II '"I J If i fm 'Tiuli' , faMr^ 'i ■,- X ' ' : .~ Vs$; m W MLmPt NMfl ■ am Jr j J | tf ^‘fe « 3 ''wL. 1 ^1* - > >»a.. ^■•6 C1 and R2-C2. The time tl for which T1 is turned off is given by R2>C2*ln2 and the time t2 for which T2 is turned off is given by R3-Cl*ln2, where ln2 is the natural logarithm of 2 (approximately 0.7). Figure lb shows the waveforms at various points in the circuit. Waveform A shows the collector voltage of Tl. The base waveform of Tl is shown in B and the exponential rise from a nega- tive voltage can clearly be seen. Waveforms C and D show the voltages on T2 collector and base. The period of the total waveform is 1 1 + 12 and the frequency, being the reciprocal of the period, is 1 ti + t2 An astable multivibrator may be made either symmetric or asymmetric i.e. with tl equal to t2 or tl not equal to t2, and it may be useful at this point to introduce the concept of duty-cycle and mark-space ratio. The duty-cycle of a rectangular waveform is the time for which the waveform is positive divided by the total period times 100%. Thus the duty-cycle of the waveform at the collector of Tl is ti tl + t2 x 100%. On the other hand the duty-cycle of the T2 collector waveform is t2 ti +t 2 x 100%. The mark-space ratio of the waveform is dimply tl/t2, or t2/tl looking at the collector of T2. Monostable multivibrator Unlike the astable multivibrator, which .s an oscillator with no stable or ‘rest’ state, the monostable, as its name implies, has one stable state. It also has an unstable state into which it can be cpped by a trigger pulse, and it will return to the stable state after a preset lane. Figure 2a shows the basic circuit of a monostable. In the stable state T2 is :-med on by current flowing into its rise through R2. Its collector voltage is Figure la. Basic circuit of an astable multi- vibrator, used as a clock generator in the multiplex encoder. Figure 1b. Waveforms at various points in the astable circuit. almost zero so the base of Tl is grounded and Tl is turned off. C2 is charged to (Vb — 600 mV). If a short positive pulse is applied to the base of Tl, Tl will turn on, grounding the left hand end of C2. The right hand end of C2 will take the base of T2 negative, turning off T2. The collector voltage of T2 will rise and Tl will be kept turned on by current flowing into its base through R4, even after the trigger pulse has disappeared. T2 will remain turned off until C2 has charged via R2 and the collector-emitter junction of Tl to +600 mV, when T2 will turn on again and Tl will turn off. The time for which T2 is turned off is exactly the same as for the astable multivibrator i.e. R2*C2-ln2. Figure 2b shows the various waveforms 5-32 — elektor may 1977 slotless model car track in the monostable circuit. Waveform A represents the trigger pulse and B shows the base voltage of Tl. Waveform C is the collector waveform of T2 and waveform D the base waveform. Figure 2c shows a modification to the monostable multivibrator to allow the pulse width to be varied, which is essential in this application. PI controls the voltage to which C2 is connected while Tl is turned on, and hence the negative voltage level applied to the base of T2 when T2 is turned off. This in turn varies the time it takes C2 to charge back up to +600 mV, and hence the time for which T2 is turned off, i.e. the pulse width. This is shown in fig- ure 2d. Waveform C shows the negative base voltage applied to T2 for different settings of PI, and waveform B shows how this affects the pulse width. NOR-gate The final main circuit element used in the multiplex encoder is a NOR-gate. This collects together the pulses from the different monostables and converts them into a pulse train. Figure 3a shows a two input NOR-gate together with its truth table. It can be seen from the truth table that if either input goes high then the output will go low. A multiple input NOR-gate can thus be used to collect pulses from the outputs of several monostables. If all inputs are normally low then if any input goes high a corresponding low-going pulse will appear at the NOR-gate output. Figure 3b shows a multi-input NOR-gate using resistor-transistor logic (RTL) fabricated from discrete transistors. If all inputs are low then all transistors will be turned off and the output will be pulled high by the common collector resistor. However, if any input goes high the corresponding transistor will turn on and pull the output low. Differentiator and integrator Two other simple networks are exten- sively used in the multiplex encoder. These are the differentiating network shown in figure 4, and the integrating network shown in figure 5. The most useful property of the differentiating network is that of producing short pulses from step inputs. If a step input. . . . elektor may 1977 - 5-33 slotless model car track such as the leading edge of a square- wave, is fed to the network then the output voltage across R will initially be the same as the input voltage. How- ever, as C charges the voltage across R will fall exponentially according to the equation -t V out = Vb • eRC When fed with complex AC signals the differentiator functions as a simple high-pass filter, attenuating the low frequency components of the signal. As might be expected, the integrating network functions in a more or less opposite manner. When fed with a step input the voltage across C is initially zero and rises slowly as C charges through R according to the equation -t V 0 ut = V b (l -eRC) The integrator is thus useful as a sort of delay network, and indeed is used as such in the astable and monostable multivibrators. When fed with an AC signal the inte- grating network becomes a simple low pass-filter. Multiplex Encoder — block diagram Having discussed the various elements that make up the multiplex encoder, the block diagram is given in figure 6a. The clock generator (an astable multi- vibrator) produces a 5 ms pulse every 25 ms, which triggers monostable MF1. This produces a pulse whose width can be varied by PI, and when it resets it Figure 2a. Basic circuit of a monostable multivibrator. Figure 2b. Waveforms in the monostable circuit. Figure 2c. PI can be used to adjust the mono- stable pulse length. Figure 2d. Showing how different positions of PI vary the time for which T2 is cut off. Figure 3a. Symbol and truth table of a two- input NOR-gate. Figure 3b. A multi-input NOR-gate using resistor transistor logic. Figure 4. A differentiating network, used to generate short pulses from step inputs. Figure 5. An integrating network, used in the synchronising circuit of the decoder. triggers MF2. When MF2 resets it triggers MF3 and so on up to MF9. Each clock pulse thus produces a train of nine servo control pulses. Each monostable is equipped with a control potentiometer to vary its pulse width and thus control its assigned servo. Since each monostable pulse starts as the previous one finishes it is not poss- ible to collect the monostable pulses directly and feed them to the trans- mitter. Because the pulses follow so closely the NOR-gate output would simply be a continuous low level, or at best a low level with extremely short positive spikes at the switchover point between one monostable and the next. Such a spiky waveform would require a very large transmitter bandwidth. To overcome this difficulty the outputs of the clock and the even-numbered monostables are fed into one NOR-gate, while the odd numbered outputs are fed into a second NOR-gate. At this point it may be useful to refer to figure 6b. Waveform (a) is the clock output, and waveforms (b) to (j) are the monostable outputs, (k) and (1) are the outputs of the two NOR-gates and it can be seen that (1) is almost an inverted version of (k). The next step is to differentiate the two NOR-gate outputs to produce a series of short spikes (m and n). Referring to figure 6b it can be seen that these spikes define the leading and trailing edges of the monostable pulses. Combining waveforms (m) and (n) in a third NOR- gate gives waveform (o), which is simply slotless model car track elektor may 1977 — 5-35 Figure 6a. Block diagram of the multiplex encoder Figure 6b. Timing diagram for the multiplex encoder. Figure 7. Complete circuit of the encoder. The transistors are not discrete devices, but 1C transistor arrays. I I a train of pulses the same lengths as the original monostable pulses with short negative-going spikes between them. However, this waveform is still too ‘spiky’ for the transmitter, so the nega- tive going pulses are used to trigger yet another monostable, which ‘stretches’ the spikes to a constant 400 jrs width. Both normal and inverted versions of the multiplexed signal are available at ihe output of MF10. It will be noted that, since the maxi- mum monostable pulse width is 2 ms the entire pulse train occupies a maxi- mum of 18 ms which, since clock pulses occur every 25 ms, leaves a gap of at least 7 ms at the end of each pulse train, more if the monostables are set to a shorter pulse length. This gap is used to synchronise the receiver decoder with the encoded signal, so that each servo receives only the correct pulses. I Complete circuit The complete circuit of the multiplex encoder is given in figure 7. The clock generator is constructed around T1 and T25, and the only difference between •jus and the basic astable circuit is the inclusion of diodes D1 and D2 and resistors R31 and R32, which help to speed up the leading edge of the rec- tangular waveform. The monostables are formed from transistors T2 to T10, and it will be noted that these are in fact only ‘half monostable’ circuits, in fact simply the pulse width determining portion of the circuit. Two of the NOR- slotless model car track 5-36 — elektor may 1977 T MPX t Yimriririr innr o; ®! 151 lil r ©! □ — f ©! ■ i — l ®! □ ©i 151 ©i K8 ©1 □_ — * — — — - 5 ...15V gates are made up from transistors Til to T20, while the third NOR-gate comprises T21 and T22. The 400 /us pulse-stretching monostable consis ts of T23 and T24, and the MPX and MPX signals are available at the collectors of T23 and T24 respectively. Capacitors Cl to C25 are included to reduce the susceptibility of the clock generator, monostables and NOR-gates to spurious pulses, which could upset the timing sequence. Note that these capacitors should be ceramic types. All other non-electrolytic types should be MKM polycarbonate or other good quality components, and resistors should have a tolerance of 5%. The transistors used in the encoder are not discrete devices but are contained in five CA 3086 transistor arrays. Since the encoder is intended to be universal in application it has been made possible to use any or all of channels 4 ... 9 for switched functions instead of pro- portional controls by including single pole change-over switches at points A and B. Also, the IC pinning is laid out so that if a maximum of only four chan- nels is required for any application then T6 to T10 (IC2), T16 to T20 (IC4) and their associated components may be omitted. However, for the slotless car track all the components shown are required to control up to four cars, and resistors R15 to R21 should be soldered to position A on the p.c. board. The encoder is provided with a simple stabilised supply to enable it to operate from an unregulated DC voltage from 8 Vto 12 V. Decoder principle Referring now to figure 8, it is apparent that the decoder circuit should be capable of extracting from the multi- plexed signal the original control wave- forms K.1 to K9 so that these can be fed to the appropriate servo channels 1 to 9. This can be accomplished by a simple decade counter having ten outputs. If the counter is initially at zero then on the leading edge of the first (Kl) pulse in the multiplex signal the counter will clock and output 1 will go high until Figure 8. Showing the principle of the decoder. Figure 9. Practical circuit of a decoder using a CMOS counter IC. Figure 10. Showing how the joystick control potentiometers are connected to the multi- plex encoder. One joystick control is required for each car — four in all. Figure 11. Circuit of the infra-red transmitter, which was described last month. slotless model car track elektor may 1977 — 5-37 the leading edge of the second (K2) pulse, when output 2 will go high until the third pulse and so on. From figure 8 it can be seen that the pulses thus obtained at the counter output corre- spond to the original control signals generated by the monostables in the encoder. However, some means must be found of ensuring that the counter is in the reset state at the beginning of each pulse sequence, otherwise the pulses might be wrongly decoded and fed to the wrong servo channels. Fortunately thi s is easily achieved by connecting the MPX signal to the reset input of the counter via an integrating network with a switch S connected across the integrator capacitor. During the positive periods of the pulse sequence the integrator capacitor will charge, but if switch S is closed during the spaces in the pulse waveform then C will be discharged and the integrator output voltage will not reach the threshold level of the counter reset input. However, during the space at the end of the pulse sequence C will charge until the voltage on it exceeds the reset threshold level, and the counter will reset ready for the next pulse sequence. Figure 9 shows a practical circuit for a decoder operating on this principle. Here the function of the switch is performed by diode D1 . While the input signal is high the capacitor can charge through resistor R, but if the input goes low C will discharge rapidly through Di. The decoder circuit will be discussed in more detail in the next article in this series. Encoder — p.c. layout A printed circuit pattern and com- ponent layout for the multiplex encoder are given in figure 12. Figure 10 shows the connections made to the joystick controls, one of which is required for each car. For example, the first joystick controls car number one and comprises PI which controls steering and P2 which controls speed. The second joystick con- trol comprises P3 and P4 which control car number 2 and so on. MPH olio ollootto nfinoooQ Q tjouuuuiy nnonopo oooonoo DOUUUUU' slotless model car track Parts list for figures 7 and 12. Resistors: R1,R22,R23,R26,R28.R29, R30,R31,R32,R34,R36 = 4k7 R2,R3,R4,R5,R6,R7,R8, R9,R10,R1 1 = 47 k R24,R25= 10 k R1 2,R1 3,R14,R1 5.R16.R1 7, R18,R19,R20,R21 = 33 k R27 = 68 k R33 = 1 k R35 = 100 k PI . . . P8 = 5 k joystick controls P9 = 5 k lin or 4k7 fixed (see text) P10.P1 1,P1 2, PI 3,P1 4, PI 5. PI 6, PI 7, PI 8.P1 9 = 50 k (47 k) preset Capacitors: Cl . . . C25 = 1 n ceramic C26,C27,C28,C29,C30,C31 , C32,C33,C34,C35,C40,C42, C45.C46 = 100 n MKM C36,C37,C38 = 10 n MKM C39 = see text C41 ,C44 = 10 p/22 V tantalum C43 = 47 p/22 V tantalum Semiconductors: T1 . . . T25 = 5x CA 3086 transistor-array T26 = BC 547 B D1 . . . D4 = 1N4148 D5 = 6V8/400 mW zener Figure 13, Printed circuit board and com- ponent layout for the infra-red transmitter. Figure 12. Printed circuit board and com- ponent layout for the multiplex encoder. 5-38 — elektor may 1977 0«e O XB O" 7 O** O k!S o** o o o o o oooooooo VB VB 11 yz *7. ye re ys *5L ya ra ya C4S _ CIO C 9 CB °~°“°0|[0 OIIOOUO ObOb Ob ApAO Ap nciQOOOQ. a tim pi ica f 01 m I r- ] O OUUOO u CT 1 C 7 olloollo CEO 22 2 * 5 * cai cis C 17 C 15 2 2 22 S So53 C 1 B OQQQOQQ 2 UUUOOO O f -o O- R22 o- R 23 slotless model car track elektor may 1977 — 5-39 Parts list for figures 1 1 and 13. Resistors: R1 = 18 k R2 = 10 k R3,R6,R10,R1 1 * = 2k2 R4.R5 = 5k6 R7= 180 il R8 = 1 k8 R9 = 470 n R12.R13* = 108 R14 = 100 n R15= 22 k PI ,P2,P3* = 5 k preset Capacitors: Cl = 470 p/25 V C2= 100 n MKM C3,C4 = 1 n MKM C5,C6* = 560 p ceramic C7 = 4p7/25 V C* = 100 n MKM (2x) Semiconductors: T1 ,T4 = BC 547 B T2,T3,T5 = BC 557 B T6.T7* = BC 141 D1 . . . D8 = 1N4148 D9 . . . D16 = LD 241 D17* . . . D24* = LD 241 Miscellaneous: 2 x heatsinks TO 5 type NB. * = see text If the unit is used to control four cars in this way, P9 will be redundant. If this channel is not used at all, P9 can be replaced by a fixed resistor between points X9 and Z9. Testing and fine adjustment of the multiplex encoder will be given in a later article. Infra-red transmitter Finally, the p.c. board and component layout for the infra-red transmitter, the circuit of which was given last month, is given in figure 13. The circuit itself is reproduced in figure 1 1 . To adjust the infra-red transmitter, first set P2 and P3 to their central positions and allow the circuit to warm up for about ten minutes. Insert an ammeter between points A and B and adjust P2 to give a T6 collector current of about 180 mA. Repeat this procedure with P3 for T7 (meter connected between points A and B'). With some specimens of the LD 241 LED it may not be possible to reduce the collector current to 180 mA. In this case R10 and Rll should be increased to 3k3. .. (To be continued.) Modifications to Additions to Improvements on Corrections in Circuits published in Elektor LED VU/PPM April 1 977, E24, p. 4-32. It was perhaps not made sufficiently clear that the display itself (figure 5) works off an asymmetrical power supply (+15 V). No negative (—15 V) supply is required: points ‘O’ and ’ on the display board (figure 7) are both connected to supply common. The rectifier section (figures 2 and 6) does require a symmetrical (+/-15 V) supply, of course. Speech shifter July/August 1976, E15/16, p.742. In some cases the TBA 1 20 has been found to be so asymmetric that the oscillator signal is audible at the output. The sol- ution in this case is to add a 22 k or 25 k preset potentiometer between pins 7 and 9 of the IC; the slider is con- nected to supply common through a 10 k resistor. / /TBA120 • Precision timebase • Microprocessors • 'Wireless' intercom • Elektor SUMMER CIRCUITS issue 5-40 - elektor may 1977 op-amp frequency compensation . . . the why and the how ®|p=©lfiRl|p lrlf©€j|U©ffil©y ©®ifiRi|p©ifii®©lsii®ifii. . . Islfo© why ©®dl Islhi© h@w When an operational amplifier is used in a negative feedback circuit its frequency response requires 'compensation', a high-frequency rolloff that may be 'internal' or 'external'. With inadequate compensation the circuit will usually misbehave or even oscillate. This article will explain the reasons for frequency compensation, describe the usual 'simple' approach and then show how an 'external' type can be fitted with an improved compensation-arrangement. The latter approach results in a circuit that responds far better to large and fast excursions of the input signal. Why does an operational amplifier need compensation? The story starts with the observation that parasitic capacitances in the IC itself cause the ‘open-loop’ (i.e. without-feedback) response of the device to roll off more or less sharply above a certain high frequency. This is illustrated by the drawn line in figure 1 - the ‘uncompensated’ response. The actual curve is bounded by asymptotes, 6 dB/octave (20 ilB/decade) above f i , 12 dB/octave above the second turnover point {2 and even 18 dB/octave above f 3 in cases where there is a third turn- over. The open-loop gain is constant from DC to fi (the real curve 3 dB ‘down’ at fi), equal to the value A 0 j. Figure 1 also shows the desired gain- with-feedback-operating, A c j (in deci- bels). If the slope of the open-loop response at the intersection with the horizontal through A c j exceeds 1 2 dB/ octave, the actual feedback will start to become positive, as the total phase-shift will have exceeded 180°. With the values assumed in the figure the op-amp would certainly be in business for itself! The only way to ‘dump’ enough open loop gain before the phase-shift in the IC exceeds 180° is to provide HF rolloff, starting early enough — so that the intersection between the open-loop and closed-loop response curves occurs at 6 dB/octave. A step-network that ‘flattens’ again at fi (drawn curve in figure 2) is the standard trick. It will result in the dashed curve of figure 1 . The situation is in fact that the ‘loop gain’ falls below the amount that would enable oscillation, if the feedback were to become positive, at a point where there is still 90° phase-margin. Note that many integrated op-amps have their frequency compensation built-in. An internal capacitor then displaces one of the ‘stray’ rolloffs so far downward in frequency that it dominates in the response, automati- cally providing the figure 1 dashed curve. Perhaps the best known example of this is the ‘741’. Those op-amps that are intended for use with external compensation are supplied with data on how this should be done, given the required values of closed-loop gain, phase-margin etc. For most appli- cations the instructions err on the ‘safe’ side. That concludes the review of the basics of frequency-compensation. It is now time to take a closer look - preferably inside the IC! It will be convenient to assume the usual op-amp circuit of differential input stage, second stage with gain and some form of wideband unity-gain output stage (usually with local feedback and biassed in class B). The rolloff time-constant is normally inserted between the first and second stages or as a ‘Miller’ integration net- work in the second stage itself. It is not difficult to see that an op-amp with the figure 1 dashed response, obtained by a ‘slow’ second stage, will have its input stage driven progressively harder above fj, due to the failing feedback (6 dB/octave above f 1 , 12 dB/ octave above f 3 ). There is a distinct danger that rapidly- changing high-amplitude signals will cause the input stage to momentarily saturate, at the steepest part of the waveform - usually the zero-crossing. This results in bursts of gross distortion -. in audio amplifiers — known as Transient Intermodulation or TIM (see Equin part 1, Elektor 12, April 1976, p. 448). The solution to this problem is to insert the compensation network at the ampli- fier input. Figure 3 a shows how this is done for a non-inverting amplifier and figure 3b gives the inverting circuit. The figure also gives rules for determining the resistor and capacitor values required. Some op-amps will misbehave with nothing connected to their ‘com- pensation’ pins; it is not always immedi- ately apparent why — so that no general rule can be given. A trick that usually works is to insert a series RC-pair that reduces the open-loop gain by 6 dB or so, in a step, at some frequency above the highest input but well below f 1 , at the usual compensation position. The insertion of the compensation ahead of the input stage removes the cause of slew-rate limiting and TIM ; the drive level of the input stage proper no longer rises with frequency during the sloping part of the open-loop response. There is however a price to be paid, quite apart from the extra mess around the input pins. Noise from the input stage is no longer attenuated by the compensation network — it receives the full open-loop gain up to fi . The kind of circuit in which TIM is a problem (high level) is however not usually so critical in respect of noise. Furthermore, the low source impedance at higher frequencies ‘seen’ by the input stage will tend to reduce its noise level anyway. H Figure 1. The drawn line shows the op-amp's frequency response without 'compensation'. The dashed line shows the compensated or rolled-off response. At its point of inter- section with the horizontal dotted line through A c | (the so-called 'closed-loop gain', i.e. the amplication obtained with feedback operating), this response curve slopes at 6 dB/octave (20 dB/decade) — and the system is unconditionally stable. Figure 2. A so-called 'step-network' compen- sation will cause the drawn line in figure 1 to follow the 'compensated' response curve shown dashed in figure 1. Figure 3. Basic circuit and 'design rules' for improved compensation of a non-inverting (a) and an inverting amplifier (b). op-amp frequency compensation . . . the why and the how elektor may 1977 - 5-41 A.J.B.M. Peters Readers may remember that a design for an automatic callsign generator was featured in Elektor a little over a year ago (Elektor 1 1, February 1976). This new design offers the same facilities with considerably simpler circuitry, though at the expense of a slightly more complicated programming procedure. It may be remembered that the previous design for a callsign generator used CMOS shift registers whose outputs were connected via diodes to two programming lines. This made for very simple programming but made the cir- cuit fairly complicated. The program- ming of the new design is accomplished by storing the callsign in a 1 00 bit read only memory consisting of a diode matrix. A dot is stored in the matrix by inserting one diode in the required position. A dash, which has a duration equal to three dots, requires three diodes. A space within a character is of one dot duration and occupies one blank space (no diode) in the matrix. A space between letters is the same duration as a dash and thus occupies 5-42 — elektor may 1977 morse call sign generator three blank spaces in the matrix. To generate the callsign the contents of the matrix are read out row by row. 100 bits may seem excessive, but it is possible for a single figure (digit 0) to occupy 19 spaces in the matrix. This, combined with long European call signs, soon uses up the spaces in the memory. British callsigns of 4 or 5 characters will, of course, not use as much of the memory capacity. The complete circuit of the callsign generator is given in figure 1 . The diode matrix is in the top left corner of the diagram. Readout is accomplished by addressing the rows and columns of the matrix using two 7490 decade counters and 7442 decoders. The rate at which the callsign is repeated is determined by IC5, a 555 timer connected as a monostable multi- vibrator. Assume that initially the monostable is in the triggered condition. The output, pin 3, is high, so both the counters 1C3 and IC4 are held in the reset condition. Output 0 of IC1 (column 0) and output 0 of IC2 (row a) are thus both low and all other outputs are high. One input of NOR gate N1 is low and the other is high, since no diode is connected in position ‘aO’ in the matrix as this is the rest position. The output of N1 is thus low. When the monostable (1C5) resets, the reset inputs of IC3 and IC4 go low and IC3 begins to count pulses from the clock generator built around S2. As the counter counts the column outputs 0 to 9 of 1C1 go low in turn. Whenever a position is reached where a diode is connected from a column output to row ‘a’ then the second input of N1 is pulled low and the output goes high. At the end of the first row the D output of IC3 will go low, causing IC4 to advance one step. One input of N2 will now be low and as IC3 counts from 0 to 9 again the information on row ‘b’ will be read out via N2. This is repeated until all the rows of the matrix have been read out. The diodes connected to the outputs of N1 to N10 form an OR gate to route the information to the inputs of an audio tone generator SI and a relay driver Tl. When a dot or dash is present the tone generator is activated and the relay is energised. During spaces between character elements there is no tone and the relay drops out. The tone generator may be used to modulate a transmitter or the relay may be used for CW keying. When a count of 100 has been reached all the rows of the matrix will have been read out. The D output of IC4 goes low on count 100. This negative-going edge is differentiated by the 1 n capacitor and 1 0 k resistor to produce a short pulse which triggers the 555, inhibiting the counters until the 555 resets again. The repetition rate of the callsign can be varied by means of PI . Programming requires a fair number of diodes, the exact quantity depending on the actual callsign. To programme the generator, start with row ‘a’ of the matrix. Leave position ‘a0’ blank as this is the rest position. Work along row ‘a’ and connect a diode for each dot with its anode to row ‘a’ and its cathode to the particular column you have reached. For a dash a diode must be connected to each of three successive columns. For a space the appropriate number of columns must be left blank. When the end of row ‘a’ is reached then return to the start of row ‘b’ and continue. The callsign example shown in the dia- I gram is the author’s, DE PA0ARR, I which in morse is This is laid out in the matrix as follows: I music cleaner elektor may 1977 — 5-43 quency of 25 Hz was chosen, but again this can easily be altered if so desired. The ultimate slope of the filter is also important. Some filters have a slope of as little as 6 dB/octave. This means that, to be effective in suppressing rumble or noise the turnover point of the filter must be within the frequency range of the wanted signal leading to partial loss of the signal or, to use another metaphor, ‘throwing the baby out with the bathwater’. In response to popular demand, here, at last is a treble and rumble filter that can be used in virtually any hi-fi system to get rid of the snaps, crackles, pops and (g)rumbles without getting rid of half the music. There has been much speculation about who first coined the phrase ‘The wider you open the window, the more the muck blows in’. However, there is no disputing its truth with respect to audio systems. The bandwidths of pickup cartridges, amplifiers and loudspeakers are now so great that the imperfections of records and turntables are often glaringly exposed, even by a relatively inexpensive cartridge, amplifier and loudspeakers. Quite apart from the argument as to whether these phenomenal bandwidths actually serve any useful purpose apart from churning up one’s innards at the bass end and annoying the neighbour’s dog at the treble end, there is obviously a need to ‘chop off the top and bottom end of the audio spectrum under certain circumstances. Turntables, for example, are not always as rumble-free as one might wish, and being mechanical devices tend to get worse with age. Records, too are rarely perfect, suffer- ing often from rumble and even more frequently from surface noise, which latter tends to become worse with wear. Then, of course, one must consider the enthusiast who possesses treasured col- lections of older records or non-noise- processed tapes. Design requirements Unfortunately, few amplifiers, except the most expensive ones, possess effec- tive treble and bass filters, simply because the requirements for such filters are rarely properly formulated. Firstly, the turnover (-3 dB) points of the filters should be chosen carefully. Many treble filters cut off at much too low a frequency, leading to loss of part of the wanted signal. Similarly, rumble filters often cut off at too high a fre- quency. The treble filter in the present design has a choice of two cutoff points. A cutoff point of 25 kHz is chosen to prevent ultrasonic signals from reaching the power amplifier, as these can lead to the (now) well-known transient intermodulation distortion (TIM). For elimination of record and tape noise a cutoff frequency of 10 kHz was chosen. The circuit can easily be adapted to give many other different cutoff frequencies according to personal taste, calculation of the required component values being a relatively simple matter. For the rumble filters a cutoff fre- Choice of Response To avoid this both the filters used in the present design have an ultimate slope of 18 dB/octave. There still remains, however, the choice of filter design. The magnitude/frequency re- sponse of the so-called Butterworth type of filter is maximally flat in the passband. The Chebishev filter attains a sharper cutoff at the turnover point than does the Butterworth, but at the expense of a magnitude frequency response that is not flat in the passband. However, neither of these types are concerned with phase distortion of the signal, which is important when dealing with complex waveforms such as music. To minimise phase distortion the phase Figure 1. Theoretical circuit of the third order lowpass Bessel filter used for the treble filter. Figure 2. Theoretical circuit of the third order highpass Bessel filter used for the rumble filter. 5-44 — elektor may 1977 music cleaner shift produced by a filter should vary linearly with frequency. This condition is fulfilled by a filter having a Bessel type of response. This produces minimum phase distortion, though the cutoff at the turnover point is not as sharp as either the Butterworth or Chebishev filters. Three-pole Bessel filters were thus chosen for this design. Having decided on the best type of filter response one is then faced with the problem of practical realisation. There are various realisations, some of which require infinite open-loop gain from the active part of the circuit. While these requirements can be ap- proximated using IC operational ampli- fiers, such ICs are not ideal from the point of view of distortion and noise. The configuration finally chosen was that using a voltage follower. This is easy to realise with low-noise audio transistors, and the overall gain of the filter within the passband is unity, which means that it can be inserted in any amplification chain without affecting the gain. Figure 1 shows the theoretical circuit of the lowpass (treble) filter, while figure 2 shows the theoretical circuit of the highpass (rumble) filter. The three resistors in the lowpass filter are identical, and the values of the three capacitors required for a given turnover frequency can be calculated from the three equations given. Once the capaci- tor values for a given frequency have been found then the turnover frequency may be altered simply by changing the values of the three resistors, e.g. if the resistors are halved the turnover fre- quency is doubled, if they are doubled the frequency is halved. For the highpass filter the three capaci- tor values are equal and the resistor values are calculated from the equations given. In this case halving the capacitor value will double the turnover fre- quency, and doubling them will halve it. The equations for obtaining the filter parameters are reproduced from ‘Electronics’ August 18th 1969. Practical Circuit To obtain a treble and rumble filter in the same unit the two circuits of figures 1 and 2 are simply cascaded. The com- plete practical circuit of one channel of the filter is given in figure 3. Here the theoretical voltage followers have been replaced by transistors connected in a ‘super emitter-follower’ configuration. The rumble filter is built around T1 and T2. R3 and R4 perform a dual function, forming part of the rumble filter but also providing the base bias forTl. The effective value of these resistors as far as the filter is concerned is R3 paral- lelled with R4 or j^ 3 * . R3 + R4 The treble filter is built around T3 and T4. Selection of the turnover point is carried out by a three way switch. With the switch open only R7, R8 and R9 are in circuit, and the turnover point is about 10 kHz. With the switch closed these resistors are connected in parallel with RIO, R1 1 and R12 respectively, and the turnover point goes up to about 25 kHz. Capacitors C4 and C8 are included to guard against the possibility of r.f. instability. The circuit has a high input impedance and low output impedance, so that it may be connected virtually anywhere in an audio system. For use with an existing amplifier the unit may be connected to the tape socket if this is music cleaner elektor may 1977 - 5-45 Parts list to figures 3 and 4 As the p.c. layout is for a stereo system all components except SI are duplicated. Resistors: R1 = 100 k R2 = 68 k R3,R4 - 820 k R5,R1 3 = 5k6 R6.R14 = 2k2 R7.R8.R9 = 3k9 (see text) R10,R1 1 ,R12 = 2k7 (see text) R1 5 = 47 k Capacitors: Cl ,C2,C3 = 68 n C4,C8 = 4n7 C5 = 3n9 C6 = 5n6 C7 = 1 n C9 = 22 m/35 ... 40 V CIO = 10 m/35... 40 V C11 = 100 n Semiconductors: T1,T3= BC547B.BC107B or equivalent. T2.T4 = BC557B.BC1 57B or equivalent. Miscellaneous: SI = six pole on/off with normally closed contacts. Figure 3. Complete circuit of the rumble and treble filter. Figure 4. Printed circuit board and com- ponent layout for the filter unit. Figure 5. Gain/frequency response curve of the filter unit. not already in use. The input to the filter is taken from the ‘tape’ output of the amplifier and the output from the filter is fed back to the ‘tape’ input. Depressing the ‘Tape Monitor’ button will then bring the filter into circuit. Distortion introduced by the filter is extremely low so that it may be used with the highest quality systems. Printed Circuit Board A printed circuit board and component layout for the stereo filter are given in figure 4. Provision is made on the p.c. board for mounting a pushbutton switch for SI. Alternatively SI may be mounted remote from the board provided the leads are not more than a few centimetres long. If only one turnover point is required for the treble filter then SI and the associated resistors may be omitted. Response Curve The gain/frequency response curve of the filter unit is shown in figure 5. As mentioned earlier, with SI depressed (open) the treble filter turnover point is around 10 kHz, while with SI closed it is around 25 kHz. M 5-46 — elektor may 1977 active loudspeaker-crossover filters (1) Few things can so hold the attention of the serious audiophile as do loudspeakers. This applies with particular strength to those whose fingers always have the experimenter's itch — so that they cannot or will not without reserve accept somebody else's idea of a loudspeaker system. This can lead to the expenditure of considerable sums, if only on wooden panels, and it will sometimes also lead to frayed tempers at home . . . One of the ways of sinking cash into an existing system is to replace the 'passive' separating ('crossover') filters by 'active' types. This of course involves the provision of a separate power amplifier for every driver in the system. This article on Active Crossover Filters (ACF's) will describe a universal filter circuit, capable of producing a vast number of filter characteristics. High-quality loudspeaker systems are invariably designed on the basis of ‘divide and rule’ principles. The incoming audio spectrum is split up into two, three or even four sub-spectra, each of which is then passed to a loud- speaker specially designed for that particular frequency range. The change- over from one loudspeaker to the next higher in frequency range is ac- complished by a complementary filter-pair whose roll-off response-flanks ‘cross over’ each other at a point some decibels below the ‘full power’ level. The filter-pairs are therefore called ‘crossover filters'. A loudspeaker system that uses such filters is usually called a ‘multiway’ system. When the filter sections are inserted between the single power amplifier and the individual ‘drivers’ (i.e. loudspeakers proper), the system is said to have a passive filter. Figure 1 illustrates a typical three-way system. The low-to- midrange crossover frequency is fi and the midrange-to-high crossover occurs at f 2 . The representatives of the animal kingdom shown have had their typical calls ‘borrowed’ to provide a classifi- cation of the drivers into the categories low-range (woofer) midrange (squawker) and high-range (tweeter). The big idea behind the multiway approach is the fact that an optimally- designed ‘woofer’ is — for basic design reasons — a sub-optimal loudspeaker at higher frequencies. This does not mean that a ‘new’ design method may not someday produce a first-class full-range driver; it simply hasn’t been done yet. The problems to be faced are quite formidable — and a computer is only useful to quickly do the sums that a human being already knows how to do. A multiway system is necessarily more complicated and more expensive to produce than a single-driver system. That is a clear disadvantage. There is however a second objection to the multiway approach - a more funda- mental objection: how does one tackle the fact that frequencies near the crossover point are radiated by both drivers? The two radiating diaphragms cannot be at the same position in space — although they often can be spaced quite closely — so that ‘interferences’ between the two radiated waves can cause irregularities in the response characteristics and in the radiation pattern of the system. ‘Dividing’ is one thing; ‘ruling’ is quite another . . . Most of the interference effects can be avoided when the two frequency- adjacent drivers are mounted concentri- cally — one within the other. This is usually no problem, since an optimal tweeter can be made smaller than a woofer. The past has known designs — many of them still very popular - in which a tweeter of one kind or another has been built into a woofer (or, more accurately, a woofer-midrange) cone- loudspeaker. The crossover can be mechanical in nature (as in the ‘good old’ Philips 971 OM), or a more advanced twin-driver-plus-electrical-crossover sys- tem can be employed (as in the famous Tannoy Monitor Gold and certain Goodmans and Isophon units). Passive or active? Having decided that a good loudspeaker system, at the present state of the art, is going to need at least one crossover filter, we have to decide whether this filter should be a ‘passive’ or ‘active’ design. (For our purposes, an ‘active’ filters is one in which the inductors have been eliminated by the application of capacitors and amplifiers ). Figure la illustrates a typical passive- filter three-way system . The passive filter is built up with inductors, capacitors and any matching networks that may be necessary (e.g. to reduce the drive to a too-sensitive tweeter). Figure lb illustrates the bare bones of a three-way passive filter. One difficulty is immediately apparent. The woofer section requires an inductor in series with the driver voice-coil. The considerable inductance involved means that there will be power loss in the copper-resistance of a many-turn air- cored coil, or else that there will be distortion due to the non-linearity of a low-loss coil that has a ferromagnetic core. Neither of these effects should however be viewed out of proportion: the often-cited effect of the series- resistance on the woofer’s electrical damping is completely swamped by the effect that the voice-coil resistance has - and one can design iron-core inductors with a level of distortion that is insignificant compared to that of the actual driver. active loudspeaker-crossover filters (1) elektor may 1977 — 5-47 Another source of difficulties is more awkward to eliminate. Normal electrical wave-filters assume a pure-resistance load-termination. When you connect a loudspeaker to such a filter the final characteristic may not be quite what you intended - it may even be wildly off. The trick of connecting an RC network across the speaker terminals to compensate the high-frequency rise in impedance (due to the coil’s inductance) certainly works and should be better known; but the fun really begins when the speaker impedance contains signifi- cant components ‘reflected’ from the mechanical ‘circuit’. That usually happens in the neighbourhood of the driver’s fundamental resonance; it can be a very expensive nuisance in the case of midrange and tweeter units that have a resonance (as is usual) at or just below their high-pass crossover frequency. Now, a well-designed commercial ‘passive filter system’ will invariably work very well but that success is due to a combination of design experience and available facilities beyond the reach of the ‘do it yourself’ audiophile. Although it would be possible to say a great deal more about passive filter arrangements and matching networks, this article is supposed to be about active arrangements. Having implied, above, that the amateur is better off tackling his problem with an active system, we must now try to explain how. Active Crossover Filters Figure 1 c shows the block diagram of a three-way active (‘electronic’) crossover filter. It is immediately clear that each of the loudspeakers requires its own power amplifier. This need not be so expensive as one might think, since the total power required (and hence the amount of mains transformer, reservoir capacitor and heat sink) is not increased by subdividing the amplifier. As a rule, the woofer will need the most powerful amplifier (perhaps SO . . . 70% of the total), with the midrange unit handling perhaps two-thirds of the remainder. Much will obviously depend on the individual drivers used. When drivers are obtainable with varying rated im- pedances, the power distribution over the output stages can be achieved by using a single supply voltage together with a low-impedance woofer (say 4 ohm), a mid-range unit of higher Figure la. Block diagram of a three-way system with passive crossover filter. Figure 1b. As an example: the KEF type DN 12 SP 1004 three-way passive filter. Figure 1c. Block diagram of an active-filter three-way system. Figure Id. An active-filter two-way system. 5-48 — elektor may 1977 active loudspeaker-crossover filters (1 impedance (say 8 ohm) and a tweeter ot still higher impedance ( 1 5 ohm ). A major advantage of the active-filter approach is the ease with which sensi- tivity differences between the drivers can be eliminated. In figure lc this is accomplished by adjustment of the presets PI, P2 and P3. Figure Id gives a simpler two-way circuit, suitable for use with smaller diameter woofers that are also well-behaved throughout the mid- frequency range. Still another possibility is shown in figure le, a ‘hybrid’ three- way system. In this case the woofer to midrange crossover is done with an active filter and two power amplifiers; the frequency ranges for the midrange and tweeter drivers are however separated by a passive filter set. What are the other advantages of the active filter approach? - the design is far more flexible; a change of crossover frequency or drive level can be quickly and conveniently achieved by changing one or two R’s and C’s or adjusting a preset potentiometer. - there is no complication in the filter design caused by the awkward termination (the loudspeaker impedance). - it is relatively simple to produce complicated filter characteristics whenever this is thought desirable or necessary. - since the power amplifiers will usually be installed in the loudspeaker cabi- net, the individual drivers can be pro- tected from overload by suitable choice of the power rating of the amplifier concerned. The filter circuits Figure If shows a set of filter charac- teristics, as would be required for a three-way system. The frequencies fl and f2 are the ‘-3 dB’ points, at which the response curves of a complementary filter-pair actually ‘cross over’ each other. Half of the power at a crossover frequency is transmitted through each filter of the pair. For a three-way system fi will frequently lie between 300 and 600 Hz (sometimes as low as 100 Hz, or as high as 800 Hz). The other crossover will then usually be found between 2 kHz and 8 kHz — typically near to 5 kHz. The single crossover in a two-way system is usually between 1 kHz and 3 kHz. (typically . around 2 kHz). The slope of the various filters well into their respective ‘stop-bands’ is a multiple of 6 dB/octave (i.e. 20 dB/deeade). The figure If curves are drawn for 12 dB/octave (1,4 ,5, 8) and for 18 dB/octave (2,3 ,6, 7). If we assume that either slope may be used for each of the four filters, then there are sixteen possibilities for a three-way filter. It is not always desirable to make the filters of a crossover-pair with the same slope - a so-called asymmetrical crossover may be needed when the response of one of the loudspeakers is not flat through the crossover point. Table 1 lists the possibilities. active loudspeaker-crossover filters (1) I elektor may 1977 - 5-49 The last four alternatives apply to two- way systems. We will refer in this article to the single crossover as fi . An electric wave-filter is characterised not only by the ‘ultimate slope’ of the rolloff curve, well into the ‘stop band’ but also by the ‘sharpness of transition’ between the pass-band and the stop- band. A number of Famous Names are associated with a classification of filters into categories with increasing sharpness (once again: note the distinction between sharpness and steepness). Almost all loudspeaker crossover filters are of the Butterworth ‘maximally flat amplitude’ type. We will therefore illustrate the workings of the practical circuits by Butterworth responses. When the ‘pass-band’ is defined as the frequency range up to the —3 dB point (low-pass) or from the —3 dB point up- wards (high-pass), then Butterworth gives the lowest possible ‘pass-band attenuation’ that can be obtained with- out allowing ‘ripples’. The figures 2, 3 and 4 give the design information for Butterworth low-pass filters (‘a’ figures) and Butterworth high-pass filters (‘b’ figures), for ultimate slopes of 18 dB/octave (figure 2), 12 dB/octave (figure 3) and 6 dB/octave (figure 4). The two sets of component numbers refer to the two different crossovers. We will come back to this when referring to the parts list. The active element in the circuits of figures 2, 3 and 4 is a voltage follower. The best known AC voltage follower is the so-called ‘emitter follower’. Since a voltage gain of unity can only be closely approximated by an amplifier with extremely high current gain, the total circuit diagram of figure 5 shows ‘super emitter followers’ using two transistors each. The derivation of the component values always assumes the use of an ideal voltage follower; any attempt to ‘make allowances’ is fraught with great uncertainties — and the assumption that a one-transistor follower is ideal is just too optimistic! This is not the place to go into the Figure 1e. A hybrid active/passive three-way system. Figure If. A few frequency-response plots, with slopes of 12 and 18 dB/octave and one or two crossovers, as an aid to interpretation of table 1. Figure 2. Circuit diagram and values for a Butterworth low-pass (a) and high-pass (b) 18 dB/octave filter. Figure 3. Circuit diagram and values for a Butterworth low-pass (al and high-pass (b) 12 dB/octave filter. Figure 4. Circuit diagram and values for a low-pass (a) and high-pass (b) 6 dB/octave filter. 5-50 — elektor may 1977 active louospeaKer-crossover filters (1) details of the derivation of design formulae. One practical consequence of the derivations must however be noted here. That is the fact that it is not always possible to design filters in which all the frequency-determining R’s and C’s have convenient values. We have chosen circuits with either three equal C’s (high pass) or three equal R’s (low-pass), the other components hopefully coming fairly close to standard E12 values. Filters with low ‘Q’ values (such as Butterworth) will, fortunately, not immediately go haywire when some of the components are a few percent out. That is not to say that a fusspot with access to 1% R’s and C’s should not indulge a craving for ‘precision’ . . . So much for the general aspects of active crossover filter design. It is now time to try working out a specification. One way to tackle this problem is to use a check-list. - Active filters only (figure 1 c or 1 d) or hybrid (le)? - Three-way or two-way? - Which speakers? - How steep the filters? - Which amplifiers? Do not try to find complete ‘paper’ answers to these questions. A great deal will depend on one’s individual taste and on whatever happens to be available. Note that the idea was to find something to play with! There is one fundamental guideline, however. Loudspeaker are meant to be used for listening to music, not the other way round . If it sounds right, then never mind what it looks like on paper. Assuming that one’s musical taste is reasonable, any discrepancy between the theory and the actual result will usually be due to an oversight or incompleteness in the theory. It will simplify this story if we introduce two further ‘boundary conditions'. Let us assume that (1) we are going to do the job properly no skimping on parts - and (2) that the reader already knows how to design his enclosure. The question that should be tackled first is the choice of the loudspeaker to be used. This usually will involve a dig into the manufacturer’s literature - or at least a good look into a distributor’s catalogue. Unless one knows precisely what one wants, it is a good idea to select a combination recommended by the manufacturer, replacing only the inevitable passive filter by circuits covered in this article. Information on how to construct special woofer enclosures, such as folded horns or ‘transmission line’ types, can often be found in the literature. The basic choice between two-way and three-way systems is not inevitably one of cost, with three-way always better if I you can afford it. On the contrary, I some of the best-sounding systems around use a woofer-midrange unit plus I a tweeter. These woofer-midrange units I do however tend to need rather more I than a simple closed-cabinet if they are I to do a really good job at the deep-bass I end. The frequencies and ultimate slopes of I the crossover filters can be taken, at I least as a starting point, from the parameters of the passive filter I recommended by the speaker manufac- I turer. If one is combining speakers from | various sources, then some experiment ' may be necessary (great fun!). There are one or two guidelines here, more ‘don’ts’ , than ‘do’s’. In the first place, beware of the ‘power handling capacity’ ratings of > tweeters. It is in the nature of things I that their smaller coil systems cannot I handle the massive amounts of input I power that will not damage woofers. I The temptation to suppliers is to quote I a high power rating for a tweeter in I combination with a specified high-pass I filter. The ‘power density’ of normal I music spectra certainly becomes I significantly lower as the frequency increases; but this no longer applies I when the amplifier is driven into I distortion (accidentally or on purpose). I active loudspeaker-crossover filters (1) elektor may 1977 — 5-51 Table 1. The different possible combinations of symmetrical or asym- metrical crossovers and 1 2 or 1 8 d B/octave slopes filters slopes at filters slopes at fl to be f 2 tc > be combine from refer to > 600 n) Re= 12 V/20 mA relay 5-54 — elektor may 1977 albar mkll-a bfs Figure 3. Transmitter for the two-transducer version. Figure 4. Printed circuit board and com- ponent layout for both figure 1 and figure 2. (EPS 9815-11. Figure 5. Printed circuit board and com- ponent layout for the transmitter in the two- transducer system. (EPS 9815-2). Figure 6. The number of transmitters (or receivers, for that matter) can be extended at will, provided the transmitters are arranged to run at the same frequency. The procedure is as follows: - set P2 at maximum resistance (fully anti-clockwise). - as before, move a hand to and fro in front of the transducers, and set PI on the transmitter board for maximum level of the tone in the earpiece. - adjust P2 until the correct sensitivity is obtained. One final tip: in some cases one j transducer is better than the other, so it is worth while interchanging them to see which one is better for the receiver. Three-transducer version . . . In theory it is possible to extend the two-transducer version by adding one or I more extra transmitters. For this it is essential that they should all oscillate at the same frequency. I Although we have not tried this in practice, the principle is relatively I simple. Output ‘Y’ from one of the ' transmitters (the ‘master’ oscillator) is used to drive or synchronise the other(s). There are two ways to do this. Provided the cable capacitance of the link ■ between the transmitters is not too high, components Rl, R2, PI and Cl of > the ‘slave’ transmitter can be omitted and output ‘Y’ from the master is fed to I pins 2 and 6 of the TC on the slave transmitter. Alternatively, all com- ponents are mounted on the ‘slave’ I board, but the lower end of C2 is disconnected from supply common and the ‘Y’ signal is fed in via this capacitor to pin 5 of the IC. PI on the ‘slave’ I transmitter is now adjusted for ‘zero I beat’ in the earpiece. IL albar mkll-a bfs market elektor may 1977 — 5-55 Parts list for receiver (figures 2 and 4) Resistors: R1 = 22 k R2,R7 = 10 k R3.R19 = 560 R4 = 5k6 R6 « 56 k R8 = 33 k R9,R1 1 = 100 k RIO = 270 n R12.R15 = 47 k R13 = 100 a R14 = 2k2 R16,R V = 4k7 R17 = 1 k R18 = 1 k5 P2 = 4k7 Capacitors: Cl = 22 p/1 6 V C4 = 1 n C5 = 22 p/10 V C6 = 100 n C7 = 2p2/63 V C9 = 100 p/16 V CIO = 10 p/16 V Cl 1 ,C14 = 47 p/16 V C12 = 10 . . . 100 p/16 V C13 = 470 p/16 V C15 = 330 p C16 = 22 p/6 V Semiconductors: I Cl = 741 T1 ,T3 . . . T5 = BC547B or BC107B T2= BC557B or BC177B D1 = 10 V/400 mW D2 . . . D5= 1N4148 Sundries: US = Murata MA 40 L 1 R or MA 40 L IS LS = headphone or earpiece (R > 600 n) Re = 1 2 V/ 20 mA relay Parts list for transmitter (figures 3 and 5) Resistors: R1,R3 = 1 k R2 = 4k7 PI = 4k7 Capacitors: Cl = 2n2 C2 = 100 n C3 = 33 n C4 = 4n7 C5 = 10 p/16 V Semiconductors: 1C = 555 Sundries: LI = 6.8 mH US = Murata MA 40L 1R or MA 40L IS Conductive spray paint Chomerics’ 4900 conductive paint is a one-component, air-drying coating which will provide effective EMI attenuation when applied in a 2-mil coating on non- conductive enclosures. It can also be used improve the shielding characteristics of metal enclosures by reducing the contact resistance of flanges; to create or improve grounding surfaces; to fill minute voids in porous metal castings; and to make conductive patterns on nonconductive substrates. This paint is a solvent-based system consisting of a silver filler in an acrylic resin. It is specifi- cally intended for spray- application, but can also be applied by dipping or brushing. Regardless of application method, 4900 must be agitated often to prevent the dense filler from settling out and causing resin-rich areas in the coated surface. Coverage is approximately 70 sq. ft. per pound in a 1-mil thickness. When used as a shield, the 2-mil thickness should be built up in 2-3 applications, allowing 15-20 minutes after each application for solvent evaporation. Optimum properties are developed after a 2-hour room-temperature drying period, which can be accelerated to 1 hour at 170°F. When elevated-temperature drying is used, the coating should first be allowed to dry tack-free for 20 minutes at room temperature to avoid solvent entrapment. For evaluation and propotype purposes, 4900 is available in 6 oz. aerosol cans. However, for best control of uniformity and thickness, conventional spray equipment should be used. A thicker, non-sprayable version of 4900 is also available, and is especially useful for brushing on areas difficult to coat with a spray. Chomerics, 66 Rue la Boetie, 75008 Paris, France (449 M) Recording audiometer An audiometer for statutory audiometry in connection with health investigations has been developed by Briiel & Kjaer. Designated Type 1800, the recording Bekesy-type audiometer provides a pure tone signal (continuous or pulsed) at 7 different frequencies from 500 Hz to 8000 Hz and records the patient’s response automati- cally. The audiometer is basically an x - y recorder - the x- axis representing the test-frequency and the y- axis representing the hearing-threshold of the patient (in the range - 10 dB to + 90 dB HL) as registered with the help of a patient-operated handswitch. The patient is instructed to hold the button on the handswitch pressed down as long as he hears the signal in the matched earphones, thus causing the signal- level to decrease, and to release the button when he no longer hears the signal. In this way he tracks his own threshold level. Remote control facilities and lamp indication of patient response outside normal range makes the audiometer ideal for group testing. Briiel & Kjaer, 23 Linde alii , DK-2850 Naerum, Denmark (452 M) HV5 preamplifier HV 30 15 Watts into 8H HV 50 25 Watts into 8 fl HY120 60 Watts into Sfl HY200 120 Watts into Bfl HY40Q E40Watts into 4(1 power supplies The HY5 >J a mono hybrid amplifier ideally suited fo* all applications All common input functions fmag Cartridge tuner, etc I nr* catered for internally, the desired function is achieved cither by a multi way switch or direct connec- tion to the appropriate pins The interval volume ana tone circuits merely reamre connecting to external poicntiu meters (not included* The HY5 is compatible with all I Lf power amplifiers and power supplies. To ease con- struction and mounting a P C, connector is supplied with each pre-amplifier FEATURES Complete pre amplifier in single pack - Multi function equalization ;iw noise Low distortion - High overload — Two simply combined lo» stereo APPLICATIONS Hi-Fi Mixers - Disco - Gu tar and Organ - Puolic address SPECIFICATIONS INPUTS Magnetic Pick-up 3 mV. Ceiamic Pick-up 30 mV Tuner lOOmV. Microphone 10 mV, Auxiliary 3-100 mV, input impedance 47 kfl at 1 kHz. OUTPUTS, Tape 100 mV, Mam output 500 mV R M S ACTIVE TONE CONTROLS Treble 1? dB at 10 kHz 8ass r at 100 Hz DISTORTION 0 1% at 1 kHz Signal Noise Ratio 68 dB OVERLOAD 38 dB on Magnetic Pick up. SUPPLY VOLTAGE - 16 SO V Price £ 5.22 + 65 p. VAT P&P free. HY5 Mounting Board B1 48 p + 6 p VAT P&P free. The HY3U is an exciting New kit from I.L.P it features a virtually indestructible I.C. with short circuit and thermal protection The kit consists of I C . heatsink. P.C. board. 4 resistors. 6 capacitors, mounting kit. together with easy to follow construction and operating instructions. This amplifier is ideally suited to the beginner in audio who wishes to use the most up to date technology available FEATURES Complete Kit — Low Distortion — Short. Open and Thermal Protection - Easy to Build APPLICATIONS Updating audio eouipment - Guitar pract-ce amplifier - Test amplifier audio oscillator SPECIFICATIONS: OUTPUT POWER 15 W RMS mto 8 SI DISTORTION 0.1% at 15 W INPUT SENSITIVITY 500 mV FREQUENCY RESPONSE 10 Hz-16 kHz -3 dB SUPPLY VOLTAGt t 18 V Price £ 5.22 + 65 p. VAT P&P free. The HY50 leads I.L.P.'s total integration approach to power amplifier design The amplifier features an integral heatsink together with the simplicity of no external components During the past three years the amplifier has been refined to the extent that it must be one of the most reliable and robust High F idelily modules in the World FEATURES Low Distortion Integral Heatsink Only five connections 7 Amp output transistors - No exter nal components APPLICATIONS Medium Power Hi-Fi systems - Low power disco - Guitar amplifiei SPECIFICATIONS INPUT SENSITIVITY 500 mV OUTPUT POWER 25 W R M S into 8 U. LOAD IMPEDANCE 4 16 11 DISTORTION 0 04?i at 25 W at 1 kHz SIGNAL.'NOISE RATIO 7b dB FREQUENCY RESPONSE 10 Hz-45 kHz -3 dB. SUPPLY VOLTAGE t 25 V SIZE 105x50x25 mm. Price £ 6.82 + 85 p. VAT P&P free. The HY120 is the baby of I.L.P.'s new high power range, designed to meet the most exacting requirements including load line and thermal protection this amplifier sets a new standard in modular design. FEATURES Very low distortion - Integral heatsink - Load line protection - Thermal protection - Five connec non — No external components. APPLICATIONS Hi-Fi - High quality disco Public address - Monitor amplifier Guitar and organ SPECIFICATIONS INPUT SENSITIVITY 500 mV. „ __ ^ OUTPUT POWER 60 W R.M.S. into 8 H. LOAD IMPEDANCE 4-16 0. DISTORTION 0 04 '. at 60 W at 1 kHz. SIGNAL'NOISE RATIO 90 dB FREQUENCY RESPONSE 10 Hz-45 kHz 3dB. SUPPLY VOLTAGE • 35 V. SIZE 1 14x50x85 mm. Price £ 15.84 + £1.27 VAT P&P free. The HY200 now improved to give an output of 120 Watts has been designed to stand the most rugged conditions such as disco or group while still retaining true Hi-Fi performance FEATURES Thermal shutdown Very low distortion Load line protection - Integral heatsink No external components APPLICATIONS Hi-Fi - Disco - Monitor Power stave Industrial Public Address SPECIFICATIONS: INPUT SENSITIVITY 500 mV OUTPUT POWER 120 W R.M.S. into 8 fl. LOAD IMPEDANCE 4 16 O DISTORTION 0 05% at 100 W at 1 kHz SIGNAL/NOISE RATIO 96 dB FREQUENCY RESPONSE 10 Hz 45 kHz 3 dB SUPPLY VOLTAGE 45 V SIZE 114x100x85 mm. Price £ 23,32 + £1.87 VAT P&P free. The HY400 s I L.P.'s '8ig Oaddy of the ronge producing 240 W mto 4 IV It has been designed tor high power disco or public address applications If the amplifier is to be used at continuous high power levels a cooling fan is iec- om mended. The amplifier includes all the qualities of the rest of the family to lead the market as a true high power hi-fidelity power module FEATURES Thermal shutdown Very low distortion Load line protection No external components APPLICATIONS Public address — Disco - Power slave - Industrial SPECIFICATIONS OUTPUT POWCff 240 W R.M.S. into 4 11 LOAO IMPEDANCE 4 16 II DISTORTION 0 1 % at 240 W at 1 kHz SIGNAL/NOISE RATIO 94 dB FREQUENCY RESPONSE 10 Hz-45 kHz 3dB SUPPLY VOLTAGE 45 V INPUT SENSITIVITY 500 mV SIZE 1 14x100x85 mm Price £ 32.17 + £ 2.57 VAT P&P free. PSU36 suitable for two HY30's £ 5.22 plus 65 p VAT P&P free PSU50 suitable for two HY50's £ 6.82 plus 85 p VAT P&P free PSU70 suitable for two HY120& £ 13-75 plus £ 1.10 VAT P&P free PSU90 suitable for one HY200 £ 12.65 plus £ 1.01 VAT P&P free PSU180 suitable for two HY?00's or one HY400 £ 23.10 plus £ 1.85 VAT P&P free. TWO YEARS' GUARANTEE ON ALL OF OUR PRODUCTS I.L.P. Electronics Ltd Crossland House Nackington, Canterbury Kent CT4 7AD Tel (0227) 63218 Please Supply Total Purchase Price I Enclose Cheque □ Postal Orders □ Money Order □ Please debit my Access account □ Barclaycard account □ Account number Name & Address Signature Our bi-monthly newsletter keeps you up to dots with Isteet guerenteed prices - our latest epeciel offers (they save you pounds) - details of new projects end new lines. Sand 30p for the next eix issues (5p discount voucher with each copy). MAPUN ELECTRONIC SUPPLIES P.O. BOX 3 RAYLEIGH ESSEX SS6 8LR Telephone: Southend (0702) 715155 Shop: 284. London Road, Westcliff -on-Sea. Essex (Closed on Mondayl Telephone: Southend (0702) 47379 IT'S A FANTASTIC BESTSELLER! 216 big (11" x 8") pages! Over a thousand illustrations! Over 30 pages of complete projects to build! Thousands and thousands of useful components described and illustrated! No wonder it's a bestseller! DON'T MISS OUT! SEND 50p NOW! POST THIS COUPON NOW FOP YOUR COPY OF OUR CATALOGUE PRICE SOp Please rush me a copy of your 216 page catalogue by return of post. I enclose 50p, but understand that if I am not completely satisfied I may return the catalogue to you within 14 days and have my 50p refunded immediately.