Pseudo-Differential Filter Design Using Novel Adjustable Floating Inductance Simulator with Electronically Controllable Current Conveyors

ELEKTRONIKA IR ELEKTROTECHNIKA,ISSN 1392-1215,VOL.23,NO.2,2017 1Abs ac —This pape p esen s new loa ing induc ance simula o based on h ee elec onically con ollable cu en con eyo s (ECCIIs) and di e en ial ol age bu e (DVB). The induc ance simula o o e s simple elec onic con ol o induc ance alue as well as simple con ol o losses by change o pa ame e s o ac i e elemen s. Based on his loa ing induc ance simula o , example o econnec ion-less econ igu able pseudo- di e en ial i s -o de il e , allowing change o ans e unc ion be ween high-pass and in e ing all-pass esponse, is discussed and s udied. Mo eo e , he p oposed s uc u e can be easily ex ended o a pseudo-di e en ial second-o de band-bass il e . Beha iou o p esen ed solu ions is e i ied by PSpice simula ions and also expe imen ally in equency band up o 10 MHz. Ob ained esul s con i med expec ed ea u es. Index Te ms—Di e en ial ac i e il e ; di e en ial mode; pseudo-di e en ial-mode; elec onic con ol; induc ance simula o ; econ igu abili y. I. INTRODUCTION In gene al, ully-di e en ial (symme ical) ci cui s ha e many applica ions due o speci ic ea u es hey o e , i.e. a enua ion o common-mode noise as well as be e ejec ion o signals incoming om powe supply sou ces and possibly also be e dynamics in compa ison o single-ended ci cui ies [1]–[4]. Howe e , inc eased complexi y o ci cui y is cos o hese bene i s. Elec onic con ollabili y o pa ame e s o analogue and mixed-mode ci cui s is use ul in single-ended as well as in case o di e en ial-mode ope a ion. The e o e, ci cui solu ions allowing hese ea u es a e subjec s o in e es o many esea che s [4]. Digi al con ol o impo an pa ame e s o he di e en ial- Manusc ip ecei ed 23 Decembe , 2016; accep ed 18 Ma ch, 2017. Resea ch desc ibed in his pape was inanced by Czech Minis y o Educa ion in ame o Na ional Sus ainabili y P og am unde g an LO1401. Fo esea ch, in as uc u e o he SIX Cen e was used. Resea ch desc ibed in he pape was suppo ed by Czech Science Founda ion p ojec s unde No. 16-11460Y. il e was also in es iga ed in he pas [5]. Di ision o di e en ial sys ems ( o example see [1]–[8]) o uly-di e en ial and so-called pseudo-di e en ial-mode o ope a ion, is pe ec ly explained in [8] which b ings new insigh in o his esea ch opic. No e ha complexi y o pseudo-di e en ial-mode ci cui s is signi ican ly educed in compa ison o ully-di e en ial-mode solu ions due o pa ially g ounded subpa s o di e en ial sys ems, i.e. inpu and ou pu po s a e di e en ial whe eas hei inne ci cui y consis s o g ounded (analog g ound) passi e componen s and e minals. Many om p e iously epo ed pseudo- di e en ial ci cui solu ions we e ocused on biquad a ic (second-o de ) ac i e il e s, [6]–[9] o ins ance. Elec onic con ollabili y [6] as well as mul i- unc ionali y (p o iding se e al ans e esponses) [7] we e also s udied in he pas . In his pape we a e in oducing a no el solu ion o loa ing induc ance simula o allowing ea u es no epo ed in p e ious wo ks. These ea u es a e bene icial especially in elec onic econ igu a ion o ans e unc ion and elec onic con ol o pseudo-di e en ial-mode il e pa ame e s. He e p oposed il e ing applica ions can be di ided o wo g oups: (i) i s -o de econ igu able il e , (ii) second-o de band- pass il e . Compa ison o selec ed solu ions o i s -o de ac i e il e s in single-ended, ully-di e en ial, o pseudo- di e en ial o m [10]–[14] is a ailable in Table I. In gene al, many o hese ha e mul i unc ional ea u es, howe e , ea u es such as econnec ion-less econ igu abili y (possible o ob ain a ious ans e esponses wi hou change o inpu /ou pu e minal) we e s udied only in single-ended il e ing solu ions (see o example [15]). To he bes o au ho s’ knowledge, ci cui s allowing econ igu a ion o ans e esponses, i.e. ce ain mul i- unc ionali y o he il e be ween a leas wo di e en ypes o ully-di e en ial o pseudo-di e en ial il e s we e no epo ed in he open li e a u e ye . He e p esen ed s uc u e o loa ing induc ance simula o allows such elec onic econ igu a ion, i i s Pseudo-Di e en ial Fil e Design Using No el Adjus able Floa ing Induc ance Simula o wi h Elec onically Con ollable Cu en Con eyo s Roman So ne 1,2, No be He encsa 2, Jan Je abek2, Aslihan Ka ci1,2, Ja osla Ko on2, Tomas Dos al3 1Depa men o Radio Elec onics, Facul y o Elec ical Enginee ing and Communica ion, B no Uni e si y o Technology, Technicka 12, B no, 616 00, Czech Republic 2Depa men o Telecommunica ions, Facul y o Elec ical Enginee ing and Communica ion, B no Uni e si y o Technology, Technicka 12, B no, 616 00, Czech Republic 3Depa men o Technical S udies, College o Poly echnics Jihla a, Tols eho 16, Jihla a 586 01, Czech Republic [email p o ec ed].cz h p://dx.doi.o g/10.5755/j01.eie.23.2.17996 31 ELEKTRONIKA IR ELEKTROTECHNIKA,ISSN 1392-1215,VOL.23,NO.2,2017 implemen a ion is app op ia e, as i is shown in his pape . TABLE I. COMPARISON OF REPORTED PSEUDO/FULLY- DIFFERENTIAL FIRST-ORDER MULTIFUNCTIONAL FILTERS. Wo k No. o pas./ac . elemen s A ailable unc ions T ans e esponse con igu abili y [10] 4-6/1 AP, BR No [11] 2-6/1-2 LP, HP, AP No [12] 4/1 AP No [13] 4/2 AP, LP No [14] 1/3 AP No Fig. 3 2(6)/1(4) HP, iAP Yes No e: (i) DT – (in e ing) di ec ans e ; (i) AP – (in e ing) all-pass; LP – low-pass; (i) HP – (in e ing) high-pass; BR – band ejec . II. DESCRIPTION OF FLOATING INDUCTANCE SIMULATOR P oposed loa ing induc ance simula o shown in Fig. 1 employs h ee elec onically adjus able cu en con eyo s o second gene a ion (ECCIIs) [16] desc ibed by he ollowing in e - e minal ans e s: IZ=B.IX,VX=VY,IY= 0 and one di e en ial ol age bu e (DVB) [16] p o iding he ollowing ope a ion: V+V=Vou . No e ha he cu en gain Bo each ECCII is con ollable linea ly by DC ol age (BVSET_B). Comme cially a ailable de ices EL2082 and AD830 a e examples o p e iously discussed elemen s (ECCII, DVB) wi h well- i ing beha iou up o se e al ens o MHz. Assuming Rx1 =Rx2 =R3a e equal and designa ed as RL_in , he admi ance desc ip ion (based on me hod o unknown nodal ol ages: I=Y×V) o he ci cui in Fig. 1 p esen s se o equa ions: 1 1 1 1 1 1 1 1 2 2 2 2 , inp inp Inp inp I V a B a B I V a B a B                                (1) whe e   2 _ in _ in 3 _ in 2 L L L a sC R B R   . Then, assuming B1=B2=B1,2, impedance o equi alen induc ance emula o has he ollowing heo e ical o m   _ in _ in 2 _ in _ in 3 _ in 1,2 1,2 ( ) 2, eq eq s L L L L s sL R sC R B R B B       (2) whe e CL_in and RL_in a e passi e elemen s o he induc ance simula o and i is clea ha he p oposed ci cui beha es as lossy induc ance. Howe e , he lossy pa Rs_in in (1) can be easily elimina ed by selec ing cu en gain B3(ECCII3) equal o 2, losses a e nega i e o B3> 2. Ne e heless, lossy pa o he p oposed induc ance simula o can be also bene icial o applica ion as in oduced la e . Mo eo e , i is ob ious ha alue o Leq_in can be con olled elec onically by B1and B2. P oposed induc ance simula o in Fig. 1 was designed o u he implemen a ions wi h he ollowing alues o in e nal elemen s: RL_in = 560 ,CL_in = 470 pF as a pa icula example. The PSpice simula ion esul s o he induc ance simula o in g ounded o m a e shown in Fig. 2. Induc ance alue Leq was changed om 2.63 H o 0.13 H by VSET_B1,2 be ween 0.1 and 2 V, i.e. gains B1and B2a e be ween 0.1 and 2. III. INDUCTANCE SIMULATOR IMPLEMENTATION IN PSEUDO- DIFFERENTIAL FILTERING APPLICATIONS Bene i s o he p oposed solu ion we e e i ied in se e al il e ing applica ions shown in he ollowing sec ions. Induc ance simula o is used as loa ing elemen in RL o RLC ladde s uc u es. Theo e ical expec a ions a e suppo ed by simula ion esul s and also by labo a o y expe imen s. Fig. 1. P oposed loa ing induc ance simula o based on ECCIIs and DVB including desc ip ion o beha iou o ac i e elemen s (b own colou ). Fig. 2. Simula ion esul s o impedance cha ac e is ics (magni ude) o p oposed induc ance simula o (Fig. 1) in lossless mode while uning B1,2. A. Recon igu able Fi s -O de High-Pass/In e ing All- Pass Fil e P oposed ci cui in Fig. 3 has a pseudo-di e en ial con igu a ion, which, in acco dance o [8], means ha i has di e en ial inpu /ou pu e minals, bu inne ci cui i sel is no ully symme ical. Special ea u es o he in oduced induc ance simula o in Fig. 1 a e e y use ul o econnec ion-less econ igu abili y o ans e unc ion in pseudo-di e en ial i s -o de high-pass/in e ing all-pass (HP/iAP) il e . These ea u es a e no a ailable i any s anda d solu ion o simula o o lossless induc ance is employed as Leq in s uc u e shown in Fig. 3. Fig. 3. Implemen a ion o he loa ing induc ance simula o in pseudo- di e en ial-mode econ igu able i s -o de HP/iAP il e (Vinp/ou _s – symme ical inpu /ou pu ol age). Di e en ial-mode ans e unc ion ( ull desc ip ion (1) 32 ELEKTRONIKA IR ELEKTROTECHNIKA,ISSN 1392-1215,VOL.23,NO.2,2017 applied) o he ci cui in Fig. 3 has symbolical o m     _ _ _ 3 _ in _ in _ in 3 1 2 2 _ in _ in ( ) 2 2 . 2 ou s HP iAP inp s L L L L L V K s V B s C R R B R B B s C R                 (3) T ans e unc ion (3) o e s high-pass (HP) esponse, i B3= 2   1 2 2 _ in _ in ( ) 2 , HP L L s K s R B B s C R   (4) whe e cu -o (pole) equency is de ined by   1 2 12_ in _ in . p L L R B B R C    (5) The easies example o in e ing all-pass (iAP) esponse is a ailable o R=RL_in ,B1=B2=B1,2 = 1, B3= 3, om (3) we ob ained simpli ied o m o ans e _ in _ in 1/ ( ) ( ) 2 . 1/ ( ) L iAP L s C R K s s C R   (6) The ollowing ela ion mus be alid o B3i iAP esponse is eques ed 1 2 32 . 2 B B B   (7) Fig. 4. Magni ude esponses o he pseudo-di e en ial-mode econ igu able HP/iAP il e econ igu ed by con ol o Vse _B3 and se e al examples o uning o he cu -o equency o HP esponse by Vse _B1,2. Bo h pole/ze o equency has simple o m  p1 = 1/RCL_in . Also no e ha pole equency can be con olled elec onically by Vse _B1,2 and ype o ans e unc ion is econ igu ed by loss pa o induc o simula o (B3gain). Figu e 4 includes compa ison o expe imen al and simula ed esul s o he il e om Fig. 3, when wo king esis o was se R= 1 k. Resul ing aces we e ob ained o ideal cu -o equency equal o 216 kHz (simula ed 226 kHz, measu ed 227 kHz) ob ained by Vse _B1,2 = 0.1 V, Vse _B3 = 1.95 V (HP esponse) o 2.15 V in case o iAP esponse ((7) o B3). The example o  p1 uning in case o he HP esponse by con ol o B1,2 can be obse ed also in Fig. 4. In measu emen s he change o Vse _B1,2 be ween 0.1 and 0.5 V ensu es adjus men o p1 om 227 kHz o 959 kHz, espec i ely. B. Pseudo-Di e en ial-Mode Second-O de Band-Pass Fil e This applica ion is easily a ailable by adding a capaci ance be ween ou pu e minals as shown in Fig. 5. T ans e unc ion o ideal RLC ladde s uc u e employing lossless Leq conside ed in ully-di e en ial o m (1) is de ined as 2 1 ( ) . 1 1 2 BP eq sRC K s s s RC L C    (8) R R Leq CDVB 1 RL Ro 50 Ω 50 Ω OPAMP RI AD8138 AD830 DUT ne wo k analyze E5071C Fig. 5. Implemen a ion o he induc ance simula o in second-o de RLC ladde di e en ial band-pass il e . Schema ic includes expe imen al se up (Vinp/ou _as – asymme ical ol age; Vinp/ou _s – symme ical ol age). Conside ing he implemen a ion o lossless Leq_in (i.e. B3= 2), ans e unc ion (8) changes o 21 2 2_ in _ in 1 ( ) , 1 22 BP L L sRC K s B B s s RC R C C    (9) whe e cen e (pole) equency and quali y ac o ha e he ollowing exp essions: 1 2 1,2 1 2 2 2 _ in _ in 1,2 _ in _ in 1 2 1, p p B B B L L L L B B C C R B C C R          (10)   1 2 1,2 1 2 _ in _ in 1,2 _ in _ in 2 2 2. B B B L L L L B B C R Q Q R C B C R R C        (11) No e ha his second-o de BP il e allows uning o he pole equency wi h heo e ically no e ec on bandwid h, because BW [Hz] = 1/(4RC), and also change o quali y 33 ELEKTRONIKA IR ELEKTROTECHNIKA,ISSN 1392-1215,VOL.23,NO.2,2017 ac o wi hou change o gain du ing he uning p ocess. No e ha bo h il e s (Fig. 3 and Fig. 5) ha e +6 dB pass- band gain (see cons . 2 in (3) and (9)). Howe e , expe imen al se up (ou pu di ide ROand RLo ne wo k analyse ) pe o ms di ision by 2, he e o e, ideal pass-band gain o he whole sys em is 0 dB. Fig. 6. Magni ude esponses o he di e en ial-mode second-o de BP il e du ing uning o he cen e equency. Compa ison o expe imen al esul s, simula ions, and heo e ical aces o magni ude and phase esponses o he pseudo-di e en ial-mode BP il e is gi en in Fig. 6. Passi e componen alues and ac i e pa ame e s we e se as ollows: RL_in = 560 ,CL_in =C= 470 pF, R= 1 k,B3= 0 and B1=B2=B1,2 = 1 (Vse _B1,2 = 1 V). Theo e ical, expe imen ally es ed, and simula ed BP achie ed p2 = {605, 606, and 596} kHz, wi h BW = {169, 163, and 163} kHz and Q= {3.6, 3.8, and 3.7} espec i ely. Tuning p ocess (adjus men o B1,2) can be also obse ed in Fig. 6. In measu emen s, adjus ing o Vse _B1,2 be ween 0.1 V and 2 V o e s ange o p2 uning om 199 kHz o 880 kHz. All de ails a e no ed in igu es. P esen ed expe imen al esul s in Fig. 4 and Fig. 6 we e ca ied ou using ec o ne wo k analyse E5071C wi h es se up gi en in Fig. 5. Figu e 7 shows he ab ica ed p o o ype including con e e s om single-ended o di e en ial mode and ice e sa se ing o pe o med expe imen al es s. Fig. 7. Fab ica ed expe imen al p o o ype o pseudo-di e en ial il e . IV. DISCUSSION AND SUMMARIZATION OF RESULTS Analysis o discussed induc ance simula o e ealed ollowing ad an ages: a) possible implemen a ion in di e en ial applica ions ( loa ing elemen ), b) pseudo- di e en ial cha ac e o ci cui y (inne g ounded capaci o ), c) simple elec onic con ollabili y o induc ance alue and losses including special o m o losses (2) ha allows in e es ing applica ions. No e ha hese ea u es a e no simul aneously a ailable in many s anda d solu ions. The alue o designed induc ance simula o was es ed om 2.63 H o 0.13 H (Vse _B1,2 adjus ed om 0.1 V up o 2.0 V), i.e. unabili y a io is app oxima ely 1:20. In acco dance o he s a e-o - he-a discussed in Table I, he i s applica ion example o he designed special induc ance implemen a ion ( econ igu able i s -o de il e ) ep esen s unique de ice, because up o now (pseudo)di e en ial-mode econ igu able il e s ha e no been widely in es iga ed. Ope a ion o he HP/iAP was e i ied by change o Vse _B3 om 1.95 V o 2.15 V (HP→iAP). Tuning ange o he pole equency yields alues om 227 kHz o 959 kHz (Vse _B1,2 se om 0.1 V o 0.5 V). The second example o applica ion, he pseudo- di e en ial second-o de BP was es ed in equency ange om 199 kHz o 880 kHz (Vse _B1,2 se om 0.1 V o 2.0 V) wi h almos cons an bandwid h app oxima ely equal o 165 kHz. V. CONCLUSIONS P esen ed loa ing induc ance simula o was es ed in wo no el applica ions namely pseudo-di e en ial-mode econ igu able i s -o de HP/iAP il e and second-o de BP il e de i ed om passi e p o o ypes. Due o limi a ions o comme cially a ailable de ices used in induc ance simula o , a ge ed equency ange was in hund eds o kHz. Expe imen al esul s a e in good ag eemen wi h simula ions and heo y. P oposed induc ance simula o o e s in e es ing ea u es in gi en applica ions and special o m o i s losses (in lossy mode o ope a ion) b ings use ul implemen a ions also in pseudo-di e en ial-mode oscilla o s. Fu u e po en ial o di e en ial-mode ci cui s is unde g ea in e es s o many esea che s. The e o e, we expec u he de elopmen o hese ci cui s especially wi h econnec ion-less econ igu a ions. REFERENCES [1] O. Casas, R. Pallas-A eny, “Basics o analog di e en ial il e s”, IEEE T ans. Ins umen a ion and Measu emen , ol. 45, no. 1, pp. 977–985, 1996. [Online]. A ailable: h p://dx.doi.o g/10.1109/ 19.481347 [2] A. A. El-Adawy, A. M. Soliman, H. O. Elwan, “A no el ully di e en ial cu en con eyo and applica ions o analog VLSI”, IEEE T ans. Ci cui s and Sys ems – II: Analog and Digi al Signal P ocessing, ol. 47, no. 4, pp. 306–313, 2000. [Online]. A ailable: h p://dx.doi.o g/10.1109/82.839666 [3] S. M. Al-Shah ani, “Fully di e en ial second o de il e ”, in P oc. 47 h IEEE In . Midwes Symposium on Ci cui s and Sys ems (MWSCAS), Hi oshima, Japan, 2004, pp. 299–301. [Online]. A ailable: h p://dx.doi.o g/10.1109/MWSCAS.2004.1354353 [4] P. Beg, I. A. Khan, S. Maheshwa i, M. A. Siddiqi, “Digi ally p og ammable ully di e en ial il e ”, Radioenginee ing, ol. 20, no. 4, pp. 917–925, 2011. [5] R. Beg, S. Maheshwa i, M. A. Siddigi, “Digi ally con olled ully di e en ial ol age- and ansadmi ance-mode biquad a ic il e ”, IET Ci cui s, De ices and Sys ems, ol. 7, no. 4, pp. 193–203, 2013. [Online]. A ailable: h p://dx.doi.o g/10.1049/ie -cds.2012.0244 [6] M. A. Ib ahim, H. Kun man, “A no el high CMRR high inpu impedance di e en ial ol age-mode KHN-biquad employing DO- DDCCs”, AEU – In e na ional Jou nal o Elec onics and Communica ions, ol. 58, no. 6, pp. 429–433, 2004. [Online]. 34 ELEKTRONIKA IR ELEKTROTECHNIKA,ISSN 1392-1215,VOL.23,NO.2,2017 A ailable: h p://dx.doi.o g/10.1078/1434-8411-54100266 [7] A. K. Singh, P. Kuma , “A no el ully di e en ial cu en mode uni e sal il e ”, in P oc. 57 h In . Midwes Symposium on Ci cui s and Sys ems (MWSCAS), Texas, USA, 2014, pp. 579–582. [Online]. A ailable: h p://dx.doi.o g/10.1109/MWSCAS.2014.6908481 [8] J. Ko on, N. He encsa , O. Sladok, J-W. Ho ng, “Pseudo-di e en ial second-o de band- ejec il e using cu en con eyo s”, AEU - In e na ional Jou nal o Elec onics and Communica ions, ol. 70, no. 6, pp. 814–821, 2016. [Online]. A ailable: h p://dx.doi.o g/ 10.1016/j.aeue.2016.03.009 [9] J. Ko on, N. He encsa , J. W. Ho ng, “Di e en ial second-o de ol age-mode all-pass il e using cu en con eyo s”, Elek onika i Elek o echnika, ol. 22, no. 5, pp. 52–57, 2016. [Online]. A ailable: h p://dx.doi.o g/10.5755/j01.eie.22.5.16344 [10] M. A. Ib ahim, S. Minaei, H. Kun man, “DVCC based di e en ial- mode all-pass and no ch il e s wi h high CMRR”, In . Jou nal o Elec onics, ol. 93, no. 4, pp. 231–240, 2006. [Online]. A ailable: h p://dx.doi.o g/10.1080/00207210600562181 [11] M. S. Ansa i, G. S. Soni, “Digi ally-p og ammable ully-di e en ial cu en -mode i s -o de LP, HP and AP il e sec ions”, in P oc. In . Con . Signal P opaga ion and Compu e Technology (ICSPCT), Ajme , India, 2014, pp. 524–528. [Online]. A ailable: h p://dx.doi.o g/10.1109/ICSPCT.2014.6884963 [12] J. W. Ho ng, C. M. Wu, N. He encsa , “Fully di e en ial i s -o de allpass il e s using a DDCC”, Indian J Eng Ma e Sci, ol. 21, no. 4, pp. 345–350, 2014. [13] I. A. Khan, M. I. Masud, S. A. Moiz, “Recon igu able ully di e en ial i s o de all pass il e using digi ally con olled CMOS DVCC”, in P oc. 8 h IEEE GCC Con . and exhibi ion (GCCCE), Musca , Oman, 2015, pp. 1–5. [Online]. A ailable: h p://dx.doi.o g/10.1109/IEEEGCC.2015.7060082 [14] N. He encsa , J. Je abek, J. Ko on, K. V ba, S. Minaei, I. C. Gokna , “Pole equency and pass-band gain unable no el ully-di e en ial cu en -mode all-pass il e ”, in P oc. IEEE In . Symposium on Ci cui s and Sys ems (ISCAS), Lisbon, Po ugal, 2015, pp. 2668– 2671. [Online]. A ailable: h p://dx.doi.o g/10.1109/ISCAS. 2015.7169235 [15] R. So ne , J. Je abek, N. He encsa , T. Zak, W. Jaikla, K. V ba, “Modi ied cu en di e encing uni and i s applica ion o elec onically econ igu able simple i s -o de ans e unc ion”, Ad ances in Elec ical and Compu e Enginee ing, ol. 15, no. 1, pp. 3–10, 2015. [Online]. A ailable: h p://dx.doi.o g/10.4316 /AECE.2015.01001 [16] R. So ne , J. Je abek, L. Langhamme , J. Polak, N. He encsa , R. P okop, J. Pe zela, W. Jaikla, “Compa ison o wo solu ions o quad a u e oscilla o s wi h linea con ol o equency o oscilla ion employing mode n comme cially a ailable de ices”, Ci cui s, Sys ems and Signal P ocessing, ol. 34, no. 11, pp. 3449–3469, 2015. [Online]. A ailable: h p://dx.doi.o g/10.1007/s00034-015- 0015-7 35