advertisement Elektor March 1979 - UK 3 Eurotronics Eurotronics - a world wiJe circuit and design idea competition, with over £10,000 worth of electronic equipment to be won! Note that the closing date for the competition is 31st March, 1979. ☆ ☆ See this issue for further details international circuit and design idea competition The professional scopes you’ve always needed. When it comes to oscilloscopes, you'll have to go a long way to equal the reliability and performance of Calscope. Calscope set new standards in their products, as you'll discover when you compare specification and price against the competition. The Calscope Super 1 0, dual trace 1 0 MHz has probably the highest standard anywhere for a low cost general purpose oscilloscope. A 3% accuracy is obtained by the use of stabilised power supplies which cope with mains fluctuations. The price £219 plus VAT. The Super 6 is a portable 6MHz single beam model with easy to use controls and has a time base range of Ips to 1 0Oms/cm with 1 0mV sensitivity. Price £ 1 62 plus VAT. CALSCOPE DISTRIBUTED BY Marshalls Electronic Components. Kingsgate House. Kingsgate Place. London. N.W.6. Audio Electronics. 301 Edgware Road, London W.2. Tel: 01-724 3564 Access and Barclay card facilities (Personal Shoppers) Maplin Electronics Supplies Ltd. P.O. Box 3 Rayleigh. Essex. Tel: 0702 715 155 Mail Order CALSCOPE UK 4 - elektor march 1979 elektor Volume 5 47 decoder Elektor Publishers Ltd., Elektor House, 10 Longport, Canterbury CT1 1PE, Kent, U.K. Tel.: Canterbury (0227) 5443a Telex: 965504. Office hours: a30 - 12.45 and 13.30 - 16.45. Bank: 1. Midland Bank Ltd., Canterbury, A/C no. 11014587 Sorting code 40-16-11, Giro no. 3154524. 2. U.S.A. only: Bank of America, c/o World Way Postal Center, P.O. Box 80689, Los Angeles. CA 90080, A/C no. 12350-04207. 3. Canada only: The Royal Bank of Canada, c/o Lockbox 1969, Postal Station A, Toronto. Ontario, M5W 1W9. A/C no. 160-269-7. Please make all cheques payable to Elektor Publishers Ltd. at the above address. Elektor is published monthly. Number 51/52 (July/August) is a double issue. SUBSCRIPTIONS: Mrs. S. Barber Subscription 1979, January to December incl.: U.K. U.S.A./Can. other countries surface mail airmail surface mail airmail £8.50 $21.00 $31.00 £8.50 £14.00 Subscriptions normally run to December incl. Subscriptions from U.K. other countries surface mail airmail £ 6.25 £ 10.50 U.S.A./Can. surface mail airmail £6.25 $15.00 $23.00 Back issues are available at original cover price. Change of address: Please allow at least six weeks for change of address. Include your old address, enclosing, if possible, an address label from ASSISTANT MANAGER: R.G. Knapp ADVERTISING MANAGER: N.M. Willis National advertising rates for the English-language edition of Elektor and international rates for advertising in the Dutch, French and German issues are available on request. EDITOR U.K. EDITORIAL STAFF W. van der Horst I. Meiklejohn TECHNICAL EDITORIAL STAFF J. Barendrecht A. Nachtmann G.H.K. Dam J- Oudelaar P. Holmes A.C. Pauptit E. Krempel sauer K.S.M. Walraven G. Nachbar P. de Winter Technical telephone query service, Mondays only, 13.30 - 16.45. For written queries, letters should be addressed to dept. TQ. Please enclose a stamped, addressed anvelope or a self-addressed envelope plus an IRC. ART EDITOR: F. v. Rooij Letters should be addressed to the department concerned: TQ - Technical Queries ADV = Advertisements ED = Editorial (articles sub- ADM = Administration mitted for publications etc.) EPS = Elektor printed circuit SUB - Subscriptions board service The circuits published are for domestic use only. The submission of designs or articles to Elektor implies permission to the publishers to alter and translate the text and design, and to use the contents in other Elektor publications and activities. The publishers cannot guarantee to return any material submitted to them. All drawings, photographs, printed circuit boards and articles published in Elektor 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 circuits, devices, components etc. described in this magazine. The publishers do not accept responsibility for failing to identify such patent or other protection. Dutch edition: Elektuur B.V., Postbus 75, 6190 AB Beek (L), the Netherlands. German edition: Elektor Verlag GmbH, 5133 Gangelt, W-Germany French edition: Elektor Sari, Le Doulieu, 59940 Estaires, France. Distribution in U.K.: Seymour Press Ltd., 334 Brixton Road, London SW9 7 AG. Distribution in CANADA: Fordon and Gotch (Can.) Ltd., 55 York Street, Toronto, Ontario M5J 1S4. Copyright ©1979 Elektor publishers Ltd. - Canterbury. Printed in the UK. ■ lABCl What is a TUN? What is 10 n? What is the EPS service? What is the TQ service? What is a missing link? Semiconductor types Very often, a large number of equivalent semiconductors exist with different type numbers. For this reason, 'abbreviated' type numbers are used in Elektor wherever possible: e '741 stand for pA741, LM741 , MC641. MIC741, RM741 , SN72741 , etc. • TUP' or 'TUN' (Transistor, Universal, PNPor NPN respect- ively) stand for any low fre- quency silicon transistor that meets the following specifi- cations: UCEO, max 20V 100 mA 100 100 mW fT, min 100 MHz of zeros are avoided wherever possible. The decimal point is usually replaced by one of the following abbreviations: p (pico-l = 1 0 _l * n (nano-) ■ 10‘* M (micro-) = 10' 6 m (milli-) = 10'* k (kilo-) - 10> M (mega ) - 10‘ G (giga-) ■ 10* A few examples: Resistance value 2k7: 2700 n. Resistance value 470: 470 fl. Capacitance value 4p7: 4.7 pF, or 0.000000000004 7 F . . . Capacitance value lOn: this is the international way of writing 10,000 pF or .01 pF, since 1 n is 10~* farads or 1 000 pF. Resistors are ’/• Watt 5% carbon types, unless otherwise specified. The DC working voltage of capacitors (other than electro- lytics) is normally assumed to be at least 60 V. As a rule of thumb, a safe value is usually approxi- mately twice the DC supply Some 'TUN'S are: BC107, BC108 and BC109 families; 2N3856A, 2N3859. 2N3860. 2N3904, 2N3947. 2N41 24. Some TUP's are: BC177 and BC178 families; BC179 family with the possible exeption of BC159 and BC179; 2N2412, 2N3251 . 2N3906, 2N4126, 2N4291 . • 'DUS' or 'DUG' t Diode Univer- sal, Silicon or Germanium respectively) stands for any diode that meets the following specifications: DUS DUG UR. max IF. max IR, max Ptot, max Cp, max 25V 100mA IpA 250mW 5pF 20V 35mA 100 pA 250mW lOpF Some 'DUS’s are: BA127, BA217, BA218. BA221 . BA222, BA317, BA318. BAX13. BAY61, 1N914, 1N4148. Some ‘DUG's are: OA85, OA91 . OA95, AA116. • 'BC107B'. 'BC237B', 'BC547B' all refer to the same 'family' of almost identical better-quality silicon transistors. In general, any other momber of the same family can be used instead. BC107 (-8. -9) families: BC107 (-8. -9), BC147 (-8. -9). BC207 (-8. -9). BC237 (-8. -9). BC317 (-8. -9). BC347 (-8. -9). BC547 (-8. -9). BC171 (-2. -3), BC182 (-3. -4). BC382 (-3. -4), BC437 (-8. -9). BC414 BC177 (-8, -9) families: BC177 (-8. -91.BC157 (-8, -9), BC204 (-5. -6). BC307 (-8. -9). BC320 (-1,-2), BC350 (-1.-2). BC557 (-8. -9). BC251 (-2.-3). BC212 (-3. -4), BC512 (-3. -4). BC261 (-2, -3). BC416. Resistor and capacitor values I When giving component values, I decimal points and large numbers Test voltages The DC test voltages shown are measured with a 20 kn/V instru- ment. unless otherwise specified. U, not V The international letter symbol 'U' for voltage is often used instead of the ambiguous 'V'. 'V' is normally reserved for Volts'. For instance: U b = 10 V, not V b = 10 V. Maips voltages No mains (power line) voltages are listed in Elektor circuits. It is assumed that our readers know what voltage is standard in their part of the worldl Readers in countries that use 60 Hz should note that Elektor circuits are designed for 50 Hz operation. This will not normally be a problem; however, in cases where the mains frequency is used for synchronisation some modifi- cation may be required. Technical services to readers # EPS service. Many Elektor articles include a lay-out for a printed circuit board. Some - but not all - of these boards are avail- able ready-etched and predrilled. The 'EPS print service list’ in the current issue always gives a com- plete list of available boards, e Technical queries. Members of the technical staff are available to answer technical queries (relating to articles published in Elektor) by telephone on Mondays from 14.00 to 16.30. Letters with technical queries should be addressed to: Dept. TQ. Please enclose a stamped, self addressed envelope; readers outside U.K. please enclose an IRC instead of a Missing link. Any important modifications to. additions to, improvements on or corrections in Elektor circuits are generally listed under the heeding 'Missing Link' at the earliest opportunity. contents 0 More and more people are reading Elektor A recent independent survey shows that on average 2.62 people read each copy of Elektor magazine each month. If you are one of the 70,000 who are not buying your own copy but reading someone elses we would like to draw your attention to the following:— Due to our growth of circulation and popularity of articles we find our stocks of back issues rapidly decreasing to such an extent that very soon the only readers with all 1979 copies will be those who purchased theirs on publication. However, it is still possible to avoid this by taking out a subscription now for the rest of the year. Of course there are many other good reasons for subscribing not to mention the fact that you would receive your own copy prior to general availability and if you are new to Elektor a subscription for the whole of 1 9 7 9 entitles you to a copy of ELEKTOR every issue published so far this year at reduced prices. A post paid order form is detailed overleaf for your convenience. J say m c sm ££££££££* STRUTT Electricals Mechanical Engineering Ltd 3C, Barley Market St., Tavistock, Devon. PL19 05F Tel. 0822 5439 Telex 45263 Printed passive components A new method of producing capacitors, coils and resistors on foil has been developed by Siemens. Introduced as ‘Sicufol’ circuits, these passive components are produced as flat zones. The rectangular, spiral and meandrous shapes of these zones and the material used determine the capacitances, inductances and resistances. The first modules produced with this technology were used in television sets. The ‘Sicufol’ principle is based on plastic foils made of polyimide or Teflon, which are first coated with copper in webs with lengths of several hundred meters. Chrome-nickel layers are then deposited, from which capacitors and resistors are later produced. The windings of the coils are etched directly out of the copper layer. The passive circuits formed next to each other on the webs in this way are cut off to produce finished modules. Protective layers made of insulating material provide both protection and reinforcement. In contrast to conventional circuit board technology, the components are deposited on both sides of the actual carrier, the plastic webs being coated on both sides with Cu and Cr-NL External ancillary devices such as potentiometers, filters and active integrated circuits are subsequently inserted; their contacts already constitute part of the finished ‘Sicufol’ circuits. The designers have paid particular attention to very close tolerances of the resistance variations with temperature and climatic conditions. It is claimed that the fluctuations measured are without exception under 0.5%. The temperature coefficient determining the mechanical strength is lowest for a chromium content of between 53% and 58% and on the average less than : 0 ppm/K. Layers comprising 20% chromium and 80% nickel have a temperature coefficient of 600 ppm/K. The users of ‘Sicufol’ circuits have at their disposal a noteworthy range of R, C and L ratings. Surface resistances of between 20 and 300 ohm are possible, the resistance can be increased 3800-fold by using meanders. The load carrying capacity is 0.5 W/cm 2 . Capacitors with a polyimide carrier foil as dielectric have a surface capacitance of 150 pF/cm 3 . Finally, inductances of up to 10 jtH can be produced using two spiral structures positioned directly one above the other on each side of the foil. Such an arrangement quadruples the inductance. The low thickness of the insulating plastic carrier permits a variety of through-plating methods, which can greatly facilitate the accommodation of new circuit designs. Furthermore, conductors made of copper can be soldered, even in a flow solder bath just like circuit boards if a polyimide foil is used (300°C permanent tempera- ture stability). The solderability and the additional reinforcement ensure that the components customary in circuit board technology can also be used without restriction for ‘Sicufol’ circuits. Siemens AG Zentralstelle fur Information Postfach 103 D-8000 Munchen 1 Federal Republic of Germany (385 s) Energy from space However much like science fiction it may sound, giant solar space power stations providing energy for domestic and industrial uses on Earth may be possible. If pilot studies to develop solar arrays into space power sources prove economically feasible, then by the end of the century giant orbiting solar power stations could be in use. Covering many miles, these would be constructed in space from basic units and materials transported from Earth by advanced launch vehicles. By using solar cells which convert the sun’s rays to electrical energy, the large arrays of today will, over the next decades, emerge as the early stepping stones which could lead to the utilisation of space initiated by the forthcoming Space Shuttle era. Under contract to the European Space Agency, Dynamics Group Electronic and Space Systems organisation at Bristol are already designing the huge 33 sq. metre (365 sq. ft.) 4 kW solar array that will power a space telescope for NASA. A 6 kW lightweight flexible fold-out array is also being developed for communication satellites of the next decade and proposals are in hand to augment the Space Shuttle power using solar power modules of up to 60 kW. Other proposals exist to develop arrays up to 500 kW as space power sources. These could form modules for further development up to 2 MW to provide a space power station pilot scheme which, if proved economically feasible, would provide a basic source of energy for use on Earth. British Aerospace Dynamics Group, Filton House, Bristol, England. (439 S) Desirable residence for sale ‘Situated in El Salvador in pleasant surroundings, modern detached fully equipped batchelor quarters with excellent facilities. Having an unusual roof design and an exceptionally good reception area, would suit a radio or electronics enthusiast. £ 6,000,000’. Actually, El Salvador, the smallest of the American states is the owner of this communications satellite ground station. Its 32 metre diameter parabolic antenna is operating telephone and telex, both data and pictures, through 1 20 channels, to North America and Europe via the Intelsat IV-A satellite. It is capable of being extended although some of the channels now being used are leased to neighbouring countries. The installation, built by Siemens of West Germany, was completed and operational in 1 8 months at a cost of 23 million DM. (437 S) 3-02 - elektor Laser gyro Ferranti are to develop a laser gyro and to incorporate it in an experimental model of a strap-down inertial navigation system. The aim is, first, to produce a system for installation in transport aircraft, followed, if tests and trials are satisfactory, by the development of a laser-gyro navigator for combat aircraft. Interest in the laser gyro exists because, potentially, it could provide a more robust and mechanically simpler device for detecting movement than the existing spinning-mass gyro employed today. Because of the greater simplicity it is also expected that the total cost of ownership (initial capital cost plus maintenance costs) of a laser-gyro inertial navigation system throughout its service life would be much less than the equivalent conventional gimbal gyro system. Currently the Ferranti Inertial Systems Department has developed a gimbal- supported inertial platform incorporating conventional spinning- mass gyros. However, a simpler method of sensing motion would be to eliminate the gimbal support and mount the gyros permanently in a fixed- position — the so-called ‘strap-down’ configuration. In a gimbal system the platform is always maintained in a fixed orientation in space — the datum reference position. With a strap-down system the displacement of the gyro with respect to the starting datum is assessed and stored by a computer associated with the system. The problem with the strap-down method is that the gyros used have to be mechanically robust and continue to perform with great accuracy even when subject to increased vibration and the higher angular and linear rates of accelerations implicit in being ‘strapped- down’. The spinning-mass gyros in the current designs of inertial platform are shielded to a considerable extent from these harsh conditions by virtue of being gimbal supported. The laser gyro, if it can be made to work with sufficient accuracy, offers a solution to this problem because it has no moving parts and is of rugged construction. Principle of the laser gyro The type of laser gyro Ferranti is developing is triangular in plan. It comprises a single block of vitreous ceramic material in which three cavities of circular cross-section have been drilled. The three cavities form a triangle in one plane. A mirror is positioned at the angular junction between each pair of cavities. The vitreous ceramic selected has a low temperature coefficient of expansion to ensure the mirrors remain accurately aligned with the cavities under all operational conditions. The cavities are filled with a lasing material and Ferranti have selected a helium neon gas mixture for this purpose. A cathode is situated midway along the length of one cavity and anodes are positioned towards the furthermost ends in the two remaining cavities. The helium neon mixture is excited to lase by electrical discharges occurring between the cathode and the two anodes. As a result, two independent coherent electromagnetic waves are generated - one travelling round the triangular resonating cavity in a clockwise direction, and the other in a counter-clockwise direction. If the assembly is now spun about a vertical axis perpendicular to the plane of the cavities the effective path around which one electromagnetic wave is travelling will appear to lengthen, and the other to shorten. To JL a detector placed at one of the reflecting mirrors the frequency of the electromagnetic wave propagated in one direction will appear to increase while that of the other wave will decrease. The difference in frequency between the two waves is directly proportional to the rate of rotation of the chamber about its perpendicular axis. The detector produces output pulses at a frequency proportional to the rate of rotation. Angular rates of acceleration can be determined by assessing changes in output pulse frequency. This device has all the characteristics of a gyro - hence the laser gyro. The major technical problems that have to be solved before the laser gyro becomes a practical proposition are : a) the generation of a truly linear output signal at extremely low rates of angular rotation (the two wave patterns tend to ‘lock together’ under these conditions) and b) to devise manufacturing techniques that enable the required high accuracies to be consistently achieved and maintained on a production basis. Should the results attained in the new development programme be satisfactory it is expected that, from the mid to late 1980s onwards, laser gyros incorporated in ‘strap-down’ systems will be specified for many airborne applications. Inertial Systems Department, Ferranti Limited, Silverknowes, Ferry Road, Edinburgh, EH4 4AD. (440 S» RCA enter the video disc market Video recording is probably the biggest new innovation in home entertainment since the colour television and RCA are not alone in their view that video discs will be a multi-million dollar business in the 1980’s. This company plan to achieve the earliest possible wide scale distribution of their ‘Selecta Vision’ video disc in the United States. The RCA system uses a grooved disc that is played with a diamond stylus at 450 revolutions per minute and contains one hour of programme per side. The player is designed for use with any television set, keeping the overall cost to a minimum. An unusual feature is the plastic sleeve resembling a record album cover, which deposits the disc on the turntable when inserted into a slot on the front of the machine. The disc is elektor march 1 979 — 3-03 —BSSSSCl; removed by reinserting the empty sleeve back into the player. In this system the sleeve is a protection against environmental hazards such as warping, dust and scratches and also prevents the disc being touched by hand. RCA’s initial catalogue will contain 250 titles, including feature motion pictures, as well as childrens, how-to, sports, cultural, educational and musical programmes. RCA International Ltd. RCA House, 50, Curzon Street, London, WIY8EU England. (438 S) pP-control for church organ A microprocessor register control system has recently been fitted to the organ in the 600 year old St. Lorenz Church in Nuremberg. The system, known as the ‘Registronik’, was specially developed for the 138-register organ by two engineers of Siemens Geratewerk, Erlangen. Up to forty register combinations can be stored for instant recall, so that the stops required for different pieces of music can be set before the performance. Whole recitals or the entire music for a church service can be preprogrammed in this way, leaving the organist free to operate the manuals and pedals so that he can give his full attention to the music itself. The control system is a very compact design, the stop keys being grouped together on a panel no larger than a notebook. The associated electronics are accomodated in a small metal cabinet in an adjoining room. The organ was previously equipped with an electromechanical four-combination register control system and, using conventional methods, forty combinations would have required more space than the rest of the organ. The organ which, except for the Passau organ, is the largest in the Federal Republic of Germany will certainly attract many people to Nuremberg. (441 S) Little L A new range of ceramic chip inductors, available from Steatite Insulations Ltd, will not pass through the eye of a needle but the dimensions are sufficiently small for these electronics components to be used in thick film switch circuits. These miniature ceramic chip inductors are an entirely new development designed for hybrid technology with a fixed inductance. With nominal dimen- sions of only 2.5 mm x 2.5 mm x 1.9 mm, they have inductance ranges from approximately 4 nH to 1 jiH, with tolerances of ± 5 and 1 0%. The coil winding is applied as a single layer and a new process has been developed so that the winding ends can be welded to metallised surface terminals on the ceramic body. They can also be soldered together with other components and are claimed to be impervious to attack from all general cleaning fluids. Higher inductance units are currently under development utilising magnetic bodies together with variable inductance components. Steatite Insulations Ltd., Hagley House, Hagley Road, Birmingham, BI6 8QW, England (442 S) Are microprocessors nasty? Apparently, microprocessors are nasty things — or so one would assume from the amusing title of a recent press release! However, the organisers of the All Electronics Show are also making every effort to bring the manufactures and potential users into (hopefully profitable) contact. A unique gathering .... A unique gathering of the manufactures of microprocessors (or ‘chips’, as they are familiarly known) has been arranged in the West End of London at which, free of charge, visitors may discuss how this new technology will affect them. Among those present are Texas Instruments, who invented the integrated circuit, and Fairchild Semi- conductor who are now seen as the technology leaders. The organisers of the event (The All- Electronics Show at Grosvenor House in Park Lane between 27th February and 1st March) claim that there is now virtually no industry in Britain which will not be affected by microprocessors. They add that every director and senior manager of a company manufacturing anything should get face-to-face with the manufacturers and distributors of a technology which the Government is supporting to the tune of £ 70 million per annum. If you would like to ask even the most basic of questions, just send a stamp and addressed envelope to Sam Clarke, Dept. 3, 34-36 High Street, Saffron Walden, Essex and a free admission ticket - together with a special listing of the appropriate exhibitors — will be sent to you. The All-Electronics Show, Ars Electronica Ltd., 34-36 High Street, Saffron Walden Essex (443 SI Si**"' 06^ rs 0( 0 , «,» a"*' 9 *' l Old 0 ’' 3-04 - elektor march 1979 crosstalk canceller crosstalk canceller Separating left from right. The weakest link in the hi-fi chain as far as crosstalk is concerned is the pickup. A Japanese firm, however, has recently introduced a special unit which, it is claimed, can dramatically improve the performance of cartridges in this respect. The following article takes a look at this interesting development. Stereo reproduction has been with us now for a good twenty years or so, and the principles involved are well- known. Two separate sound channels are modulated in two different planes in the grooves of a disc. This 'left and right channel’ information must be fed through separate amplifier channels to separate loudspeakers. If this is achieved, the result is more or less accurate ‘positioning’ of individual instruments within an overall sound image or stereo ‘picture’ as it is called (see figure 1). The position of a par- ticular instrument, or of vocals, within the stereo image is determined by the proportion (and phase) of the corre- sponding electrical signal present in each channel. The greater the difference in signal strength, the more the resulting sound will appear to be shifted towards the loudspeaker of the channel where it is strongest. Talking about crosstalk In the ideal situation the left and right channel information is only combined as the output sound waves of the two loudspeakers encounter one another in the listening room. In practice, however, it is unfortunately the case that neither channel is completely free of some small part of the other channel. The effect of this is to reduce the differences between the two channels so that they tend to sound more similar. As figure 2 illus- trates, the process is analogous to mixing a touch of black with a white paint and vice versa. The result is a light and dark grey, which provide much crosstalk canceller slektor march 1979 - 3-05 s less of a contrast than pure black or white. In audio terminology this is known as crosstalk. The greater the crosstalk, the greater the similarity between the two channels. This results in a smaller stereo picture (figure lb), giving the sound a greater approxi- mation to mono reproduction. Crosstalk can be caused in amplifiers by capacitive and/or inductive coupling between the wiring and layout of the two channels, or via a common supply line. Poorly designed balance controls are another common cause of crosstalk. In general, however, crosstalk produced in the amplifier is of secondary import- ance, since there is another point in the hi-fi chain which is much more critical. Furthermore it is possible to eliminate amplifier-induced crosstalk virtually completely provided one is willing to go to the expense of employing a separate supply for both channels (both in the pre-amp and power amp), separate pcbs etc., in short, with the exception of the case and mains plug, using completely separate mono amplifiers for each channel. As mentioned, the weak link in the chain for crosstalk is not the amplifier, iu: the cartridge. In the particularly crucial range of frequencies between several hundred and several thousand Hz, channel separation is typically not =-ch better than 25 dB. At higher frequencies this figure is even lower, however it is also less important. The reasons for the comparatively poor channel separation of pickup cartridges are fairly complex and would require a considerable explanation. Suffice to say that the cartridge is by far the worst culprit when it comes to producing crosstalk. 'Phasing out' crosstalk Ideally one would like to increase the channel separation of cartridges to around 40 dB (a figure of roughly 40 to 50 dB is the best that can be obtained in the actual cutting process of a disc and in the tracking performance of the stylus). However, improvements of this order now seem possible with a new approach by the Japanese Company, Denon (see figure 3). The circuit which, according to Denon, virtually eliminates pickup crosstalk is called the Phono Crosstalk Canceller (PCC). The unit, which is incorporated into a number of Denon amplifiers (there are plans to market the unit separately), is pro- vided with four potentiometer con- trols, two of which are employed to (audibly) eliminate crosstalk from the left-hand channel to the right-hand channel, whilst the other two are set for minimum crosstalk from the right- hand to the left-hand channel. The adjustment procedure is performed with the aid of a test record. The operation of the crosstalk canceller Figure 1. Crosstalk results in the stereo sound 'picture' being 'compressed', so that the individual instruments/vocal tracks appear to be pushed closer towards a central point. Figure 2. Graphic representation of the effects of crosstalk. The larger the cross- hatched area, the greater the crosstalk and nearer the resultant sound is to mono. The crosstalk is shown as being asymmetrical, i.e. greater from one channel to the other than Figure 3. Channel separation as a function of frequency with (a) and without Ibl crosstalk suppression. can best be illustrated with reference to the series of phasor diagrams in figure 4. If a voltage is summed with a second voltage which is of equal magnitude but in antiphase to the first, the result is of course zero voltage. The vectors rep- resenting two such voltages are shown in figure 4a, and the sum of these two vectors would be a third vector of zero length, i.e. no vector at all. Assume now that two voltages A and B have a phase difference^, the cor- responding vectors will also form an angle < p (4b); the sum of these two voltages will be represented by the vector C, and the difference (A-B) by the vector D. If, in figure 4c, A represents the signal voltage of one stereo channel, whilst B represents the crosstalk voltage from the other channel, then it is possible to resolve B into a crosstalk component C, which will be in phase with the signal voltage, and a crosstalk component D, which will be 90° out of phase with the signal voltage. It is thus further possible to eliminate the crosstalk signal by introducing voltages C' and D' (which produce B') and summing these with the single channel information + crosstalk signal. The same is equally true for figure 4d, where the crosstalk signal B is 180° out of phase with that of figure 4c. Were the value of and the magnitude of the crosstalk vector constant, i.e. 3-06 - elektor march 1979 crosstalk canceller 4a 4b Figure 4. Phasor diagrams illustrating how the crosstalk signal can be eliminated by summing it with a signal of equal amplitude but opposite phase. Figure 5. Block diagram of the Denon Phono Crosstalk Canceller. Figure 6. Circuit diagram of the PCC-1000. independent of frequency, the sup- pression of crosstalk would be com- plete. Unfortunately, however, as a glance at figure 3 makes clear, that is not the case. Nonetheless, it is a fact that suppression of crosstalk is most important in the mid-range of fre- quencies, where both < p and the length of the crosstalk vector (channel sep- aration in dB) are more or less constant. Thus the principle remains valid. In practice there are pickup elements with a crosstalk voltage as shown in both figure 4c and figure 4d, which means that the C' and D' voltages not only must be independently variable, but must be available in both phase versions. For independent control of the C' and D' voltages, two potentiometers per channel are required, and since the crosstalk from one channel to another is not necessarily the same as that going the other way (in fact it generally is not), four potentiometers in all are needed. Block diagram of the PCC-1000 Figure 5 shows the block diagram of the Denon Phono Crosstalk Canceller PCC-1000. Each channel comprises two 180° phase-shifters, one 90° phase- shifter, two single-ganged potentio- meters with fixed, earthed centre taps, and a summing circuit. The wiper volt- ages of the potentiometers in each chan- nel are fed to the summing circuit of the other channel. The fixed, earthed centre taps ensure that the compen- sation voltages are available in both polarities depending upon the type of pickup being used. Circuit The circuit diagram of the PCC-1000 is shown in figure 6. The unit is connected between the tape output and tape input (monitor input) of the preamp. Since the preamp tape sockets are no longer available for their original application, the PCC-1000 itself provides the necess- ary connections. Switch S2 is the new tape monitor switch. Switch S3 selects (1) ‘left channel only’, (3) Tight channel only’, and (2) ‘nor- mal’, (i.e. stereo). The bypass switch for the unit is SI. Working our way back from the out- puts, transistors TR5, 7 and 9 (6,8 and 10) plus accompanying passive com- ponents form the summing amplifier to which the three input signals (see figure 5) are fed via R29, R31 and R33 (R30, R32, R34). The junction of these resistors is at virtual earth. The voltage at the collector of TR1 (TR2) is in antiphase with the input voltage. The wiper voltage of VR1 (VR2) determines the amplitude and phase (polarity) of the compensation voltage C' (see figures 4c and 4d). The RC network C7/R15 (C8/R16) introduces a phase shift of 90 , whilst TR3 again causes a 180 phase shift between in- and output voltage. The level and polarity of the D' compen- sation voltage is determined by the setting of VR3 (VR4). That basically is all that there is to the circuit. The remaining components are part of the power supply stage. In conclusion The PCC-1000 crosstalk suppressor from Denon, is certainly an interesting and potentially valuable idea. However the question is whether the improve- ment in channel separation is matched by an equal improvement in the quality of the resultant sound. The problem of evaluating hi-fi equipment is fraught with the dangers of subjectivity, and it has not been unknown for a reviewer to remark upon the better stereo imaging of a particular pickup for example, when that pickup in fact had a poorer channel separation than others which were under review. This is a subject which we hope to come back to in the future. At least one can say that for anyone interested in ‘state-of-the- art’ hi-fi the crosstalk canceller is definitely well worth a listen. M 3-08 - elektor march 1979 robust lab power »upply robust lab power supply An essential feature of any electronics enthusiast's lab is a reliable power supply unit. The basic requirements for such a unit are that it provides a fully stabilised continuously variable output voltage and should be fully protected against eventual fault conditions such as output short circuits. The circuit described here meets all the above points, is both simple and inexpensive, and should provide years of trouble- free service. Until only a few years ago, power supply units almost exclusively employ- ed discrete regulator circuits. However with the advent of cheap universal precision voltage regulator ICs, it has become possible for the amateur to build an inexpensive PSU enjoying the specifications which previously were the preserve of expensive professional equipment. The basic function of a voltage regulator is twofold. Firstly, to maintain a con- stant output voltage in spite of variations in the input voltage (i.e. the mains). Its performance in this respect is termed line regulation and is expressed as the percentage change in the input voltage which is passed on to the output voltage. Thus, with a line regulation of 0.1% — the figure for the circuit described here - a change of 10 V in the mains supply will produce a variation of not more than 0.1% of 10V = 0.01V in the output voltage of the regulator circuit. The second function of a regulator is to maintain a constant output voltage despite variations in the current drawn by the load. Load regulation is ex- pressed as the percentage change in the output voltage for a specific change in the output load current (or when the load current is varied over its complete range). Thus, in this circuit the output voltage will not vary by more than 1% for fluctuations of up to 5 A in the current drawn by the load. Double stabilisation As can be seen from the block diagram of the PSU (see figure 1), the design of the circuit is slightly unusual in that it incorporates a pre-stabiliser stage be- tween the unregulated supply and the output regulator proper. There are several reasons for adopting this ap- proach. Firstly, the actual stabiliser does not have to cope with large line voltage variations, whilst secondly, and more importantly, the dissipation of the circuit is spread over two stabilisers. Finally, the pre-stabiliser is required to limit the input voltage of the regu- lator IC used in the circuit. Apart from the preliminary stage, the design of the PSU is fairly conventional: mains step-down transformer plus recti- fier, smoothing capacitor, the two series-connected regulator stages and finally, a meter circuit to measure the output voltage/current. The first stabil- iser circuit is provided with current limiting, whilst the second is protected against short-circuits and thermal over- load. In view of the fact that, as we shall see, the supply is also protected against both negative voltages and large positive voltage transients, the circuit is virtually ‘idiot-proof and definitely merits the accolade ‘robust’. Circuit The complete circuit diagram of the PSU is shown in figure 2. Two versions of the circuit - one designed to supply a maximum of 5 A, and a simpler version which provides 2.5 A — are presented. In both cases the output voltage can be varied between 5 and 20 V with the aid of potentiometer P 1 . The differences between the two versions of the circuit are detailed in table 1. What may initially appear to be a rather curious problem with regulator circuits is the fact that the lower the output voltage, the greater the dissi- pation. The reason for this is not diffi- cult to explain however, since the less power ‘dissipated’ in the form of an output voltage, the more power is ‘left over' and hence must be dissipated in the output transistors of the regulator circuit itself. Thus it only makes sense to limit the input voltage of the circuit whenever only low output voltages are required. To this end the transformer is provided with both 1 2 V and 24 V secondary windings whjch can he switched with the aid of S2. In view of the lower current, the dissipation of the 2.5 A version of the circuit is not a problem, hence the secondary voltage of the transformer can safely be left at 24 V. The rectifying and smoothing stages of the circuit are completely conventional. The presence or absence of the second smoothing capacitor, C2, determines the size of the ripple voltage at the output. For the 2.5 A version of the circuit C2 robust lab power supply elektor march 1979 — 3-09 specifications: output voltage: 5 ... 20 V continuously adjustable output current: depending upon the model; 2.5 or 5 A current limiting: at a single fixed value, plus short-circuit protection at 5 A LED current limit indicator meter: single switched meter for both voltage and current load regulation: < 1% line regulation: < 0.1% (referenced to line voltage variations) noise voltage: < 75 pV ripple voltage: (100 Hz): < 500 pV Figure 1. Block diagram of the 'robust' lab PSU. Figure 2. Complete circuit diagram. Two versions of the circuit are possible: one provides a maximum output current of 5 A, the other is a simpler version with a maximum output of 2.5 A. can be omitted; but it must be included in the 5 A version for the specifications listed above to apply. The pre-stabiliser stage is formed by T1 and T2 which are connected as a con- ventional series stabiliser pair. The reference voltage is derived from zener diode D 1 . The circuit only functions as a stabiliser when S2 is in position ‘a’, however, since only then is it necessary to limit the input voltage of the output regulator or to split the dissipation of the circuit over two stages. Assuming that S2 is in position ‘a’, the voltage at point B will be limited to between 25 and 26 V (approx). Current limiting is provided by T3 and R5. When the voltage across R5 exceeds robust lab power supply elektor march 1979 — 3-11 approx, 0,7 V, T3 turns on, turning off T1 and T2, and lighting LED D2. With the component values shown, the current limit comes into operation at approx. 100 mA, although this can be varied by altering the value of the current sense resistor R5. One should bear in mind that the dissipation of this resistor will be a maximum of 0.7 times the current limit in amps. The maxi- mum current should not be allowed to exceed 2 A, since at that stage, should a short-circuit occur at the output, some- thing like 60 W is already going to be dissipated in T2. The current limit facility can be dis- abled by closing S3, when the circuit will simply be short-circuit proof, the current being limited in such an eventu- ality to the maximum current of 5 A. The output voltage regulator is formed by the 1C type n A 78HG from Fairchild. This device provides a stabilised output voltage which can be continuously varied between 5 and 24 V and which in normal use is virtually impossible to damage. The IC also provides short- circuit and thermal overload protection. The principal specifications of the HA. 78HG are listed in table 2, while pin- out details for the four-lead TO-3 pack- age are shown in figure 2. The output voltage is set by means of potentiometer P 1 , which together with R7 forms a variable voltage divider. The IC controls the output voltage such that the voltage at the ‘control’ input (pin 3), which is derived from the voltage divider, is always 5 V. D3 and D4 are included to protect the IC from output voltages which might exceed the input voltage, a situation which can occur when using the PSU to charge batteries for example. In the 2.5 A version of the circuit one of the diodes can be omitted. D5 protects the circuit against any negative voltage transients which might find their way to the output of the circuit. The PSU employs a single meter to display both voltage and current, and switch S4 selects one mode or the other. R6 is a shunt resistor for the current measurement and the meter scale can be calibrated by means of P2. Calibration for voltage measurements is not re- quired. Instead of a moving coil meter it is possible to use the universal digital meter published in the January issue (Elektor 45), in which case some component values need to be changed. Resistors R7 and R8 of the digital meter should be changed for a wire link and a lk 1% resistor respectively, while in the PSU circuit R8 should be altered to 19 k (18 k and 1 k in series, both 1%). The above meter can be powered directly from point A of the PSU circuit. 4 Construction To ensure a long and trouble-free operating life it is worthwhile spending robust lab power supply elektor march 1979-3-13 Table 1. Differences between 2.5 and 5 A versions of the PSU 2.5 A 5 A FI 250 mA fuse 500 mA fuse Trl 2x12 V/3.5 A 2x12 V/7 A transformer transformer S2 omitted SP switch, 7 A B1 B40C2500 B40C5000 D4 omitted 1N5406 Figure 5. Wiring details for the construction of the PSU. Figure 6. This graph illustrates the relation- ship between output current and output voltage for the two positions of S2. Specifications of pA 78HG max, dissipation: 50 W (at 25°C) max. input voltage: 40 V max. voltage difference between in- and output: 25 V max. current: 7 A control voltage: 4.8 V ... 5.2 V load regulation: < 1 % line regulation: < 1 % quiescent current: <10 mA ripple suppression: > 60 dB noise voltage: 75 pV a certain amount of attention on the constructional details of the PSU. The printed circuit board for the project is shown in figure 3. However, since a comparatively large number of compo- nents must be cooled, most of these are mounted off-board. Particular attention must be given to the cooling arrange- ments for the bridge rectifier, B 1 , transistors T1 and T2, and the voltage regulator, IC1 . In the case of the IC, one must reckon with a dissipation of some- thing like 50 W, hence a beefy heat sink is a must. Similarly, adequate precautions should be taken with T2, since it will dissipate anything up to 60 W. The bridge recti- fier and T1 are not quite such a problem and could be cooled by, for example, mounting them on the case. However, do not forget to electrically insulate the components being cooled (mica washers and silicon grease) - with the possible exception of IC 1 wich may be mounted without insulation. The shunt resistor, R6, will have to be self-wound. The simplest method of doing this is to wind 36 cm of 0.6 mm ( 24 s.w.g.) diameter enamelled copper wire on a 1 W resistor (e.g. 10 k). The inductance of such a home-made resistor can be neglected in this type of application. Care should be taken to ensure that the connecting wire is rated to carry the full 5 A the circuit can supply. Wiring details are shown in figure 5. Note that in contrast to normal usage the case should not be connected to circuit earth, but rather is earthed via a seperate socket. In this way the circuit can be used for both positive and negative voltages. The only component in the circuit requiring initial adjustment is P2, which is used to calibrate the meter for current. The simplest way of doing this is to set the PSU for, say, 10 V out, and then load the circuit with 45 Watt car headlamp bulb, with an ammeter in series (full scale deflection at least 4 A). With S4 in position ‘I’ the meter deflec- tion can be adjusted until it shows the same reading. A slightly less reliable method is to inhibit the circuit’s current limit facility, and with the output short-circuited, adjust P2 for a full- scale deflection (corresponding to a current of 5 A). Operation Using the PSU is quite straightforward, the only problem which might present itself is switch S2 (which, of course, is only present in the 5 A version of the circuit). As was already mentioned, this switch should be set to position ‘b’ for high-output-current low-output-voltage applications. Should the switch be left in position ‘a’ in such circumstances, the thermal shutdown facility of the IC will quickly come into effect, since the device will be dissipating more than the permitted 50 W. The dissipation is reduced to acceptable proportions by changing over the position of the switch, with the result that the maximum output voltage is limited to approx. 12 V. The graph in figure 6 illustrates the relationship between output voltage and current for the two positions of S2. N 3-14 — elektor inarch 1979 modulator ring modulator and envelope follower A ring modulator is a circuit which was originally employed in telecommunications systems for the modulation and detection of transmission signals. More recently however, the ring modulator has found an interesting application in the field of electronic music and is in fact now a common feature in many synthesisers. A ring modulator is basically a four quadrant multiplier, that is to say, a circuit which will multiply two input voltages, regardless of whether they are positive or negative and ensure that the product voltage is of the correct po- larity. Thus a positive voltage multiplied by a negative voltage will yield a negative voltage, a negative voltage times a negative voltage will give a positive voltage, and so on. The question is: why is such a circuit of interest to the electronic music enthusi- ast? The answer can be found by con- sidering the following mathematical expression for the product of two sinewaves: sin o • sin 0 = 'A cos (a-0) - Vi cos (a+0). Since a cosine is simply a sinewave with a 90° phase shift, it can be seen that multiplying two sinewaves results in two new sinewave signals whose frequencies are the sum and difference respectively of the two original signals. Note that this is only true for sinewave signals and not for other types of waveform. However the same effect will be produced by com- binations of sinewaves. Thus, for example, if a combination of two sinewaves is multiplied with a third sinewave, each of the constituent sinewaves in the original signal will produce its ‘own’ sum and difference products. The multiplication of two sinewave input signals is illustrated in the oscilloscope photo of figure 1 . The sinewave of the upper trace is multiplied with a second sinewave of higher fre- quency to produce the product waveform shown on the lower trace. 'Klangs' The most significant feature of the ring modulator is its ability to exploit the harmonic relationship of different notes. This can best be explained with the aid of a further example. Assume that we feed two sinewave signals with fre- quencies of 2.5 and 4.5 kHz respectively to one of the ring modulator inputs. The ratio of these two frequencies is 5:9, which means that, in musical terms, the resultant note is roughly equivalent to a lower seventh (the actual fre- quencies are somewhat on the high side, however they are only chosen to illus- trate an example). If now a third sinewave with a frequency of 500 Hz is fed to the other input of the ring modulator, what will appear at the output? The 2.5 kHz signal multiplied with the 500 Hz tone produces two new signals of 2 and 3 kHz respectively. Similarly, the 4.5 kHz and 500 Hz signals will produce two new signals of 4 and 5 kHz. Thus at the output of the ring modulator will be four signals with frequencies of 2, 3, 4 and 5 kHz, i.e. a major chord. The musical relationship of the lower seventh has therefore been transformed into a different musical relationship, that of a major chord. However, the above example is not typical, since it will be the exception rather than the rule that musically related frequencies at the input of the ring modulator will also produce a musically coherent chord at the output. In the vast majority of cases the har- monic relationship of the sum and difference signals produced at the out- put of the ring modulator will be uncorrelated, resulting in a dissonant, unmusical sound. This is particularly true if, instead of sinewaves, other types of waveform are used as input signals. Figure 2 illustrates what happens when a sinewave is multi- plied with a squarewave input. It is well known that non-sinusoidal periodic waveforms can be considered as con- sisting of a sinusoidal fundamental with the frequency of the signal in question, plus a number of harmonics of the fundamental, i.e. sinewaves whose fre- quency are a multiple of the fundamen- tal frequency. Thus, for example, a sawtooth waveform of 1 kHz consists of sinewaves of 1 kHz, 2 kHz, 3 kHz . . . etc. The character of the resultant note depends upon the relative strength of the constituent harmonics. If a sawtooth is fed to one input of a ring modulator elektor march 1979 — 3-15 ring modulator and a pure sinewave of, e.g. 300 Hz, is fed to the other input, then each harmonic of the sawtooth will be multiplied by the 300 Hz sinewave, and a series of signals with frequencies of 0.7 kHz, 1.3 kHz, 1.7 kHz, 2.3 kHz, 2.7 kHz, 3.3 kHz etc. will be produced at the output. Thus the ring modulator has converted the original 1 kHz sawtooth and the 300 Hz sinewave into a complex note composed of musically unrelated harmonics. If now the 300 Hz sinewave is replaced by a second sawtooth signal, the harmonic structure of the resulting output signal is even ‘denser’ and more complex. At the lower end of the spectrum alone, the output signal will contain the following frequencies: 100, 300, 400, 500, 600, 700, 800,900, 1100, 1200, 1300, 1400, 1500, 1600 and 1700 Hz. Each of these tones will have a characteristic ampli- tude, with particular frequencies tending to predominate whilst others are relatively attenuated. Due to its extremely complex harmonic structure, the timbre of the resultant signal resembles that of a large bell or gong, or the sound of metal striking metal (hammer on an anvil etc.). This type of percussive effect is called a klang, and is frequently used by composers of electronic music. The potential of the ring modulator can best be exploited if both input signals are varied in frequency (e.g. modulated by a low frequency signal). The result is sounds which exhibit tremendous vari- ations in both pitch (as far as one can still talk of the ‘pitch’ of such sounds) and tone colour, and which run the gamut of tonal possibilities between pure harmonies and the shrillest of dissonances. Extremely interesting effects can also be obtained by combining ‘normal’ sounds with a noise signal, by using the ring modulator in conjunction with various types of filter, and by using a combination of several ring modulators. The well-known modem composer Stockhausen once wrote a work for Hammond organ and four ring modulators! Frequency doubler The ring modulator can also be em- ployed in more ‘conventional’ musical applications as a frequency doubler or octave shifter (doubling the frequency is of course equivalent to shifting the pitch of the signal up an octave). To achieve this effect one simply feeds the same signal to both inputs of the ring modulator. It is clear that the difference frequency of the two input signals will in this case be 0 Hz, i.e. there will be no difference signal at the output, whilst the sum signal will have double the fre- quency of the original input signal(s). If the ring modulator is used as a fre- quency doubler for non-sinusoid al and polyphonic signals considerable inter- modulation between the constituent harmonics will produce an interesting range of effects. A further possibility is to process one of the two input signals through a phasing or echo unit. Finally, it is also possible to use the ring modulator in a somewhat less conven- tional mode, namely as a voltage con- trolled amplifier. The control voltage is fed to one input, whilst the signal to be modulated is fed to the other. The ring modulator as an instrument The above remarks give only a brief outline of the ‘musical’ applications of the ring modulator. However it is plain that, as an instrument, it is ideally suited for those interested in the field of experimental music, the persons looking for totally new tonal effects. The ring modulator is a ‘difficult’ instrument, which, if its capabilities are to be exploited to the full, demands a considerable degree of skill and knowl- edge on the part of the operator. Nonetheless, the ring modulator is a basic feature of most reasonably-sized synthesisers, as well as being a common accessory in the equipment of guitarists, ‘keyboard players’ and other instrumen- talists. Figure 1. The signal shown on the lower trace of this photo is the result of the ring modulator multiplying the sinewave of the upper trace with a second sinewave of much higher frequency. Figure 2. This photo illustrates what happens when a squarewave (upper trace) and sinewave signal are multiplied together in the ring modulator. The result is shown on the Figure 3. Block diagram of the Elektor ring modulator. The ring modulator is not a ring modulator After the foregoing - of necessity - somewhat lengthy digression, we can now concentrate on the technical aspect of the circuit. First, however, it is worth clearing up a slight confusion which unfortunately exists regarding the true name of the circuit in question. The term ‘ring modulator’ in fact des- cribes a particular type of circuit, which happens to function as a four quadrant multiplier (at least as far as AC voltages are concerned), and in the early years of electronic music was used to denote specifically this effect. In the intervening period, however, new and better circuits have been designed to accomplish the same end, and these are now used almost exclusively when it comes to musical applications. The name ring modulator remained, however, since most musicians are interested only in what comes out of the ‘black box’ and not what it contains. A more accurate name for the type of multiplier used in the majority of modem ring modulator circuits is ‘double-balanced modulator’. This is a somewhat delicate circuit, consisting of a combination of voltage-controlled current sources. Fortunately the complete circuit of a double-balanced modulator is now available in IC form, so that all that is required to construct a ‘ring modulator’ suitable for musical applications is the addition of a few ancillary components. Circuit The block diagram of the Elektor ring modulator is shown in figure 3. It will be seen that the ring modulator (shown marked with an X sign) has three avail- able inputs. Both the A and B inputs accept signal levels of up to approxi- mately 1.5 volts peak to peak and are therefore compatible with the Elektor Formant and other synthesisers. Input C has a preamplifier with a maximum input level of 1 0 mV and is sufficiently sensitive for the majority of guitar pickups and microphones. An additional feature of this circuit is that the B and C inputs can be used simultaneously as they are mixed prior to the input of the ring modulator IC. A further addition to the circuit (although not a functional part of the ring modulator) utilises two op-amps, A2 and A4, to form a peak rectifier and amplifier providing an envelope follower circuit which has an output level of up to 10 volts peak to peak. This gives a waveform output which is relative to the envelope of the low level instrument input (input C) which may be used in conjunction with synthesisers. The complete circuit diagram of the ring modulator is shown in figure 4. The heart of the circuit Is IC 1 , the double- balanced modulator, which is responsible for multiplying the input signals. The IC used is the LM 1496N from National (or MC 1496P from Motorola). A number of external resistors are required for the 1C to function satisfactorily. It is necess- ary to limit the amplitude of the two output signals, otherwise there is the danger that the input signals will not be sufficiently suppressed and will appear at the output. For this reason the input Figure 4. Complete circuit diagram of the ring modulator. The actual process of 'ring modulation' is carried out in the double- balanced modulator IC1. Figure 5. Printed circuit board for the circuit of figure 4 (EPS 79040). signals are held to a reasonable level (max. approx. 150mV p p) with the aid of the divider networks R1/R3 and R8/R 1 1 . This has the effect of ensuring that the input signals are approx. 50 dB down at the output. The suppression can be optimised by means of adjusting potentiometers P2 and P3. R6 and R7 set the correct DC offset voltages at pins 8 and 10 of the IC, whilst the remaining associated resistors round the IC ensure the correct DC bias currents. The output voltage is tailored to match the standard Formant level of 1.5 Vpp. Op-amp A3 functions simply as an output buffer. The C input, which uses op-amp A1 as a pre-amplifier stage, has been designed for guitar pickups, microphones etc. The input level to this stage is adjustable by means of P 1 while the output is fed to the ring modulator via R9 and also to the envelope follower via C5. Diodes D1 and D2 are included to clamp excessively large voltages at this point. The remaining two op-amps, A2 and A4, are used in the envelope follower. Together D3, C6 and A2 form the peak rectifier. The rectified signal is then fed through a lowpass filter with a turnover frequency of 10 Hz. Finally, op-amp A4 ensures that the output voltage of the envelope elektor march 1979 — 3-17 follower can vary between approxi- mately 0 and 1 0 volts. Construction and setting-up The circuit can be mounted on the p.c.b. shown in figure 5. In addition to the MC 1496P, a further equivalent for the LM 1496N exists, namely the S 5596 from Signetics. Unfortunately, however, this IC has a different pin-out, and hence cannot be used with the p.c.b. of figure 5. The circuit requires supply voltages of + 15 V and -15 V. Current consumption is extremely low, being in the region of a few dozen milliamps. Setting up the circuit is quite straight- forward: an input signal is fed to input A and potentiometer P3 is adjusted such that as little of the input signal as possible can be heard at the output. The same procedure is then carried out for input B and poten- tiometer P2. Finally, the entire setting- up procedure should be repeated, whereupon the circuit is ready for use, and the experimental musician can embark upon what will hopefully be a i fruitful ‘voyage’ of musical discovery . . . M 3-18 -elektor march 1979 variable pulse generator variable Many digital applications require the use of a pulse generator of which not only the frequency, but also the duty-cycle can be varied. A problem with certain simple variable pulse generators is that altering the duty-cycle also affects the frequency of the output signal. The circuit described here, however, which uses only a handful of components, is free from this drawback; both frequency and duty-cycle are independently variable. The frequency range extends from approx. 1 kHz to 20 kHz, whilst the duty-cycle can be varied from almost 0 to 100%. K. Kraft Figure 1. The circuit of the variable pulse generator employs only two ICs, yet both the frequency and duty-cycle ere independently variable. Figure 2. This timing diagram illustrates how the duty-cycle of the output signal is deter- mined by the reference voltage of the com- parator (U re f). Furthermore, by varying the RC-constant of the integrating network in sympathy with that of the multivibrator it is possible to make the duty-cycle independent of the frequency. pulse generator The complete circuit diagram of the variable pulse generator is shown in figure 1. As can be seen, the circuit is extremely simple indeed. The pulses are generated by an astable multivibrator round N 1 . This provides a symmetrical squarewave (duty-cycle = 50%), the fre- quency of which can be varied by means of PI a. The squarewave is cleaned up by N2 and is available via an extra external output. To allow the duty-cycle to be varied without affecting the frequency of the squarewave, the circuit employs an integrating network (Plb/R2/C2) and a comparator (IC1). The RC-constant of the integrating network (C2 = 1/6 • Cl) is chosen such that the voltage across C2 may vary between approx. 20 and 80% of the supply voltage, Ub- Whenever this voltage exceeds the reference voltage on the inverting input of the comparator, the output of the latter changes state. The result is therefore a squarewave signal (U x ) whose duty-cycle is deter- mined by the reference voltage (U re f) of the comparator. This process is clearly illustrated in the timing diagram of figure 2. By varying the voltage at the inverting input of the comparator it is therefore possible to adjust the duty- cycle of the squarewave as desired without affecting the frequency. There now remains the question of what happens to the duty-cycle if the fre- quency of the squarewave is varied. Normally the duty-cycle would be influenced by the frequency change, however due to the use of a twin-ganged potentiometer (Pla/Plb), in the circuit shown here, the RC-constant of the inte- grating network will vary in sympathy with that of the multivibrator. If the frequency, f, of the multivibrator is increased to x-f, the period of the resulting squarewave will be reduced by a factor x. However since the RC- constant of the integrating network is likewise reduced by a factor x, the duty- cycle of the squarewave at the output of the comparator will remain unchanged. It is not difficult to see that altering the RC-constant of the integrating network will not affect the shape of the charge curve of C2, so that the pulse diagram of figure 2 is also valid for any fre- quency x-f. The ratio T1/T2 and thus the duty-cycle (= T1/T2 x 100%) is therefore constant. The values of R3, R4 and P2 are chosen variable pulse generator elektor march 1979 - 3-19 Eurotronics ☆ ☆ ☆ ☆ ☆ I'A t . ☆ . international w Hrcuit and idea compet: asr such that the reference voltage at the inverting input of IC 1 may vary between 13 and 87% of the supply voltage. As already mentioned, the voltage across C2 can vary between 20 and 80% of supply. Thus it is possible to vary the duty-cycle of the output signal between virtually 0 (i.e. no output signal) and 100% (DC voltage). The two remaining Schmitt-trigger gates of 1C2 are used at the output, N3 to further square up the output signal and N4 to provide an inverted version. Thus if a squarewave with a duty-cycle of 30% is present at the output of N3, the output of N4 will provide a squarewave of identical frequency but with a duty- cycle of 70%. With the component values as shown in figure 1, the frequency range of the circuit extends from approx. 1 kHz to 20 kHz. The frequency range can be altered if desired; the essential parameters of the circuit are given by the following equations: Cl = 6 x C2 Pla = Plb and R1 = R2 (Pla+ Rl) • Cl • 0.4 It is also possible to control the ampli- tude of the output signal by connecting a 22 k potentiometer between the output of N3 or N4 and ground. The output signal can then be taken from the wiper of the potentiometer. The supply voltage for the circuit need not necessarily be stabilised, however if any sort of demands are to be placed upon the stability of the frequency, amplitude or duty-cycle, it is best to employ a voltage regulator. Since the entire circuit consumes no more than roughly 20 mA, a regulator from the 78L-series is the obvious choice. De- pending upon the supply voltage, the 78L05, 78L06, 78L08, 78L09 and 78L0 10 should prove suitable. M Elektor is promoting the first world- wide circuit and design idea competition for electronics enthusiasts, with over £ 10,000 worth of electronic equipment to be won. It is the intention that this competition should stimulate elec- tronics as a hobby on a world-wide scale, by the resulting exchange of cir- cuit ideas. Entries are not limited to fully-developed and tested circuits: original design ideas, that could be im- plemented in circuits (given time and sufficient experience), can also be entered. Obviously, both circuits and design ideas must be original. Complete circuits Entries should be interesting, original circuits that can be built for less than £ 20.00 — not counting the case and printed circuit board. Circuits used in commercially available equipment, de- scribed in manufacturer’s application notes, or already published are not considered ‘original’. The complete circuit should be sent in, together with a parts list, a brief expla- nation of how it works and what it is supposed to do, a list of the most im- portant specifications and a rough estimate of component cost. The latter can be based on retailer’s advertisements. A jury, consisting of members of the editorial staffs for the English, German and French issues of Elektor and the Dutch edition, Elektuur, will judge the entries according to the criteria listed above. The best designs will be published in the four Summer Circuits issues, with a combined circulation of over 250,000 copies. All entries included in this final round will be rewarded with an initial ‘fee’ of £ 60.00. Design ideas Readers who cannot submit a complete circuit (for lack of time, know-how or hardware) may enter an interesting and original design idea. However, the same basic rule holds: the idea should be for a feasible circuit that can be built for an estimated component cost of less than £ 20.00. The idea should be described as fully as possible. Perferably, a block diagram and - if at all possible - a basic (untested) circuit should be in- cluded. The jury will select the best ideas for inclusion in the final round. These ideas will be rewarded with a ‘fee’ of £ 20.00. The final round The readers of Elektor and its sister publications will select the winners! This is where the half-a-million-or-more readers of the Summer Circuits issues come in (yes, we know that each copy is read, on average, by 2.6 people . . . ). The readers are requested to select the 10 best circuits from those published. Everybody who co-operates in this final vote may also win a prize. The prizes Over £ 10,000 worth! The ten entries selected by our readers will receive a total of £ 1 0,000 worth of prizes. Dream prizes for any enthusiastic electronics hobbyist! The closing date for the competition is 31st March, 1979. Entries should be sent to: Elektor Publishers Ltd., Elektor house, 10 Longport, Canterbury, CT1 1PE, Kent, U.K. Both the envelope and the entry should be clearly marked ‘Eurotronics circuit’ or ‘Eurotronics design idea’. ( \ General conditions • Members of the Elektor/Elektuur staff cannot enter the competition. • Any number of circuits and/or design ideas may be submitted by any • Entries that are not included in the final round will be returned, pro- vided a stamped, addressed envelope is included. • The decision of the jury is final. 4 3-20 — elektor march 1979 tup-tun-dug-dus TUPTUNDUGDUS Wherever possible in Elektor circuits, transistors and diodes are simply marked TUP' (Tran- sistors. Universal PNP), TUN' (Transistor, Universal NPNI, 'DUG' (Diode, Universal Ger- manium) or 'DUS' (Diode, Universal Silicon). This indicates that a large group of similar devices can be used, provided they meet the minimum specifications listed in tables la and 1b. type | Uceo 1 •c hfe 1 Ptot 1 ‘ T TUN TUP NPN ' PNP 88 < < 100 mA 100 mA 100 100 100 mW 100 mW : 100 MHz | 100 MHz type UR | IF l R Ptot | CD DUS DUG Si Ge 25 V 20 V 100 mA | 35 mA 1 /ja 1 100/iA j 250 mW 250 mW 5 pF 10 pF Table 2. Various transistor types that m Table 3. Various tr The letters after the type number denote the current gain: a' (0. hf e ) = 125-260 a' = 240-500 a' = 450-900. is diodes that meet the DUS oi DUS BA 127 BA 217 BA 218 BA 221 BA 222 BA 31 7 BA 318 BAX 13 BAY 61 1N914 1N4148 OA 85 OA 91 OA 95 I AA 1161 Table 5. Minimum specifications for the BC107, 108, -109 and BC177. -176. -179 families (according to the Pro-Electron standard). Note that the 8C179 does not necessarily meet the TUP specification He. max - 50 mA). BC 107 BC 108 BC 109 100 mA 100mA 100 mA 300 mW 300 mW 300 mW 150 MHz 150 MHz 150 MHz BC 177 BC 178 BC 179 45 V 25 V 20 V 100 mA 100 mA 50 mA 300 mW 300 mW 300 mW 130 MHz 130 MHz 130 MHz Table 6. Various equivalents for the BC107, ... families. The data are those given by the Pro-Electron standard; individual manu- NPN PNP Case Remarks BC 107 BC 108 BC 109 BC 177 BC 178 BC 179 •o BC 147 BC 148 BC 149 BC 157 BC 158 BC 159 4 Pmax ■ 250 mW BC 207 BC 208 BC 209 BC 204 BC 205 BC 206 •© BC 237 BC 238 BC 239 BC 307 BC 308 BC 309 0 BC 31 7 BC 318 BC 319 BC 320 BC 321 BC 322 •cmax “ 150 mA BC 347 BC 348 BC 349 BC 350 BC 351 BC 352 ® BC 407 BC 408 BC 409 BC 417 BC 418 BC 419 ■<$>.„. Pmax - 250 mW BC 547 BC 548 BC 549 BC 557 BC 558 BC 559 3 500 mW BC 167 BC 168 BC 169 BC 257 BC 258 BC 259 <3! 169/259 50 mA BC 171 BC 172 BC 173 BC 251 BC 252 BC 253 ■0 251 . . . 253 low noise BC 182 BC 183 BC 184 BC 21 2 BC 213 BC 214 •0 *cmax = 200 mA BC 582 BC 583 BC 584 BC 512 BC 513 BC 514 0 •cmax “ 200 mA BC 414 BC 414 BC 414 BC 416 BC 416 BC 416 0 low noise BC 413 BC 413 BC 415 BC 41 5 0 low noise BC 382 BC 383 BC 384 0: BC 437 BC 438 BC 439 B Pmax = 220 mW BC 467 BC 468 BC 469 a:! Pmax " 220 mW BC 261 BC 262 BC 263 ■0 low noise sinewave generator In last month's issue of Elektor we published a design for a spot I sinewave generator which offered truly 'top-notch' specifications. For the purposes of many amateurs, however, the accuracy provided by such a circuit (e.g. harmonic distortion of less than 0.0025%) is an expensive luxury. The circuit described here represents the other side of the coin — a simple cost-effective sinewave generator whose frequency can be continuously varied over virtually the entire audio spectrum, while being easy to construct and requiring no calibration. Just for good measure the circuit also offers the choice of a squarewave output. The idea of incorporating a limiter in the feedback loop of an oscillator circuit as a means of stabilising the amplitude of the output signal was originally employed in the spot sinewave generator published in last month’s issue of Elektor. That circuit was designed for extremely low harmonic distortion and high amplitude stability. The circuit described here, although employing a similar principle to its ‘big brother’ of last month, is intended to meet a differ- ent set of criteria. It was felt that many amateurs would appreciate a simple low- cost sinewave generator with continu- ously variable frequency which would require no complicated calibration procedure. Block diagram The block diagram of the simple sinewave generator is shown in figure 1 . A selective filter is followed by a limiter circuit which clamps the signal level to + and -Ub- The output of the limiter is fed back to the input of the selective filter, thereby providing the conditions for oscillation. The circuit will only continue to oscillate if the loop gain of the system is greater than unity. To ensure that this requirement is always satisfied an amplifier is included in the feedback loop. The circuit is forced to oscillate at the centre frequency of the filter, since only at that frequency will the in- and output signals of the filter be in phase. The reason why the amplitude of the output voltage of such an oscillator remains constant was described in detail in the above-mentioned article on the spot sinewave generator, to which readers are here referred. A continu- ously variable oscillator frequency is obtained by making the centre fre- quency of the selective filter tunable. The Q of the filter will inevitably vary with the centre frequency, thereby affecting the suppression of higher Figure 1. Block diagram of the sinewave generator. harmonics in the output signal, and hence the purity of the sinewave. However, some degradation in waveform purity is the inevitable consequence of having a continuously variable output frequency, and as has already been emphasised, the harmonic distortion of the circuit was deemed to be of second- ary importance - within certain limits of course - in a project designed for minimum cost and maximum simplicity. The output of the limiter is a clipped sinewave, i.e. a trapezoidal waveform, which is then shaped by a Schmitt trigger providing a squarewave output signal with a duty-cycle of 50%. The output of the Schmitt trigger is buffered and is available both with and without a DC offset. Circuit The complete circuit diagram of the sinewave generator is shown in figure 2. The selective filter formed by op-amps A2 . . . A4 is of the 'variable state’ type, and comprises two integrators and a summing amplifier. The centre fre- quency of the filter can be varied by means of PI. It is apparent that how well the two gangs of this potentiometer are matched will determine the ampli- tude stability of the output signal for changes in frequency. The output of A3 (which is also fed back to Al) supplies the sine wave output of the generator. A 1 is configured as an amplifier with a gain of ten and its output is clamped to approximately 12 volts peak to peak by means of zener diodes D5 and D6, before being fed back to the input of the selective filter. The clipped sinewave is also fed (via R9) to the Schmitt trigger formed by transis- tors T2, T3 and T4, with the resulting squarewave being buffered by transistors T5 and T6. The circuit operates off a 24 V supply. A virtual earth point is created via Rll, R12, R13 and Tl, so that a roughly symmetrical supply of ± 12 V is obtained. Construction The component layout and track pattern of the p.c.b. for the sinewave generator is shown in figure 3. One can either provide the case with three separate output sockets, or else use a single socket with a three-way waveform selector switch. The circuit requires no calibration. The only preliminary measure which, in view of component tolerances, may prove necessary is to experiment slightly with the value of R20 and R21. The fre- quency range of the sinewave generator is approx. 20 Hz to 25 kHz. The ampli- tude of the sinewave output is constant over the range 150 Hz to 6 kHz. Above this point there is a very slight rise in amplitude, and below 150 Hz the ampli- tude will tend to fall slightly. Harmonic sinewave generator elektor march 1979 - 3-23 Figure 2. Complete circuit diagram. The am- plitude of the sinewave output signal can be varied by means of P2. Figure 3. Track pattern and component layout of the p.c.b. for the sinewave generator (EPS 79019). Only the transformer and potentiometers are mounted off-board. Parts list Resistors: R1.R17- 100k R2- 10 k R3,R10,R19 = 2k2 R4.R7 = 330 k R5,R6,R8,R22 - 27 k R9 = 22 k R11,R12 = 6k8 R13- 1 k R14 = 120k R15 = 47 k R16.R26 = 1k5 R18 = 82 k R20.R21 = 39 n (see text) R23.R27 = 5k6 R24 = 2k7 R25 = 8k2 Pla/b = stereo potentiometer, 10 k log. P2 = potentiometer, 47 k log. Capacitors: Cl - 220 p C2 = 470 p/40 V C3.C4 = 100 n C5 = 22 p C6.C7 = 1 n C8 - 10 p/16 V Semiconductors: T1.T5 = BC 5578 T2.T3.T4.T6 = BC 547B IC1 = TL 084 01 . . . D4 = 1N4001 05,06 = 5V6/400 mW Miscellaneous: SI = mains on/off switch, 400 mA FI = fuse, 400 mA transformer, 18 V/50 mA distortion is less than 1% over the entire frequency range. The amplitude of the squarewave output is virtually constant over the entire range (8 V pp ). Current consumption is roughly 12 mA. M applikator Universal Counter Microprocessors and digital memories are not the only product areas to be influenced by the ever increasing chip densities achieved in LSI (Large Scale Integration) devices. Other more 'traditional' logic functions, which once were implemented using a mountain of dis- crete TTL- or CMOS ICs can now be realised with a single LSI chip. This point was well illustrated by the design for the Elektor ’1/4 GHz Counter' published in June 1978. The heart of the above circuit was a six-decade counter/display driver 1C, the MK 50398N from Mostek. This one 1C did the job which previously would have required a boxful of 7490's, 7475's etc. As 'intelligent' as it was, however, the MK 50398N was by no means the last word on this subject, and sure enough, a new family of LSI counter/timer ICs has recently been announced, whose performance represents an advance on anything seen so far. The device in question is the ICM7216A/B/ C/D from Intersil. As the type number suggests, there are four different versions of the chip; the ICM7216A and B are fully integrated universal counters/display drivers for common-anode and common-cathode dis- plays respectively (see table 1), whilst the ICM 721 6C and D are frequency counter only versions of the above. The chips combine a high frequency crystal oscillator, a decade timebase counter, an eight decade data counter plus latches, and associ- ated elements for decoding, multiplexing and driving 8 large LED displays. Gate times of .01, 0.1, 1 and 10 seconds in the frequency counter mode allow input frequencies of up to 10 MHz. In the case of the 7216A and B the maximum input frequency in other modes is 2 MHz, whilst times can be measured over 1. 10, 100, and 1000 cycles to give period measurement from 0.5 ps to 10 seconds. Additional features include: decimal point and leading zero blanking automatically con- trolled by the chip, hold and reset inputs, low power/display off mode, and internal test and display test functions. A single nominal 5 V supply is the sole power requirement. Universal Counter Of the four available versions of the chip, the ICM7216A and B are the most interesting in view of the truly impressive range of different functions which they offer. In addition to frequency counting, the devices can operate as a period counter, frequency ratio counter, time interval counter, and unit counter. The internal organisation of the chip is illustrated by the block diagram of figure 1. The input signal is fed to input A, whereupon control logic determines whether it is fed to the clock input of the main counter, or whether (when measuring period, for instance) it is used to gate the internal clock signal of the 1C to the main counter. The number of clock pulses counted by the latter is then an index for the length of the period. When a 'store' pulse is received from the con- trol logic, the contents of the main 10 s coun- ter are transferred to the data latches. The 8 digit count data is converted into 7-segment code and multiplexed to the segment outputs. As already mentioned, the counter can measure more than simply frequency and period. When operated in the 'unit counter' mode, it counts the total number of pulses in the input signal (instead of the number of pulses per second). The chip will also measure the frequency ratio between two input sig- nals, a facility which in certain situations can offer advantages over simple measurement of frequency. If, for example, an extremely stable reference frequency is available, in principle it is possible to measure an unknown input frequency with greater accuracy using this method. Furthermore, there are instances where the ratio of two frequencies is more important than their exact value, or where the frequencies vary but the ratio remains con- stant (e.g. in certain music applications). When functioning in the frequency ratio mode one input signal is fed to input A, the other to input B. To ensure good accuracy, the fre- quency of the signal fed to input A should be greater than that at input B. If the frequency ratio is greater than 10 s , all 8 digits of the display can be used. The fifth type of measurement of which the chip is capable is time interval. The chip counts the number of microseconds which elapse between a 1 0 transition occurring at input A and a 1-0 transition at the B input. To complete the measurement cycle input A must again go negative after B goes negative Under the heading Applikator, recently introduced components and novel applications are described. The data and circuits given are based on information received from the manufacturer and/or distributors concerned. Normally, they will not have been checked, built or tested by Elektor. 1 979 - 3-25 - see figure 2. Input A goes negative at time tc, completing the measurement cycle and the elapsed time T is displayed. Multiplexed interface In the ICM 721 6A and B efficient use is made of inputs and outputs by multiplexing. Not only are the segment driver outputs multi- plexed, but the function, range, control and external decimal point inputs are time multi- plexed also. The input functions which are required are selected by connecting the appro- priate digit driver output to the inputs con- cerned. The functions selected by each digit for these inputs are listed in table 2. As can be seen, the digit driver outputs are numbered from do to d7. The digit driver for the ex- treme right-hand (least significant) digit is do, whilst d7 is the driver output for the extreme left-hand digit. Depending upon which of these outputs is connected to the function input, the chip will operate in one of the above-described modes. The connections to the function input may either be hard-wired or selected via a multi-way switch. Which digit driver output is connected to the range input determines both the gate period of the main counter when used in the fre- quency mode, and the number of periods sampled in a period measurement. Both sets of values are listed in table 2. The greater the gate period or number of periods, the more accurate the measurement — although the latter will then take longer and there is a greater risk of an overflow occurring. The control input provides a number of additional facilities. By connecting d3 to the control input and simultaneously taking the hold input to V + , the displays can be blanked, the chip remaining in the display-off mode until hold is returned low (V - ). With the displays blanked the current consumption of the chip is reduced to around 2 mA. In the 'display test' mode all segments are enabled continuously, whilst in the general test mode the main counter is split into groups of two digits which are clocked in parallel. The con- trol input is also responsible for determining the clock frequency. As can be seen from the table, in addition to the 'standard' 10 MHz crystal, there is the possibility of using a 1 MHz crystal or employing an external clock oscillator. The latter facility can prove useful when using the 1C in larger systems. In addition to the multiplexed control inputs, the 1C has two 'conventional' inputs in the form of a hold input, which, as has already been mentioned, is normally held low, and a reset input. When the former is taken high the count is stopped and the contents of the main counter are latched, whilst the counter itself is reset. Taking 'hold' low again initiates a new measurement. The reset input is the same as the hold input, with the exception that the latches for the main counter are enabled, resulting in an output of all zeros. In addition to the five measurement functions already mentioned (frequency, period, fre- quency ratio, time interval and unit count) the ICM7216A and B offer the possibility of measuring the internal clock frequency. This facility is a little unusual and appears to be somewhat superfluous, since the gate time of the main counter is also determined by the internal clock oscillator and hence the count displayed cannot be used to determine if the Under the heading Applikator, recently introduced components and novel applications are described. The data and circuits given are based on information received from the manufacturer and/or distributors concerned. Normally, they will not have been checked, built or tested by Elektor. Mmuauiu-. Under the heading Applikator, recently introduced components and novel applications are described. The data and circuits given are based on information received from the manufacturer and/or distributors concerned. Normally, they will not have been checked, built or tested by Elektor. applikator elektor march 1979 - 3-27 ommmiL:. 3-28 - elektor march 1979 the ICU — a 'mini' microprocessor the ICU* a ’mini’ microprocessor Who's afraid of the one-bit (rP? One of the problems of microprocessors is that they require a considerable amount of specialised knowledge in order to operate them. An additional problem is that there are many applications for which their inherent sophistication renders them unsuitable. Recently, Motorola seem to have succeeded in killing two birds with one stone by bringing out a new 1C, the Industrial Control Unit, MC14500B. This is a one-bit processor which has been specially designed for simple control applications, and is extremely easy to program. The following article takes a look at this 'mini'-microprocessor, which should prove of particular interest to those readers who have had problems getting to grips with its 'big brothers'. Main characteristics of tha ICU: can perform 16 logic functions 1-bit bi-directional databus 1-bit memory four flag outputs meets Jedec-B specifications for CMOS ICs supply voltage range 3 ... 1 8 V clock frequency DC ... 1 MHz intended primarily for use in industrial control systems Thanks to the inherent flexibility of the stored program concept, which allows system changes to be rapidly im- plemented simply by altering the program, the microprocessor is being used in an increasing number of control applications. Washing machines, sewing machines, ovens, are only a few of the consumer products which now feature microprocessors, whilst in industry, microprocessors have appeared on the shop-floor in a wide variety of adaptive control processes. However for certain control applications involving decision oriented tasks, (e.g. intruder alarms, signal controllers for model railways, slide changers, PROM programmers, to name but a few cases which are of interest to the amateur electronics enthusiast) such is the relative simplicity of the system, that the expense and sophistication of a microprocessor would be the equivalent of the sledgehammer used to crack a nut. For this reason, a new type of chip has been developed which is designed to offer the advantages of programmability, without the unnecessary complexity of a microprocessor - the Industrial Control Unit (ICU) or one-bit processor, of which the MC14500B from Motorola is an example. The ICU can be viewed as a simplified form of microprocessor, capable of performing logic operations on one-bit input data and transferring the results to an output device. The great advantage of the ICU is the fact that it is uncomplicated and easy to program. The inexperienced user can rapidly familiarise himself with the basic system and learn how it can be tailored to meet his particular requirements. Thus the ICU also represents an excel- lent introduction to (micro-)processor based systems in general. The following article provides a basic description of the MC14500B ICU, and with the aid of several simple examples, shows how it can be programmed to perform a variety of control functions. Those readers whose interest in the MC14500B is aroused by the article, are referred to the Motorola handbook for the device, which contains a more detailed description than we have room for here. Characteristics The Motorola MC14500B is a single chip, one-bit static CMOS processor, which is housed in a 16-pin DIL pack- age. The IC executes one instruction per clock period and is timed by a single- phase internal clock oscillator, the frequency of which can be varied up to 1 MHz. Alternatively the clock signal can be controlled by an external oscil- lator. The electrical characteristics of the ICU conform to JEDEC B-Series specifications forCMOS B-Series devices. The IC has an operating supply voltage range of 3 to 18 V (in practice, assuming the device is not being operated in an electrically noisy environment, a supply voltage of 5 V can often be chosen, allowing the IC to be used in conjunc- tion with TTL), and is capable of driving at least one Low Power TTL 1C (the Data- and Write outputs can drive two normal TTL ICs). The ICU differs from conventional logic ICs in its ability to be programmed to perform more than one type of logical operation. The instruction set of the ICU, which is shown in Table 1, consists of 1 6 four-bit instructions. The ICU is a one-bit processor, i.e. data is manipulated one bit at a time and is routed to and from the ICU via a 1-bit bi-directional data bus. To perform a logical operation requiring more than one bit of data (e.g. a logical AND), an internal register called the Result Register (RR) is used. To perform the logical AND function, the first data-bit is loaded into the Result Register by means of a LOAD instruction. The ICU is then supplied with an AND instruc- tion, and the second data-bit is read onto the bi-directional data bus whereupon the logical AND operation is performed on the data present in the Result Register with the data present on the data bus. The result of this operation becomes the new content of the Result Register (which always receives the result of any of the ICU’s logical operations, hence its name). Since a third instruction (STORE) would be required to transfer the result of the above logic function to an output device, it can be seen that three separate the ICU - li' microprocessor elektor march 1979 — 3-29 operations are needed to simulate a two- input AND gate (see figure 1). Like a microprocessor, the ICU operates on the stored program principle. The instructions to be executed by the ICU are normally stored sequentially in the system memory. However, in addition to the ICU’s op-codes, the memory must also contain the address of the data to be loaded into the Result Register, or the address of the latch onto which the content of the Result Register is to be stored. The addresses are decoded by input and output selectors — as shown in the block diagram of figure 2. Once again, let us take the example of a logical AND function being performed on two input signals A and B, which are present at inputs 1 and 5 respectively of the input selector, and assume that the result of this operation is to be stored on output 9 of the output latch. When an address is presented to the input or output selector the corresponding input or output is connected to the data line of the ICU. Thus the AND operation is performed as follows: 1) The system memory provides the ICU with the LOAD (LD) instruction and supplies the input selector with the input address. The logic level of this input is then transferred via the ICU’s one-bit data bus to the one-bit Result Register. 2) The ICU fetches the next instruction from system memory — the AND instruction — whilst the input selector is supplied with the address of in- put S. The addressed input data is demultiplexed onto the ICU’s data line and then is logically ‘ANDED’ with the data in the Result Register. The original contents of the Result Register are lost. 3) Finally, the ICU is provided with the STORE (STO) instruction, whilst the output selector is presented with the address of output latch 9. The data in the Result Register is then trans- ferred to this output latch via the bi-directional data line. instruction Figure 1. This figure illustrates the difference between dedicated logic devices and the (programmable) ICU. The conventional logic gate requires only a single operation to ex- ecute the function. The ICU on the other hand requires three separate operations to perform the same task. Figure 2. Block diagram of a basic ICU system that is clocked by hand. Tabla 1. Instruction set of the ICU. Instruction Code Mnemonic Action #0 0000 NOPO No change in registers. R -» R, FLGO <-_TL_ #1 0001 LD Load Result Reg. Data -» RR #2 0010 LDC Load Complement Data -» RR #3 0011 AND Logical AND. RR . D -* RR #4 0100 ANDC Logical AND Compl. RR . D - RR #5 0101 OR Logical OR. RR + D - RR *6 0110 ORC Logical OR Compl. RR + D -* RR #7 0111 XNOR Exclusive NOR. If RR = D. RR «- 1 #8 1000 STO Store. RR -» Data Pin, Write •- 1 #9 1001 STOC Store Compl. RR — Data Pin, Write «- 1 #A 1010 IEN Input Enable. D -» IEN Reg. #B 1011 OEN Output Enable. D -» OEN Reg. #C 1100 JMP Jump. JMP Flag *-_TT_ #D RTN Return. RTN Flag *-_TL, Skip next inst. #E 1110 SKZ Skip next instruction if RR = 0 1111 NOPF No change in Registers RR -► RR, FLGF 3-30 - elaktor march 1979 the ICU - a ‘r mini' microprocessor Program Memory and Program Counter One of the advantages of an ICU-based system is that the number of inputs and outputs can be expanded virtually ad infinitum, providing the memory is sufficiently wide to address the I/O ports. The Program Counter (PC) supplies the system memory with the address of the instruction to be ex- ecuted. The PC will normally count up to its highest value and then ‘wrap- around’ to zero and start counting up again. Thus the sequence of instructions in memory can be repeated, providing a ‘looping’ program. Minimum system Figure 3 shows the circuit of a mini- mum ICU-system based on the block diagram of figure 2. The system has 8 inputs and 8 output latches. 1C 3 is a specially designed type of multiplexer- demultiplexer, which, like the ICU, has a bidirectional data line. Thus, the ICU can not only write data onto an output latch, it can read data off it as well. Both the desired instruction and the appropriate address are set up on a dual-in-line switch. In this case, the clock signals are provided by hand, by depressing the pushbutton switch SI. Care should be taken to ensure that only one clock pulse is given at a time, lest the ICU execute a particular instruc- tion several times in succession. The state of the Result Register, data line, etc. can be displayed by means of LEDs, which are connected via buffers to the most important points in the circuit. If things threaten to get out of hand, S2 takes the reset pin of the ICU high, clearing all registers and resetting the FLAG outputs. With a four-bit address word, up to 16 locations can be selected; in this case 8 inputs and 8 outputs. The highest address bit, A3, is used to control the Chip-Enable (CE) inputs of IC2 and IC3, and hence switch between input and output lines. In conjunction with the Write signal of the ICU, A3 ensures that only one device, be it IC2, IC3 or the ICU, has access to the data line at any given time. The basic design of the system shown in figure 3 bears a close resemblance to that of larger microprocessor-based systems, which also contain a data bus and address bus shared by a large number of ICs. With the practical circuit of figure 3, let us see how one would actually execute the above example of a logical AND function. As we have seen, the program I for an AND operation is: 1. LOAD A 2. AND B 3. STORE C | The op-code for these instructions can l be found by looking up the instruction | set of the ICU in Table 1 . j First, however, we must ensure that A is | present at input 1 , B at input 5 and that C is stored on output 9. In this case the I addresses of these locations are simply | the binary equivalent of 1,5 and 9, i.e. 0001, 0101 and 1001 respectively. Thus the DIL switch should be set as follows: 1) 0001 0001 - whereupon a clock pulse is applied (A is | transferred to RR) 2) 00110101 - another clock pulse (A • B = C) 3) 1 000 1001 — another clock pulse (C is transferred to out- put 9) Step by step, therefore, the procedure is as follows: The four right-hand DIL switches are set to address (in binary) 0001, with the result that on the next clock pulse, input 1 of IC2 is connected to the data line of the ICU. By means of e.g. a wire link, this input should be connected to logic 1 . Once the four left-hand switches have been set to the instruction code 0001 (= LOAD), SI should be pressed, taking the clock signal at pin 14 of the ICU low and extinguishing the clock signal indicator LED. On the negative going edge of the clock signal (the clock signal was previously high) the ICU latches the LOAD instruction into the Instruction register, and upon The third instruction differs from the SI being released and the clock signal first two slightly. Once the op-code of being returned high, the data on the the instruction (1000 = STO) and the data line is read into the Result Register, correct address have been programmed Since the data line is high, a ‘1’ is read on the DIL switch and SI pressed, the into the Result Register, causing the ICU executes the instruction immedi- corresponding LED to light. ately. The only instructions which the The second instruction (0011 = AND) ICU will execute on the negative going and second address can now be pro- edge of the clock pulse are STORE and grammed on the DIL switch. Since the Flag instructions. The Flag instructions second instruction is an AND operation, (JMP, RTN, FLGO, FLGF) are used to input 5 should also be taken high. As provide external control signals by soon as the clock signal is taken low, setting pins 9 through 12 of the ICU. input 5 will be connected to the data These output flags remain active for a line and the ICU will read the AND full clock period after the negative- instruction into its Instruction Register, going edge of the clock signal. The data line LED should light at this In the case of the STORE instruction, stage. When S 1 is released and the clock the moment the ICU reads the instruc- signal returns high, the ICU will execute tion into its Instruction Register the the instruction, and since 1 • 1 = 1, the content of the Result Register is put on Result Register will contain a T. the data line and the Write line (pin 2) is enabled (taken high), with the result that the logic state of the data line is latched onto the appropriate output (FF9). On the positive-going edge of the clock pulse the Write line goes high again. The state of the various indicator LEDs at each stage in the above program are shown in Table 2. The above program is a simple example of how the ICU can be programmed to simulate a conventional logic gate. Further examples of programs designed to imitate the function of logic ICs (4-input AND, NAND, OR, EXOR etc.) are shown in figure 8. As was pointed out earlier, the ICU requires a number of operations to perform the single function executed by a conventional logic device (e.g. 10 program steps are needed to perform Figure 3. Circuit for a minimum ICU system which is clocked by hand. The instructions to be executed and the addresses of the data to be manipulated are set up on the dual-in-line switch. The circuit has seven inputs (the eighth is connected to the Result Register of the ICU) and 8 latched outputs. The state of the data line Result Register and clock signal, as well as selected inputs and outputs can be displayed by means of LEDs. The number of I/O ports can be extended by increasing the number of address bits on the address bus. With 8 address bits a total of 256 inputs and outputs could be used. Figure 4. Block diagram of a minimum ICU system. The basic building blocks of such a system are: The ICU, which controls the flow of data within the system and performs logical operations on the data in its Result Register and data appearing on its 1-bit bidirectional data line. Program memory, which stores the instruc- tions to be executed by the ICU. The Program Counter, which ensures that the instructions are presented to the ICU in the correct sequence. Input and Output Selectors, which decode the operand addresses stored in memory and determine which input or output is connected to the ICU’s data line. Table 2. This table shows the state of the system indicator LEDs at each stage of the AND program. The negative going and positive going edges of the clock signal (1-0-1) are differentiated to further clarify the sequence of events. 3-32 — elektor march 1979 the ICU - a 'mini' microprocessor 5 the function of a D-flip-flop). Thus the ICU will inevitably take longer to execute a given logic operation than its dedicated counterpart. For example, with a clock frequency of 330 kHz, the above program for a 2- input AND will take approx. 10 /is. However this trade- off in terms of speed is the consequence of the simplicity and, more importantly, the flexibility of the ICU, which allows it to be programmed for a wide variety of different functions. The ability to step through a program one instruction at a time is particularly convenient for the beginner who is attempting to familiarise himself with microprocessor programming. However such a procedure is naturally time consuming. Particularly for longer programs, the obvious solution (as is shown in the block diagram of figure 4) is to store the program instructions in memory and employ a program counter to ensure that they are presented to the ICU in the correct sequence. The clock pulses can then be provided by the ICU’s internal clock oscillator, the fre- quency of which can be set to up to 1 MHz (i.e. 1 instruction every 1 /is). The circuit of a suitable instruction and address memory is shown in figure 5. The instructions and operand addresses are entered by hand into two 256 x 4-bit RAMs (type 2 112). The program counter, which consists of two 4029’s (presettable binary/decade counters) counts from 000 up to 256 and then starts from 000 again. The counter increments by one after each clock pulse and ensures that the contents of the next memory location are presented to the ICU. Thus the program instruc- tions are executed in the correct se- quence. Instructions are programmed into memory as follows: The initial contents of the memory are first erased by closing the DATA switch, setting switch S2 in figure 3 to the ‘run’ position, and pressing the ‘write’ switch. This has the effect of writing logic ‘0’ into every location in memory. The frequency of the internal clock oscillator (determined by the 56 k external resistor) is 330 kHz, which means that the entire memory is erased in approx. 1 ms . Switch S2 of figure 3 should then be set to the ‘single-step’ position, thereby stopping the clock oscillator. The first program instruction accompanied by the operand address is now set up on the data lines of the 211 2’s and written into the memories by pressing the ‘write’ switch. A clock pulse is then provided by hand, incrementing the program counter in preparation for the next instruction to be entered. Once all the instructions have been stored in memory the program can be run simply by switching S2 to the ‘run’ position. Of course it is also possible to continue to provide the clock pulses by hand, stepping through the program one instruction at a time, and checking the state of the program by means of the LEDs. To this end the second DIL switch and the ‘load’ button have been included. When entering a program the program counter is incremented at each instruc- tion, thus a program containing e.g. I 5 instructions would leave the program counter at 005. If one then wants I to step through the program by hand, it would involve supplying 256-5 = 251 clock pulses until the program counter ‘wrapped round’ to the start of the program. With the aid of the DIL-switch and ‘load’ button, this becomes unnecessary. One simply sets up the address of the first instruction (e.g. 0000 0001) on the DIL-switch and presses the ‘load’ button. The program counter is then set to that address. By adding additional LEDs with ac- companying buffers (ULN 2003) it is a simple matter to display the state of the program counter and contents of the memory at any given moment. If desired, one could also use 7-segment displays, or - perhaps the best idea of all - link the entire system up to an existing microprocessor system; figure 6 provides a simple interface circuit for linking an ICU system to the SC/MP. The display and the keyboard of the SC/MP system can then be used to read data into and out of the ICU system memory. A further advantage of this arrangement is that the cassette dump routine of the SC/MP can be used to store programs for the ICU system. The ICU represents a useful extension to the SC/MP system, since it can be used to assume a number of simple control tasks, freeing the microprocessor for more complex operations. Enabling Instructions — IEN OEN Before proceeding to examine an example of a program which can be run on the above-described system, it is first necessary to take a look at two ICU instructions which are of prime import- ance: IEN and OEN. All microprocessor systems provide the facility to perform conditional jumps, i.e. depending upon the result of a test, the processor jumps from one part of the program to another. Such a provision allows the processor to make logical decisions. For example: If signal A is high, the red lamp should light; if A is low, the green lamp must light. The conventional way of solving this problem is to test signal A and depending upon the outcome the processor will either continue normally (i.e. the program counter is incremented by one and loads the address of the previous instruction + 1 ) or else perform a jump to another part of the program (i.e. the program counter is incremented not by one, but by, e.g. 10, 100, etc.). The section of program which has been ‘jumped over’ will contain the instruc- tion ‘turn on the red lamp and extinguish the green lamp’, whilst the section of program to which the pro- cessor jumps will contain the instruction ‘extinguish the red lamp and turn on the green lamp’. To implement such a jump, however, requires a more complicated chip structure than is present on the ICU, thus an alternative approach is necessary. The solution chosen is to have the ICU execute the program in the correct sequence, however, depending upon the result of the test, prevent the ICU from actually carrying out a block of instruc- tions. This is the function of the IEN and OEN instructions which respectively inhibit input data from effecting the system’s output and latch the system’s outputs into their current state by inhibiting the Write signal. Figure 5. With the addition of memory and a program counter the ICU »y*tem can perform independently, running programs at a speed determined by the frequency of the ICU's internal dock oscillator. In the case of the circuit shown in figure 3, the clock frequency it approx. 300 kHz, which means that any version of the 2112 may be used. Figure 6. It is a simple matter to interface the ICU to the SC/MP system. With a little skill the appropriate connections can be via 1C sockets, which replace the DIL switches. With the inverter in address line 13, the address of the ICU system memory is 2000 ... 20 FF. Figure 7. These two flow diagrams illustrate how the ICU is capable of performing con- ditional jumps. Assuming that the program in question contains two blocks of instructions, numbered I and II, a normal microprocessor system could jump over a block of instruc- tions simply by forcing the appropriate address into the program counter. The ICU, however, runs through all the instructions in the program, but simply fails to execute the block of instructions to be 'jumped'. IEN This instruction causes the ICU to latch the data on the data line into its ‘input enabling register’. If the input enabling register is loaded with a logic ‘O’, all further input data is interpreted by the ICU as logic ‘O’, until the IEN register is loaded with logic ‘1’ (by a subsequent IEN instruction). (Note that an LDC or ORC instruction will cause the Result Register to be loaded with a logic ‘ 1 ’, regardless of the state of the inputs. A certain amount of care is required when using the IEN instruction). OEN The operation of the OEN instruction is similar to that of the IEN instruction. The ICU latches the data on the data line into its output enabling register. If that data is logic ‘O’, the Write signal from the ICU is inhibited, thereby dis- abling the output latches. Once the OEN register has been loaded with a logic ‘0’ the system outputs will remain in their current state until the OEN register is loaded with a logic ‘ 1 ’ (by a subsequent OEN instruction). Thus the ICU will effectively jump over whole blocks of instructions, since they will have no effect upon the system outputs. An example of a conditional program jump using the OEN instruction is shown in figure 7a. A is first loaded into the Result Register, and the comp- lement of A is stored in a temporary scratch-pad memory. The OEN instruc- tion performs the test ‘A = ‘0”. If A does in fact equal logic ‘O’, the OEN instruction results in the outputs being 3-34 - alektor march 1979 the ICU — a 'mini' microprocessor Table 3. hex binary 01 02 03 04 05 06 07 10 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 30 31 32 33 34 ORC RR 60 IEN RR A0 OEN RR BO LD Cl 11 XNOR B1 7E STOC B1 9E AND Cl 31 STO C2 8F XNOR 82 7D STOC B2 9D AND Cl 3F STO C2 8F XNOR B3 7C STOC B3 9C AND C2 3F STO C3 8F XNOR B4 7B STOC B4 9B AND C3 3F STO C4 8F XNOR B5 7 A STOC B5 9A AND C4 3F STO C5 8F XNOR B6 79 STOC B6 99 AND C5 3F STO C6 8F XNOR B7 78 STOC B7 98 AN DC RR 40 IEN RR A0 OEN RR B0 0110 1010 1011 0001 0111 1001 001 1 1000 0111 1001 0011 1000 0111 1001 001 1 1000 0111 1001 001 1 1000 0111 1001 0011 1000 0111 1001 0011 1000 01 1 1 1001 0100 1010 1011 0000 0000 0000 0001 1110 1110 0001 1111 1101 1101 1111 1111 1100 1100 1111 1111 1011 1011 1111 1111 1010 1010 1111 1111 1001 1001 1111 1111 1000 force R R to ‘1 ' enable inputs enable outputs load "1 ' EXNOR with first bit of counter store result in first bit of counter generate carry store carry in scratch-pad EXNOR previous carry with second store result in 2nd bit of counter generate new carry store new carry in scratch-pad >it of counter 0000 force RR to '0' 0000 disable inputs 0000 disable outputs blocked, so that instructions 4 and 5 have no effect and are effectively jumped over. Since the second OEN instruction is executed with the comp- lement of A (in this case logic ‘1’), the Write signal is now enabled and instruc- tions 7 and 8 are executed. If A was not logic ‘0’ but logic ‘1’, instructions 4 and 5 would have been executed, whilst instruction 7 and 8 would have been ignored. When the Master Reset (RST) pin is taken high, all registers, including IEN and OEN are cleared (i.e. reset to ‘0’). This means that at the start of each program it is necessary to load the IEN and OEN registers with a logic T. This may appear to present a problem, since it cannot always be guaranteed that a logic ‘1’ will be available at one of the inputs. However by using the ORC instruction and the Result Register it is possible to force a logic ‘ 1 ’ into the result register (since the data which is inverted and ‘ORed’ with the Result Register is the initial content of the Result Register, then one or other signal must always be logic ‘1’, RRorRR= 1). The IEN register can then be loaded with logic ‘1’ by means of an IEN RR instruction. Thus every program should commence with the following start routine: ORC RR IEN RR OEN RR (in the minimum system described above the Result Register is connected to input 0, i.e. address 0000). It is also a useful precaution to termin- ate each program by loading logic ‘0’ into the IEN and OEN registers. This prevents the ICU executing any other instructions which may be stored sub- sequently in memory. The appropriate routine is ANDC RR (which forces a logic ‘0’ into the Result Register, fol- lowed by OEN RR and IEN RR. Sample Program Table 3 lists an example of a program which can be run on the basic ICU system which has been described above. IC3 is used as a counter, and each time the program, which is 34 steps long, is executed the contents of the counter are incremented by one. The counter is 8 bits wide, however one bit is used as a ‘scratch-pad’ memory to store the carry bit, so that in actual fact the maximum count is 2 7 = 128. With a clock fre- quency of roughly 300 kHz and a program counter which counts up to 256 before wrapping around to 000, the program will perform one complete loop and the counter will increment by one approx, every 0.9 ms (256x3.33 ps). LED 8, which displays the content of bit 8 of the counter will therefore flicker on and off every 1 28 x 0.9 ms = 109 ms, i.e. something just under 10 Hz. The counter is actually incremented as follows: when adding two bits together there are only four possible outcomes, 0+0 = 0, 1+0=1, 0+1 = 1, and 1+1 = 1. Only in the last case (1 + 1 = 1) will a carry be generated. The logic function represented by the above truth table can be simulated by means of an XNOR instruction, which outputs a ‘1’ if and only if both inputs are the same, followed by STOC, which inverts the result. The carry is generated using the AND instruction; a condition for generating a carry is that the result of the XNOR function is ‘1’, and that one of the two inputs was also ‘1’. The addition is carried out with one bit of the counter, which is combined with the carry from the preceding addition. At the start of the count there is no carry from the preceding bit, so that a ‘1’ must first be presented to input 1. If a ‘0’ were at this input, the circuit would simply add zeroes and the output of the counter would never change. The operation performed by the program can be expressed by the fol- lowing equations: Sn = Bn + Cn and Cn + 1 = 5n • Bn (where S = Sum, B = Bit, C = Carry and n may equal 1 ... 7) Thus Bit I ‘1* Carry 1 + ‘1* Sum 1 ‘0’ As a further illustration of the type of programming required, figure 8 lists several short program sections that can be used to simulate standard logic functions. Lit.: Motorola Industrial Control Unit Handbook. (Available from Motorola distributors.) mmgm B . * — - wT ham iiiwi fSit iiitgit AO !_l S LD A AND "J ) OC AND B BO n / STO C rn i - 5 AO— s LD A and i c °— 1 .-1 J AND D DO 1 STO E MM • ' v- . rut . AO — i i v LD A NAND 1 OC AND B bo— rL_y STOC C « 5 Photo 1. Prototype board for the minimum »o» li>^ c s r"? Table 3. Listing of a sample program which uses IC3 as a counter to make an LED flash off and on. Inverter AO 08 STOC B ORC RR execute a wide variety of logic functions. The logic gates shown here can all be simulated by the corresponding set of ICU instructions. As the example of the D flip-flop shows, even quite complicated logic operations can be OEN RR LD old clock DO— , STO temp D O — OQ ld clock D Flipflop c|ock ST0 0 | d ctock CQ | ANDC temp LD D ST Q elektor march 1979 PSUs PCBs PSUs an PCBs building power supplies the easy way The printed circuit board has been The ideal situation is, of course, to have cuit diagrams and it will be apparent completed and tested. It is working fine a printed circuit board for the power that many can be modified to suit and now ready to fit into the case. But supply as well as for the project and this specific requirements. The 78** regu- what about the power supply? Is it still is possible via the Elektor Print Service, lators are interchangeable provided the that ‘Christmas tree’ tacked onto the In many Elektor circuits the power transformer can supply 3 volts above transformer terminals? supply has been included on the main the regulated voltage (e.g. the 7815 It happens to most of us (or so it would printed circuit board. However, there requires 18 V from the transformer), seem) judging from the comments in are a number of others that are entirely Remember also that the working volt- our reader’s letters. All too often the separate and the purpose of this article ages of the capacitors must be adequate I power supply is forgotten until the last is to group these together as a handy (otherwise they could become momen- moment, especially if the test equip- reference. tary action switches — once!), ment includes a variable power supply. We have included the most useful cir- elektor march 1979 — 3- +15 V 250 mA and -15 V 250 mA Issue E42, October 1978, page 10-38. Originally designed for the Elektor TV scope but very useful where op-amps are Board number EPS 9968-5a. Capacitors: C1,C2= 470 n/35 V C3.C4 - 100 n C5.C6 ■ 1 m/25 V tantalum Semiconductors: IC1 = 7815 IC2- 7915 D1 . . . D4 = 1 N4001 Miscellaneous (noton p.c. board Trl = mains transformer, 2x18 V/250 mA SI = double-pole mains switch FI = fuse. 100 mA +15 V 1 A Issue E31, November 1977, page 1 1-37. Board number EPS 9218b, limited stocks still available, price £ 1 .05. g -loos' foSSV ~o OHHXHHO C« C3 Capacitors: Cl = 2200 m/40 V C2,C4 = 100 n C3-470 m/16 V Semiconductors: IC1 =7815 ^ cHH>o oHhoo oHhoo O “ O (-[SJ i i8uuuuu' 6 T elektor march 1979 Symmetrical ± 5-15 V 1 A Issue El 5/ 16, July/ August 1976, page 7-63. Board number EPS 9637, limited stocks still available, price £ 0.80. 5 15V 1 A +12 V.+33V Issue E19, November 1976, page 1 1-15. (Albar). Board number EPS 9437. Capacitors: Cl . . . C4= 100 n C5 = 2200 m/40 V C6 = 47 m/10 V C7= 100 p Semiconductors: T1 = BD 241 A, MJE3055 D1,D2 = 1N4002, BY 188 IC1 = 723 Miscellaneous: Tr = Transformer, 24 V/1.5 A NiCad Accumulator. 18 V PSUs on PCBs elektor march 1979 — UK 15 Resistors: R1.R4 = 2k7 R2 = 8k2 R3= 100 n R5 = 0.18 0/2 W (see text) R6= 180 n PI = 2k 5 P2 = 1 k Capacitors: Cl = 2200 m/25 V (see text) C2,C3= 100 n C4 = 1 n C5= 10 m/16 V C6= 1000 m/25 V C7 = 1 m/25 V tantalum Semiconductors: IC1 = 723 IC2 = 79G T1 = BD 137, BD 139 T2 = 2N3055 B1 = B40 C5000 40 V 5 A bridge rectifier (see text) 82 = B40 C800 40 V 800 mA bridge rectifier Miscellaneous: Trl = Transformer 1 2 V, 3 ... 4 A secondary (see text) Tr2 = Transformer 15V, 0.5 A secondary (see text) FI ,F2 = 300 mA slo bio fuse UK 16 — elektor march 1979 missing link Modifications to Additions to Improvements on Corrections in Circuits published in Elektor 1/4 Gigahertz counter Elektor 38, June 1978, p. 6-01. The sensitivity of the high fre- quency input amplifier (figure 9) can be unproved by decreasing R79 to 47 n and inserting a 500 fl (470 n) carbon preset potentiometer in series with it; R78 is decreased to 15 k and R84 is replaced by a wire link. To reduce clipping of the low frequency input amplifier (figure 11) a DUS can be placed in series with both zener diodes D32 and D33. In the parts list D36 is given as a 2V7 zener diode; this should in fact be 4V7 as the circuit diagram shows. Central alarm system Elektor 42, October 1978 p. 10-20. On the p.c. board for the master station (figure 9), pin 3 of IC6 should ideally be connected to supply common or positive supply (e.g. pin 1) - we have often stressed in the past that unused inputs of CMOS ICs should not be left floating! In practice it will not normally make any difference, but if the IC is running hot this is almost certainly the reason. Similary, on the alarm board (figure 8): if either of the two alarm inputs ‘X’ and ‘Y* is not used, it must be connected to supply common. A floating input at this point can easily result in ‘false alarms’! 1x100031'. High accuracy control of ultra-fine torque For applying and checking ultra- fine torque values to a guaranteed accuracy of ± 2%, a range of torque guages is available from MHH Engineering. The guages have bi-directional operation and three scales in either direction for easy reading from any angle - one on the dial and two (standard and minor image) on the sleeve. Slave pointers operating in either direction are standard on both dial and sleeve. A three-jaw chuck accepts a wide variety of spindles and a mechanical stop prevents overloading. The instruments are ideal for checking the torque on pre-loaded bearings, gear trains, watch, camera and instrument spindles and similar very low torque applications. The torque values covered are from 10 gf cm to 15 kgf cm (0.1 ozf in to 15 Ibf in). These guages are manufactured by the Tohnichi company of Japan. MHH Engineering Company Ltd. Bramley, Guildford, Surrey, GUS OAJ. England (1087 M) Switching power supply with fan A new dual-output switching power supply from Gould Electronic Components Division, the MGD500, incorporates a fan-cooling system which enables the unit to provide a high power density and also offers the facility for mounting in any plane. Designed specifically to power systems using emitter-coupled logic, the MGD500 provides two independently controlled outputs which track together, and gives a total power output of 526 W from a package measuring 5x8x10.5 in (12.7x20.3 x 26.7 cm). The two d.c. outputs provided by the Gould MGD500 are 5.2 V (* 0.2 V), adjustable from 0 to 80 A, and 2.2 V (± 0.1 V), adjustable from 0 to 50 A, with a common positive terminal. Input voltage can be either 230 V or 115 V(± 15%) a.c., ata frequency of 47 - 63 Hz, and is selectable from the front panel. Regulation is within 0.2% on both outputs for a worst-case combination of i 15% input and 0 - 100% load change. Ripple on both outputs is 10 mV r.m.s. maximum and 50 mV peak-to- peak (30 MHz bandwidth), and temperature coefficient is less than 0.01% per degC. The Gould MGD500 is protected against overcurrent, overvoltage and overtemperature, and has a hold-up time of 28 ms at full load in the event of mains failure. Insulation voltage is 2.1 kV d.c. between input and ground and 500 V d.c. between output and ground, and insulation resistance is not less than 50 Mn at 500 V d.c. The transient response is such that the output returns to within 1% of its original value within 500 us of a 50% load change. There is no limit on parallel operation with other units, and units may be operated in series to a maximum total voltage of 250 V. The output voltage can be remotely programmed with a ± 5% variation for system margin checking. Gould Electronic Components Division, Raynham Road, Bishop ’s Stortford, Herts. CM23 5PF England (1034 M) Telephone line simulator Telephone lines were originally designed for voice transmission rather than high speed data communication, so it is not surprising that the latter is sub- jected to signal degradation and errors when transmitted over existing lines. To assist in the design of modems and data transmission equipment a new instrument capable of simulating telephone line distortions has been manufactured by Axel Electronics Incorporated, New York. Switch-selectable simulation of standard line worst-case characte- ristics is possible, and in addition the instrument can superimpose such steady-state disturbances as variable random noise, phase jitter, frequency shifts and harmonic distortion, as well as transient phenomena including impulse noise, phase and amplitude hits, and amplitude drop-outs. All simulated disturbances can be selected and varied individually or simultaneously. Wandel & Golterman (U.K.) Ltd. 40-48, High Street, Acton, London W3 6 LG, England. (1089 Ml Solar power panel Ferranti Electronics Limited has developed a new solar power module for industrial, pro- fessional and domestic applica- The MST300 series have been designed for long life under extremes of environmental and climatic conditions. The standard module contains thirty-six silicon cells, each 3 inches in diameter, series connected to give an output of 1.1 amps at 14.4 volts. It measures 560 mm x 480 mm and is only 130 mm deep. Its aluminium construction offers good heat sink capability and makes the module ideal for use The module is hermetically sealed to prevent moisture entering the resin filled space containing the silicon cells. Injection of this resin ensures that all air is removed from between the cover and base plate, an important feature of the design as air pockets have been known to cause premature module failure. A cover of fibre reinforced polyester provides protection against the environment including such extremes as sand blasting or ultra violet degredation. An additional bonus of the design concept is the ease with which it is possible to change the module dimensions. Where necessary, additional silicon cells can be incorporated to meet specific customer requirements. Ferranti Electronics Limited Fields New Road Chadderton Oldham OL9 8NP (1092 M) elektor march 1979 - UK 17 IUlLlUIA-. Ultra low noise preamplifier The SL561C from Plessey Semi- conductors, is a high gain, low noise preamplifier designed for use at frequencies up to 6 MHz. Upper and lower cut-off frequencies can be selected by single capacitors and the gain selected between 10 and 60 dB with a single resistor. Operation at low frequencies is eased by the small size of the external capacitors and the low ^ noise. The noise voltage is less than 1 nV/s/Kz and the current consumption of the circuit is 2 mA from a single S volt supply. Applications include use with photo-conductive IR detectors, magnetic tape heads and dynamic microphones. It will also replace an op-amp in many applications where a DC response is not required. Plessey Semiconductors Limited, Chenney Manor, Swindon, Wiltshire, SN2 2QW, England High profile DIL switch A special high profile dual in-line switch for front panel and through panel mounting has been announced by Erg Components. The basic version is a 4-pole 2-way DIL switch that can be finger actuated, or switched by using a small probe. All wiping action contacts are gold plated and specifically designed for high reliability mV/jiA switching, with a maximum voltage/current capability of 240 V AC/2 A (non- switching), 30 V/0.25 A (switching). Custom design versions are also available. Erg Industrial Corporation Ltd. Luton Road, Dunstable, Bedfordshire, LU5 4U, England. New range of miniature buzzers Not miniature electronic wasps from Scotland but a range of small buzzers being produced by Highland Electronics under the type number GA 100/K. Emitting a 400 Hz signal with an output of 70 - 83 dB (A) at 22 cm., they have no moving contacts and do not cause any electrical or R.F. interference. Current consumption is from 16 to 25 mA making them suitable for use in portable or battery operated equipment requiring an audible warning. The case measures 22 x 15 x 10 mm and is made from a high quality plastic, colour-coded according to the four operating voltages of 2.5, 6, 12 and 24 volts. Highland Electronics Ltd. Highland House, 8, Old Steine, Brighton, East Sussex, BN1 1EJ, England. (1091 M) Audio power transistors A range of cpibase and single- diffused TO-220 packaged power transistors now available from Micro Electronics Ltd. are designed for audio output and switching applications, with current ratings of 2 A to 7 A and power ratings of 20 W to 50 W. The devices are available in four chip configurations for both amplification and control circuits. The transistors with epibase chips are designed for good linearity of d.c. current gain, have a higher frequency response (up to 3 MHz), and feature a good safe operating area. Both n-p-n and p-n-p complementary versions are available. These characteristics make the Micro Electronics epibase devices ideally suited to use in the output and driver stages of high-fidelity amplifiers. The range of single-diffused transistors available from Micro Electronics Ltd. are characterised by a distinct safe operating area, and are more suitable for use in voltage regulators, solenoid drivers, and low-speed switching and control circuits. Micro Electronics Ltd., York House, Empire Way, Wembley, Middlesex. England (1093 Ml Digital Gauss/flux meter RLF Industries has announced a new magnetic-measuring instrument, used for measuring both flux density and total flux. The model 906 Portable Digital Gaussmeter/Fluxmeter is designed for use in the laboratory, in production, or in field locations. When operating in its gaussmeter mode, the instrument uses probes based on the Hall effect. A wide range of transverse and axial probes ia available from RFL. Built-in axial and transverse reference magnets provide overall accuracy better than 3%. Full- scale ranges of 1000 and 10,000 gauss provide resolution of 1 and 10 gauss, respectively. Capability for 1 00% overrange provides for measurement up to 20,000 gauss on the instrument’s 3-1/2 digit liquid-crystal display. As an integrating fluxmeter, the Model 906 provides ranges of 10 5 and 10’ maxwell-turns. Probes required for measurements in the fluxmeter mode are available from RFL, or the user may fabricate his own as the need The model 906 operates from either 115- or 230-volt, 50-60 Hz power, or from its internal gel-cel (TM, Gould, Inc.) battery which is included for field operation. A charging circuit is included. The Gaussmeter/Fluxmeter is provided with an impact-resistant plastic case equipped with a carrying handle which serves also as a tilt bail to position the instrument for easy viewing. Size is 8-1/2" wide by 3" high by 9-1/2" deep (21.6 x 7.6 x 24 cm). Weight is 4 lbs, 6 oz. (1.95 kg). RLF Industries, Inc. Boonton, New Jersey 07005, U.S.A. (1094 M) Quad bi-fet switch PMI’s new quad Bi-FET switches deliver high-performance but without the problems that plague competing CMOS products. Because these devices employ Bi-FET rather than CMOS processing they are naturally immune to static blow out. No special handling is ever required. Nor is there any danger of SCR- type latch up problems. Available in both normally-open (SSS7510) and normally-closed versions (SSS75 1 1), each is packaged in a 16-pin, hermetic DIP. They are pin-for-pin replace- ments for the Analog Devices AD7510DI/AD7511DI. ‘On’ resistance at 75 ohms maximum is low. In addition, high •off’ isolation, low crosstalk and low leakage currents makes these quad switches ideally suited for a variety of applications. The SSS7510/SSS7511 are excellent choices for pro- grammable active filters, pro- grammable-gain amplifiers and position controllers. They are also suitable for chopper, demodulator and general-purpose switching and multiplexing applications. Digital inputs are TTL and CMOS compatible. Break-before-make switching is guaranteed - without the need for external pull-up resistors. Precision Monolithics, Inc. 1500 Space Park Drive Santa Clara, California 95050, U.S.A. (1090 M) (1095 M) Elektor March 1979 - UK 20 advertisement Here's why you should buy an ICE instead of just any multimeter Sfr Best Value for money. Jf-Used by professional engineers, D.I.Y. enthusiasts, hobbyists, service engineers. Jf- World-wide proven reliability. Jf. Low servicing costs. Jf. 20K/volt sensitivity and high accuracy. Jf. Large mirror scale meter, jf. Fully protected against overload, jf. Large range of inexpensive accessories, jf. 12 month warranty, backed by a full after sales service at E.B.Sole U.K. Distributors. Prices from £16.60 — £32.00 + VAT I.C.E. | — ELECTRONIC BROKERS LIMITED - ] Please send me full 49-53 Pancras Road. London NW1 2QB. Tel: 01-837 7781. Telex: 298694 colour leaflet and prices on whole ICE range including j Addri ELK 1 HIGH PERFORMANCE POWER SUPPLY KIT lero- 30v 5mA • 1 5A current I Kit of paits/F/G pcb tinned. components £11+8% Schools and Colleges £10 +8% P8iP 7 Op 40kHr Ultrasonic Transducers £2/pair I15pl Audio Amp I.C.p A706^BA641B 5w 12v 4U £1 (15pl RCA 3055 tv EOPT 2N3773 CR25 Resistc 50p(15pl e Amp/OSC, Board £1.30 (!5pl te Deck Top £9 (60p) ^marked £5/100 8 pin Octal Relays 12v 24v 1 lOv DC coil 1 1 pin Octal Relays £1 ea I25pl 24v 48v DC coil 1 15v 240v AC coil 250 R 1 k2 20k 220k 1 m4 60p 11 Dpi Transformers 12v 10A £5 (90pl 24v 10A £6 (£11 9. 3A £1.60 (60pl. 20v 1A Toroid 3 diax 1 £2.25 (30p) 6.0 6. 250mA £1,10 (20p). 6v 500mA £1 .10 (20pl. 18v 2.5A £2.25 I35p). 12 + 1 2v 36V A £2.25 (60pl. 12v 4 A £2.30 (65p), 15.0 15. 36VA £2.25 (60pl. t ELECTROLYTICS 4700p 25v 50p (15p) 4700m 40. SOp (15p) 15.000m 50. £1.25 (60pl 25p (12pl 80p I12pl E.S. £5 150pl Ip 600. Paper Cap 65p(15p) 35p Ideal for Electronic Ignition £2 (25p) CONVERGENCE POTS 512 1012 5012 20012 £5/100 |40pl 1000 <30pl W.W. RESISTORS 0.312 1 20p null pins) 20p|10pl 30A mains Iter £4 (60pl P8iP shown in brackets in 47,"x4'Vx2'' min. order £2 l.m. £350 (80p) ly Switch Adjustable Add 12%% VAT to items mark. 50p(15p) Others 8% KEYTROIMICS 332 LEY STREET, ILFORD. ESSEX Shop open Mon -Sat 9 30 a m 2 p.m Telephone 553 1863