Effects of is negligible on for the cavity parameters considered . Additionally, theory and simulation are in Table I since is included and is limited by . within 1% since the effects of reduces to if .

Fig. 11. (a) Theoretical (solid lines), simulated (dashed lines), and (dotted line) versus frequency as is changed. theoretical maximum from simulation versus ring radius with cavity dimensions (b) Extracted . from Table I, except with

Fig. 14. Plot of frequency versus capacitance at various dielectric constants. increases and reduces TR. As the dielectric constant increases,

and when , a reduction of about 11% in TR. Based on (18), pF for , pF for , and pF for . D. Surface Inductance
Fig. 12. is higher when the ring gap resistance is simulated with PEC compared to when ring gap resistance is simulated with copper . Simulation with is within 1% of theoretical and with is within 3%4% of theoretical . simulation

valid. Also, compared previously to Fig. 8, theory and simulation are in better agreement, within 1%, now that the effects of are included in theory. Moreover, this gure shows that is limited by , which reduces to only if is negligible. In Fig. 11(a), the ring radius was changed to vary . Fig. 11(b) shows that is strongly dependent on . increases with increasing (increasing circumference) since the surface area of the capacitance increases. Even though has an air gap, the dielectric material of the cavity will partly change due to fringing electromagnetic elds. Fig. 14 conrms this by showing the effects of various on resonant frequency and TR. The results in the gure are based on HFSS simulation with cavity parameters from Table I as ranges from 2.67 to 0.63 pF. Simulation shows that when

The current that ows on the top surface from the inner post to the outer cavity wall results in a surface inductance . This current ow depends on the ratio of and , and on . If current ow is concentric, will be minimal. The arrangement of the varactors can alter the current path, thus changing . 1) Consider the case of . When is large and the varactors are spatially distributed, current ows equally in all directions through the varactors and current ow is concentric. Fig. 15(a) illustrates this with . Fig. 15(b) shows current ow when is small. Current is forced to ow through the varactor resulting in a longer current path. Thus, increases as decreases, which is veried in Table II. In simulation, as is varied, is also changed to keep constant for different values. 2) Consider the case of . Since most of the current is distributed in the surface ring gap, current ow is concentric regardless of . Fig. 15(c) and (d) shows that current ow is concentric when or when . 3) When and are comparable, then current is distributed between the varactors and surface ring gap. If

ANAND et al.: THEORY AND DESIGN OF OCTAVE TUNABLE FILTERS WITH LUMPED TUNING ELEMENTS

4359

Though not discussed earlier, the effects of is included in all the theoretical analysis and plots in Section III-C. Even though is dependent on various parameters, based on the above discussion, is always minimal as long as is sufciently large. The effects of is further reduced by increasing . As mentioned previously and shown in (6), varies slightly with . For , the variation in is less than 0.4% and for , the variation in is less than 3.5% over the frequency range with parameters from Table I. E. Design Methodology Based on the theoretical analysis, the compromise in and tuning range is highly dependent on . With the aid of simulation software, appropriate is designed by changing the surface ring gap dimensions or the dielectric constant (Figs. 11(b) and 14). For example, a resonator is designed for high , 16040, but with a limited TR of 1.2 in [12]. This is analogous to the curve in Fig. 11 with pF, which has a TR of 1.27. Alternatively, to design a resonator with maximum tuning range, has to be minimized so that it is small relative to . To maximize , and can be increased up to the size limitations and can be set by the optimal ratio. From (7) and (2), an appropriate number of varactors can then be mounted to lower the frequency to the desired range, resulting in an optimized for a maximum TR design. This design, however, may result in that is too large. For example, a cavity with parameters from Table I, except with mm for optimal ratio, requires varactors on the surface ring gap to tune from 0.6 to 1.2 GHz. If fewer varactors are to be used, then the cavity dimensions may be changed to get the desired frequency range. The height and radius of the cavity are typically limited by the constraints on maximum device size. Varactors with larger capacitances are not recommended since the increase in capacitance comes at the cost of reduced , which directly limits . The reduction in tuning range and cost of high dielectric constant material restricts as a exible parameter for design. The optimum ratio may need to be sacriced to get the desired frequency range for a design with less than 20 varactors. From (7) and (6), the resonant frequency depends on the ratio. From Fig. 16, increasing from the optimum ratio of 3.6 to 25 can decrease the frequency from to . The previous example where varactors were needed to tune from 0.6 to 1.2 GHz can alternatively be designed by increasing the ratio from 3.6 to 24. Fig. 17 shows a reduction of 5% to 7% in for compared to . IV. EXPERIMENTAL VALIDATION A. Resonator (23) resulting in (24) A SIW combline resonator with solid-state varactors as tuners is designed and fabricated to validate the theoretical derivations. Fig. 18(a) shows the designed resonator. In this design, metallic vias are inserted in a PCB substrate to create the outer wall of the cavity. The diameter of the vias ( 1 mm) and the spacing between vias ( 2.5 mm) are designed with the recommendations given in [25] to keep losses minimal. Another center metallic via

Fig. 15. HFSS simulation showing surface currents on the varactor loaded is small. surface. (a) Current is equally distributed in eight varactors and (b) Current ows through just one varactor, which increases current path and . (c) When is dominant, current ows concentrically through regardor (d) . less of

TABLE II VERSUS

is large and the varactors are equally spaced on the surface ring gap, current ow is concentric. For small , current ow depends on ring radius and . As gets larger, gets larger and leads to a more concentric current ow, decreasing . Also, however, as gets larger, the path of the current ow through the varactor gets longer, increasing . Additionally, as is tuned, changes and the impedance ratio of and changes, making vary with . Thus, when and are comparable and is small, is highly dependent on , ratio, and . The surface inductance with is included in (17) by replacing

4360

IEEE TRANSACTIONS ON MICROWAVE THEORY AND TECHNIQUES, VOL. 61, NO. 12, DECEMBER 2013

TABLE III CAVITY DIMENSIONS USED IN FABRICATION

Fig. 16. Plot of ratio versus normalized frequency , where is the ratio. The resonant frequency decreases resonant frequency for the optimum as is changed to 25. by

Fig. 17. Resonator can be designed with with optimal ratio equal can be compromised to design to 3.6. Alternatively, the optimal a resonator with the same frequency range where fewer varactors are needed with reduced .

Fig. 19. Comparison of measured, simulated, theoretical (with and without and ) versus frequency.

the varactors. The two rings also allows for back-to-back varactor placement for improved linearity [20]. Since the additional ring is in series with the original ring, (2) and (3) are modied to (25) (26) and thus the same frequency range, In order to keep the same needs to be doubled for a two-ring gap design compared to a one-ring gap design. The rst part of the theoretical analysis was based on the lumped model presented in Fig. 2(c). In order for (7) and (16) had to be negligible. A SIW resonator to be valid, and is fabricated on a Rogers TMM3 substrate with the dimensions , the effects of and given in Table III. By choosing is minimized. Fig. 18(c) and (d) shows the fabricated resonator and a close-up of the arranged varactors with the isolated bias point for the varactors (not yet soldered on). A 10-M resistor is included in the dc bias line to reduce RF losses. An additional lumped resistor was used during the measurement [see Fig. 24(c)] to further reduce RF losses. Fig. 19 compares is plotted versus frequency as measured, HFSS, and theory. the varactors are biased from 0 to 30 V. This gure validates

Fig. 18. (a) Designed resonators top view and (b) bottom view showing Skyworks CPW feed lines. (c) Rogers TMM3 substrate with vias and SMV1405 varactors. (d) Close-up view of the two surface ring gaps showing the varactors (not soldered on yet) and the dc bias point.

with radius of shorts the bottom and top ceiling of the cavity. Fig. 18(b) shows the coplanar-waveguide (CPW) feed lines for this design with length . Since solid-state varactors are used, a structure with two ring gaps is needed to create a bias point for

ANAND et al.: THEORY AND DESIGN OF OCTAVE TUNABLE FILTERS WITH LUMPED TUNING ELEMENTS

4361

Fig. 20. Measured maximum theoretical and is limited by exceeds .

versus frequency for to along with (solid line) at pF. For larger , ; as decreases, becomes comparable to and Fig. 21. (a) Two resonators with inter-resonator coupling and weak external and post-to-post distance coupling. Both the width of the coupling iris can be changed to change . (b) One-port resonator structure used to extract by varying the coupling angle when the feed line and coupling gap are constant. size

SUMMARY

OF

TABLE IV MEASURED PERFORMANCE OF FABRICATED RESONATORS WITH VARYING

that the simplied lumped model from Fig. 2(c), without and , is reasonable in predicting and . The extracted pF and nH from measured data are then included in theory and plotted in the same gure. The slope of the curve with and is even closer to measured and simulation data. The measured TR is about 2.1, simulated TR is about 1.94, and theoretical TR including the effects of is 2.01. Measured is lower due to losses in dielectric material (loss tangent is 0.002), fabrication, and assembly. Another set of resonators, with similar dimensions as Table III, were fabricated with ranging from 1 to 20 to show the effects of on and TR. To compare the measured with , Fig. 20 includes a plot of theoretical at (maximum ). In the gure, is limited by when is large, but exceeds for or less: has become signicant compared to . In fact, in the extreme case of , is 0.315 pF, which is less than pF. When the number of varactors decreases, ratio increases and TR decreases. Table IV summarizes the results of frequency range and TR. Table IV shows that the TR decreases from 2.17 when to 1.78 when . Table IV also summarizes the results of and ratio at various s. When , and thus, dominates, (10) states that , which is close to the measured ratio of 2.38. It should be noted that of the tunable resonator depends heavily on the tuner technology. In this case, solid-state varactors have higher at lower frequencies. RF-MEMS tuners can be used to get high at higher frequencies. For example, the authors have demonstrated a surface ring gap resonator with of 240 at 6.6 GHz [14].

Fig. 22. Simulated design curves for: (a) versus coupling angle .

versus post distance

and (b)

B. Filter To demonstrate the application of the resonator, a two- and three-pole tunable lter are designed and fabricated with the resonator parameters from Table III. A similar procedure as outlined in [21] is used to get the external quality factor and inter-resonator coupling from HFSS simulation. HFSS simulation is used to simulate various values of by changing the width of the coupling iris and the post distance while the feed line is kept far from the post [see Fig. 21(a)]. Fig. 22(a) shows versus when mm is simulated around 1.1 GHz. A resonator with one port shown in Fig. 21(b) is used to simulated . As seen in the gure, another opening is created at the end of CPW feed lines with a coupling gap size and coupling angle of . Fig. 22(b) shows versus

4362

IEEE TRANSACTIONS ON MICROWAVE THEORY AND TECHNIQUES, VOL. 61, NO. 12, DECEMBER 2013

Fig. 23. (a) Designed and (b) fabricated octave tunable two-pole lter with and . measured (c)

when mm and 1.1 GHz. The required values of lter response are

cm is simulated around and for a two-pole Butterworth (27) (28)

Fig. 24. (a) Designed and (b) fabricated octave tunable three-pole lter with and . measured (c)

where %, , , and . Fig. 23(a) and (b) shows the designed and fabricated two-pole lter with cm, , and mm in a 5 3 0.5 cm volume. The lter tunes from 0.5 to 1.1 GHz with measured insertion loss from 4.46 to 1.67 dB and measured return loss from 8.3 to 27.8 dB [see Fig. 23(c)]. A 4 0.1% FBW is maintained through out the tuning range. An Agilent PNA-X is used to measure the input third-order intermodulation intercept point (IIP3) with the two tones separated by 100 kHz. The IIP3 ranged from 17 to 30 dBm when the varactors were biased at 030 V. The required values of and for a three-pole Butterworth lter response are (29) (30) %, , , and . Fig. 24(a) where and (b) shows the designed and fabricated three-pole lter with cm, , and mm in a 5 3 0.5 cm volume. This lter has a tuning range of 0.581.22 GHz with a constant 3-dB FBW of 4 0.2%.

Fig. 24(c) shows that the insertion loss varies from 6.2 dB at 0.58 GHz to 2.05 dB at 1.22 GHz. Since both the two- and three-pole lters were designed at 1.1 GHz, the simulated and values are best matched to the desired values at 1.1 GHz. As seen in Figs. 23(c) and 24(c), the return loss degrades as the lter is tuned to lower frequencies. In [26], a varactor is mounted on the CPW feed line to change and another varactor mounted between the two resonators to change . Though [26] had a different resonator technology, the same concept of tunable and can be implemented in the presented lter design without added fabrication complexity. This allows and to be tuned as the frequency response of the lter is tuned to improve return loss away from 1.1 GHz. However, adding more varactors will introduce more insertion loss. Table V compares some of the recent tunable lters. As mentioned earlier and summarized in Table V, the complexity of the presented surface ring gap resonator (and other planar varactor tuned lters) is reduced compared to some of the vertically aligned piezoelectric or RF-MEMS-based lters. Moreover, to the best of authors knowledge, the presented surface ring gap resonator exceeds the TR of all other solid-state varactor based

ANAND et al.: THEORY AND DESIGN OF OCTAVE TUNABLE FILTERS WITH LUMPED TUNING ELEMENTS

4363

TABLE V COMPARISON OF TUNABLE FILTERS

lters for a given varactor capacitance ratio exceeds [see (8)] for this work. V. CONCLUSION

: measured TR

This paper has presented the modeling and design of an octave tunable combline lter using surface mount lumped tuning elements. Detailed theoretical analysis on the tuning range and of these resonators/lters are presented. A systematic design methodology is also proposed. To validate the theory and the design procedure, various tunable resonators were fabricated with up to 214 at 1.1 GHz. Additionally, a two-pole tunable lter with tuning range of 0.51.1 GHz at a constant FBW of 4 0.1% and measured insertion loss of 1.67 dB at 1.1 GHz, and a three-pole tunable lter with tuning range of 0.581.22 GHz at a constant FBW of 4 0.2% and measured insertion loss of 2.05 dB at 1.22 GHz have been demonstrated. REFERENCES
[1] M. A. El-Tanani and G. M. Rebeiz, High-performance 1.52.5-GHz RF-MEMS tunable lters for wireless applications, IEEE Trans. Microw. Theory Techn., vol. 58, no. 11, pp. 16291637, Jun. 2010. [2] C.-C. Cheng and G. M. Rebeiz, High- 46-GHz suspended stripline RF MEMS tunable lter with bandwidth control, IEEE Trans. Microw. Theory Techn., vol. 59, no. 10, pp. 24692476, Oct. 2011. [3] S.-J. Park, K.-Y. Lee, and G. M. Rebeiz, Low-loss 5.155.70-GHz RF MEMS switchable lter for wireless LAN applications, IEEE Trans. Microw. Theory Techn., vol. 54, no. 11, pp. 39313939, Nov. 2006. [4] H. Joshi et al., Highly loaded evanescent cavities for widely tunable high- lters, in IEEE MTT-S Int. Microw. Symp. Dig., Jun. 2007, pp. 21332136. [5] S. J. Park et al., High- RF-MEMS 46 GHz tunable evanescentmode cavity lter, IEEE Trans. Microw. Theory Techn., vol. 58, no. 2, pp. 381389, Feb. 2010. [6] X. Liu et al., High- tunable microwave cavity resonators and lters using SOI-based RF MEMS tuners, J. Microelectromech. Syst., vol. 19, no. 4, pp. 774784, Aug. 2010. [7] X. Liu et al., A 3.46.2 GHz continuously tunable electrostatic MEMS resonator with quality factor of 460530, in IEEE MTT-S Int. Microw. Symp. Dig., Jun. 2009, pp. 11491152. [8] M. S. Arif and D. Peroulis, A 6 to 24 GHz continuously tunable, microfabricated, high- cavity resonator with electrostatic MEMS actuation, in IEEE MTT-S Int. Microw. Symp. Dig., Jun. 2012, pp. 13. [9] V. Sekar, M. Armendariz, and K. Entesari, A 1.2-1.6-GHz substrateintegrated-waveguide RF MEMS tunable lter, IEEE Trans. Microw. Theory Techn., vol. 59, no. 4, pp. 866876, Feb. 2011. [10] S. Fouladi et al., Combline tunable bandpass lter using RF-MEMS switched capacitor bank, in IEEE MTT-S Int. Microw. Symp. Dig., Jun. 2012.

[11] S. Fouladi et al., High- narrowband tunable combline bandpas lters using MEMS capacitor banks and piezomotors, IEEE Trans. Microw. Theory Techn., vol. 61, no. 1, pp. 393402, Jan. 2013. [12] S. Sirci et al., Varactor-loaded continuously tunable SIW resonator for recongurable lter design, in Proc. 41th. Eur. Microw. Conf., Oct. 2011, pp. 436439. [13] S. Sirci et al., Analog tuning of compact varactor-loaded combline lters in substrate integrated waveguide, in Proc. 42th. Eur. Microw. Conf., Oct. 2012, pp. 257260. [14] A. Anand et al., A novel highoctave-tunable resonator with lumped tuning elements, in IEEE MTT-S Int. Microw. Symp. Dig., Jun. 2013. [15] G. F. Craven and R. F. Skedd, Evanescent Mode Microwave Components. Boston, MA, USA: Artech House, 1987. [16] High Frequency Structure Simulator 15.0 (HFSS 15.0). Ansoft Corporation, Canonsburg, PA, USA, 2013. [Online]. Available: http://www. ansys.com/Products [17] F. T. Ulaby, Transmission lines, in Fundamentals of Applied Electromagnetics. Upper Saddle River, NJ, USA: Pearson Educ. Inc., 2004, ch. 4, pp. 4142, media ed.. [18] P. Vizmuller, Useful formulas, in RF Design Guide: Systems, Circuits and Equations. Norwood, MA, USA: Artech House, 1995, ch. 2, p. 238. [19] J. Martinez et al., Capacitively loaded resonator for compact substrate integrated waveguide lters, in Proc. 40th. Eur. Microw. Conf., Sep. 2010, pp. 192195. [20] M. A. El-Tanani and G. M. Rebeiz, Two-pole two-zero tunable lter with improved linearity, IEEE Trans. Microw. Theory Techn., vol. 57, no. 4, pp. 830839, Apr. 2009. [21] D. G. Swanson, Narrow-band microwave lter design, IEEE Microw. Mag., vol. 8, no. 5, pp. 105114, Oct. 2007. [22] X.-G. Wang, Y. Cho, and S. Yun, A tunable combline bandpass lter loaded with series resonator, IEEE Trans. Microw. Theory Techn., vol. 60, no. 6, pp. 15691576, Jun. 2012. [23] J. Long et al., A tunable microstrip bandpass lter with two independently adjustable transmission zeros, IEEE Microw. Wireless Compon. Lett., vol. 21, no. 2, pp. 7476, Feb. 2011. [24] A. R. Brown and G. M. Rebeiz, A varactor-tuned RF lter, IEEE Trans. Microw. Theory Techn., vol. 48, no. 7, pp. 11571160, Jul. 2000. [25] K. Wu, D. Deslandes, and Y. Cassivi, The substrate integrated circuitsA new concept for high-frequency electronics and optoelectronics, Telecommun. Modern Satellite, Cable, Broadcast. Service, vol. 1, pp. P-IIIP-X, Oct. 2003. [26] H. Joshi et al., High- fully recongurable tunable bandpass lter, IEEE Trans. Microw. Theory Techn., vol. 57, no. 12, pp. 35253533, Dec. 2009.

Akash Anand (S12) received the B.S. degree in electrical engineering from the University of California at Davis, Davis, CA, USA, in 2009, and is currently working toward the Ph.D. degree at University of California at Davis. His research interest includes RF/microwave tunable lters, RF and analog integrated circuit (IC) design, and recongurable RF systems.

Joshua Small (S11M12) received the Bachelors degree in electrical engineering from Morgan State University, Baltimore, MD, USA, in 2005, and the Ph.D. degree from Purdue University, West Lafayette, IN, USA, in 2012. He is currently a Postdoctoral Research Engineer with the Department of Electrical and Computer Engineering, University of California at Davis, Davis, CA, USA. His research interests includes microstructures for investigating novel highrecongurable radio front ends.

4364

IEEE TRANSACTIONS ON MICROWAVE THEORY AND TECHNIQUES, VOL. 61, NO. 12, DECEMBER 2013

Dimitrios Peroulis (S99M04) received the Ph.D. degree in electrical engineering from The University of Michigan at Ann Arbor, Ann Arbor, MI, USA, in 2003. Since August 2003, he has been with Purdue University, West Lafayette, IN, USA, where he is currently a Professor leading a group of graduate students on a variety of research projects in the areas of RF-MEMS, sensing, and power harvesting applications, as well as RF identication (RFID) sensors for condition monitoring of sensitive equipment. He has been a Principle Investigator (PI) or a co-PI in numerous projects funded by government agencies and industry in these areas. He has been a key contribRF MEMS tunable lters utor on developing very high quality in mobile form factors. Furthermore, he has been investigating failure modes of RF MEMS and MEMS sensors through the Defense Advanced Research Projects Agency (DARPA) MEMS/Nanoelectromechanical systems (NEMS) Science and Technology (S&T) Fundamentals Program, Phases I and II) and the Center for the Prediction of Reliability, Integrity and Survivability of Microsystems (PRISM) funded by the National Nuclear Security Administration. He has coauthored over 200 journal and conference papers. Dr. Peroulis was the recipient of the National Science Foundation (NSF) CAREER Award (2008) and the IEEE Ultrasonics, Ferroelectrics, and Frequency Control Society (Ferroelectrics section) Outstanding Paper Award (2012). His students have been the recipients of numerous Student Paper Awards and other student research-based scholarships. He is a Purdue University Faculty Scholar and has also been the recipient of ten teaching awards including the 2010 HKN C. Holmes MacDonald Outstanding Teaching Award and the 2010 Charles B. Murphy award, which is Purdue Universitys highest undergraduate teaching honor.

Dr. Xiaoguang Leo Liu (S07M10) received the B.S. degree from Zhejiang University, Zhejiang, China, in 2004, and the Ph.D. degree from Purdue University, West Lafayette, IN, USA, in 2010. He is currently an Assistant Professor with the Department of Electrical and Computer Engineering, University of California at Davis, Davis, CA, USA. He has authored or coauthored over 30 papers in peer-reviewed journals and conferences. His research interests include RF-MEMS and other recongurable RF/microwave components, software-dened radios, and terahertz components and systems. Dr. Liu was the recipient (as a student) of the IEEE Antenna and Propagation Society Graduate Research Fellowship in 2009. He was selected as a University of Californiat at Davis Hellman Fellow for the 20132014 academic year.