Fullscreen HTML5Med Res (??MB)HTML4

Big HO locomotives like this American outline unit present few problems in installation but smaller British, European and some Australian locomotives will be a real shoehorn job. Part 4: the receiver/decoder modules This month we present the receiver/decoder for the Protopower 16 Command Control system. Each locomotive on the layout needs one of these decoders and the circuit is laid out on two PC boards to enable it to be shoe-horned into the locomotive body. Design by BARRY GRIEGER As discussed in previous articles in this series, the Com­mand Control system impresses a serial data stream onto the track voltage. The serial data stream has blocks of 16 pulses, one pulse for each of the 16 locomotives which can be used on the system. These pulses have an amplitude of 5V peak-to-peak and so form a very “robust” data stream which will not be subject to interference from the commutator hash of typical model locomo­tives. The job of the receiver/decoder is to separate the particu­lar width-modulated pulse for its own locomotive from the block of 16 pulses and then turn that pulse into direction and voltage signals to drive the locomotive’s motor. Essentially, the receiver/decoder can be regarded as a speed and direction control built into each locomotive and get­ting its “throttle” settings from the serial data stream. The speed control part of the circuit will supply up to 1A to the locomotive motor at up to about 13-14V DC. To understand how the receiver/ decoder works, we need to refer to the block diagram of Fig.1 and then to the complete circuit of Fig.2. Fig.1 just shows the main circuit Run your model railway with Command 60  Silicon Chip Fig.1: this block diagram shows the major circuit functions of the receiver/decoder board. The data stream superimposed on the track voltage is demultiplexed by the up/down counter to recover the widthmodulat­ed pulse for the particular locomotive and this pulse is then fed to the servo decoder. functions. On the lefthand side at the top of the diagram you will see the track voltage being fed to a bridge rectifier. The data pulses pass through the bridge rectifier unchanged. The voltage from the bridge rectifier then goes in four separate directions. First, it feeds a 3-terminal 5V regulator (REG1) to drive the three ICs on the receiver board. Second, it is fed via diode D1 to the H-bridge circuit to drive the motor in either direction. Third and fourth, the pulses superimposed on the track voltage are fed in two directions, to the Sync Detector and a Schmitt trigger/buffer (IC1f). The Schmitt trigger/buffer squares and cleans up the signal before feeding it to the CD input of IC2, an up/ down counter which acts as a de­ multiplexer. If you remember, in the encoder described in the February 1998 issue, we had a multiplexer to insert the 16 throttle settings into the pulse waveform. Now, in the decoder circuit, we need the opposite form; a “demultiplexer” to get the throttle information out of the pulse signal. The sync detector, for its part, finds the sync “gap” bet­ween blocks of 16 pulses and feeds the detected sync pulse to the “load” input of IC2. By a mysterious process which we’ll describe later, the up/down counter (demultiplexer) then magically ex­tracts the wanted pulses for the particular locomotive and feeds it to IC3, the servo decoder. This servo decoder turns the width-modulated pulses into direction and speed signals which drive the H-bridge and this in turn, drives the locomotive motor forward or backward at any speed between stop and “flat chat”. Well, the broad overview is just that, a broad overview and it doesn’t really tell you how the same track voltage can provide the power for the motor and the ICs as well as the speed and direction information. To really understand the nitty-gritty of the circuit operation, we need to have a detailed look at Fig.2. Circuit description Again, you will see the bridge rectifier, BR1, on the lefthand side of the circuit and it is fed with the track vol­tage, via the wheels and current pickups of the locomotive. Remember that the track voltage is 11V DC with a 5.9V pulse signal superimposed on top, giving a total track voltage of about 16.9V peak. The track voltage passes through the bridge rectifier virtually unchanged, apart from the voltage losses in the bridge diodes of about 1.3V. So after the bridge rectifier we have about 10V DC with a 5.6V pulse signal still superimposed on top. This “composite” track voltage is then fed via diode D1 Control May 1998  61 Fig.2: three ICs perform the crucial functions to drive the locomotive motor with a pulse-width modulated (PWM) signal via the H-bridge transistors. These also provide forward and reverse operation. to the H-bridge circuit which drives the motor. There is a small amount of filtering provided by capacitor C13 but it is mainly there to remove commutator hash from the motor. At the same time, track voltage from BR1 is fed directly to REG1, the 78L05 3-terminal regulator, to provide 5V DC to power the three ICs. The fact that the 5.6V pulses are riding on top of the DC input to the regulator makes little difference to its performance. Counting the pulses As well as drawing DC power from the track voltage, the receiver circuit must decode the data stream. So, following the bridge rectifier, the track voltage is fed via a 10V zener diode which effectively removes the 10V DC and just leaves the 5.6V pulses to 62  Silicon Chip be fed to a voltage divider consisting of resistors R1 & R2. From there, the signal voltage goes to the inputs of two 40106 Schmitt trigger inverters, IC1a & IC1f. IC1f squares up the pulse signal and feeds it to the CD (count down) input of IC2, the 40193 pre­settable up/ down counter which functions as the demultiplexer. IC2 has four data inputs (pins 1, 9, 10 & 15) which can be hard-wired (high or low) to set the wanted channel. Upon the application of a “load” pulse to pin 11, IC2 counts down by 16 from the preset channel so that the decoded output is present at the Borrow terminal (pin 13). Now before we go too far, we’ll clear up a possible area of confusion. We have said that the Command Control system uses a serial data block of 16 pulses and so it does. But the 40193 is a binary counter so it counts up from 0 to 15 or down, from 15 to 0. So while we might be talking about the overall system having 16 channels, IC2 actually counts down from a count of 15 to as far as zero, if channel 1 is required for the particular receiver/ decoder. If we’re talking about a locomotive on channel 4 or the 4th pulse in the data stream, we preload the counter using the four data inputs so that IC2 gives an output when the 4th data pulse is reached. What actually happens is that IC2 counts down until it reaches a count of 4, whereupon the “Borrow” output at pin 13 goes low. It goes high again as soon as the input at pin 4 goes high. Hence the output pulse at pin 13 lasts as long as the relevant 4th pulse in the data stream fed to pin 4 and so we have recovered the wanted data pulse and it is inverted by IC1d before being fed to IC3, the servo decoder. Fig.3: these waveforms show how IC2 recovers the correct width modulated pulse from the data stream. The top trace shows the data signal fed to pin 4 of IC2. Below that, the wide negative-going pulse is the “load” signal fed to pin 11 of IC2. The bottom trace is the output of IC1d, at pin 8. Note that the narrow positive-going pulse of the bottom trace is an inverted version of the wanted 4th pulse in the data stream on the top trace. Fig.5: these waveforms show the operation of the servo decoder, IC3. The top trace shows the input pulse for forward motion. The middle trace shows pin 5 pulsing low at the same rate as the input pulse while the bottom trace, pin 9, stays high. We can see this sequence of events in the waveforms of Fig.3. The top trace shows the signal fed to pin 4 of IC2. Below that, the wide negative-going pulse is the “load” signal fed to pin 11 of IC2. The bottom trace is the output of IC1d, at pin 8. Note that the Fig.4: these waveforms show the operation of the sync detector or “sync stripper”. The top trace is the inverted data stream at pin 2 of IC1a. The middle trace is the integrated pulse waveform at pin 5 of IC1c with its series of little “teeth” followed by a big tooth. The bottom trace at pin 6 of IC1c shows how the little teeth have been completely erased while the big tooth becomes a wide negative-going pulse, somewhat narrower than the big sync pause in the top waveform but still wide enough for our purpose. Fig.6: these waveforms show IC3’s operation for reverse motion. The top trace is the input (note its greater width than in Fig.5). Pin 5 (middle trace) now stays high while pin 9 (bottom trace) pulses low. narrow positive-going pulse of the bottom trace is an inverted version of the 4th pulse in the data stream on the top trace. Sync pulse detection Before we can look at how IC3 works, we need to understand how the sync or “load” pulse fed to pin 11 is obtained from the pulse stream. This is achieved by inverter IC1a, diode D2, R3 & C3, together with inverter IC1c. We noted previously that the track May 1998  63 Fig.7 (above): this is the STOP condition for the receiver/ decoder. The input pulse (top trace) is close to the nominal neutral condition at 244µs wide so that both pin 5 (middle trace) and pin 9 (bottom trace) stay high and keep all the H-bridge transistors turned off. Fig.8 (right) shows some of the waveforms across the motor when it is driven signal is fed via zener diode ZD1 to a voltage divider consisting of resistors R1 & R2 and then to Schmitt trigger inverters IC1a & IC1f. IC1f does precisely the same job as IC1a but it then drives a network consisting of diode D2, resistor R3 and capacitor C3. R3 and C3 can be regarded as a pulse integrator, with R3 feeding a slight positive charge to C3 for each pulse on the data line but C3 is then discharged by diode D2 as each pulse drops to zero. However, when the much longer positive sync pulse arrives from IC1a, capacitor C3 is able to charge to a much higher vol­ tage before being discharged by diode D2. The result is a wave­form with 15 little “teeth” followed by a big “tooth” represented by the integrated sync pulse. This waveform is fed to Schmitt forwards. The top trace is the output pulse at pin 5 of IC3 and the middle trace is the waveform at the commoned collectors of Q5 & Q7. The bottom trace is the waveform on the other side of the motor, at the commoned collectors of Q4 & Q8. Note that the middle trace shows the remnant pulses which are superimposed on the track voltage. trigger IC1c which ignores the little teeth and squares up the big tooth to form the recon­stituted sync pulse which becomes the “load” signal for counter IC2. The waveforms of Fig.4 show the above process in action. The top trace is the inverted data stream at pin 2 of IC1a. The middle trace is the integrated pulse waveform at pin 5 of IC1c with its series of little “teeth” followed by a big tooth. The bottom trace at pin 6 of IC1c shows how the little teeth have been completely erased while the big tooth becomes a wide nega­tive-going pulse, somewhat narrower than the big sync pause in the top waveform but still wide enough for our purpose. The above process is sometimes referred to as “sync strip­ping” whereby the small pulses are “stripped out” of the wave­form, leaving just the sync pulse. The waveforms of Fig.4 give a graphic illustration of this process. Servo decoder So now we have the actual pulse information for the locomo­tive, it needs to be turned into speed and direction Capacitor Codes ❏ ❏ ❏ ❏ ❏ ❏ Value IEC Code EIA Code 0.1µF 100n  104 .015µF  15n  153 .01µF  10n  103 .0047µF  4n7  472 .001µF  1n0  102 Resistor Colour Codes ❏ No. ❏  1 ❏  1 ❏  1 ❏  1 ❏  1 ❏  7 ❏  2 ❏  2 64  Silicon Chip Value 1MΩ 100kΩ 68kΩ 3.3kΩ 2.2kΩ 1kΩ 620Ω 470Ω 4-Band Code (1%) brown black green brown brown black yellow brown blue grey orange brown orange orange red brown red red red brown brown black red brown blue red brown brown yellow violet brown brown 5-Band Code (1%) brown black black yellow brown brown black black orange brown blue grey black red brown orange orange black brown brown red red black brown brown brown black black brown brown blue red black black brown yellow violet black black brown This is the finished receiver/decoder board, shown here larger than actual size. Note the way in which all the resistors are mounted end-on. Some of the resistor pigtails then become convenient test points in case you have to troubleshoot the board. The H-bridge board has the four Darlington output transistors laid flat and stacked to minimise height. Fig.10: the artwork for the two PC boards, shown twice full size. Fig.9: the wiring details for the receiver/decoder and H-bridge PC boards. No not forget to install the links under IC1 & IC2 before soldering these chips in and take care to ensure that all polarised parts are correctly oriented. Note that the PC boards are shown twice actual size, for the sake of clarity. Fig.11: the four Darlington output transistors are laid flat and stacked on the H-bridge board to reduce the height. May 1998  65 The receiver/decoder and H-bridge boards are a neat fit inside the shell of this locomotive. Each loco will need the boards installed in a particular way to fit everything in. It is most important to make sure that there are no shorts to the motor or locomotive chassis. signals to drive the motor. This job is done by IC3, the ZN409CE servo decoder. There are no servos in this circuit but the ZN409CE was originally designed to drive the servo motors used in radio-controlled aircraft, cars, boats and so on. For those not familiar with how a servo drive circuit works, you can find a full description in Bob Young’s “Radio Control” column in the November 1997 issue of SILICON CHIP. You can also refer to a servo circuit employing the ZN409CE in the “Circuit Notebook” pages of the December 1997 issue. Now when the ZN409CE is used to drive a servo it compares the incoming pulse at pin 14 with an internally-generated pulse which is varied by a potentiometer driven by the servo motor. The potentiometer’s wiper is connected to pin 3. When the two pulses match, the servo comes to a stop in the desired position. Where our circuit varies is that the potentiometer (VR1) is not driven but is set to match the internally generated reference pulse to the incoming pulse when the throttle setting is for STOP. This corresponds to a nominal pulse 66  Silicon Chip width of 244µs which corresponds in turn to the crystal-derived pulse frequency of 2048Hz. Thus, when the input pulse is narrower than the reference pulse of 244µs, the locomotive motor is driven forward; when it is wider than 244µs, the locomotive motor is driven in reverse. Motor drive Pins 5 & 9 of IC3 are the outputs and these drive transis­tors Q2 & Q1 which provide level shifting and signal inversion to Q3 & Q6. In turn, Q3 & Q6 drive the H-bridge transistors Q4, Q5, Q7 & Q8. To drive the locomotive in the forward direction, pin 5 of IC3 pulses low at the same rate as the pulse train at pin 14 (the input) while pin 9 stays high. Tracing that through, this means that Q1, Q6, Q7 & Q8 stay off while Q2, Q3 & Q5 are pulsed on. Q4 is also turned on, by dint of the pulse signal from the collector of Q3 but Q4 is turned full on because of the 1µF filter capaci­tor at its base – see the waveforms of Fig.8. To drive the locomotive in the reverse direction, pin 5 of IC3 stays high while pin 9 pulses low. This means that Q2, Q3, Q4 & Q5 are turned off while Q1, Q6 & Q8 are pulsed on. Q7 turns on fully because of the 1µF filter capacitor at its base – see the waveforms of Fig.8. Finally, the waveforms of Fig.7 show the STOP condition. Here the top trace is the input pulse to pin 14 of IC3 and the other two traces are the outputs at pins 5 & 9. Both are high, leading to the condition where all the transistors in the H-bridge are off. Fig.8 shows some representative waveforms across the motor when it is being driven forwards. The top trace is the output pulse at pin 5 of IC3 and the middle trace is the waveform at the commoned collectors of Q5 & Q7. The bottom trace is the waveform on the other side of the motor, at the commoned collectors of Q4 & Q8. Note that the middle trace shows the remnant pulses which are superimposed on the track voltage. By the way, we have referred to pin 5 pulsing when the motor is going forward and pin 9 pulsing when the motor runs in reverse. At the same time, whenever the motor is being driven forward, Q9 and Q10 turn on to drive the locomotive’s headlight. No doubt some enterprising modellers will want to extend the headlight drive to drive the headlights and tail lights of diesel locomotives to cater for both directions. For the time being at least, this is beyond the scope of this article. PC board assembly Two PC boards are used to accommodate the receiver/decoder circuitry. The main board measures 53 x 30mm and is coded 09105981 while the smaller board for the H-bridge transistors measures 25 x 26mm and is coded 09105982. The main board is quite crowded and you will need to solder it carefully to avoid solder splashes shorting out adjacent conductors. Before installing any components on either board, check the copper patterns carefully for any open circuit tracks, bridges or undrilled holes. Fig.9 shows the component layout for both PC boards and the interconnecting wiring between them. Before soldering any components in, install the short links under IC1 & IC2. Then insert all the resistors which are in­ stalled “end on” to conserve space. The diagram of Fig.9 actually does show how the bodies of the resistors are oriented. For example, the body of resistor R10, from pin 5 of IC3, is nestled up to transistors Q2 and Q9. It is important to orient the resistor bodies in the same way as depicted on Fig.9 because the accessible resistor pigtails then become test points if you have to troubleshoot the receiv­er/ decoder. Hopefully, you won’t have to do any troubleshooting but if it comes to the crunch, it’s nice to have those test points accessible. Make sure that you check the value of each resistor as it is installed. Use your multimeter to physically check each value because it is almost impossible to check resistor colour codes once the resistors are all installed and obscured by other com­ponents. Next, install the zener diode, bridge rectifier, the two diodes and the capacitors. Note that all the polarised components must go in the right way otherwise the circuit won’t work or it may be damaged. All the electrolytic capacitors on the PC boards are tantalum types, specified because of their small size. C6, the .018µF capacitor connected to pins 1 & 2 of IC3, must be an NPO ceramic type. If you can’t obtain .018µF, you can use a value of .015µF but it still must be NPO. Do not substitute other capacitor types here, such as Fig.12: this diagram shows how to hook up a temporary throttle potentiometer and reversing switch to the encoder PC board (published in March 1998) so that the receiver/decoders can be tested. MKT polyesters, because their temperature coefficient is just not good enough. Next, insert the 78L05 regulator and the transistors. Finally, the three ICs may be inserted and soldered. Do not use sockets as there is not enough room on the board. Finally, there are two long insulated links to be installed, one on top and the other on the underside of the board. H-bridge board The H-bridge board has only a few components on it but there is a preferred order of assembly. First, insert May 1998  67 Another American HO locomotive installation. The receiver/decoder is at one end while the H-bridge board is at the other. Note that these are early prototype boards and differ from those shown in Fig.9. the end-on resistors, followed by the two tantalum capacitors and the three small-signal transistors. Then mount the four power transistors, Q4, Q5, Q7 & Q8. Mount Q4 & Q5 first. You will need to bend their leads at rightangles, close to their bodies. They should both sit flush with the PC board, with their metal mounting surfaces facing down. This done, bend the leads of Q7 & Q8 at rightangles in the same way and mount them so that they sit flush on top of Q5 & Q4, respectively. Q7 & Q8’s metal mounting surface should face up, as shown in the photos. Finally, mount the 2.2µF electrolytic capacitor. Testing To test the boards you will need to temporarily intercon­nect them with short lengths of hookup wire and you will need to program the receiver board so that it can be addressed by the Command/Power Station, described in the February and March 1998 issues. The programming involves tying four pins on IC2, the 40193 programmable up/down counter. Table 1 shows how the pins are tied high (H) or low (L) and we are using channel designations 1 to 16 rather than the coun­ter’s binary sequence of zero to 15 (as noted previously). If you are doing a batch of these receiver/decoder boards, you will need to make sure that each one is programmed with its own code. Most importantly, you need to label the board with its code as soon as it has been done otherwise you will become very confused later on. So either use a 68  Silicon Chip pencil to write the channel number on one of the ICs or use a little stick-on label to accom­plish the same thing. Once you have programmed the board, you need to hook it up to a locomotive motor. We strongly suggest that you do not in­stall the receiver/ decoder into a locomotive before it has been tested. That would be asking for trouble. Use a spare locomo­tive motor if you have one or any small permanent magnet motor which draws a few hundred milliamps or so. The motor should have a 0.1µF capacitor connected across it to suppress commutator hash, as shown in the circuit of Fig.2 and the wiring diagram of Fig.9. You also need to wire up a temporary “headlight” so that you know Table 1: Program Pins On IC2 C h. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Pin 9 Pin 10 Pin 1 Pin 15 L L L L L L L L H H H H H H H H L L L L H H H H L L L L H H H H L L H H L L H H L L H H L L H H L H L H L H L H L H L H L H L H which direction the motor is turning. We used a red LED in series with an 8.2kΩ resistor to simulate a headlight load for Q10. Using a temporary throttle Since we have not yet described the handheld throttles (that comes next month) it is now necessary to hook up a tempo­rary throttle to the Command/ Power Station, so that it can drive any of the receiver/decoder channels. Fig.12 shows how this is done. You will need a 10kΩ potentiometer and an SPDT switch, wired as shown to the encoder board. The wiper (centre tag) of the potentiometer is connected to the appropriate pin of the 16-pin header socket. We did this by wiring the pot wiper to a length of wire with a cutoff pin from a defunct IC. The pin can then be inserted into any pin socket on the 16-pin header. Table 2 shows the channel numbers and their respective pin numbers on the socket. Power station changes By the way, there are a couple of changes to be made to the Power Station wiring, as noted in the Errata at the end of this article. Having made the appropriate throttle connection and having programmed the receiver/decoder board and connected it across the track outputs of the Command/Power Station you are ready to proceed. Rotate the throttle potentiometer to its minimum setting and set the Forward/Reverse switch to forward. Turn on the Power Station. The motor may buzz or rotate. Don’t worry about that for a moment, just measure the voltage from the 3-terminal regu- lator. It should be close to +5V. This is most convenCh. Pin No. iently measured 1 6 across the three 2 4 ICs: between pins 3 2 7 & 14 of IC1, pins 4 8 8 & 16 of IC2 and 5 1 between pins 6 & 6 7 10 of IC3. 7 3 You should also be able to measure 8 5 about +2.2V be9 11 tween pin 6 (0V) 10 13 and pin 2 of IC3. 11 15 If the motor is ro12 9 tating or buzzing, 13 16 adjust trimpot VR1 14 10 until it stops. Then 15 14 flick the Forward/ 16 12 Reverse switch to reverse and check that the motor is still stationary. Now rotate the throttle potentiometer clockwise and the motor should start running and speed up as you rotate the pot further clockwise. Rotate the throttle pot fully anticlockwise and the motor should come to a complete stop. If it doesn’t, you may need to tweak VR1 again. Now flick the Forward/Reverse switch to forward and rotate the throttle pot clockwise. The motor should now run in the opposite direction to the reverse condition and the headlight LED should come on. Table 2 Troubleshooting What if it doesn’t work as it should? Then you have to put on your thinking cap and figure out why. First, check that the programming for IC2 matches the channel you have selected on the 16-pin header on the decoder board. Second, check that the outputs of IC3, at pins 5 & 9, are working as they should. For example, when forward motion is selected, pin 9 should high, (ie, close to +5V) while pin 5 should be pulsing low. If you don’t have an oscilloscope, you can measure the DC voltage at pin 5. As you advance the throttle, the voltage at pin 5 should gradually reduce. We had a fault with one of our receiver/decoders which demonstrates how easily a typical fault can occur. Regardless of which way the Forward/ Reverse switch was set, the motor always ran in the one direction while the headlight LED did come on correctly for the forward setting. When we checked pins 5 and 9 they performed as they should but the motor steadfastly ran in the same direction anyway. We then checked the voltage at the collectors of transistors Q1 and Q2. The collector of Q1 should be low when pin 9 is high and vice versa. Similarly, the collector of Q2 should be low when pin 5 is low and vice versa. The fault turned out to be a small sliver of solder between the base and emitter of Q1. With a small, tightly packed PC board like this, you need a good magnifying glass and good light to find faults like this. Installing the boards The most important aspect of installing the receiver/decod­er boards in the locomotive is that you must ensure that there are no shorts. The existing locomotive wiring must be removed so that the wheel wipers no longer connect to either side of the locomotive motor or to the locomotive chassis. This is doubly important for locomotives with metal shells. The second most important aspect of installation is to make sure that no part of the receiver/decoder circuit, including the motor itself, is shorted to the locomotive shell, any of the wheel pickups or anything else. In many, if not most, locomotives, you will need to sepa­rate the receiver/ decoder and H-bridge boards to fit them in. For example, the H-bridge board might mount at one end while the receiver/decoder mounts at the other end. It may be wise to sleeve the boards with heatshrink tubing to make installation easier and less subject to shorts. When each locomotive installation is complete, you will need to hook it up to the Command/Power Station again to ensure that it all works as it should. Be sure to label the underside of the locomotive with its channel number. Next month, we will continue with the wiring of the throt­tles and control panel. Errata Command Control Power Station, March 1998: a change should be made to the circuit of page 55 and the component overlay diagram on page 56. R4 should be changed to 2.2kΩ. R5 on page 56 should be 1.5kΩ to agree with SC the circuit on page 55. Parts List For Receiver/Decoder (one required for each locomotive) 1 PC board, 53 x 30mm, code 09105981 1 PC board, 25 x 26mm, code 09105982 1 1kΩ miniature sealed top adjust trimpot (VR1) Semiconductors 1 40106, 74C14 hex Schmitt trigger (IC1) 1 40193 programmable up/down counter (IC2) 1 ZN409CE servo decoder (IC3) 1 78L05 3-terminal 5V regulator (REG1) 3 PN100 NPN transistors (Q3, Q6,Q10) 3 PN200 PNP transistors (Q1, Q2,Q9) 2 BD681 NPN Darlington transistors (Q5,Q8) 2 BD682 PNP Darlington transistors (Q4,Q7) 1 WO4 bridge rectifier (BR1) 1 1N4936 fast recovery diode (D1) 1 1N4148 small signal diode (D2) 1 10V 400mW or 1W zener diode (ZD1) 1 red LED (for temporary headlight) Capacitors 1 2.2µF 63VW PC electrolytic 6 1µF 25VW or 35VW tantalum electrolytic 1 0.33µF 25VW or 35VW tantalum electrolytic 1 0.1µF MKT polyester or ceramic (across motor terminals) 1 .015µF or .018µF 100VW NP0 ceramic 2 .01µF MKT polyester 1 .0047µF MKT polyester 1 .001µF MKT polyester Resistors (0.25W, 1%) 1 1MΩ 1 2.2kΩ 1 100kΩ 7 1kΩ 1 68kΩ 2 620Ω 1 3.3kΩ 2 470Ω Miscellaneous Heatshrink tubing, tinned copper wire, hookup wire, solder May 1998  69