THE PROFESSION Z>£ ■_ -z First >t ever interview giver Electronics Magazine Manu Chhabria to by ! Mr □ Optoelectronics u ix P — based multimeter □ Bus interface for LCD screens Publisher: C.R. Chandarana Editor: Surendra Iyer T echnical Adviser : Ashok Dongre Circulation : J. Dhas Advertising: B.M. Mehta Production: C.N. Mithagari Address : ELEKTOR ELECTRONICS PVT. LTD. 52. C Proctor Road, Bombay-400 007 INDIA Telex: (011)76661 ELEK IN Volume-6 Number 12 December - 1988 INTENTS Overseas editions: Elektor Electronics 1, Harlequin Avenue, Great West Hoad, Brentford TW8, 9EW U.K. Editor : Len Seymour Pulltron Publicacoes Tecnicas Ltda Av Iplranga 1 100, 9° andar CEP 01040 Sao Paulo - Brazil Editor: Juliano Barsali Elektor sari Route Nationale; Le Seau; B.P. 53 592270 Bailleul - France Editors : D R S Meyer; G C P Raadorsdorf Elektor Varlag GmbH Susterfeld-StraBe 25 100 Aachen - West Germany Editor : E J A Krempelsauer Elektor EPE Karaiskakl 14 16673 Voula - Athens - Greece Editor: E Xanthoulis Peter Treckpoelstraat 2.4 6191 VK Seek - the Netherlands Editor: PEL Kersemakers Ferreira & Bento Lda R.D. Estefania, 32.1° 1000 Lisboa - Portugal Editor : Jorge Goncalves Ingelek S.A. Plaza Republica Ecuador 2.28016 Madrid-Spain Editor : A M Ferrer In Part: Kedhorn Holdings PTY Ltd Cnr Fox Valley Road & Kiogle Street Wahroonga NSW 2076 - Australia Editor : Roger Harrison Electronic Press AB Box 63 182 11 Dandcryd - Sweeden Editor : Bill Cedrum The Circuits are domestic use only. The submission of designs or articles implies permission to the publisher to alter and translate the text and design and to use the contenls in other Elektor Publications and activities. The publishers cannot guarantee to return any material submitted to them. Material must be sent to the Holland address igiven above). All drawings, photographs, printed circuit boards and articles published in elektor publications are copyright and may not be reproduced or imitated in whole or part without prior written permission of the publishers. Patent protection may exist in respect of circuit devices, components etc. described in this magazine. The publishers do not accept responsibility for failing to identify such patent or other protection. . trtiwsft, p '&* / '' 7 ( 10 ] Lt-T Printed at : Trupti Offset Bombay - 400 013 Ph. 4923261,4921354 Copyright ® 1988 Elektuur B.V. Editorial Technology Transfer 12.07 Special Feature The jumbo success of Chhabria (An Exclusive Interview) 12.10 Computer gurus for video teaching 12.14 Optoelectronics 12.43 A very intelligent computer terminal 12.46 Audio & Hi-Fi PROJECT: LFA 150 -a fart power amplifier (Parti ) 12.28 Computers PROJECT: Bus interface for high-resolution liquid crystal screens (Part 1) . 12.48 Electronics PROJECT: Portable MIDI keyboard 12.37 PROJECT: Harmonic enhancer 12.59 General Interest PROJECT : Infra-red remote control for stepper motors 12.24 DESIGN IDEA: Mains signalling 12.34 DESIGN IDEA: Fade asleep 12.27 Radio 8r Television DESIGN IDEA: A microprocessor-based intelligent multi-function test instrument 12.55 Information Telecom news 12.18 Electronics news 12.19 New products 12.66 Guide Lines Classified ads 12.74 Index of advertisers 12.74 elektor indie december 1988 1 2.05 Front cover A recently introduced 1C makes a dream of many electrophonics enthusiasts come true: to build their own MIDI keyboard from a handful of compo- nents. The portability of the keyboard des- cribed in this issue makes it ideal for first- aid testing of MIDI equipment. Moreover, in conjunction with a microprocessor, it can be used for practising, composing, and editing musicdl pieces in places where a full- size keyboard would be cumbersome to use. TECHNOLOGY TRANSFER The Prime Minister, Mr. Rajiv Gandhi, has observed that import substitution is “one of the biggest mistakes” committed by our country. While distributing the Shanti Swarup Bhatnagar awards to outstanding scientists Mr. Gandhi regretted that we ended up substituting third or fourth generation but never reached the forefront while resorting to substitution. Mr. Gandhi's statement may appear to be a sweeping generalisation but he is not wide off the mark. But, the blame for this lopsided result does not lie at the door step of scientists and technologists alone as often technology transfer decisions are based on political considerations. Examples of this kind are too many. Telecommunication in the country is now saddled with half-a-dozen technologies with shifting stands taken at various times. The so-called Phased Manufacturing Programmes in electronics never crossed the phase of importing the kits and assembling them. We had seen unsavoury experiences regarding the technology ti ansfer for the National Silicon Facility. Indigenous technology finally got a chance in this respect. Again, we find a similar problem in technology transfer regarding the manufacture of 1 .5 micron chips. Two public sector organisations are working at cross purposes. Bharat Electronics ltd. is embroiled in a controversy over its Taloja glass shells unit. The defence ministry desires BEL to give up its diversification into consumer electronics and concentrate on defence projects. To make colour glass shells, an investment of Rs. 95 crores is required. BEL entered into joint venture with Rashtriya Chemicals and Fertilisers Ltd. another public sector unit for its Taloja expansion programme. The government shot down RCF's entry and Introduced Samtel, a private firm with the multinational Corning Glass. Government has been forced to re-examine the issue in the wake of widespread criticism. It is thus clear that true technology transfer cannot take place unless vested interests give way to national interests. elektor india december 1988 12.07 INFRA RED REMOTE CONTROL FOR STEPPER MOTORS Many audio purists balk at the use of an electronic volume con- trol, but would still like to upgrade a home-made preamplifier with remote control. This can be accomplished by using a good- quality potentiometer, a stepper motor, and the simple, yet versatile, infra-red transmitter and receiver described here. One particularly interesting application of the proposed infra-red remote control system is the actuating of the volume po- tentiometer in a high-quality audio pre- amplifier, such as the one described last month. Basically, the potentiometer is operated by a small stepper motor, whose direction of travel is controlled by means of pulses emitted by a hand-held infra-red transmitter. Infra-red transmitter The circuit diagram of this part of the remote control system is shown in Fig. 1. Simple data encoding is used to keep the cost of the circuit as low as possible. The direction of travel of the stepper motor fitted at the receiver side is determined by the width of the pulses supplied by two monostable multivibrators, MMVi (70 ps) and MMVj (470 ps). The pulse frequency, and hence the motor speed, is set with Pi in oscillator Ni-Nj. Buttons Si and S: form the volume up/down controls because they determine whether MMVi or MMV 2 drives the output tran- sistor, Ti. The pulsating infra-red light beam is emitted by series-connected IREDs Di-D-Di. Receiver and motor driver Photodiode Dj in Fig. 2 was selected for optimum sensitivity in the part of the infra-red spectrum covered by the sender diodes (see also Ref. 1.) The photo- current generated by the incident infra- red light is magnified and converted to a voltage by opamp A 2 , which drives detector At via high-pass C 5 -R 13 . This filter serves to eliminate interference Features: • Infra-red remote control link with a range of up to 8 metres. • Forward/reverse control of stepper motors. • Adjustable motor speed. • Drives a variety of 2-stator, unipolar, motors. • Maximum drive capability: 4 A per phase. • Supply voltages: receiver logic: 9.5-18 VDC; stepper motor: max. 18 VDC; transmitter: 9 V PP3 battery. • Simple, low-cost design. caused by sunlight and hum superimpos- ed on light by electric bulbs. The detec- tion threshold of comparator Ai is kept low at about 10 mV (Riu-Ru) to ensure adequate sensitivity. Feedback resistor Rn provides the necessary hysteresis to prevent jitter and spurious step pulses being generated when Ai toggles. Each received pulse triggers both MMVj and MMVj, and is compared to refer- ence pulses of 220 ps supplied by MMVj. Received pulses are applied direct -to the clock input of stepper motor driver ICs, whose DIR (direction) input is controlled by the output of MMVj. This makes the direction of travel of the stepper motor dependent on the length of the received pulses, relative to that of the reference pulses. Although the stepper motor driver Type SAA1027 (SGS/Philips Components) is capable of supplying stator currents of up to 500 mA, power drivers (Ti-Ts) are added to prevent excessive dissipation, and to allow the use of motors that re- quire more current. The flyback diodes in the power stage should be fast- recovery types (lN493x series, or BYV27). The use of the ubiquitous 1N4001 is not recommended unless the total stator current is known to remain well under 1 A. Power resistors Ru, and Rn may be used to achieve a rudimen- tary kind of current drive of the stator windings in the motor — more on this under ‘The power supply’. Provision has been made for manual op- eration, at the receiver, of the volume control. This is achieved by T’-Ru auto- matically interrupting the base current for the driver stage in the SAA1027 when no pulses have been received for about 0.1 s. Series transistor T: then interrupts the hold torque for the motor, so that the potentiometer spindle can be oper- ated manually. This type of control guarantees low overall dissipation because there is no quiescent hold cur- rent. Certain motors do require a con- tinuous hold current, however. These can still used with the present circuit simply by omitting T: and fitting a wire link between the connections provided for its collector and emitter terminals. Fig. I. Circuit diagram of the hand-held infra-red transmitter. 12.24 eiektor india december 1988 Fig. 2. Circuit diagram of the infra-red receiver, pulse decoder and stepper motor driver. The power supply The IR transmitter is powered by a 9 V PP3 battery. An on/off switch is not re- quired because the quiescent current consumption of the circuit is negligible at a few nano-amperes. This rises to a few milli-amperes when either of the two buttons is pressed. The actual current consumption then depends on the set- ting of Pi. The type of supply required for the re- ceiver depends mainly on the environ- ment in which this circuit is used. The logic circuitry can operate from a supply voltage between 9.5 V and 18 V. It will be clear that the supply for the motor is laid out in accordance with the type used. A 12 V motor is ideal because it allows powering the driver stage and the logic circuitry from a common supply, connected to terminals Um and ground (fit jumper JPi). The logic supply is decoupled with the aid of Ri;-Ce. Fit wire links in positions Rif. and Rn when the motor used requires voltage drive. Many stepper motors are 5 V types. Where a relatively powerful type is used, it is recommended to dimension the motor supply for 5 V (connect to Um Fig. 3. Printed circuit board for the trans- mitter. Together with a PP3 battery, it fits 0 5 08: 1 N 4913 ...IN 4937, BtV 27 N 1 . . . N 4 = 1C 2 = 4001 MMV 3, MMV 4 = 1C 3 = 4538 A1. A2 = IC4 = TIC 272, C A 3240 1C 5 x SAA 1027 Motorized volume control built into the high-quality preamplifier described last month. exactly in a Type 222 enclosure from Heddic. elektor india december 1988 12.25 Fig. 4. Printed circuit board for the receiver. and ground). Fit wire links for Ri6 and Rn, but do not fit JPi — connect the 12 V supply to terminals + and 0 (close to D-j on the PCB). Where a relatively small stepper motor is used, Ris and Rn are dimensioned to reduce Um from 12 V to the voltage required. This is con- venient because it allows the complete receiver plus motor driver to be powered from a single supply. Small, 200-step, 5 V motors used in disk drives are some- times offered by surplus stores. These motors give excellent results with Ri6=Ri7=39 Q; 4 W (stator current = 200 mA). It shoud be noted that the circuit can only drive unipolar motors. These nor- mally have 6 connecting wires, but there are also 5-wire types in which the centre taps of the two stator windings (COM1; COM2) have been connected internally. R m Resistors (±5%l: Ri =22R R2;Ra;R9;Rii = 1M0 R3;R4;Ri3= 100K Rs = 47K Re;R7 = 22K Rio=560K Ri2 = 390R Ri4 = 120R Ri5 = 100R Rib;Ri 7= 4W resistor; value depends on stepper motor used. Rie = 22K Rie. . R 22 incl. = 4K7 Pi = 100K preset H Capacitors: Ci = 330n C2= 10p; 16 V C3;C5;C7 = 10n C4 = 3n3 Ca = 10p; 25 V Ca;Cio=100n Cs=4p7 Cn = lOOp; 25 V Semiconductors: Di;02:D3=LD271 * D4 = BP104' D5. . .Da incl. = 1 N4933/4 + /5/6/7 or BYV27 + Ti;T2 = BC557B T 3 . . .Ta incl. = BD438 ICi;IC3 = 4538 IC2 = 4001 IC4=TLC272 or CA3240 ICa=SAA1027 + Miscellaneous: Sl;S 2 = Oigitast switch IITW or ITT/Schadow). 3 off plastic reflectors' for D 1 -D 3 . PCB Type 880161 Fig. 5. Two suggestions for coupling the stepper motor spindle to that of the potentiometer. 12.26 elektor india december 1988 Fig. 6. Modifications to transmitter and receiver to enable manual volume control when the stepper motor and potentiometer spindles are coupled as shown in Fig. 5h. Constructional hints Construction of the transmitter and re- ceiver on the printed circuit boards shown in Figs. 3 and 4 should not cause problems. The transmitter is fitted in a hand-held ABS enclusure with integral battery compartment. The 3 IREDs are fitted with ready-made reflectors to in- crease the range of the transmitter. To prevent it seeing light from bulbs or fluorescent tubes fitted to the ceiling, the photodiode in the receiver should be mounted in a short tube whose inside is painted matt black. If the diode is fitted on to the front panel of the audio equip- ment, it should be connected to the re- ceiver board by means of shielded wire. In some cases, it may be necessary to decrease the sensitivity of the receiver to prevent it being triggered by ambient light. This can be achieved by increasing the value of R 12 to, say, 560 Q. The use of a stepper motor that draws more than about 1 A necessitates cool- ing of the power transistors in the re- ceiver by clamping them together with the aid of 3 small, 2.5 mm thick, pieces of aluminium and a central M3 bolt. Figures 5a and 5b provide suggestions for coupling the stepper motor spindle to that of the potentiometer. Cog-wheel systems should not be used because they are damaged quite easily by the vibration of the stepper motor. A rubber or nylon belt as used in cassette recorders, or a strengthened O-ring, is perfect because it allows manual control of the poten- tiometer as discussed earlier. The stepper motor and the volume po- tentiometer may also be secured on to a common U-shaped piece of aluminium as shown in Fig. 5b. In this arrange- ment, the spindles are coupled direct. Manual control is still possible, however, when the transmitter is duplicated and fitted close to the receiver. The modifica- tions to the transmitter and the receiver to achieve local control are shown in Figs. 6a and 6b. In the transmitter, the (shaded) IREDs are replaced with a wire link, Ri is replaced with a 10 kR type, and the points marked A in the trans- mitter and receiver are interconnected. Diode Di is inserted between the com- parator output and the trigger inputs of the monostables. Together with the Ti in the transmitter, it forms a wired-OR function. The ‘local’ volume up/down controls, Si and S 2 , are fitted on to the front panel of the equipment. H References: 1. Long-range infra-red transceiver. Elektor India, December 1987. Con- tains a useful background to infra-red light communication. fade asleep Some children have less difficulty falling asleep if the bedroom light is left on. How- ever, this means that one of the parents has to go up after half an hour or so to turn the lights off - hopefully without waking the children up again . . . The circuit described here will fade out the lights very slowly, either completely off or to a preset mini- mum (night-light) level. . As long as SI is closed the light(s) will burn at full brightness. As soon as SI is opened, the lights start to fade out very gradually until they reach a certain level (preset by means of PI). The fade-out time is deter- mined by the value of C4 and by the setting of PI. As an example, if C4= 100 ft and PI is set at minimum it will take approximately half an hour for the lamp to fade out. If desired, C4 can be increased; however, it is not advisable to go above about 470 p. The circuit must be mounted in a well- insulated case, and PI should be a potentio- meter with a plastic spindle. The type of triac required will depend on the load, of course. It is advisable to select a type that is capable of handling a current of up to where PL is the nominal ‘wattage’ of the lamp and Um is the nominal mains voltage. A ‘cold’ filament draws a relatively heavy current! (Ptessey application) elektor indie december 1908 1 2.27 LFA-150: A FAST POWER AMPLIFIER (PART 1) from a basic idea by A. Schmeets This first of a two-part article describes the design of a power amplifier that makes use of very fast ring-emitter transistors and delivers up to 150 watts into 8 ohms. A feature of the design is the low negative-feedback factor. Power output (20 Hz- 20 kHz; THD = 0,5%) THD (1 kHz) (20 Hz- 20 kHz! Frequency response (1 W) Power bandwidth Phase error (20 Hz -20 kHz) TIM (75 W; 50 Hz: 7 kHz; 4:1) Slew rate Open-loop bandwidth Open-loop amplification Output impedance Input sensitivity Signal-to-noise ratio 1 50 W into 8 ohms 200 W into 4 ohms < 0.01% (1 W) < 0.01% (10 Wl < 0.04% (100 W) < 0.025% (1 W) 1 Hz - 1 MHz (unweighted) 1 Hz -350 kHz (unweighted) < 5° < 0.05% > 50 V//rs (unweighted) 10 kHz 2,300 < 0.05 ohm 1.1 V r.m.s. > 110 dB Although commercial high-quality audio power amplifiers have made use of multiple-emitter and ring-emitter tran- sistors for some time, these devices have not been easily obtainable for private purposes. That situation has changed, fortunately, and a number of importers can now supply them on a small- quantity basis. Multiple-emitter transistors consist of a number of identical transistors connec- ted in parallel on one chip. The ring- emitter transistor is a power transistor with a special chip structure for the base, collector and emitter regions. These transistors are the fastest and most linear devices for use in audio power amplifiers. High-quality audio power amplifiers are still based on discrete designs, although good-quality power amplifier modules have become available over the past year or so. However, where absolute top qual- ity is wanted, based on an uncoventional design, there is no other way than the use of discrete transistors. The present design hinges on low open- loop gain, which guarantees minimal transient-intermodulation (TIM) distor- tion and thus the best possible sound quality. The bandwidth is sufficiently large to ensure minimal phase shift over the entire audio range, which again aids the sound quality. Design philosophy The design of an AF power amplifier can go two ways. The first uses a very high open-loop gain combined with a very large negative-feedback factor; the second, a low open-loop gain and a consequent smaller negative-feedback factor. Most audio power amplifiers belong to the first category, because in that design it is easy to achieve low har- monic distortion. However, that design also has a serious shortcoming. When ' the input signal is fairly large and of a frequency that lies outside the open-loop bandwidth of the amplifier, there is a 1 2.28 oloktor indla decamber 1988 likelihood, owing to the high open-loop gain, of some of the internal amplifier stages becoming saturated. This results in strong bursts of intermodulation that are clearly audible and sound like crossover distortion. Note that the audio signal variations are maximum around the zero crossing points, so that satu- ration is most likeiy about these points. These problems may be avoided by re- ducing the open-loop gain. This in- creases the open-loop bandwudth, so that the likelihood of the frequency of the input signal lying outside the open- loop bandwidth is much smaller. Of course, this also causes an increase in the total harmonic distortion (THD), but that is not really a serious problem. The human ear is nowhere near as sensitive to THD as to TIM and crossover distor- tion. In other words, an amplifier with 0.3% THD and 0.003% TIM will in practice always sound better than one with 0.003% THD and 0.3% TIM. Apart from low open-loop gain, the various stages of a good-quality ampli- fier need to have a large bandwidth to ensure, if possible, an open-loop band- width greater than the audio range. It is, fortunately, possible to optimize both the bandwidth and the phase behaviour where necessary in the amplifier with the aid of lead-compensation (networks that locally increase the amplification above a given frequency). Another important aspect is the fre- quency compensation that limits the open-loop bandwidth (the so-called lag- compensation). This compensation delermines the slew rate of the amplifier and must, therefore, be applied as close to the input as possible to ensure that the input signal is limited before it is fed to the amplifier stages. In many amplifiers, the negative- feedback factor for a.c. signals is differ- ent from that for d.c. signals, which is normally achieved with the aid of a ca- pacitor. It is true that. this puts less of an onus on the stability of the circuit, but it may give rise to problems, particularly since the capacitor often has such a large value that an electrolytic type (!) is used. With correct design and good tempera- ture stability throughout the amplifier, there is no need for the two factors to be different. Practical considerations Although the foregoing, on the face of it, would lead to a near-ideal design, there are some practical problems. To start with, it is difficult to achieve low THD and low TIM in the same design: in practice, a compromise has to be sought. In the present design, this is found in an open-loop amplification of 2,300 and an open-loop bandwidth of 10 kHz. The amplification is sufficient to achieve acceptable THD figures. The goal of an open-loop bandwidth of 20 kHz or more proved impossible to achieve, however, in spite of extensive lead-compensation. Furthermore, the stability requirements meant severe limiting of phase shifts and this proved only possible by restricting the open- loop bandwidth to around 10 kHz. It should be noted, of course, that this is still an outstanding bandwidth: most commercial amplifiers with a high open- loop amplification (100,000 to 1,000,000) have an open-loop bandwidth of 30 to 50 or 60 Hz! The lag-compensating network, which determines the bandwidth, is located be- tween the branches of the first differen- tial amplifier. It would have been poss- ible to locate it between the inputs of that amplifier, but that would have meant taking back the feedback to the input also. And that in turn would result in the amplification becoming depen- dent, partly at least, on the characteris- tics of the preceding preamplifier. To make it possible for the amplifier to be DC-coupled throughout (to keep the a.c. and d.c. gains equal), a double FET was found necessary at the input: not an inexpensive solution, but one resulting in very good stability. It is true that the gain of a FET combination is on the low side, but in this particular design that does not matter. The slew rate in the practical design is kept to 50 V//ts. Again, this is on the safe side,, because in the prototypes slew rates of around 100 V//js were attainable. Fig. 1. General view of the LFA-150. +70V Fig. 2. Simplified circuit diagram of the LF'A-150. elektor india decembert988 12.29 The design The basic design may be assessed from the simplified circuit diagram in Fig. 2. It is split into two parts: a voltage ampli- fier and a current amplifier. The input of the voltage amplifier is formed by the dual FET already mentioned. The cascode circuit connected to the drains of the FETs not only enables the drain- source voltage of the FETs to be kept at a reasonable value, but also, more im- portantly, to eliminate to a large extent the internal drain-gate capacitance of the FETs, resulting in a substantial band- width. The first differential amplifier is fol- lowed by another, which is, however, constructed from discrete transistors and, moreover, is provided with a cur- rent mirror, Tm and Tn. The current mirror serves to provide a signal at B that is in phase with that at A. Network Rs-C.i provides lag compen- sation, while Cs and G> provide lead compensation. The current amplifier consists of a quiescent-current control around T20 and a symmetrical dual output stage, comprising a driver and two parallel- connected output transistors. Noteworthy in the output stage is that the output transistors are not connected as emitter followers but in a so-called compound configuration. In this, a sort of darlington is created which, due to a large amount of internal negative- feedback, combines very low distortion with a low output impedance. The stabilized power supply to the voltage amplifier is 4 V higher than that to the current amplifier, so that the voltage drop across the output tran- sistors remains small, even at maximum drive. Finally, the protection circuit serves to monitor the setting of the quiescent cur- rent level, the loudspeaker impedance, and the output current. Circuit description Each of the four unshaded parts in Fig. 3 is housed on a separate PCB. At the left is the voltage amplifier; beside it the current amplifier and protection cir- cuit; and at the top right the auxiliary power supply. Voltage amplifier. The input signal is ap- plied to differential amplifier Ti-T: via Ci (the only capacitor in the entire signal path) and low-pass filter R:-Cr. The filter has a cut-off frequency of about 200 kHz. It serves to limit the bandwidth, and thus the slew rate, before the signal is amplified. The differential amplifier is a dual FET housed in a metal case. The negative feedback voltage is applied to the gate of T:. Transistors T> and Tj and the FETs Fig. 3. Circuit diagram of the LFA-150. 12.30 elektor india december 1988 olektor india december 19B8 12.31 Parts list CURRENT AMPLIFIER 80ARD Resistors: R45=39R R46 = 2K74 + R47 = 1K0 + R48 = 47R R49;R50=56R R5o;R 67=100R; 1.5 W R5i;R53;Rsa;R60=2R2 R52;R54;Rs3;R8i "OR 2 2 non-inductive resistor R55;Re2=470R; 1.5 W R03 = 4R7; 1.5 W R04 = 22R; 1.5 W Res = 150R Ree=100R R 07 = 1 8K R 00 = 270R Ro9;R70 = 120R R 71 =47K P 4 - 1K0 multiturn preset ICermetl ' Metal film resistor Capacitors: C 25 — lOOn C2e;C27 = 680n C 28 = 27n; 250 V C29;C30 = 1 00p; 63 V C3t;C32 » 20,000p; 63 V (can-type capacitor: not on PCBI Semiconductors: Bi - BVW66 (not on PCBI D9 ;Dio= 1N4002 Dii;Oi2:Di3= 1N4148 T20=BD139 T 21 =2SC2238 T22 = 2SA968 T23;T24 = 2SA1095 T25;T26 = 2SC2565 T27 = BC556B T28;T29;T30=BC546B Miscellaneous: Ki= 10-W8y straight header. Li " 12 turns enamelled copper wire, dia. 1.5 mm: internal diameter approx. 15 mm. Re 1 = V23127-B0006-A201 (24 V change-over relay; Siemens) PCB 880092-2 . form a cascode circuit: Ti and Tr main- tain the drain potential, derived from’ divider R12-R13-R14-D1-D2, at about 20 V. The amplification of the input stage is restricted to 3.5 by resistors Rs and Rio. To keep the bandwidth at the output of the cascode circuit as large as possible, the values of collector resistors R 7 and Rs-Pi are fairly low. The preset. Pi, serves to eliminate any inequalities in the d.c. operating points. Lag compensation is provided by R5-C3. The capacitor determines the open-loop crossover point, while the resistor keeps the phase shift down. The d.c. operating point of the FETs is set by a constant-current source around Ts. Differential amplifier T 6 -T 7 , together with T* and T 9 , also form a cascode cir- cuit to keep the bandwidth as large as possible. The output of Ts is fed to the current amplifier via current mirror Tiu-Tn and terminal B. The signals at terminals A and B are, therefore, in phase with one another. Lead-compensation capacitors Cs and Cs serve to maximize the bandwidth of the second cascode circuit. Current amplifier. The current amplifier consists of drivers T21 and T22 followed by power transistors T23, T24, T25 and T 2 s, which, as already mentioned, are connected in a compound configuration. This section also provides a small voltage amplification due to resistors R57 and Rt2. The power transistors are protected by diodes Dv and Dio against any large negative voltage surges that may originate in the loudspeaker system. The d.c. operating point is provided by tran- sistor T 20 , which acts as an adjustable zener diode. This stage enables the set- ting of the voltage drop across T 21 , Rso, Rss, and T 22 , and thus that across resistors R 49 and R.w, which determine the quiescent current of the power tran- sistors. Transistor T 20 is mounted on the heat- sink for the drivers and power transistors to guarantee good thermal feedback: this ensures that the quiescent current re- mains steady even when the temperature rises. The quiescent current is about 100 mA per transistor, so that the output stages can comfortably handle small signals in class A. Boucherot network RM-C 28 ensures that the output is loaded even at high fre- quencies. Inductor Li limits current surges caused by predominantly capacitive loads at the output. The signal at the collectors of the power transistors is fed back to the gate of T 2 in the voltage amplifier via R«. The ratio R(,:R4 determines the voltage am- plification: with values as shown, this amounts to 3.5. The input sensitivity of the voltage amplifier is then 1.1 V r.m.s. Power supply. The power supply uses two mains transformers in series, Tri and Th. Note that Fig. 3 shows the power supply for a mono amplifier. Transformer Tn is a heavy-duty toroidal type with a centre-tapped secondary: each half delivers about 40 V a.c. Full- wave rectification is effected by bridge rectifier Bi and smoothing of the d.c. voltage is carried out by four 10,000 n? electrolytic capacitors: C31 and C32. The open-circuit supply voltage for the power transistors is about ±57 V; at full load, this drops to around ±51 V. The series connection of Tri and Tn provides a supply voltage of ±70 V for the voltage amplifier. This supply is regulated at ±60 V by discrete regulators T 12 to Tis and Tt« to T19 re- spectively. A differential amplifier in each regulator compares the output voltage with a zener-derived reference potential; any differences are eliminated by a darling- ton series regulator in the two supply lines. Presets P 2 and P 3 facilitate the set- ting of the respective voltage to their cor- rect level. Fig. 6. After the drivers and power transistors have been fitted to the heat sink, the cur- rent amplifier board is mounted above with the aid of spacers. elektor India dacember 1988 1 2.33 Protection circuit. The protection circuit will be described next month, but its connections to the other parts of the cir- cuit are already shown in Fig. 3. The output relay is located on the cur- rent amplifier board to ensure the shortest possible loudspeaker connec- tions. Transistors T 27 and T 30 monitor the cur- rent through emitter resistors Rm and R 59 respectively and, if necessary, actu- ate the protection circuit via TTs and T 29 . This happens when the output cur- rent exceeds 10 A. Practical design The sub-division of the circuit over four PCBs makes the construction rather easier to keep under control. The con- struction details will be given next month, but Fig. 1 gives some idea what the LFA-150 looks like. The PCBs have been designed in a way that makes it possible for three of them to be fixed together with the aid of suitable spacers. Only the PSU board is mounted by itself in the enclosure. The drivers, power transistors and T 20 are all screwed firmly to the heat sink with their terminals away from the heat sink. The current amplifier board is mounted on top of this arrangement (see Fig. 6), then the voltage amplifier board on top of that (see Fig. 7), and finally the protection board at the very top. All connections carrying large currents on the current amplifier board have been kept as short as possible. This explains the rather strange position of the output relay at the centre of the board. Fig. 7. The voltage amplifier board is mounted above the current amplifier board, again with the aid of suitable spacers. Fig. 8. Drilling diagram of heat sink. MAINS SIGNALLING by A.M. Karailiev Mains signalling is a method by which signals can be superimposed on mains wiring for remote control of electrical equipment. Typical applications are the control of street lighting, space heating, energy management systems and many other control switching applications in domestic, commercial and industrial premises. The operating principle of the proposed modulator is shown in Fig. 1. Two thyristors, Thi and Tha, are connected in parallel across a variable inductor. Thi and Thr are controlled via optical fibres connected to the control centre. 1 2.34 elektor india decembar 1988 Normally, the thyristors are open, so that the total alternating supply current passes through them and the inductor. In this state, there is no modulation, since the modulation voltage, Ua, is nought. This changes when one of the thyristors is closed, since then U 02 O.OIU- — the transmitter is sen- ding a signal into the mains network. In the simplest case, the receiver detects Ua, and switches on a certain load, or group of loads. The second thyristor is included to provide a complementary- phase signal Us that can be used for switching off the load(s). The main function of the variable inductance is to ensure that the amplitude of Us is virtu- ally independent of the load current. To achieve this, the control centre sets the required inductance with the aid of a servo-motor. Modulation type The modulation voltage should be not smaller than 0.01U- to allow a suitable rtttise margin for the receiver, and not greater than about 0.05U- to prevent it disturbing the operation of certain loads. In analogy with ordinary ampli- tude modulation, the modulation depth, or relative amplitude of Uq with respect to U«, is expressed as ... _ Us(max) The modulation method used in the above control system is less simple to qualify than would be expected. It could be called a special form of amplitude modulation, since the modulation voltage is unipolar, causing amplitude variation of either the positive or negative half cycles of the carrier voltage, but not both simultaneously. The system could also be considered as based on phase modulation , because it involves the sum of two amplitude- modulated voltages of equal frequency but opposite phase. Waveform modu- lation may be a suitable qualification because the modulation voltage, Uq, ef- fectively changes the waveform of the sinusoidal carrier voltage. Mathematical and experimental analyses of the spectrum of the modulated voltage supplied by the proposed system show that it consists mainly of ejven har- monics, among which the second, 2 f, dominates. In this regard, the change of the carrier waveform caused by the modulation could be qualified as distor- tion, and can be expressed as a distor- tion factor, k. It can be shown that this is roughly equal to the previously men- tioned modulation depth: But this is not a pure type of modu- lation. In order to improve the noise re- sistance of the receiver, the maximum positive and negative excursions of the carrier voltage have to be decreased in an alternative fashion, which has the basic elements of phase modulation. Being able to bring about a difference A\Jq between the positive and negative maximum excursions of U» by means of a modulator in a high-voltage line is essentally the same as injecting a signal of amplitude 0.5/1 Uc and of frequency 2 f — the frequency of the second har- monic. Normally, odd-numbered harmonics with k =0.03-0.04 are permanently pres- ent in many mains networks, while even- numbered harmonics appear only from time to time when large loads are switched on or off. Their total duration is relatively short at about 5% of the ‘quiet’ periods. Encoding system Encoding of the modulation signal is essential in view of the relatively high noise level on most mains networks. 1 he timing diagrams of Fig. 2a and 2b show how the control centre sends trigger pulses to the thyristors to switch a load on and off respectively. The load, or group of loads, controlled is selected by assigning a corresponding value to k. The block diagram of the receiver is given in Fig. 3, and the practical circuit elektor india december 1988 12.35 in Fig. 4. A full-wave rectifier drives an active filter. In the absence of a modu- lation voltage, all maximum excursions of the rectified voltage are of equal am- plitude, and there is no voltage at the output of the filter. When the mains net- work is modulated, the Filter supplies a sinusoidal output voltage, whose phase shifts 180° when the thyristors in the puts supplies the demodulated signal, Ua. Bistables, a load address decoder/filter and delay networks are then used to achieve reliable control of the power switch for the load. The filter is laid out in accordance with the load selection frequency: f=2/(k T) The cost of the transmitter and receiver compares favourably with existing units based on so-called ripple control of the mains voltage. Finally, Fig. 5 shows a suitable replace- ment for the variable inductor in the transmitter. It should be noted that the diodes have to be capable of handling Fig. 4. Circuit diagram of the receiver. k.1 h.1 k« I “I PI W I FI I ~~ t ■ ^ MgJ «-a I Fig. 5. Eight high-voltage diodes as a re- placement for the variable inductor in the transmitter. „ . .. r r— 5: MAINS ^ O-CZlh — I 7474 OilJS [j Ojui 880172-15 V ''' Fig. 6. Basic layout of a parallel modulator. transmitter change state (on/off con- trol). A phase detector compares the phase of the filter output signal with that of the mains voltage. One of its out- The possible number of load selection frequencies'is more than ten, but practi- cal needs normally seldom exceed about five. the total current demand of the load or group of loads. The block diagram of an alternative, parallel, modulator is shown in Fig. 6. H 1 2.36 elektor India december 1988 PORTABLE MIDI KEYBOARD A recently introduced integrated citcuit makes a dream of many electrophonics enthusiasts come true: to build one’s own MIDI keyboard around a handful of electronic components. The portability of the keyboard described makes it ideal for ‘first- aid’ testing of MIDI equipment. Moreover, in conjunction with a microcomputer, it can be used for practising, composing and editing musical pieces in places where a full-size keyboard is cumbersome to use. The Type E510 is a recently introduced integrated circuit that reduces the com- plexity of a MIDI keyboard to the extent that home construction of such a unit is at last within reach. Until recently, building one’s own MIDI keyboard was way of out reach of the average elec- trophonics enthusiast because of cost and complexity. At that time, even the simplest of do-it-yourself MIDI keyboard required building blocks such as a processor, random-access memory, read-only memory, high-precision mech- anical parts to ensure good dynamic key response (velocity), a musical keyboard, and a data entry keyboard, to mention but a few. The E510 can be used with a musical keyboard of ten octaves (128 keys) whose keys arc suitable for providing the vel- ocity information. The only auxiliary components needed are an EPROM loaded with transposition data, two binary decoders, and, of course, key contacts. The benefits of a portable keyboard are obvious: quick testing of MIDI instru- ment arrangements, practising and com- posing (parts of) musical pieces, par- ticipating in workshops, and trying out chords or tone combinations in situ- ations where a full-size keyboard simply takes up too much space. The miniature keyboard is also very useful for simulating a temporarily absent instru- ment or full-size keyboard for editing se- quences loaded in a sequencer, sounds in an expander, or scores in a computer sys- tem. Apart from its function as a versatile ac- cessory in the musical education field, the keyboard will aiso prove useful for experienced musicians whose principal instrument is, for example, the sax- ophone, the guitar or percussion — in any case, not the piano. Even if the mini keyboard serves as a mere gadget, it still deserves its very own place among far more complex MIDI equipment. MIDI KEYBOARD • Overall size geared to portable appli- cations. • Electronic circuit complies with MIDI standard (inch velocity). • Miniature keys and control circuit on compact double-sided PCB. • Range: 2 octaves and 1 note (25 keys); from C to C. • Switch-controlled transpose function over ±1 octave. • Switch-controlled MIDI channel selection (channel 1 or 2). • Simple to power from mains adaptor with DC output. • Low chip-count. MIDI keyboard: principle of operation The task of the MIDI keyboard is to detect the individual states of the keys to enable polyphonic playing. This means that a number of notes can simul- taneously appear or disappear, notes can last when others stop, and notes can ap- pear before others have disappeared. It is the aspect of polyphony that makes a musical keyboard functionally com- pletely different from, say, a computer or data entry keyboard. The ‘key state’ means that it is either re- leased (the corresponding contact is in the non-actuated, or rest position), pressed (the corresponding contact is ac- tuated), or in between these extremes. The time that lapses between the instant when a key is no longer in the rest pos- ition, and the instant it reaches the work position, is translated into a VELOCITY value. Evidently, the velocity at which the key is pressed is proportional to the ? Switch pole in the rest (non-actuated) position. ® t . 128 /us later, the pole has just left the rest position, and counter decrementing J, T commences. @ T- __A After 256 ^s, the pole has not yet reached the work contact, so counter decrementing continues (VELOCITY= VELOCITY- 1). T After n clock cycles, the pole has reached the work contact. © A Counter decrementing stops, the VELOCITY value is ^7 known, and MIDI code NOTE ON can be trans- mitted. Fig. 1. The main functions of the electronics in the MIDI keyboard are to analyse the pos- ition of the keys, and to measure the time that lapses between the opening and closing Of the contacts, for both directions of travel of the switch pole. Although in principle available on the small MIDI keyboard dis- cussed here, the latter function is, unfortu- nately, of no use because the relevant switches are of a type whose pole travel is vir- tually instantaneous rather than continuous. elector India december 1988 1 2.37 intensity with which the player strikes it. The softer the key is struck, the more time will lapse before the pole of the key has travelled from the rest contact to the work contact. This time is measured by counting down from 127 to 1 (see Fig. 1); the smaller the final count, the softer the key-touch. When it is detected that a key is no longer in the non-actuated position, nothing happens on the MIDI output of the keyboard. Counting down, however, commences or continues. Code NOTE ON is not transmitted until the pole reaches the work contact. If the minimum VELOCITY value is reached by decrementing before the pole reaches the work contact, it is assigned the lowest value, 1 . Basically the same hap- pens when a key pole leaves the work position to return to the rest position. The scanning of a MIDI keyboard thus entails the fastest possible analysis of the state of each key. In practice, this is achieved by an electronic circuit that works in combination with mechanical change-over (toggle) switches to derive key on/off and velocity information. MIDI keyboard controller Type E510 Figure 2 shows the internal structure and pinning of the programmed MIDI keyboard controller Type E510. The power supply is conventionally connec- ted to pins 8 and 16. The keyboard scan- ning signal and the timing of the serial MIDI data are derived from an on-chip clock oscillator that operates with an ex- ternal 4 MHz quartz crystal connected to pins 14 and 15 (pin 15 may be used for applying an external clock signal). The data rate at the MIDI output may be doubled by fitting an 8 MHz crystal. Pin 13 should always be connected to the positive supply line. Chip outputs AO to A6 allow the con- troller to scan up to 2 7 = 128 addresses (=keys). The MIDI data is available at output SO (pin 9). This output can be used in two ways: it can be made TTL- compatible by fitting a pull-up resistor, or it can function as a current-source by fitting a series resistor. The latter option is used here to give a MIDI-compatible current loop output. Input BE is connected to the ‘bused’ rest contacts of the switches. Similarly, input BS is connected to the work contacts of the switches. The pole of a switch addressed by the E510 is made logic low. During scan- ning, when the pole is at the rest pos- ition, the level of line BE is logic low in- stead of logic high (normal state due to pull-up). When the key pole has reached the work contact, BS goes logic low. Neither BS nor BE is low when the pole is anywhere between the rest and the THE MIDI STANDARD: A BRIEF RECAPITULATION The acronym MIDI stands for Musical Instrument Digital Interface. This standard has been designed to allow digitally-controlled musical instruments to communicate in a system (note that digital control often implies the use of a microprocessor or microcon- troller, although this is, of course, not always necessary). The MIDI interface is basical- ly a serial data link, based on a current loop. The data format is: 1 start bit, 8 data bits, and 1 stop bit. The data speed, 31.25 kilobits per second, is high relative to that used for many types of computer peripherals, but may still be too slow for real-time operations of a complexity beyond that of the most rudimentary types. The bulk of MIDI data is formed by the notes (events), played on a keyboard, or transmitted by an instrument. This recapitulation covers only MIDI events such as the NOTE ON and NOTE OFF messages. Of the three bytes in a ‘NOTE ON’ message, the second one carries the note value. With the MSB (most significant bit) set to 0 to indicate that the byte is a data type, this leaves only seven bits to carry the note value. This gives a range of 2 7 values, and these are assigned numbers from 1 to 127. The value of 60 is equivalent to the middle C. The interval between any two adjacent numbers is a semitone, so that a total compass of about ten and a half octaves is available. MIDI VALUES: NOTES 9. A 2 , r 4 . _ 3 . 6 . . A 8 . . 99 . A 2 . . 84 _ _ % _ 108 120 127 Cl C2 C3 C4 C5 C6 C7 C8 range of piano In addition to PITCH, the MIDI standard uses parameters NOTE ON and NOTE OFF (or KEY ON and KEY OFF). The first corresponds to the actuation of a key (or, in more general terms, to the start of the note), the second to the release of the key (end of the note). In reality, the relation between the duration of the note and the trans- mission of data NOTE ON and NOTE OFF is much more complex. Although the start of a note usually coincides with the transmission of code NOTE ON, the complemen- tary code, NOTE OFF, rarely marks the end of the note — usually, by the time NOTE OFF is transmitted, the note is already ended (in the case of a percussion sound without sustain), or it still sounds (long sustain). The third byte in the ‘NOTE ON’ message provides keyboard velocity information. Ranged in values from 1 to 127, the velocity is normally used to control the loudness of the notes (0=‘key off’; 1 = pianissimissimo — ppp; 127 = fortissimissimo — fff). It should be borne in mind, however, that there is no specified relationship between the velocity value and the loudness. If a MIDI instrument is not designed to handle velocity information, it adopts a default value (usually 64). 1 MIDI VALUES: VELOCITY 0 1 64 127 OFF ppp pp p mp mf f ff fff Since a single MIDI interface can be used for connecting several MIDI devices, pro- vision has been made to identify data to ensure it is correctly routed in multi-instrument set-ups. This data marking allows individual addressing of any instrument connected to a single MIDI interface. The MIDI standard specifies up to 16 channels, numbered 0 through 15 (sometimes 1 through 16), which means that any one of up to 16 instru- ments can be controlled independently and individually. In the case of the NOTE ON and NOTE OFF information, the channel number forms part of the ON or OFF code. 01234567 01234567 01234567 0 - Channel t 0 - KEY OFF 1 = Channel 2 1 = KEY ON The above diagram shows a MIDI message sent by a keyboard when a a key is actuated. The start bit is followed by an 8-bit word, in which the first 4 bits (0 through 3; least significant nibble) indicate the channel number (the keyboard described in this article can drive only two channels). The last bit, (number 7; most significant bit) is logic high to indicate that the byte sent represents status information, i.e., it is not, strictly speak- ing, adafaword. The logic level of bit 4 provides the KEY ON/OFF (NOTE ON/GFF) information: 0=OFF; l=ON. The six bits that indicate the key number follow the start bit of the second byte. Bit 7 of a databyte is always logic low. The six bits of the third byte (second databyte) hold the velocity information. Bit 7 is logic low to mark that the byte it forms part of is still a databyte. In the present case, the MIDI message is terminated with the stop bit of the third byte. 1 2.38 elektor india december 1988 work contact. The above arrangement is summarized in Table 1. The logic level at chip input CO (pin 12) determines the current MIDI channel: C0=0=channel 0; C0=l = channel 1. not analyzed pole at rest contact pole at work contact pole travelling impossible Fig. 2. Internal structure and pinning of the polyphonic MIDI keyboard controller Type E510. This chip supports the use of a 128-key keyboard with up to 10 octaves, and transmits MIDI values for VEIDCITY, NOTE ON and NOTE OFF. Circuit description The crucial components in the circuit diagram of Fig. 3 are controller ICi (E510) and decoders/demultiplexers IC 3 and IC4. EPROM IC 2 has the auxiliary function of code converter. The operation of the circuit is best understood if IC2 is initially ignored. It is assumed, therefore, that the address outputs of ICi drive IC3 and IC4 direct. On its outputs AO to A6, the E510 counts from 0 to 127. Each time the counter is incremented, another output on IC3, and then IC4, goes low. This cyclic counting up forms the scanning of the keyboard. Each time the E510 pulls one of its address lines logic low, it reads back the logic levels of lines BS and BE to determine the current state of the ad- dressed key. This state is combined with that read during a previous scan (i.e., 128 ps earlier at /xtal=4 MHz). The Fig. 3. Circuit diagram of the small MIDI keyboard. elektor india decern ber 1988 12.39 result of the combination is deduced as shown in Table 2. That the keyboard described here has 25 instead of the maximum number of 128 matters very little as far as the elec- tronics are concerned, since BE and BS simply remain logic high simultaneously for the 103 non-existing keys, and state BE = BS = 1 is effectively ignored by the E510. Although the contact travel time of the Digitast keys used in the MIDI keyboard can be measured with some precision, it will be found that this is largely constant, i.e., hardly subject to applied force. This is because Digitast keys have tactile feedback (they produce a click when pressed). The upshot of it is that the VELOCITY value transmitted by the standard version of the keyboard is of no practical use. As already noted, EPROM IC 2 func- tions as a code converter in the present circuit. The E510 counts cyclically from 0 to 127. In the absence of the EPROM, the two octaves of the keyboard would be comprised in the lowest range of the scale covered by the PITCH parameter, i.e., between note 0 and 24. Also, double addressing of the decoders in the circuit would cause a single key to provide several, different, MIDI codes simul- taneously. The task of the EPROM is, therefore, to ignore the lowest of the ad- dress codes, and to activate the two decoders (74HCT154) only once when the counting has reached a value that corresponds to audible notes in the mid- dle of the useful range. The second duty of the EPROM is to switch between two address ranges, which results in the transpose function. This is effectively done with the aid of a toggle switch with a centre contact, S 26 , that determines the logic level on EPROM address inputs A7 and A8. The EPROM converts the addresses supplied by the E510 by adding or subtracting the equivalent of one octave. For example, when the address of note 60 is applied, the EPROM converts this to an address that corresponds to note 72, one octave higher. The contents of the EPROM are listed in Table 3. A Type 2764 is used here because this is currently the least ex- pensive EPROM. Split programming extension Switch S 27 determines the channel selec- tion by controlling the logic level aplied to input CO of the E510. Instead of manually giving a channel selection command, it is also possible to do this via the keyboard by splitting this into zones. Figure 4 shows the circuit diagram of the optional extension to achieve this. Notes played to the left or the right of the split go to MIDI chan- nel 1 or 2 respectively. The split is defin- ed by pressing the PROGRAM switch together with the desired key on the Fig. 4. Optional add-on circuit to achieve programmable split zoning. 12.40 elektor india december 1988 Table 2. 0 = active count = decrement counter Parts list Resistors (± 5%): Ri . . .Rs incl. = 1 KO R6;R7 = 22QR Capacitors: Ci;C2 = 22p C3;Cs;Ce=2p2; 25 V; tantalum C4 = 100n Semiconductors: Di . . .025 incl. = 1N4148 ICi =E510 + IC2 = 2764 IESS567; IC3;IC4 = 74HCT1 54 IC5 = 7805 Miscellaneous: S 1 . . . S 25 incl. = miniature Digitast toggle ISPDTI key. 1 S 20 — miniature toggle (SPDT) switch with centre position. S 27 = miniature SPDT switch. Xt = quartz crystal 4.00 MHz PCS Type 880168 Fig. 5. Component mounting plan of the printed circuit board for (lie MIDI keyboard. elektor india december 1988 1 2.41 keyboard. The corresponding key num- ber is then latched into the 74HCT373 octal bistable. Byte comparator 74HCT688 drives input CO of the E510 logic low when the current key code is greater than that of the split, which is read from the latch. The programmable split option is not supported on the printed circuit board for the MIDI keyboard, since this was desired to re- main as small as possible. Construction The following constructional description is slightly more elaborate than usual to enable anyone, even those with only limited experience in the electronics field, to build the keyboard successfully. Prototype of the MIDI keyboard Table 3 TRANSPOSITION DATA Addresses applied to transposition EPROM: S26 count from 0 to 128 > CD > A6 A5A4 A3 A2 A1 AO note C S26 0 1 0 10 0 10 0 n° 36 2 -1 oct. 1 0 0 11110 0 n° 60 4 4-1 OCt. 1 1 0 1 1 0 0 0 0 n° 48 3 normal Output data supplied by transposition EPROM: D7 D6 D5 D4 D3 D2 01 DO hex NC NC 0 1 0 0 0 0 10 r keys 0 to 15 NC NC 0 1 1 1 1 1 l 74HCT154/IC3 NC NC 1 0 0 0 0 0 r keys 16 to 25 NC NC 1 0 1 0 0 0 28 1 74HCT154/IC4 Table 4. EPROM CONTENTS 0 1 2 3 4 5 6 7 8 9 A B c D E F 00A 10 11 12 13 14 15 16 17 18 19 1A IB 00B 1C ID IE IF 20 21 22 23 24 25 26 27 28 013 10 11 12 13 014 14 15 16 17 18 19 1A IB 1C ID IE IF 20 21 22 23 015 24 25 26 27 28 01B 10 11 12 13 14 15 16 17 18 19 1 A IB 1C ID IE IF 01C 20 21 22 23 24 25 26 27 28 Addresses not given are left blank (FF) Construction is not difficult, but re- quires great care and precision because components are fitted at both sides of the printed circuit board, which is quite densely populated. The first thing to note is that the compo- nents, with the exception of the keys, are fitted at the track side of the board. The holes in the board are intended for the wires and the keys. All other component terminals are cut to a suitable length, preformed, and soldered direct to the rel- evant copper islands. Commence construction at the track side by fitting the two wire links: one be- tween Rr> and R-, and the other, a very short one, close to solder terminal a of S 26 (in both cases, use insulated wire to prevent a short-circuit with tracks runn- ing below'). Next, mount the 25 Digitast keys at the reverse side of the board (note that a number of keys can not be soldered any more once the integrated circuits have been fitted). Study the orientation of each and every diode before fitting it! Mount the solder terminals for the wires to the board (MIDI output and power supply), and then those for Sir. and S 2 -. Depending on personal preference, these switches are either mounted direct on to the board (to the right of the Digitast keys), or on the front panel of the en- closure that houses the MIDI keyboard. Cut the terminals of voltage regulator ICs to a length of about 3 mm from the enclosure, bend them over, and place their ends on the spots provided. Insert an insulating mica washer between the metal tab of the regulator and the PCB surface. Secure the regulator with a short M3 bolt and nut. Take great care to avoid short-circuits be- tween component terminals and nearby tracks. Make sure that the leads of the quartz crystal are left long enough to en- able the metal enclosure to be bent towards the PCB without touching the solder joints below. Bend D 25 slightly away from the crystal enclosure. There is no objection to confident and experienced constructors soldering the integrated circuits direct on to the board. If you arc hesitant about doing this, however, use low-profile IC sockets. Since the E510 may have to be removed for use later in a full-size touch-sensitive keyboard (see below), it is recommended in all cases to fit this IC in a socket. Finally, be sure to use good-quality strain reliefs for the MIDI output and supply cables. From mini to full-size The electronics in the MIDI keyboard is suitable for connecting to a ‘real’ keyboard, i.e., one of standard size and having change-over key contacts of a quality that ensures equal VELOCITY values over the full keyboard range. The function of a sustain pedal can be created by inserting a push-to-break but- ton in the pulled-up BE line to the E510. This switch, when pressed, prevents the E510 from detecting that actuated keys have returned to their rest position, in which case BE is logic low. K 12.42 elcktor india december 1988 OPTOELECTRONICS by K. Roberts, BA Optoelectronics is one of the fastest growing branches of electronics and British research and development is leading the world in many of its facets. Although this technical excellence is not (yet) matched in the commercial sector, companies operating in the opto-electronics field are increasingly exploring export markets and seeking international collaboration. Already, many export more than half their production and some export nearly all of it. Optoelectronics may be defined as the technology that makes use of the interac- tion between photons (small packets of light energy) and electrons. The study and science of this interaction is called photoelectronics, often, particularly in the USA, contracted to photonics. Broadly speaking, optoelectronic prod- ucts may be categorized into sensors (responders to light), emitters (of light), and users (of light), which arc often a combination of the first two. Sensors comprise, among others, photo- cells, also called light-dependent resistors (LDR); solar cells; photo- diodes; and phototransistors. Emitters comprise ordinary light bulbs; light-emitting diodes (LED); gas dis- charge tubes; lasers; electroluminescent displays; and cathode ray tubes. Users comprise optocouplers, sometimes called opto-isolators; infra-red alarm and remote control systems; security in- stallations; and metrological devices. Sensors An LDR (photocell) consists of a thin polycrystallinc film of cadmium- sulphate sandwiched between two metal 880023-2 Fig. I. Typical infra-red sensor. contacts. The potential across these con- tacts is directly proportional to the cur- rent flowing through the contacts. The conductivity of the cadmium-sulphate increases greatly (by a factor of about 10 5 ) when it is subjected to elec- tromagnetic radiation of a wavelength between, roughly, 3xl0 _lt m and 3xl0~ 5 m. This results in a photocur- rent, superimposed on the small dark current, flowing in an external circuit. Fig. 2. Some typical phototransistors with in the centre an infra-red photodiode. A solar cell is a photovoltaic device that converts light directly into electrical energy. It is essentially a p-n junction: by far the largest number of solar cells cur- rently manufactured are made from crystalline silicon. Others arc made from amorphous silicon, copper sulphide- cadmium sulphide, gallium arsenide, or cadmium-selenium. See Ref. 1. A photodiode, either of the depletion- layer or of the avalanche type, has itsp-w junction exposed to external light. The depletion layer type, operated below its break-down voltage, produces excess Fig. 3. Basic construction of silicon solar cell. electron-hole pairs when radiation in the UV-IR region falls on to the junction. The pairs in or near the depletion layer cross the junction and produce a photo- current. In the avalanche type, operated above its break-down voltage, current multiplication of the electron-hole pairs generated by incident illumination en- sues owing to avalanche break-down. A phototransistor is a bipolar junction transistor whose junctions are exposed to external light. It is normally operated in the common-emitter configuration. When light in the UV-IR region falls on to the junction, a base current is pror duced, and the normal current- amplifying action causes a greatly amplified collector current. The phototransistor is, of course, far more sensitive than the photodiode. Emitters A light-emitting diode (LED) is a p-n junction that emits light as a result of recombination of excess electron-hole pairs. The emission is normally a fairly narrow bandwidth of visible (red, orange, yellow or green) or infra-red light. The colour is a function of the semiconductor material used for the junction. LEDs typically require for- ward operating voltages of about 2 V and forward currents of 10 to 20 mA. A gas-discharge, or fluorescent, tube elektor India december 1988 1 2.43 normally contains a small amount of argon together with a little mercury. It has two electrodes (filaments) that are coated with a mixture of barium and strontium oxides. The resistance between the electrodes is high until the gas is ionized. Gas ionization is usually brought about by the application of a very high voltage (of the order of 1500 - 2000 V) across the two electrodes. The high voltage is normally induced across a starter choke by the sudden disruption of the current through the choke. The laser was developed by Theodore Maiman in 1960. Light emitted by a laser differs from normal light in two important respects: it is coherent, i.e., all the photons are in phase; and it is of one frequency only. There are many types of laser: small out- put He-Ne lasers, primarily intended for use in laboratories; argon-ion lasers for medical applications; carbon-dioxide lasers for industrial uses; dye lasers for use in spectroscopy; high output Nd/YAG lasers for surgical applications excimer lasers for use in chemical analysis and semiconductor processing; and, most common of all, semiconduc- tor, or injection, lasers. The semicon- ductor laser is of prime importance in modern (fibre optic) communications, optical memories, and compact disc players. See Ref. 2 and 3. Electroluminescent displays make use of the ability of phosphorus to emit light continuously when a voltage is applied to it. The most commonly found appli- cation of this phenomenon is in the screen of a cathode ray tube as used in many hundreds of millions of TV sets the world over, not to mention the millions of computer monitors and oscilloscopes. Such a display consists of a sandwich of a luminescent-phophorus layer and two transparent metal films. When an a.c. voltage is applied to the films, the phosphorus glows through the films. big- <>• Basie construction of electrolumi- nescent display. anode terminal capillaries ' -\ h HHr' — d r partly transparent mirror He-Ne gas mixture outer glass shell cathode glass-metal weld cathode terminal Brewster window , (only in polarized lasers) totally reflecting mirror mirror holder length of resonant cavity 87128-9 Fig. 7. Cross-sectional view of a He-Ne laser (courtesy of Siemens). 12.44 elektor india decembor 1988 Fig. 8. Outline drawings of two types of optocoupler; the one on the right uses a darlington phototransistor for increased sensitivity. Users of light Optocouplers, sometimes called opto- isolators, are devices that consist basi- cally of a light-emitting diode and a phototransistor that are optically coupled within a light-excluding case. Optocouplers may be used with digital or analogue signals. They are normally specified by their isolation voltage (also called common mode rejection -CMR), speed (propagation delay), and forward coupling, normally called current transfer ratio-CTR. A typical, good- quality optocoupler has a CMR of about 2 kV, a propagation delay of around 5 ns, and a CTR, expressed as the ratio of the output current to the input cur- rent, of 30°7o. Alarm and security systems often de- pend on the combination of an optoelec- tronic sensor and emitter, usually operating with infra-red light. Such systems may use a single or dual light- beam transmitter-receiver, in which the receiver is switched on to actuate an alarm when the beam from the trans- mitter is broken by an object. There are also systems that operate by reflecting the light beam back to an integral light emitter-sensor with the aid of a prismatic mirror (which simplifies align- ment as compared with a plane mirror). Infra-red remote control systems for use with TV receivers and audio systems, to name but a few, also use a transmitter (hand held) and receiver (located in the equipment to be controlled). The trans- mitter is usually controlled by a five-bit (for 32 codes) or a six-bit (for 64 codes) keypad. The code is transmitted by a number of IR LEDs. The coded signals are received by a photocell and fed to a decoder in the receiver. The decoder usually provides both digital outputs (channel changing, loudspeaker muting) and analogue outputs (volume control). The future As already stated in the introduction to this article, Britain is in the forefront of optoelectronic research and develop- ment. Unfortunately, many companies in the industry find it increasingly diffi- cult, even more so than other electronics concerns, to find suitably trained and qualified staff. In line with the increased outward look- ing of the optoelectronics industry is the widespread participation in European research programmes such as EUREKA, ESPRIT, and RACE in order to spread the cost of R&D. There is, furthermore, a growing co-operation between op- toelectronics companies and universities, actively encouraged by the government. Of the many universities, particularly Heriot-Watt and Southampton Univer- sities are in the vanguard of optoelec- tronics research. Moreover, the Royal Signals and Radar Establishment - RSRE-at Malvern is one of the world’s leading defence research establishments. The Government has set up Defence Technology Enterprises (DTE) to exploit commercially the research undertaken at places such as RSRE. The Government has also co-operated in the setting up of the Optical Sensor Collaborative Association - OSC A . Apart from well-established companies such as Plessey, GEC, STC, Barr & Stroud, Ferranti, and many others, a fairly new powerful force in the British optoelectronics industry is British Telecomms’ joint venture with the American giant Du Pont, called BT&D Technologies. Most of the British industry is moving towards metallorganic chemical vapour deposition - MOCVD - manufacturing techniques that will make possible true mass production of complex optoelec- tronic devices. There also appears to be a bright future for non-linear optical switches made from lithium niobate. Barr & Stroud, a Pilkington company, markets a range of lithium niobate optical ICs, including phase modulators, intensity modulators, and directional couplers operating in the IR region. Although the UK is not so strong in laser manufacture. Philips Components, for- merly Mullard, supplies most of the CD lasers required by its parent company Philips of Holland. The largest producer of thermionic valves in Europe is EEV, a GEC company. Apart from thermionic valves, Fig. 9 Basic outline of an infra-red remote-control transmitter-receiver. elektor india decern ber 1988 12.45 this company, which employs 2,700 per- sonnel and has annual sales of some £80 million, manufactures liquid-crystal dis- plays (LCDs), charge-coupled devices (CCDs), and image intensifies made in MOCVD. Another worldleader is the Midlands company of Hadland, which designs and manufactures high-speed electronic cameras and image converters. One of their cameras is said to run at over 600 million frames per second! Pundits reckon that the world market for optoelectronic devices will grow from under £300 million in 1987 to around £800 million by 1992. Given the state of our research and development, there must be many rich pickings there for the British optoelectronics industry. OPTOELECTRONICS BRIEF A very Intelligent Computer Terminal by Bill Pressdee Over the last decade the thrust of inno- vation in information technology has moved gradually from mainframes towards minicomputers and specialist workstations. It is not surprising, therefore, that intelligent terminals systems development is now of con- siderable interest. Lynwood Scientific Developments (1) of London, which claims to be the first in- dependent European company to embed a microprocessor in a video display unit, is a world leader in the design and manufacture of sophisticated display technology. Moreover, it has an enviable reputation in the development of specialized systems for meeting clients’ specific needs and for forming joint teams with them. Nevertheless, it has a low market profile, possibly because it is involved in military projects with a high security classifi- cation. In fact, one is more likely, perhaps, to come across an original Lyn- wood terminal to which a major elec- tronics manufacturer has attached its logo. The philosophy adopted by Lynwood in manufacture involves concentrating op- erations into three specialist units: short run special products, such as those built for the Ministry of Defence; high volume runs; and production of logic boards and sub-assemblies. The company is unusual in today’s world of volume output in that it maintains an ex- tensive capability for specialized systems complementary to its hardware develop- ment. The j300 enables graphics to be shown together with file data — a possible extension to a criminal records facility. 12.46 elektor india december 1988 Sophisticated communications software The successful Lynwood Alpha and Beta terminals have now been superseded by the latest range: the j 102, j300, j500 and j700. Based on the Alpha terminal, these new displays represent an updating with a large number of systems uses in mind. The new Lynwood j300 high definition, intelligent display has an embedded Zilog Z8001 processor and 256k of dis- play memory that enables sophisticated video and communications software to be used. Programs can be supplied for two or three emulators to operate within the same terminal concurrently, using split or virtual screens. Other programs can support a variety of communi- cations protocols. Through multiple ports, simultaneous communications can be maintained with separate host computers. Attached peripherals which make up the work- station for a system — such as various types of reader, letter quality printers, security devices and dispensers — can be controlled by the terminal, and the dis- play can be used for local calculations or to execute terminal resident tasks. The unit is compact and great attention has been paid to sound ergonomic design. Considerable care has also been taken in the placement of components which, coupled to the optional provision of fibre optic interfaces, is a clue that the company also supplies a version of the j300 to full Tempest security specifi- cation. Banking applications The largest terminal is a 482 mm colour display with a resolution of 1280 by 1024 4-bit pixels. Graphics functions are ex- ecuted by a powerful controller assisted by the terminal’s Motorola 68010 pro- cessor incorporating a 2 Mb memory. Optionally, a Motorola 68020 and a 4 Mb memory can be provided. A large frame buffer can hold three 1280 by 1024 displays and the screen image may be built from selected parts of the frame buffer. This arrangement is designed for command and control applications where a static 16-colour geographic background can be overlaid with another 16-colour plane of dynamic in- formation. A system developed by Lynwood, in- itially in conjunction with the United Bank of Kuwait, has recently been adopted by a number of other banks in the Middle East. This sophisticated on- line teller system can display at a teller position all relevant account details and authorized signatures appropriate to an account. The records are created via the signature capture station, using a facsimile reader in conjunction with a j300 which has a security badge reader so that an audit trail of signature entry and authoriza- tion can be recorded. The informatiin is sent as an ASCII alpha string to a database held on a separate processor that interfaces to whatever host pro- cessor the bank may use. In another application Lynwood is pro- viding integrated dealer workstations to various banks including the United Bank of Kuwait. These provide a num- ber of screens controlled by a single keyboard and by virtue of its excellent emulation and communications pro- grams, information from a plethora of sources can be co-ordinated and cor- related. They may be from in-house computers such as IBM, DEC, DG, Tandem, ICL and NCR or external networks, from Telex, via service gateways, and external services, such as Reuters, Telerate, Topic and Datastream while a page cache can be held within the workstation. Text and graphics can also be displayed on a com- mon screen and controlled from the keyboard. Machine-readable passports Another interesting use for the terminals is a machine-readable passport system developed in conjunction with De La Rue, the leading currency and passport printer, and marketed by De La Rue Identity Systems l2) of Basingstoke. It enables laborious manual infilling to be replaced by a flexible issuing system which is much faster and automatically logs the number of passports issued. At the point of entry to a country, immi- gration throughput is greatly improved and monitored information may be automatically recorded leading to im- proved security and control of visitors. Lynwood terminals have enjoyed con- siderable popularity with metropolitan police forces in the United Kingdom and other parts of the world. The terminals used in the London police system have been upgraded to interface with the X25 networking protocol of METNET. The facility for multi-emulation enables the terminals to tap a number of resources to provide a criminal record in- formation system with inputs from the Police Command network, the Police National Computer, and HOLMES, the Home Office serious crime investigation matching profiles dossier. These can be correlated on a single video display unit (VDU). At present, photographs of criminals have not yet been introduced into such system but it is only a matter of time as the technology is there to be used. Terminals such as the j300 contain suf- ficient memory and processing power to have a significant effect on the overall system design. Not only can they help to reduce some of the problems, mainly concerned with real time, that are associ- ated with certain types of computer sys- tem, but they can also provide a much more effective man-machine interface that can be tailored to the needs of in- dividual users. To harness the hardware, the company has developed a range of software that is flexible in its approach to both com- munications and emulations. In ad- dition, the company has examined specific applications to determine whether generalized packages can be developed that are project independent and capable of handling the hardware in a very flexible manner. One of these, TRAPS, is a terminal- resident software package that enables a format-driven dialogue to be generated at the terminal. With the use of its pro- cessing power and memory, the most ef- fective dialogue is produced while minimizing the demands on the host processor and communications network. Format-driven dialogues are particularly successful because they enable the user to be led through a set procedure with the input data being validated at all stages. Keystrokes and entry time are re- duced by using prediction tables that an- ticipate characterstring matching within the validation. References. (l) Lynwood Scientific Developments Ltd, Unit Five, Bowling Green Lane, London EC 1R 0BD. I2) De La Rue Identity Systems Ltd, De La Rue House, Basing View, Bas- ingstoke RG21 2EL. elektor india decembar 1988 1 2.47 BUS INTERFACE FOR HIGH- RESOLUTION LIQUID CRYSTAL SCREENS Part 1 Although large LC display modules are currently available in many shapes and sizes, their special serial input calls for the use of an interface to enable connection to a computer bus. The interface described here is versatile, yet relatively simple to configure and program as a bus-connected device in a number of popular computer systems. Although the application discussed concentrates mainly on the 400x64 dot matrix LCD module Type LM40001 from Sharp, the interface board is also suitable for a number of similar units in Hitachi’s LM series. Liquid crystal display (LCD) units for text and graphics applications are usually supplied as a module consisting of a glass-protected, reflective backplane (the actual display), and a controller board attached to it at the rear side. The controller translates the data applied to its serial input into backplane waveforms, which result in dot patterns that form legible characters or graphic shapes. In most cases, LC controllers have an on-board character ROM. The serial format used for controlling LC display modules is usually of a type that bears no resemblance whatsoever to that adopted for, say, the well-known RS232 port. When a large, intelligent LC dis- play module, such as the LM40001, is to be used in conjunction with a computer, an interface circuit is required as de- scribed in this article. Among the computers that can be con- nected to the present interface are • 6502-based systems (C64, C128, Acorn computers); • Z80-based systems (CP/M and MSX computers); • IBM PCs and compatibles; • the Elektor Electronics BASIC com- puter (l> . Significantly, the LCD interface can be controlled entirely in BASIC. Liquid crystal screens Although Sharp’s Type LM40001 is, strictly speaking, a liquid crystal display module, it is better qualified as a liquid crystal screen because of its large viewable area (220x35 mm), and its ability to process data as graphics infor- mation (individual dots can be ad- dressed). This is in contrast to most 12.48 eloktor India decembor 1988 smaller LCD units, which are usually only capable of displaying text and numbers on 1, 2 or sometimes 4 lines, depending on the size. An LC screen is essentially a dot-matrix display unit without predefined characters. The interface described here, in conjunction with the existing backplane controller on the LC screen module, makes it possible to combine dot patterns into legible characters, just as on a TV screen, or a dot-matrix printer. Although the prototype of the interface was developed, tested and used in con- junction with the LM40001 from Sharp, it can also be connected direct to Hitachi’s Types LM200, LM021, LM212 and LM211. These, and similar units from other manufacturers, are occasion- ally offered inexpensively at rallies and in surplus stores (but make sure you ob- tain the relevant data sheets). Principle of operation The block diagram of Fig. 1 shows that an address decoder is required to ‘map’ the LC screen in the computer’s mem- ory. Depending on the type of processor in the computer, this address can be in actual memory (e.g., 6502-based Fig. 1. Block diagram of (he universal LC screen interface. systems), or in the I/O segment (e.g., Z80-based systems). The configuration logic, shown as a separate block in Fig. 1, is required to ensure correct timing and combination of the pulses for the interface. The 8 Kbyte RAM block is divided in two 4 Kbyte segments by a dedicated LCD controller chip, the Type HD61830B from Hitachi. In text mode, each 4 Kbyte screen memory holds the data for ten text windows. In graphics mode, the same memory holds one graphics screen. The difference in storage capacity between the text mode and the graphics mode is brought about by the fact that any ASCII character (a complex dot pattern) can be called up by only one byte, whereas, in graphics mode, that same single byte produces only a horizontal row of 8 dots. The internal division of the screen mem- ory in displayable windows is shown in Fig. 2. An external control signal pro- vided by a latch divides the memory in two equal halves. The start address determines which 400 memory locations appear on screen (‘display window’). Hence, there are 10 text screens (4096/400). The on-screen location of the next character loaded is determined by the cursor address, which is automati- cally incremented by one after the con- troller has displayed the current charac- ter. Starting at cursor address 0, and assum- ing that the start address is not changed, characters following number 400 will not be displayed, but are still loaded in the screen momory. They become visible only when the start address is moved up accordingly (‘scrolling’). Old data then disappears from the screen, but remains in the screen memory. Memory location 0 is overwritten, however, with new data Fig. 2. Each screen memory of 4 Kbytes is subdivided in ten windows of 400 bytes. when the screen memory is full. Scroll- ing per line or per screen is simple to ef- fect by incrementing the cursor start ad- dress in steps of 8 or 50, respectively, assuming that the LC screen is pro- grammed to display 8 lines of 50 characters. In graphics mode, 50 bytes are required for 400 dots horizontally. The vertical resolution is 64 dots, so that one graphics screen corresponds to 50x64=3200 bytes. This means that the screen memory (4 Kbyte) can hold one graphics screen with 896 bytes left. The controller used in the interface cir- cuit is a relatively complex chip. It has a built-in character ROM, and takes care of the parallel-to-serial conversion of the data provided by the computer interface circuit. When an (optional) external EPROM is added, the user has a choice of three character fonts. The last block in Fig. 1 is the contrast control circuit. A discrete 4-bit DAC is driven via a register, and provides a 16- levei contrast setting. The directly ad- dressable register is also used for switching between the two 4 Kbyte screen memories, and the two EPROM- resident fonts, which are optional. Circuit description The complexity of the circuit shown in Fig. 3 is only apparent, and caused mainly by the ability of the interface to be driven by various types of computer. Connector Ki links the computer’s CPU to the LC screen interface. Circuits IC« and ICt, together with 8-way DIL switch blocks, form a presettable 16-bit address decoder for mapping the card in the computer’s memory. When the bit pattern set by means of the DIL switches match es that on the address bus, output P=Q goes low. The least significant three address lines are not connected to the address decoder, and appear as X0, XI and X2 on the internal bus of the in- terface. This arrangement allows the combining of system-dependent signals with the address decoding. In the case of the IBM PC, for instance, X0 carries signal AEN. Si milarly, with MSX systems, X2 carries IOREQ, and X0 bus signal Ml (XI is not used, jumper R is not fitted). The interface occupies 8 memory locations, 5 of which are used for addressing registers — see Table 1. Circuit ICs functions as a bidirectional databus buffer. Together with ICt, a decoder for internal signals, it is enabled when the interface is selected via the computer’s address bus. Gates Ni to N6 convert and combine the control signals provided by the microprocessor bus. Table 2 provides bus connection infor- mation, and lists the configuration of jumpers A to T, in accordance with the computer system used. Interface output WAIT is provided to elektor india december 1988 1 2.49 Fig. 3. Circuit diagram of the universal interface for high-resolution liquid crystal screens. The configuration of the jumpers is in accordance with the type of computer used for driving the circuit. ensure correct operation of the con- troller when this is connected to a rela- tively ‘fast’ computer bus. Bistable FFi is a monost able multivibrator which pulls WAIT low via FET T:. Its monotime is a bout 45 0 ns, as set with network Rt-C:. WAIT is an open-drain line that can be connected to an existing wired-AND network as used in IBM PCs 12.50 elektor india decemOer 1988 (8088/80 86) and MSX micros (Z80). Ob- viously, WAIT is not used in systems whe re it is n ot required. The RESET input of the control ler, ICs, is connected to the CPU RESET line via a low-pass filter, R12-C1, which serves to suppress spurious pulses. Circuit ICn is the previously discussed latch for the contrast setting circuit. Its outputs, Q0 to Q3, drive the discrete DAC, R(.-R‘)-Ti. Only two of the remain- ing four outputs of ICn are used — Q6 and Q7 as the RAM as the 4 Kbyte selec- tion lines, A12, of the screen RAM (IC9) and character EPROM (ICio) re- spectively. It is possible to store 4 screen fonts in EPROM by using the 16 Kbyte 27128, and one of the two remaining outputs on ICu as the 14th address line. The clock generator for the controller chip is formed by an R-C oscillator, Ns. The actual clock frequency is not so im- portant, but a symmetrical clock signal is a must for the HD61830B, hence the use of divide-by-two bistable FF 2 . One slightly unusual connection in the interface circuit is that of input R/-W of the controller to address line Ai. This solution was adopted to solve possible timing problems. Read and write levels should be available 140 ns before the en- able pulse, which, in turn, should have a minimum duration of 440 ns. The con- nection of Al to R/-W results in differ- ent addresses for read and write oper- ations to the interface registers — see Table 1. Construction The LC screen interface is constructed on a double-sided, through-plated printed circuit board (see Fig. 4). The track layout is not given here because this PCB is virtually impossible to make other than from films, while through- plating equipment is usually only available in a professional workshop. The size of the ready-made PCB is such that it can be attached to the controller board of the LM40001 unit with the aid of 4 spacers. To be continued next month. BBBQBBflB BBBBBBBB Fig. 4. Component mounting plan of the double-sided, through-plated PCB for the LC. screen interface. This high-quality PCB is available ready-made through the Readers Services. Resistors (±5%): Ri;R5;Ria = 10K R 2 ;R 3 ;R 4 = SIL resistor array 8x10K Re = 8K2 R7 = 3K9 Rs = 2K2 Re;Rio;Rn = 1K0 R 12 -I 8 K Capacitors: Cl =330n C 2 = 27p C3 = 56p C 4 . . . C 14 incl. = 100n Semiconductors: Ti = BC547B T2 = BS!70 ICi;IC2 = 74HCT132 IC3-74CT139 IC4 = 74HCT74 IC5 = 74HCT245 !Ce;IC7 = 74HCT688 ICs = HD6l830B (Hitachi)' iCe = 6264 or 8264 8K x8 static RAM ICio=2764 {optional character set EPROM, ICti = 74HCT377 Miscellaneous: Si . . .Sa incl.jSe. . .S 16 lncl.= 8-way DIL switch block. Ki = 40-way PCB header with eject handles: male; with angled pins for PCB mounting. K2 = 10-way SIL header. PCB Type 880074 LC display Type LM40001 Hitachi Electronic Components (UK) Ltd. • Hitec House 221-225 Station Road • Harrow • Middlesex HA1 2XL. Tel.: (01 861) 1414. Tlx.: 936293 hitec g. Fax: (01 863) 6646. Hitachi Electronic Components Europe GmbH (Headquarters). • Hans-Pinsel Strasse 10A • D-8013 Haar nr. Munich. Tel.: +49 894614/0. Tlx.:5-22593 hitc d. Fax: +49 89463151. Sharp Electronics (UK) Ltd. • Sharp House • Thorp Road • MANCHESTER M10 9BE. Tel. (061 205) 2333. Fax: 061 205 7076. elektor india december 1988 12.51 SIMPLIFIED TIME-SIGNAL RECEIVER The automatic synchronization facility of many microprocessor- based clocks ensures reasonable long-term accuracy even when the relevant time-signal transmitter is received for only a couple of minutes each day. Obviously, this feature relaxes the design requirements of the receiver, which can be kept relatively simple. Such a receiver is described here: it has a digital pulse output, excellent sensitivity, and can be tuned to time-standard stations transmitting in the VLF band between 50 and about TOO kHz. i- Completed prototype of the simplified time signal receiver, eonneeted to tlie associated active aerial. Time-signal transmitters such as Rugby MSF, HBF and DCF77 operate in the VLF (very low frequency) band, at fre- quencies between 50 and 100 kHz. The VLF band is characterized by very predictable propagation characteristics, but received signals often suffer from in- terference generated by electrical ap- paratus such as TV sets and dimmers. The receiver should, therefore, have good or very good selectivity. The fre- quency conversion principle (heterodyne receiver) must be dismissed, however, when the practical design is to remain as simple as possible. Circuit description Selectivity of the present VLF receiver is determined solely by the aerial and two tuned circuits. High-gain RF amplifiers are used, and a special, non-linear, demodulator extracts the time signals from the still relatively noisy RF input signal. The circuit diagram of Fig. 1 show's that the RF signal from the transmitter is picked up by an active aerial circuit, whose output signal is filtered by tuned circuit L 1 -C 2 , and amplified by dual- gate MOSFET Ti. A further tuned cir- cuit inserted in the drain line of this tran- sistor ensures adequate receiver selectivi- ty. The drain signal is rectified by Di to provide automatic gain control (AGC) on gate 1 of the MOSFET. The AGC has a relatively slow response because fading is generally slow on VLF. Circuit ICi is a Type S042P balanced mixer/oscillator from Siemens. In the present application, it functions as a four-quadrant multiplier, so that its out- put signal is proportional to the square of the input signal provided by Ti. The modulation frequency on DCF77 is rela- tively low, so that a single R-C network, Rs-Cis, is sufficient for removing the RF component from the rectified time signals. These are filtered and shaped in a further (active) rectifier, ICia, whose output signal is a measure of the instan- 12.52 elektor india december 1988 taneous amplitude of the time signals. The discharge time of Ci6 is relatively long (P:-Rio), so that the voltage on this capacitor is largely constant for the dur- ation of the time pulses. Comparator IC:b compares the instantaneous ampli- tude of the rectified voltage to a part of the absolute amplitude, set with Pi. The output of the receiver supplies time pulses as they are modulated, i.e., a time pulse corresponds to a logic low level. This makes the present time signal re- ceiver compatible with the Intelligent Time Standard published in Ref. IJ >, but only if DCF77 is being received. The circuit diagram shows the capacitor values needed for reception of DCF77 on 77.5 kHz. The tuned circuits can be modified for reception of, for instance, Rugby MSF at 60 kHz, by multiplying the value of Ci, C- and Cs by a factor (77.5/60) :as 1.67, and using the closest practical capacitor value. Construction and alignment The receiver is composed of two boards: active aerial and receiver/demodulator. The active aerial is identical to that used for the DCF77 receiver and locked fre- quency standard (sec Ref. "’). The unit is constructed on the small printed cir- cuit board shown in Fig. 2. The aerial, Ls, is formed by about 200 closew'ound turns of 0.2 mm dia. enamelled copper wire on a 30 mm long cardboard or pax- olin former. This is slid on to a 12-20 cm long ferrite rod of 10 mm diameter. The rod and associated former used for building the prototype receiver were parts salvaged from a discarded MW/LW radio. Populating the receiver/demodulator board shown in Fig. 3 should not pres- ent problems. A 15 mm high tin plate or brass screen is fitted across Ti as shown on the component overlay. The screen has small clearances for Ti and Rs, and is secured to the PCB with the aid of two soldering terminals. Note that a number of parts are fitted upright. Use one metre or so of screened micro- phone wire to connect the active aerial to the main receiver board. First, concentrate on setting up the ac- tive aerial. Power up and check the DC settings at the points indicated in the cir- cuit diagram. Set a sine-wave generator to the receiving frequency (e.g., 77.5 kHz), and connect a coupling loop and a series resistor to the output of the instrument. Wind the coupling loop on to the ferrite rod, and connect an AC- coupled oscilloscope to the output of the active aerial. Slide the former until the signal amplitude is a maximum. Reduce the output of the generator, and move the coupling loop away from the rod. Again slide the former on the rod to find the resonance point. If this is found with the former partially off the rod, the number of turns of Ls should be re- duced. Experiment with the value of Gu and the setting of P 2 until the com- pleted active aerial has a selectivity of about 10 kHz, and the former is about flush on the rod. When this cannot be achieved, the ferrite rod may have incor- rect RF properties, and there is no alternative but to try out another type. After adjustment, the former is secured on the ferrite rod by means of wax or sellotape. Do not use a metal support for fixing the rod. Switch off the generator, increase the sensitivity of the scope, and rotate the rod in the horizontal plane until the RF signal from the time signal station is ob- served on the oscilloscope screen. The signal is relatively small, but should have an amplitude between 5 and 50 mV PP . Connect the probe to the drain of Ti, and carefully peak Li and L: for maxi- mum amplitude. If clipping or oscil- lation occurs, reduce the gain of Ti by adjusting Pi. Readjust the active aerial and the tuned circuits with a high- impedance voltmeter connected to pin 2 of ICi. The time signals can be heard on high- impedance (600 Q) headphones connec- ted between the positive supply and test point TP2. Finally, mount the active aerial in a position well away from Fig. 1. Circuit diagram uf the VLF time signal receiver. elektor india december 1988 12.53 sources of interference. The length of the screened cable between the active aerial and the main receiver board should not exceed 15 m or so. References: (1) DCF77 receiver and locked fre- quency standard. Elektor India, February 1988. 121 Intelligent time standard. Elektor India, March 1988. EPS B7513-B ♦ l U 2 i § s 8 s 8* I * tv ACTIVE AERIAL: Resistors (±5%l: R47 = 100K R4B = 68R R49 = 150R P 2 = 2M5 preset H Capacitors: C44 = 2n2 C45=4p7; 16 V; radial C46=100p; 16 V; radial Semiconductors: Tt3 = RF256C (Cricklewoodl Ti4 = BF256A (Cricklewoodl Tis = 8C550C Inductor: Ls = see text. Ferrite rod: e.g. Cirkit Type FRA (stock number 35-141471. Miscellaneous: PCB Type 87513-2 MAIN RECEIVER BOARD: Resistors (± 5%): Rl = 220K R2 = 68R R3=56K R4;Re = 100R R5=100K Re = 1M0 R7;Ra = 8K2 Rio= 1 50K Rt 1 = 18K Ri2 = 47R Pi = 50K preset H P 2 = 1 0OK preset H Capacitors: Cl = 5n6 C2 = 1 80p C3= InO C4;C ! 7 - 47^; 16 V; radial Cs;Cn;Ci4=100n Ce = 1 0Op C7 = 270p Ca = 680p Ce = 47 fi; 16 V; axial Cio= lOn Cl 2 = 220n Cl3 = 4n7 Ct5-lp0; 16 V; radial Cie=220p; 16 V; radial inductors: Li;l 2 = 22mH variable; Toko 10PA series, Type CAN1896HM (Cirkit stock number 35-18960). L3 = 2mH2 radial choke; Toko Type 181LY-222 (Cirkit stock number 34-22202). Semiconductors: Dl;D2 = AA119 D3= 1N4148 Ti=BF982 (C-l Electronics) ICi=S042P (Bonex; C-l Electronics; Universal Semiconductor Devices Ltd.) IC2 = CA3240 Miscellaneous: PCB Type 87513-1 Fig. 3. Printed circuit board for the time signal receiver/demodulator. 12.54 elektor india december 1988 A MICROPROCESSOR-BASED INTELLIGENT MULTI FUNCTION TEST INSTRUMENT by Dr. D.P. Mital School of Electrical and Electronic Engineering, Nanyang Technological Institute, Singapore. A high precision, intelligent, test instrument is described that offers ten measuring functions: DC and AC voltages, DC and AC current, resistance, capacitance, frequency, digital counter, data logging and phase measurements. The system is intelligent enough to automatically range itself for proper measurements, and is capable of providing statistical functions such as calculation of mean and standard deviations, which are useful for low-frequency measurements. The researcher or technician involved in project development usually requires many instruments to perform various common measurements. Often, a lot of precious time is wasted in connecting and disconnecting these instruments. Also, handling and storage problems soon arise in the work area as the num- ber of instruments increases. Recently, there has been considerable in- terest in integrating many functions in one compact unit (Ref. (21 )- The in- dustry has been very receptive to this idea. This article presents a multifunc- tion unit which has some intelligence, and integrates many common measuring functions. Accuracy, reliability and low cost are also important considerations. The proposed system is developed around the Type 8086 16-bit micropro- cessor from Intel, and is capable of per- forming operations which include that of a common multimeter, capacitance meter, phase meter, frequency meter, digital counter and data logger. Autoranging and repetitive modes for averaging are also available. The measur- ing function and, optionally, the range, is selected by means of a 16-key mem- brane keyboard, and measurement results arc sent to a 40-character LC dis- play. Since the system has intelligence, on-line data and results may be stored semi-permanently, and retrieved when required. This feature makes the system highly suitable for real-time interactive measurement and control applications. A multiple number of readings can also be recorded in a fixed time interval. For reasons of speed and efficiency, the system software has been written in 8086 assembly language. The software was written and debugged using an IBM PC/AT and an HP6400 development sys- tem. 880173- 11 Fig. 1. Block diagram of the multifunction test instrument. INTERRUPT SERVICE ROUTINE voltage service routine current service routine resistance service routine capacitance service routine frequency service routine phase service routine calculation subroutine calculation subroutine calculation subroutine calculation subroutine calculation subroutine calculation subroutine 1 — — i display subroutine display subroutine display subroutine display subroutine display subroutine display subroutine 880173-12 Fig. 2. General structure of the system software. cloktor India december 1988 1 2.55 * = corresponding resistance for (he particular range M0173- 14 1 1 Fig. 4. Schematic diagram of the capacitance meter section. Az Jin Signal Pulsewidth to micro - XOR shaping measurement processor 880173- 15 Multifunction system Figure 1 shows the block diagram of the proposed system. Analogue input signals are sampled and converted to digital data. When this is done, the mi- croprocessor strobes the relevant block, and data processing commences. Results arc read on the LC display. The keypad, which operates in an interrupt structure, provides the user with minimum control of the instrument as explained above. Current and voltage meter Signals for V and 1 measurements are first scaled internally to prevent their ex- ceeding the maximum input specifi- cation of the A-D converter (±5 V). After scaling, the DC signal is fed direct to the sampling circuit, while the AC signal is fed to an RMS-to-DC converter, and then to the sampling circuit. The ADC awaits the strobe signal from the microprocessor to start conversion. Con- verted data is immediately read via the data bus. A two-pole, microprocessor- controlled, change-over switch selects the appropriate measurement function. Signals exceeding 200 mV are at- tenuated, and later amplified to TTL level. The ranges for voltage and current measurments are 20 mV to 1000 V and 20 pA to 2 A respectively. Resistance meter For resistance measurement, two voltage references are used. This is done to ex- tend the measurement range. One refer- ence voltage is used in the autoranging mode. The resistance sampling circuit shares the first stage opamp with the voltage sampling circuit. The output voltage of the comparison block is pro- 12.56 elektor india december 1988 Fig. S. Block diagram of the phase meter. portional to the unknown resistance, for which six ranges are available. The cir- cuit diagram of the multimeter section is shown in Fig. 3. Capacitance meter Capacitance measurement is based on pulse-width modulation techniques. Again, six ranges are provided. The basic circuit diagram of Fig. 4 shows that the charging and discharging properties of capacitors are used to determine capacitance with the aid of two timers Type 555, and two bistables. The timers arc configured to function as monostable multivibrators. Fig. 6. Block diagram of the frequency meter. Phase meter Phase detection first requires the conver- sion of sinusoidal input signals to rec- tangular waves as shown in the block diagram of Fig. 5. The phase difference between the input signals can be com- puted by the CPU because it is pro- portional to the pulse-width of the signal supplied by the XOR gate. Frequency meter Frequency measurement is similar to phase detection. The block diagram of Fig. 6 shows that the input signal is made rectangular before being applied to a combinational circuit and a pro- grammable divider. These two circuits control a 32-bit counter. The divider, which is initially set to a factor of -1-2, stops the counter on the rising edge of the second pulse. This completes the first phase of the measurement. The CPU calculates the period and estimates the number of pulses needed from the input signal for 0.1 s of sampling time. The divider is then programmed accor- dingly. The count procedure is similar for the second sample of the imnput signal, but this time the counter stops after a known number of pulses, as de- termined by the CPU and executed by the divider. The measurement principle adopted allows signal frequencies to be determined with high precision. Event counter Event measurement is essentially similar to that for frequency, and shares a part of the relevant circuitry. Only counter start and stop signals are needed, which are provided by the measuring circuit. Data logger Data logging is purely a software func- tion. Displayed data is stored in memory when the strobe key is actuated on the keypad. Key recall may be pressed for data to appear on data lines and on the LC display. The system is capable of storing hundreds of data readings sequentially. Last-in data can be recalled first (LIFO stack). The flowchart of the data logger function is given in Fig. 7. Software development The following description is intended to give a basic idea of the operation of the control software for the instrument. It is assumed that this is set to capacitance measurement. Pressing the capacitance key causes the interrupt servicing routine to be activated. After the interrupt source is determined, the service routine passes control to the capacitance measurement program, which arranges the necessary switching and sampling of data. Repetitive reading may be used for calculation of mean and standard devi- ations. The calculation routine is called to process the available data, followed by the display routine to provide legible results. Fig. 7. Flowchart of the data logging routine Every measuring function has a control program, which is called up by pressing the appropriate key. Control programs perform the necessary switching in the measurement circuits, and manipulate available data. First, the range flag is checked, and the hardware is controlled accordingly by autoranging software to ensure optimum accuracy before data is accepted. The control software is also in- volved in the repetitive mode of oper- ation. A simplified flowchart of the con- trol programs is shown in Fig. 8. Autoranging is not used during measure- ment of frequency and phase, since in these modes data samples are taken twice: first for estimating the order of magnitude, and then for the actual measurement. The flowchart of the fre- quency control program is given in Fig. 9. The operation of the test instrument is elektor indla december 1988 1 2.57 Fig. 8. General flowchart of the control programs. further co-ordinated by software modules written for each function. These modules each consist of a func- tion service routine, a calculation routine and a display routine. System hardware The wiring diagram of Fig. 10 shows that the hardware of the CPU card basically consists of a 8086 CPU with 64 Kbyte of memory (RAMs and EPROMs), buffers (74LS245 and 8286), latches (74LS373 and 8282) and decoders (74LS138). A total of 64 I/O ports is used. These are addressable from 00h to 3Fh as shown in Table 1. The 4x4 key membrane keypad utilizes a Type 74C922 encoder. The 1 line x 40 character LC display is of the dot-matrix type. It is connected to data I/O lines DB0-DB15 through 74LS245 buffers, which form port numbers 10n and 1 1 h. Table 1. Port addresses Vfl/R BOARD 00-07 A-D data input 00 stautus check 01 A-D control 02 Range control 04 Function control 06 FREQUENCY BOARD 08-OF Counter setting 08 Divider setting 09 Ready input 0A Start output oc Counter data input 0E CAPACITANCE & PERIPHERAL BOARD 10-17 Display control 10 Display data 11 Input data 12 C range control 14 All addresses in hexadecimal. Fig. 9. Flowchart of the frequency measure- ment program. measurements were on average within 0.5 Vo accuracy. Table 2 shows the results of a few comparative measurements taken with the proposed multifunction instrument. The author believes that it is worthwile to spend time on further development of the multifunction instrument, whose basic lay-out has been discussed here. The design of the instrument shows that test & measurement equipment is heading in the same direction as much other electronics equipment, i.e., towards high-level integration. Experimental results and con- clusion Results of all measuring functions of the instrument were compared with existing, high-precision, laboratory equipment. In all cases, deviations from the standard equipment remained well within 2%. Phase, frequency and event count 1 2.58 elektor india december 1988 Table 2. Experimental results Voltage Standard Measured Current Standard Measured Resistance Standard Measured Capacitance Standard Measured 5 V 4.98 V 50 V 49,44 V 1 20 V 119.12V 4.8 V 4.68 V 12V 11.92V 50 V 48.81 V 1.5 kQ 1,506 kQ 10 kQ 10.04 kS! 47 kQ 47.02 kQ 10 pF 10 pF 1 000 pF 1001 pF 0.1 pF 0.099 pF For AC measurements RMS values are recorded. Fig. 10. Wiring diagram of the microprocessor board in the multifunction instrument. Acknowledgements: References: Journal, Vol. 37. no. 2, 1986. the author would like to thank Mr Bay (l> Trautman J. and Desjardin L.: A 131 Walter A. and Singh A.: The 8086 Way Yee and Miss Chong Mong Tan for portable low-cost high performance Microprocessor Architecture, software developing and testing the system, and digital multimeter. HP Journal, Vol. 34, and interfacing techniques. Prentice- professor Brian Lee, Dean of the School no. 2, 1983. Hall, 1986. of Electrical and Electronic Engineer- (2) Steever S. et al.: Seven-function |J) Macro assembler. (IBM personal ing, for providing facilities for carrying systems multimeter offers extended res- computer language series) Microsoft, out this work. olution and scanner capabilities. HP 1985. HARMONIC ENHANCER by W. Teder An harmonic enhancer, or exciter, generates harmonics from, and superimposes these on to, a music signal that has none, or few, of these overtones. In that sense, it is a sound-correcting device that adds warmth to a sound. The principle of the operation of an ex- The basic set-up in Fig. 2 may be cuit and followed by an expander circuit, citer is shown in Fig. 2. Part of the modified and refined in various ways. It This method obviates the serious distor- original signal is fed to a variable clip- is, for instance, possible to make several tion caused by short signal peaks and per, whose cut-off frequency can be set of the filter parameters adjustable exter- also ensures that the harmonic content from 1-5 kHz. The filter output, whose nally, but for most relatively simple does not vary too greatly with the input amplitude should not exceed 10% of needs this sophisticated approach is not level. Whatever refinements or modifi- that of the original signal, is then recom- really necessary. Moreover, the filter cations are introduced, they lead to a bined with the original signal. might be preceded by a compressor cir- unit with may operational possibilities, elektor india december 1988 12.59 all of which have to be set up carefully. The enhancer described here is intended as an experimental unit for use by the constructor to become acquainted with the basics of the harmonic enrichment effect. None the less, the unit may, of course, be expanded as required at a later date. The harmonics caused by the clipping are mainly odd-order ones. After they have been recombined with the original signal, the resulting sound is only little louder (about + 1 dB), but, as already stated, it is warmer, more mellow. The new sound may, however, just be dis- torted if, for instance, the cut-off fre- quency of the clipper is set low, i.e., at 1 kHz, and the level of the harmonics is much higher than 10% of that of the original signal. Used with electric guitars, this may not be unacceptable, but it certainly would be with a good audio amplifier. Circuit description The circuit of the basic enhancer is shown to the right of the dashed line in Fig. 3. It is based on two integrated cir- cuits, ICi and IC 2 . The signal is applied to low-noise ampli- fier ICi via Ci and Ri, which form a high-pass section with a cut-off fre- quency of 2.4 kHz. Further attenuation of frequencies below 1 kHz is provided by high-pass section Rj-Cj. After amplification (gain is set by Pi), the high-frequency part of the signal is clipped asymmetrically by R5-D1. The distorted (i.e., rich in harmonics) signal is applied to the inverting input of IC2 via P2 and a further low-pass section, R7-C4. The effect signal may be switched off by Si. Also applied to the inverting input of IC 2 is the original signal (via Rj). Since a too high amplification of the ef- fect signal leads to audible distortion, a peak meter has been added. This con- sists of opamps Ai and A 2 , which form a window discriminator, and T-i. The reference voltages for the window (pins 3 and 6 of IC 2 ) are derived from divider Rn-Ri 2 -Rn. The output voltage of ICi is monitored via Rio. If this lies outside the window potentials, capacitor C 7 is charged via R15 and transistor Ti switches on peak warning light D4. Ca- pacitor Cs in the collector circuit of Ti extends the operation of D4, so that even short peaks are indicated. The set- ting of Pi is optimum when D 4 flickers during signal peaks. The cut-off points of high-pass sections R1-C1 and R7-C7 have been determined emperically with the aid of an electric guitar. The harmonic content can be controlled satisfactorily when the unit is used with an electric guitar. If the enhancer is for use with hi-fi or PA equipment, the Fig. 1. General view ol’ the Harmonic Enhancer. Fig. 2. Block schematic of the Harmonic Enhancer. values of Ci, C 3 and C 4 should be halved. It is, of course, also possible to experiment with a high-order variable filter at the input circuit of ICi. If the threshold of operation of D 4 is found too high, the value of R 12 may be reduced as required. This is conveniently done with a 500-ohm potentiometer in series with a 470-ohm fixed resistor. If the enhancer is intended for use as a guitar effects unit, the amplifier to the left of the dashed line in Fig. 3 is rec- ommended. Strictly speaking, this is an impedance inverter, based on a FET, which has been designed specifically for Semiconductors: Di;D2;D3:Ds;D6= 1N4148 D4= red LED Ti =BC264A 1 (Philips Components! T2 = BC550C T3 = BC560C T4 = BC547B ICi = NE5534P IC2 = TL071CP IC3 = TL072CP + Listed by Crickiewood Electronics. Miscellaneous: Lt= max. ImHO (see text] Si = miniature SPST switch. S 2 = 4-way OIL switch block. PCB Type 880167 12.60 elektor India december 1988 Fig. 3. Circuit diagram of the enhancer (to the right of the dashed line), and the impedance inverter for use with guitars. use with a guitar pick-up. The input consists of two low-pas sec- tions, L ')-€■> and Rie-Cio, which effec- tively prevent interference from HF equipment. In a non-critical environ- ment, Li-G> may be omitted. Diodes Dt and Dr. protect the input against too high voltages. The signal is taken from the low- impedance source of Ti and applied to the enhancer via C«. The circuits around T: and T 3 provide further smoothing and filtering of the power supply lines. DIP switch S 2 facilitates matching to various cable lengths, which, of course, is a boon for many musicians. With values of capacitors C17 to C20 as shown, cable lengths of 1 to 10 m may be accommodated. Finally The enhancer and input amplifier for guitars may be conveniently constructed on the PCB shown in Fig. 4. Although the circuit shows a mains- operated power supply, a ± 9 V battery supply may also be used if only the enhancer and input amplifier are used. Rechargeable 9 V batteries will give about 6 hours continuous use, while two PP9 batteries will give about 25 hours. It should, however, be borne in mind that in view of the supply current of around 20 mA it is advantageous to use a mains supply. This is even more so if other modifications are incorporated. Parts list Resistors (±5%l: Ri = 2K2 R2 = 56K R3;Ri2;Ri5= 1K0 R4 = 22K H5;Rto;Rii;Ri3;Rt7;R2i;R23;R24= 10K R6;R7;Ra = 33K R9-100R Ri4;Ri6 = 100K Rie = 820R R1 9 = 1 0K; metal film R20=1MO R22 = 3K9 Pi = 22K or 25K logarithmic potentiometer P 2 = 10K logarithmic potentiometer Capacitors: Cl = 1n2 C2 = 33p C3= 1 50n C4 = 2n2 C5;C9;Cio = 68p Cb - lOp; bipolar; radial C7= IpO; 40 V; radial C8 = lOp; 40 V; radial C 11 =47n Ci 2 ;Ci 6 = 22p; 25 V; radial Ci3;Cis=220n Cu;C2i;C22 = 1pO; 63 V Ci7=100p CiS = 220p C 1 9 = 330p C 20 = 390p C23 = 100n Fig. 4. The PCB can accommodate both the enhancer and the input amplifier. elelrtnr irtriia rinr.ArrthAr IBFtfl 1? fil NEW PRODUCTS Programmer Controller JELTRON Model 814A Programmer Controller is a self contained microp- rocessor based set point programmer and a single loop industrial controller combined in one compact case. The 814A accepts directly process variable inputs from TCs. RTDs, transmitter vol- tages and currents and Optical Radiation Temperature Detectors. All tempera- ture inputs are linearized and direct reading in degress F or C, switch selecta- ble. All transmitter units are field con- figurable in engineering units, from 999 to 9999 with full decimal point position- ing. All program parameters such as Range Limits, Set points. Ramp Times and Soak Times are entered in full en- gineering units. Times are user configur- able from 0.0 to 99.9 hours or minutes, the SWA is a single channel instrument that can store a maximum of 30 seg- ments. Individual programs can be fully independent or linked. A loop instruc- tion alows the segment or any combina- tion of segments to be repeated upto 99 times. Complete program configuration is stored in solid state non volatile EAROM. The 814A is available in wide choice of Control outputs. Reverse act- ing or direct acting control action is switch selectable. Integral Auto/Manual Station standard with bumpless transfer from auto manual. Remote/Local set point operation is standard. The remote set point input is automatically scaled to the field configured range of the control- ler. Optionally, RS-232C or RS 422 com- munication interface is availabe for supervisory control applications. Set Point, Control output and Configuration of parameters are viewed using the main display. Segment number as well as Set Point are continuously displayed. Prog- ram security is assured using a front key lock. Jeltron Instruments (India) Private Limited. • 6-3-198/2, Road No. 1 • Ban- jara Hills • Hyderabad- 500 034. Temperature Indicator Controller HOSHAKUN has developed Digital Temperature with double thumbwheel type Digital Controller. It is a rugged, compact panel mountable instrument. The bright red 12.5 mm LED Display en- ables one to read the temperature from a long distance. Set temperature is all the time visible on the front panel by the use of thumbwheel switches. Broken sensor protection and automatic cold junction compensation is standard feature for thermocouple input. Most of the as- semblies used in this instrument are plug in which offers simplicity in assembly as well as dismantling for servicing pur- pose. This instrument can be used for furnaces having heaters in delta/star con- nection such that during the start up the furnace is heated with heaters in delta form upto a certain temperature (set low) and after than it gets connected into star form up set high around which con- trol action will take place . In other words during start up, furnace is in “MORE HEAT” mode and after exceeding first control point it goes into “LESS HEAT” mode which then controls the tempera- ture of the furnace around the second control point. This also can be used for oil fired furnaces (MORE HEAT mode equivalent to both main and pilot bur- ners on and less heat mode equivalent to only pilot burner on and main burner off). HOSHAKUN • Vivek Appartments • Plot No. 15 • Tulshibagwale Colony • Sahakarnagar No. 2 • PUNE- 411 009. Programmable Batch Counter Controller Micronix offers an ideal instrument for applications where counting, controlling and sequencing operations are involved. The Unit is based on 8085 microproces- sor with battery backed memory to re- tain data during power failures. The front panel consists of an interactive 4 digit display and hermatically sealed membrane keyboard for entry of parameters. The display is used to indi- cate event count as well as Jobs done de- pending on the selected mode. It also in- dicates te system status on LED indi- cators. The number of counts per sequ- ence is 999 and no. of sequences is 8 both expandable as per user request. The unit accepts a variety of inputs such as mic- roswitch, optical or proximity switch etc. It provides a change over contact for controller action. The unit works on 230V AC and is housed in DIN standard enclosure suita- ble for panel mounting or bench top models. Micromix • I)-74, Angol Industrial Es- tate • Udyambag • Belgaum-590 008. Karnataka State. New Open-Type Terminal Connectors “IEC” has introduced a new range of Open Type Terminal Connectors -TBM Series. They are presently available in 12 ways, 10 ways and 8 ways. The Connec- tors are rated at 15 Amps, 250 V AC with a insulation resistance of more than 1000 Mohms and can withstand H.V. test of 2000 V for 1 minute. The Terminals are of Brass with Nickel plating and the housing of electrical grade bakelite or melamine on order. M/s. Asia Electric Company • Katara Mansion • 132A, DR. A. Besant Road • Worli Naka • Bombay-400 018. 1 2.66 elector India december 1988 NEW PRODUCTS Dual Counter JACSTECH, has recently developed a DUAL COUNTER for use as position encoder/indicators. It contains two seperate 4 digit counters, 2 reset sensors, and 2 seperate encoder sensors. The sensors are based on in- frared LEDs and Phototransistors. A slotted disc rotation in the encoder sen- sor field is sensed for direction and the counter is incremented by clockwise ro- tation and decremented by counter clockwise rotation or vice versa. Both- the counters are independent in func- tion. The counter can work to an input frequency of 15 KHz. The approximate dimension of the card is 100 mm x 110 mm. The system can be used for Lathe machines, displacement benches, in- dustrial cameras, paper cutting machines, printing machiens, textile machines etc. Jacstech (P) Ltd. • 1 17, M.G. Road, • PONDICHERRY- 605 001. Proximity Switches for Hazardous Locations. ACCENT has recently introduced a line of Intrinsically Safe Inductive Proximity Switches for use in hazardous locations. These are contactless limit switches and can be used to replace the conventional flameproof limit switches. These switches feature solid state circur- ity with no moving parts. Consequently, problems like roller break-age, contact bounce etc. , that are common in the case of conventional limit swithces, can be conveniently avoided. These switches are available in the 2- wire DC version with sensing range from 5 to 25 mm. They are fully encapsulated in cylindrical chrome-plated threaded brass housings. Two lock nuts are pro- vided for mounting and adjustment. These switches are recommended for use in: Coal Mines, Petroleum Refineries. Fertiliser Plants, Chemical Plants etc. M/s. Accent Controls (P) Limited • Post Box No : 16596 • Worli Naka • Bombay-400 018. 1 200 BPS Modems SYSTEMS AIDS, an Electronic Equip- ment manufacturer in Bangalore, has in- digenously developed Stand Alone Modem (SAM 1200 A) and IBM PC Compatible Internal Modems (SAM 1200 B) which combine speed and simplicity to transfer voluminuous data over long distances economically. These Modems operate at 0 - 300 and 1200 bps in the asynchronous mode over PSTN and 2 wire leased lines and are full-dup- lex which enable them to send and re- ceive data simultaneously. They also have a voice/data transmission capability which allows transmission of a data file and switching over to voice capability to discuss and anlyse the information - all with a single phone connection. The Modems conform to CCITT V. 21/ V.22 Standard. These Modems can be used in organisa- tions that need to link up their computers for exchanging information and can find extensive use in Airlines, Hotels, Stock Exchanges, New Paper Officers, Banks, Business Houses, Industries or Govern- ment Departments. M/s. Systems Aids • 80, Feet Road • (Opp. Deccan Studios) • Indiranagar • Bangalore- 560 038. PLA DM-20A The PLA DM-20A 41/2 digit Multimeter has a LCD display with resolution of 10 uV on 200 mV range in both AC & DC model. Its basic accuracy is 0.1% Maximum voltage measurable in DC range is 1000V & 700V RMs in AC range. It has resolution of 10 nA on 200 uA range in Ac mode. It has wide fre- quency range of 20 Khz in AC voltage. It is a battery operated and can be easily acco-modated in a briefcase for on field testing. M/s. Pla Electro Appliances Pvt. Ltd. • Thakor Estate • Kurla Kirol Road • Vid- yavihar (West) • Bombay-400 086. 1 2.70 elektor india december 1988 NEW PRODUCTS Digital Shock Meter SHINKEN CO. LTD., of JAPAN offers v-7103, a digital acceleration meter hav- ing a charge amplifier input, a digital dis- play with PEAK HOLD performance for vibration and shock measurements. It covers the wide frequency range of 3Hzto 20000 Hz with three high cut fil- ters which are inevitably needed for shock measurements. Though an ac- celerometer (MODEL VII-101) is fitted as standard accessory, a piezoelectric force sensor may be used instead of it for shock force measurements. Murugappa Electronics Ltd. • Agency Division • 22 Ilnd Street, • Kamaraj Av- enue • Adayar • Madras 600 020. DIN 41612 Connectors ERNI Elekroapparatc GmBH two- piece connectors for Printed Circuit Boards confirm to international stan- dards e.g. DIN 41612, IEC 603-2, VG95324 and to military standards, Mil- C 55302/131-134, Mil-C 55302/157-158. Apart from the standard Euro and Re- verse Euro type connectors, ERNI West Germany can offer a wide range of spe- cial connectors and alternative wiring techniques to provide the user with themost economical solution for his par- ticular application. These connectors are suitable for appli- cation in telecommunication equipment, industrial control instruments, etc. Where reliability is the most important feature. A.T.E. Pvt. Ltd. • (Electronics Division) • 36, SDF 2, Seepz • Andheri (East) • Bombay 400 096. Wire Cutting Machine The machine gives one predetermined wire length variable from 6 cm to 60 cm. with off timing variable from 0.5 sec. to 6 0 sec. The machine can out ordinary single core PVC flexible copper wires as well as flexiable copper braided wires. It can also cut fine copper/silver wires of fixed length as per customers require- ments. The machines are supplied against specific orders only. These machines are useful in Panel Board Wiring, Fuse manufacturing. Electronic Component manufacturing etc. M/s. R. R. Enterprises • D-13 Nav Monica • C.S.T. Road • Opp. Univer- sity Campus • Kalina • Bombay -400 098. Moulded Instrument Cases DIN Standard. G.A.M. has designed and developed Din Standard Instrument Case moulded in A . B . S . Plastic in one piece and its Cas- sette (chasis), clamps, Front Transpa- rent Cover, and front Bracket. It is suita- ble for Panel Flush mounting Instru- ments like Temp. Controller, Temp. In- dicator, Digital Timers, or Digital Coun- ters etc., Cases are available in two sizes: 96 X 1 10 mm (I.C. 110) and 96 x 96 x 150 mm (I.C. 150) and in three colours Black, Light Grey and Dark Grey. It is also available in plain and with cutouts. Moulded Glass filled polymide clamps makes case suitable for Panel Flush mountings. Transparent fornt cover is also available. M/s. Gaurang Auto Manufacturers • 44, Bombay Talkies Compound • Malad (W) • Bombay -400 064. • Tel: 6824159 Cable ties and Accessories Cable Ties are made of high strength nylon. These ties are self locking and cannot work loose. The one piece all nylon Cable Tie can be used in virtually any bundle configuration and are availa- ble in variety of sizes either non-returna- ble (CP Series) or Reusable Style (CPR Series) to cover bundle diameter of 1.6 mm to 106 mm. Each Cable tie incorporate a non-return can action locking device which ensures that once in place the tie will never come off or slacken. Cable Ties can be instal- led easily & quickly without aid of any special tool. Available in natural colour for indoor use and also available in ultra- violet weather resistant black colour. Cable Tie mounts come to two sizes and styles to meet a wide variety of mounting applications. They are available in 3/4” and 1” square with adhesive back or 1” plain back for screw mounting. All offer 4 way insertion to speed mounting oper- ations and the 1” Mount have a ‘Cable Cradle’ for added holding power. Suresh Electronics & Electronics • 3B Camac Street • CALCUTT-700 016. 1 2.72 stektor India december 1988 DO RN No 39881/83 it UC No 91 IT YOURSELF MH BY WEST 2 28 UC No 91 Price Rs. 50.00 Available at all leading book shops. LEARN-BUILD- PROGRAM the P to The Junior Computer book is for anyone wishing to become familiar with microcomputers, this book gives the opportunity to build and program a personal computer at a very reasonable cost. The Indian reprint comes to you from elektef Send full payment by M.O./I.P.O./D.D. No Cheque Please. Packing & Postage free to: EUkTOR ElECTRONiCS pvT ItcJ . 52-C, Proctor Road, Grant Road |E), Bombay-400 007.