Momentary Action with a Wireless Switch

Many different types of wireless switch modules with a relay for switching AC power loads are commercially available. However, some applications require a short On or Off pulse, such as is provided by a momentary-action (pushbutton) switch. Here we describe a solution that simulates a pushbutton switch with a standard wireless switch. A supplementary circuit converts the switch module into a remotely controllable momentary action switch.

In the supplementary circuit, S1 is the switching contact of the relay in the wireless switch module. This contact energises a 24-V power supply connected directly to the AC power outlet, consisting of a bridge rectifier (D1–D4) with a series resistor (R1), a series capacitor (C1), and a charging capacitor (C2). The two Zener diodes in the bridge rectifier (D1 and D2) limit the DC voltage on C2 to approximately 24 V.1

Momentary Action with a Wireless Switch Schematic

When the wireless switch module closes con-tact S1, 24 VDC is applied to the coil of relay RE1, which closes. At the same time, capacitor C3 charges via D5. When the contact of RE1 switches, capacitor C4 provides the charging current for C3. The charging current flows through the coil of RE2, which remains actuated as long as the current is sufficiently large. The current decreases as the voltage on C4 rises, with the result that RE2 drops out and the contact of RE2 (the ‘momentary’ contact) opens again.

S1 opens when the relay in the wireless switch module is de-energised, which causes RE1 to drop out shortly afterward and connect capacitor C4 to ground. The capacitor discharges through the coil of RE2, causing its ‘momentary’ contact to be actuated again. The timing diagram shows the switch-on and switch-off sequences of the wireless switch (S1 contact).

The duration of the ‘button press’ (engagement time of RE2) depends on the capacitance of C3 and C4. The equation Q = C × U = I × t can be used to calculate suitable capacitor values for a specific hold time (t1in the timing diagram) with a given relay current. The value shown in the circuit diagram (1000 µF) corresponds to a hold time of 1 second with a relay current (holding current IH) of 10 mA:

C = IH× t1/ U = (0.01 A) × (1 s) / 10 V = 1000 µF.

A reed relay cannot be used for RE2 because the voltage across the coil reverses. This also means that a free-wheeling diode can-not be used, but it is anyhow not necessary due to the slow discharge of C4. RE2 should be a 'Class II' relay (such as the Omron G6D-1A-ASI 24DC) to provide adequate insulation of the switch contact. RE1 does not have to be a Class II relay. Due to the presence of AC power line voltage, R1 and R2 must have a rated working voltage of 250 V (150 V ), although they can also be formed from two resistors with half this rated working voltage connected in series, each with half of the specified power rating. In this case, R1 consists of two 47 Ω / 1 W resistors and R2 of two 100 kΩ / 0.25 W resistors. Readers on 120 VAC 60 Hz power networks should change C1 into 680 nF.

The circuit can be fitted in a plastic enclosure with an integrated AC power plug, which can easily be plugged into the wireless switch module. The contact of RE2 can then be fed out to a terminal strip as a floating contact. For adequate AC isolation, a safety clearance of at least 6 mm (air and creepage paths) to other conductors must be maintained, in addition to using a Class II relay.

Read more

OBD Vehicle Protection

Vehicle immobilisers are fitted as standard to modern cars and heavy goods vehicles. Anti-theft mechanisms have become more sophisticated but so have the methods employed by crooks. Nowadays once the thief has gained access to a vehicle they will most likely use an electronic deactivation tool which seeks to disable the immobiliser, once this has been accomplished a blank transponder key/card can be used to start the engine. In many cases communication with the immobiliser is made using the OBD-II diagnostic connector.

Although the OBD-II protocol itself does not support the immobiliser, the vehicle manufacturer is free to use the interface as neces-sary for communication, either the standard OBD-II signals or unused pins in the OBD-II connector (i.e. those undefined in the OBD-II standard). Using one of these pathways the immobiliser can usually be electronically disabled.

OBD Vehicle Protection Circuit Diagram

This may be unsettling news for owners of expensive vehicles but when professional car-thieves call, armed with the latest OBD-II hacking equipment this simple low-cost low-tech solution may be all that you need. The idea is ver y simple: if all connections to the OBD-II connector are disconnected there is no possibility for any equipment, no matter how sophisticated to gain access via the vehicle’s wiring.

The OBD-II connector is usually locate d underneath the dashboard on the passenger side; once its wiring loom has been identified a switch can be inserted in line with the wires. The switch should be hidden away some-where that is not obvious. In normal opera-tion you will be protected if the vehicle is run with the wires to the socket disconnected. Make sure however that you throw the switch reconnecting the socket before you next take the vehicle along to a garage for servicing or fault diagnosis.

The diagram shows the ISO K and ISO L wires switched. To cover all bases it is wise for every wire to the socket is made switchable except the two earth connections on pins 4 and 5 and the supply voltage on pin 16. Almost ever y vehicle manufacturer has their own method of vehicle immobilisation, by disconnecting every wire it ensures that no communication is possible (even over the CAN bus). Now the innermost workings of your vehicle will be safe from prying eyes. When a hacker plugs in a deactivation tool it will power up as normal but probably report something like ‘protocol unrecognised’ when any communication with the OBD port is attempted.

Author : Florian Schäffer - Copyright: Elektor

Read more

A Doorbell for the Deaf

This circuit provides a delayed visual indication when a door bell switch is pressed. In addition, a DPDT switch can be moved from within the house which will light a lamp in the door bell switch. The lamp can illuminate the words "Please Wait" for anyone with walking difficulties.

A Doorbell for the Deaf Circuit Diagram :

Notes :

The circuit uses standard 2 wire doorbell cable or loudspeaker wire. In parallel with the doorbell switch, S1, is a 1N4001 diode and a 12 volt 60mA bulb. The bulb is optional, it may be useful for anyone who is slow to answer the door, all you need to do is flick a switch inside the house, and the bulb will illuminate a label saying Please Wait inside the doorbell switch or close to it. The double pole double throw switch sends the doorbell supply to the lamp, the 22 ohm resistor is there to reduce current flow, should the doorbell switch, S1 be pressed while the lamp is on. The resistor needs to be rated 10 watts, the 0.5 Amp fuse protects against short circuits.

When S2 is in the up position (shown as brown contacts), this will illuminate the remote doorbell lamp. When down, (blue contacts) this is the normal position and will illuminate the lamp inside the house. Switch S1 will then charge the 47u capacitor and operate the transistor which lights the lamp. As a door bell switch is only pressed momentarily, then the charge on the capacitor decays slowly, resulting in the lamp being left on for several seconds. If a longer period is needed then the capacitor may be increased in value.

Read more

3D LED Pyramid

The author 'just wanted to do a bit of microcontroller programming'. However, the project rapidly grew into this impressive and visually attractive pyramid. The circuit consists essentially of a specially-sawn printed circuit board,  23 LEDs and a microcontroller. Despite the fact that the microcontroller  is a rather modest Atmel ATtiny2313, the author nevertheless has found room in the 2 KB flash memory for 16 different light sequences.

The 23 LEDs are divided into three groups. The lower and middle sections consist of eight LEDs, while the upper section  has just seven. The microcontroller has only 20 pins, and so it is not feasible to provide a direct individual drive for each LED. The multiplexing approach adopted uses just eleven output port pins. Buffer transistors are used to increase the  current drive capability of each output.

The software was written in assembler and can, as usual, be downloaded from the Elektor web pages accompanying this  article [1]  as either source code or as a hex file. The printed circuit board layout files are also available from the same  place, as well as a link allowing purchase of ready-made boards and pre-programmed microcontrollers.

Populating the printed circuit board is straightforward: there are some surface-mount components to be soldered,  but space is not tight. For best results,  it is best to choose LEDs with the widest possible viewing angle so that the pyramid  looks its best even when seen from the side. The author used type LO 1296 orange LEDs from Osram, which have a viewing angle of 160 '. A six- way connector is provided to allow in system programming  (l5P) of the microcontroller. The  configuration fuses are set to enable use of the internal4  MHz  clock source, which is divided down to 0.5 MHz by an  internal divider.  lf the fuses are not correctly programmed the light sequences will run too quickly, too slowly, or even not at all!

When everything is working, take an 11 cm length and  a 5.5 cm  length of 1.5 mmz solid copperwire and solder one end  of the shorter piece to the middle of the longer piece to make a 'T' shape. Pullthe  printed  circuit board spiral apart  so  that the T-shaped wire assembly fits underneath, and then solder it to the two pads as shown in the photograph. Fine-bore brass tubing can also be used instead of solid copperwire.

As well as the ISP connector a USB interface is provided, whose job is solely to provide a 5 V supply. An external 5 V mains adaptor would do the job equally well. Two jumpers affect the behaviour of the light pyramid: JP1 deter-mines  whether the sixteen sequences follow one another in strict order or at random;  and JP2 determines whether the light  patterns are displayed orwhether all LEDs will be continuously lit. S1 is a reset button, which will come in handy if you  wish to experiment with modifying the software.

Author : Lothar Goede - Copyright : Elektor

Read more

Simple Car Alarm Sound Booster

For car alarms, emphasis should be put on hearing the audible alert and identifying it as belonging to your 'wheels'. Unfortunately, modern car alarm systems seem to have more or less the same alarm sound especially if they are from the same brand. Also, to comply with legal noise restrictions, the alarm sound is not always loud enough to be heard if the car is parked down the road.

Circuit diagram :

Simple Car Alarm Sound Booster Circuit Diagram

The circuit shown here is designed to help boost the alarm sound by also activating the car's horn(s) when the alarm goes off. lnternally the car alarm system often provides a signal that activates the (optional) engine immobilizer and/or volume (ultrasound) sensors. This signal usually goes Low upon sys-tem triggering and high again when the alarm system is deactivated.

The alarm activation signal is fed to the circuit through Dl . When in idle state, T1 's gate is High and consequently the FET conducts,  keeping  power  FET T2 firmly switched off. When the  system gets an  active  low signal, T1  switches  off allowing  timing  capacitor C2 to charge  via  R2. About 15 seconds later, when the voltage across C2 is high enough, T2 starts to conduct and relay RE1 is energized. This, in turn, provides the required path for the 'lights flashing' signal to energize RE2 and feed battery power to the car's horn(s).

When the alarm system is turned off the activation signal returns to High. T1 starts to conduct and rapidly discharges C2 via R3. T2 is then cut off and REl is de-energized. Diode D2 suppresses back EMF from REl.

The circuit draws less than 2 mA when idling. When activated the circuit's current consumption is virtually that of the RE1 coil. RE1 is any simple SPST or SPDT relay, capable of switching  about 0.5 A (at 12 V). The coil rating is for 12 VDC and a current requirement as low as you can find. Fuse F1 should be a slow blow type and rated about twice RE1's coil current.

The B5.170 in position T2 can sink a continuous current of about 0.5  A. However, a value of 1.2 A pulsed is specified by Fairchild  for their devices. To keep the FET's d-s current due to C2 discharging within safe limits, R2 may be increased, C2 decreased and R3 increased, all proportionally. A factor of 2 will keep the FET out of harm's way with maybe a slight change in the 15-second delay and the sensitivity of the circuit. C1 is used as a smoothing capacitor and F2 should be rated in accordance with the horn(s) maximum current draw.

Caution. The installation and use of this circuit may be subject to legal restrictions in your country, state or area.

Author : Hagay Ben-Elie - Copyright : Elektor

Read more

Simple Wire Link Bender

When you want to mount components on a PCB or a piece of prototyping board, you not only want to do this quickly, but also tidily. The bending of really tidy wire links with the correct pitch is often a tedious chore. The following is a handy aid for doing this.

Using a small piece of 0.1 inch (2.54 mm) prototyping board, you can very easily make a handy bending jig for wire links. With a jigsaw, cut the piece of prototyping board into a staircase shape as shown in the drawing. You can make it as big as you need. Make sure that the horizontal cuts are slightly  towards  the outside with respect to the holes, so that clear indentations remain in the horizontal sections.

Bending a wire link is now very easy: choose the desired pitch on the jig (dashed line), take a piece of wire and fold it sharply around the indentations corresponding to the selected  pitch. A neat wire  link  is the result, with exactly the right pitch and ready for soldering tightly into the PCB or prototyping board. With close-fitting wire links the board looks much better and they are also mounted much
more quickly.

Tof course he bender is also suitable for resistors with leads.

Author :Louter van der Kolk - Copyright : Elektor

Read more

Wireless and Wired Emergency Stop System

This circuit allows a cheap or discarded wireless doorbell set (i.e. transmitter and receiver unit) to be used as a remote emergency stop on a high-power electrical motor or motor controller system.

Circuit diagram :

Wireless and Wired Emergency Stop System Circuit Diagram

When the button on the wireless doorbell unit is pressed, the resulting 0 V signal from the receiver unit (‘motor E-Stop’) causes PNP  transistor T1 to be turned on. Via  transistor T2, latching relay Re1 then changes state. The same is achieved when the wired Motor E-Stop button, S1, is pressed. The reset button, S2, must be pressed to reverse the state of the latching relay.

The choice of T1 and T2 is not critical they are general purpose, low voltage PNP and NPN switching  transistors  respectively,  for  which  many equivalents exist.

As an EMC precaution, small capacitors (100 pF) are fitted across base resistors R1 and R2, preventing the motor  from being shut down by external  electrical noise and interference. The  set and reset coils of the latching relay  each have a flyback diode to prevent  back emf peaks damaging T1 and T2. The contacts of the latching relay can  be used to switch a more powerful  relay, or a motor driver.

Author : Jacquelin K. Stroble  – Copyright : Elektor

Read more

Micropower Crystal Oscillator

Crystal oscillators for digital circuits are normally built as Pierce oscillators with an inverter.The inverter operates as a linear amplifier and thus requires extra current. But you can also build a crystal oscillator using an  operational amplifier (op amp for short)! If a  very low frequency is involved, for instance  32.768 kHz (commonly used for clocks), you can get away with a comparatively ‘slow’ micro power op amp.

Circuit diagram :

Micropower Crystal Oscillator Circuit Diagram

In the sample circuit shown a widely avail-able TLC271 is used. On pin 8 we have the  opportunity to set the ‘bias mode’, with three  choices ranging between fast operation with  higher current consumption and slower operation at low current. For our clock crystal the middle setting will suit us fine. Pin 8 is there-fore connected to the voltage divider R1/R2. The current consumption of the entire circuit  is impressively modest and at 5 V this is just  56 µA! The oscillator also functions astoundingly well at 3.3 V. At the same time the cur-rent drops to a more battery-friendly 41 µA. A  prototype built in the Elektor Labs produced  the slightly higher values indicated in the circuit diagram.

The output signal delivered by this circuit has  admittedly scant similarity to a square wave.  Nevertheless some cosmetic surgery will tidy  this up, with treatment in the Schmitt trigger  following. To save current (naturally) we use  a CMOS device such as the 74HC14.

Author :Rainer Reusch - Copyright :Elektor

Read more

Time Transporter

Some microcontroller applications such as  those which log or track information often  require current time and date information  to be stored along with the collected data.  A Real Time Clock (RTC) chip such as the IC  DS1307 with battery back-up can be used to  supply the required information. This particular chip easily integrates into most designs  using the absolute minimum of external components. The process of programming the  chip in software is simple and is supported in  the majority of programming environments.  Intrinsic functions, header files and libraries are widely implemented for the device. A  quick trawl of the Internet will uncover lots of  programming examples.

Project Image :

 Time Transporter Image

 

So far so good except that the chip first needs to be programmed with the current time and date information. This information is maintained and updated (thanks to a keep-alive battery) even when the external circuitry is shut down. To carry out the programming requires connection to a keyboard and display but the additional hardware will only ever be needed for this one-off event.

Circuit diagram :

Time Transporter Circuit Diagram

The design suggested here solves  the problem by combining the IC, battery, crystal and peripheral components  onto a tiny plug-in PCB. The circuit consists  of a small square of prototyping perf board  onto which is mounted the IC, a crystal, battery, decoupling capacitor (C1) and two  (optional) pull-up resistors for the open collector outputs. An IC socket with extra longs  pins (or two modular connector strips) completes the design. The complete RTC module (see photo) is self contained and can be  plugged from one circuit to another using  its long pins without losing track of time and  date. The only requirement in the target sys-tem is space for an 8-way DIL socket, wiring  to the socket and software to read the time  information.

The essential advantage with this  approach is that hard and software  expenditure in the target system is  kept to a minimum, it will only ever  need to read the time and date information. The extra hardware and soft-ware required to set both time and  date are assigned to a separate sys-tem, maybe a dedicated breadboard  design.

Once programmed the ticking clock  module can then simply be transferred to the target system.

 

Author :Jochen Brüning - Copyright : Elektor

Read more

Build Your Own Rotary Encoder

The cheaper variety of rotary encoders, including those from Bourns, are mechanical devices rarely capable of generating more than 25 pulses per revolution (ppr). If more ppr are desired, an optical encoder is usually a better alternative. Devices exist in this class with up to 256 ppr but then the price is well beyond the reach of the hobbyist. A mechanical (pulley/string) transmission to increase the ppr of mechanical encoders is possible in theory but at the cost of an awkward amount of torque. Also, the simplest solution (apparently) of turning the mechanical encoder faster than usual is not viable as it will stress the device beyond its limits. Another alternative is to turn a small stepper motor into an encoder. After all, a stepper motor has permanent magnets inducing voltages in the rotor coils. Without going into too much detail, a stepper motor requires two signals with a phase difference of 90 degrees. The voltages generated per coil can then be said to represent a ‘Gray code’, that is, two voltages 90 degrees out of phase.

Circuit diagram :

Build Your Own Rotary Encoder Circuit Diagram

Smaller motors salvaged by the dozens from old printers and flatbed scanners are particularly suited to our purpose as they usually turn smoothly and have a small cogwheel attached allowing a larger wheel to be driven. A 1:10 transmission for example easily results in a rotary encoder with 150 or so ppr, which may be very suitable for tuning a receiver in 100 Hz steps. Some printers and flatbed scanners have stepper motors with 1- or 2-wheel gear reductions on the spindle. The motor used by the author gave an effective reduction of 1:13 using two wheels. A 6-mm spindle was provisionally mounted on the second cog-wheel, and turning the spindle resulted in 180 pulses per revolution.

In this circuit, voltages supplied by the coils in the stepper motors are converted into square wave signals having TTL levels. As with a ‘real’ Bourns encoder, Gray encoded signals are output at 90 degrees phase difference. The two opamps inside the TL072 case are configured as comparators. Thanks to their high gain, even small voltages are reliably processed, enabling your logic to respond when the spindle is turned slowly.

The additional hysteresis created with R1 and R2 is required in view of the ‘output’ signals typically supplied by the stepper motor. This simple circuit is the poor man’s equivalent of a very reliable, high resolution rotary encoder and may also be used to decode speed and direction of fast turning spindles on, for example, electric motors. Mechanical encoders simply aren’t suitable for that purpose.

Author :Gert Baars - Copyright : Elektor

Read more

Breakout Board for PIC10F2xx (SOT23-6)

Microcontrollers come in all sorts and sizes,  and it’s very tempting to use them every-where, even for very simple tasks. Tiny, inex-pensive microcontrollers especially suited  to very simple tasks, such as the Microchip  PIC10F2xx family, are also available. Thanks  to their compact size  and their ability to  source or sink 25 mA on their I/O pins, these  miniature microcontrollers are a good choice  for driving LEDs directly in miniature lighting  effect devices. They can also operate from a  2-V supply voltage, which allows them to be  powered directly by batteries (such a button  cells).

However, their small dimensions have  a few drawbacks, especially for developing  prototypes. The first drawback is that the IC  leads are so small that soldering is not easy,  and the lead pitch makes them difficult to  use with a breadboard or perforated proto-typing board. Another problem is that they  can only be programmed in-system, which  means that you always need an extra header  for programming (even if you can find a suit-able ZIF socket for a programmer, it will cost  you an arm and a leg).  The small PCB described here is intended to  make it easier to use Microchip PIC10F2xx  devices in the SOT23-6 package without  making the entire arrangement so big that you could just as well use a DIL version of the same IC.

Although the easiest way to solder the six-lead IC to the board is to use solder paste and  a hot-air iron, it is in fact possible to do this  with a normal soldering iron. Any excess sol-der can be removed with desoldering braid.  All leads are brought out to SIL connector  K1, which has a more conventional 100-mil  pitch and mates perfectly with a breadboard  or piece of perfboard for prototype develop-ment. What’s more, it is a one-to-one match  to the connector of a Microchip PICkit2 or  PICkit3 programmer.

The pads for the IC pins are surrounded by  larger pads that can be used as attachment  points for wires, resistors, LEDs and so on.  Once the prototype and the firmware are  finalised, the portion of the board outside  these pads can be sawn off and/or filed down  to make it easier to fit the board in a minia-ture enclosure.

Author :Luc Lemmens - Copyright : Elektor

Read more

Noise Suppression For R/C Receivers

Receiver interference is hardly an unknown problem among model builders. Preventive measures in the form of ferrite beads fitted to servo cables are often seen in relatively large models and/or electrically driven models, to prevent the cables from acting as antennas and radiating interference to the receiver. If miniature ferrite beads are used for this purpose, the connector must be first be taken apart, after which the lead must be threaded through the bead (perhaps making several turns around the core) and then soldered back onto the connector.

Project Image :

Noise Suppression For R/C Receivers Image

An interference source can also cause problems in the receiver via the power supply connection.The battery is normally connected directly to the receiver, with the servos in turn being powered from the receiver.  The servos can draw high currents when they operate, which means they can create a lot of noise on the supply line. This sort of interference can be kept under control by isolating the supply voltage for the receiver from the supply voltage for the servos. All of these measures can easily be implemented ‘loose’ in the model, but it’s a lot nicer to fit everything onto a single small circuit board. That makes everything look a lot tidier, and it takes up less space.

Circuit diagram :

Noise Suppression For R/C Receivers Circuit diagram

The schematic diagram is shown in Figure 1. Connectors K1–K8 are located at the left. They are the inputs for the servo signals, which are connected to the receiver by the servo leads. The outputs (K9–K16) are located on the right. That is where the servos are connected. Finally, the battery is connected to K17. Interference on the supply voltage line due to the motors and servos is suppressed by a filter formed by L10, R1, C1 and C2. L10 is a ferrite-core coil with an impedance of 2000 ohms at 30 MHz. In combination with C1 and C2, it forms a substantial barrier to interference in the 35-MHz R/C band.

Parts :

Resistor:

R1 = 1Ω

Capacitors:

C1 = 100nF

C2 = 22pF

Miscellaneous:

L1-L8,L10 = ferrite inductor

L9 = common-mode coil

K1-K8 = servo cable

K9-K17 = 3-way SIL pinheader

PCB Layout :

Noise Suppression For R/C Receivers PCB Layout

Signals with frequencies close to the 10.4-MHz intermediate frequency (which is used in many receivers) are also effectively blocked by this filter. L9 filters out common-mode noise on the supply line for the servos, which effectively means that it prevents the supply lines to the servos from acting as antennas. Finally, high-frequency currents on the servo signals are filtered out by ferrite beads in order to limit the antenna effects of these connection lines.

Author : Paul Goossens – Copyright : Elektor

Read more

70 A Solid-state Starter Relay

Overall, electro-mechanical scooter starter  solenoids are cheap enough but the down-side is that they’re not very reliable. The contact resistance increases over time, the coil  can be open-circuited due to the vibration,  and sometimes the power contacts weld up.  One solution is to replace them with a solid-state relay. In DC mode, we’ll need to use a  MOSFET transistor.

As is often the case in automotive systems, the supply negative is connected to the chassis ground, which means we’ll need to use a P-channel MOSFET. The current to be switched is relatively high, between 55 and 100 A (depending on engine capacity and compression), so we need a transistor with a  very low RDS(on) capable of carrying a large IDS. Since the starter is a DC motor with brushes, it generates considerable voltage spikes that are  quite destructive for the driving device, whence  the need to protect everything very well.

 

70 A Solid-state Starter Relay Circuit Diagram

A look at the wiring diagrams for various  scooters reveals that the safety switch on the  brake (which has to be applied first) supplies  +12 V, but the starter button (to be operated  next) connects to ground. One simple solution is to use an opto-isolator. While we’re on  the subject, let’s just note that this technique  means this circuit can be used for many other  applications too.

And finally, the circuit must be ‘Plug-n-Play’, i.e. usable with the original connector,  thereby limiting the circuit dimensions to  50 × 50 mm. Building a PCB capable of handling a current  of 70 A needs a few calculations. The resistance RT of a copper track with thickness E of  35 μm (0.035 mm) with length L and width W  is calculated from : RT = 1.7 × 10-5 × L / (E × W)   [Ω] , where E, L, and W are in mm, and T = 25 °C). 

The component positions mean our tracks  can be 15.25 × 44 mm, thus each track rep-resents 1.4 mΩ, or 0.7 mΩ if we use a double-sided board. At 75 A, the total voltage  drop will be around 100 mV and the power  dissipated 7.5 watts. The SUP75P03-07-E3  MOSFET from Vishay Siliconix (Farnell part no.  179 4 812) off er san RDS(on) of 7 mΩ at 75 A, i.e.  3.5 mΩ if we put two in parallel. In this case,  the voltage drop is 0.263 V and the power dissipated in each transistor is around 10 watts. The end result is that we get an overall volt-age drop of around 360 mV and a total dissipation of around 27.5 watts.

Let’s take a look now at the circuit diagram. On  the left, everything within the dashed rectangle  corresponds to the original wiring of the major-ity of Chinese scooters. R1 sets the current in  the 4N28 opto-isolator LED to around 25 mA  and R2 biases the base of the phototransistor. The phototransistor collector is connected  directly to the gates of the two MOSFETs T1  wired in parallel. At rest, the MOSFETs are held off by R3, but start to conduct when both contacts S1 and S2 are made, thanks to D3 and the  low impedance of the starter motor. Once the  starter turns, the charge on C2 ensures that the  circuit will continue to function.

Components C1, D1, C2, D2, and D3 protect  the circuit against the interference produced  by a load that is anything but purely resistive. Tests and measurements have been carried  out on a scooter using a GY6 engine type  CJ12M. The average consumption was 53 A:  49 A at bottom dead centre (minimum compression) as against 57 A at top dead centre  (maximum compression). The voltage drop  measured at the circuit terminals was strictly  identical to the theoretical value. After three  hours’ testing, at a rate of one start every five  minutes, no heating was detected.

Author :By Georges Treels - Copyright : Elektor

Read more

1-Watt LED Driver (PR4401)

The PR4401 chip from Prema can be used to drive an LED directly,  but not a high-power LED like one of the popular 1-watt types currently available on the market. The circuit shows that the drive signal at the Vout terminal of the PR4401 chip (pin 2) turns a medium-power PNP switching transistor (T1) on and off. When T1 is switched  into conduction, inductor L1 is charged. When T1 is switched off, the  inductor discharges its stored energy through the LED during flyback  with enough current to allow a one-watt LED to light up at nominal  brightness.

 

Circuit diagram :

 

PR4401 1-Watt LED Driver Circuit Diagram

 

During the ‘on’ time of transistor T1, the current through inductor L2 ramps up linearly to a peak value as expressed by. IL2(pk) = [(Vbatt – VCEsat(T1)) ×Ton ] / L2

 

Where VCEsat(T1) is the collector-to-emitter saturation voltage of T1 (here, a type BD140 is suggested).

 

During T1’s ‘off’ time, the inductor voltage reverses, forward-biasing  the LED and discharging through it at a constant voltage roughly  equal to the forward voltage of the LED, while its current ramps down  to zero. Because this cycle repeats at a high rate, the LED appears  to be always on, its brightness depending on the device’s average  current, which is proportional to the peak value. The LED current is  roughly a triangular pulse with a peak current approximately equal  to the inductor’s current because of the finite turn-off time of T1. The  estimated average current may be calculated from ILED(avg) = 1/2 × IL2peak × [Tdis  / (Ton + Toff)]

 

Where Tdis is the discharge time of inductor L2 through the LED. The  LED’s brightness can be increased or decreased by varying the inductance of L2. In practice, any value between 10 and 56 µH will work just  fine. The inductor current increases on each cycle until T1 goes out  of saturation, hence a small resistance (R1) is required at the base of  T1. Without the ‘stopper resistor’, the final current goes out of control  due to the DC gain of T1. A transistor with a high DC current gain and  low collector-to-emitter saturation voltage is the best choice if you  want to tweak the circuit for efficiency. Regarding L2, make sure the  peak current through it is below the saturation level.

 

Author :  T.A. Babu - Copyright : Elektor

Read more

Backlight Delay

Lots of devices are fitted with a liquid crystal display(LCD). Now LCD implies backlighting that rather useful option that enables us to read the message being dis-played! For devices where there’s no need to read the display continuously, the backlight doesn’t need to stay lit up all the time-several seconds is often all you need to read the display. This saves a little power and lengthens the life of the backlight.

Circuit diagram :

Backlight Delay Circuit Diagram

Devices fitted with an LCD also have a processor, and so it’s possible to employ a function to control the backlight directly from within the processor software. But some-times it’s not possible to implement this sort of function within the microcontroller, because all the controller’s pins are already in use, or because you don’t have the source codes or tools needed to modify the software.The circuit described here has been designed for just such cases.

A device using an LCD usually has at leas tone button that, in most cases, pulls one of the microcontroller inputs down to 0 V when it is pressed. If no such button exists, one can always be added. We can use the signal from this button to control the backlight. As soon as the button is pressed,the backlight is activated, then extinguished a few seconds later by the timer. Using an OR gate, it ’s possible to use several different buttons to trigger the timer.

It doesn’t take many components to build a timer like this. The OR gate consists of a pull-up resistor R1+R2 and as many diodes as there are but-tons. Thanks to these diodes, transistor T1 conducts while the button is pressed, and hence capacitor C1 is charged, the MOSFET T2 conducts, and the backlight comes on. Because R3 has a very low value, capacitor C1 charges very rapidly, so even a very brief press of one of the buttons is enough to trig-ger the timer. Once the button is released, T1 turns off, and C1 then discharges slowly through R4 alone, since T2 has a very high input impedance. When T2’s gate voltage falls low enough, it turns off and the back-light goes out. The time the backlight stays lit after all the buttons have been released is roughly R4 (O) × C1 (F) seconds.

Of course, this circuit can be used for other applications too, and can be used to switch things other than an LED for example, a relay.The value of R5 depends on the load being switched. For an LED running off a 5 V supply, a value of around 300 O will be about right.

Author : Clemens Valens - Copyright : Elektor

Read more