Fullscreen HTML5Med Res (??MB)HTML4

Just in time for winter . . . Infrared Sensing Heater Controller With summer rapidly becoming a distant memory, you will no doubt be dragging out the heaters to stay warm. Our new Heater Controller allows you to select exactly how much warmth you want from the heater and it can even be fitted with a thermopile sensor for precise temperature control. M ost electric radiators are pretty crude devices: you switch ’em on and they get hot. But in small rooms, they will quickly get too hot and then you will want to switch them off or possibly down a level or two, if they have such a switch. Most heaters pump out too much heat for a small room, even when they are on the lowest setting. You really need a heat controller. Our new Heater Controller is exactly what you want. Think of it as a “dimmer for radiators”. But it is a lot more than that. We should note that some heaters, typically oil-filled units, do have a thermostat where the heater switches off when the air temperature reaches a set value and then it switches back on when the heater cools down. The thermostat is usually mechani14 Silicon Chip cal and often makes a clicking noise whenever the switching occurs and it may also cause a neon or LED to flash on and off at the same time. That intermittent noise and light may be be disturbing if the heater is used in a bedroom while you are trying to get to sleep. It can also result in a noticeable heating/cooling cycle if you are close to the radiator. Our new Heater Controller has no mechanical switching to cause noises and it incorporates a thermopile infrared sensor so it can maintain a set temperature in a room. Alternatively, it can be built without the optional infrared sensor so you can use it to simply select the amount of heating that you desire. by John Clarke Celebrating 30 Years We should point out that the Heater Controller is not suitable for fan heaters. At the lower settings the fan will not run and that could cause the unit to over-heat and fail. How it works Our Heater Controller works by applying an integral (ie, whole) number of full mains voltage cycles to the heater. At lower power settings, fewer cycles are applied and for higher power, a greater number are applied. We have included a number of scope grabs to illustrate this switching operation. For example, at low power, it may only apply one cycle of 50Hz 230VAC out of every 15. At full power, the mains voltage is on constantly (as if the controller was not present). It uses a 15-cycle control period, siliconchip.com.au The Heater Controller can be built in two versions: with a thermopile sensor for accurate temperature control; or without, which allows you to vary the power over 15 different levels. The model shown here has the thermopile sensor so you can “dial up” the temperature you want. which corresponds to 300ms for 50Hz mains (like in Australia, New Zealand and the UK) or 250ms for 60Hz. At half power, there are seven or eight cycles of mains voltage applied to the heater out of 15 incoming cycles. The switching action will not be noticeable with heaters with wound resistance elements but it will be visible with radiator elements in silica glass tubes and very noticeable in those with halogen lamps. In fact, the flashing of halogen lamps in those heaters will drive you bonkers! Don’t do it. The mains voltage is only switched as it passes through zero volts (ie, “zero-voltage switching”). This minimises any generation of electromagnetic interference by the controller. Two versions As noted, this project can be built in one of two versions, with or without a thermopile sensor for temperature control. If you build it without the optional thermopile sensor, the knob on the front panel will allow you to vary the power siliconchip.com.au over 15 different power levels. The second version, with the thermopile sensor, provides a non-contact temperature measurement method by detecting the infrared radiation emitted by objects outside the box. This form of room temperature sensing is ideal since the Heater Controller’s circuitry runs at 230VAC mains potential and contact with an uninsulated external sensor would be dangerous. By having the thermopile located safely inside the unit, with a transparent window for insulation, it is rendered safe and its operation is not affected by self-heating due to internal dissipation, as would be the case with an internal sensor. The Heater Controller can be used with a 220-250VAC mains supply at 50Hz or 60Hz. It is not suitable for use with 110VAC supplies without some changes being made to the power supply and mains voltage detection circuitry. As you can see from the photos, the Heater Controller is mounted in a lowprofile diecast aluminium case with mains plug and socket leads at one end, along with a fuse holder. The adjustment potentiometer is on the lid. Circuit description The complete circuit for the Heater Controller is shown in Fig.1. We’ll start by describing the complete version which provides temperature control. The circuit is based around microcontroller IC1, Triac Q2 and thermopile sensor TS1. IC1 provides overFeatures & specifications all control, driving 3 Controls 230VAC heaters up to 10A/2300W. the sensor and the 3 Suitable for use with bar heaters, ceramic heaters or oilTriac which confilled convection heaters. Not suitable for halogen or fan heaters. nects mains power 3 220-250VAC mains operation, 50Hz or 60Hz. to the heater. TS1 has four pins 3 Heating power or Temperature control and it actually com3 Power control: 15 steps from low to full heating power. prises two separate 300ms (50Hz) or 250ms (60Hz) cycle. devices in the one 3 Temperature control: 15-31°C in 1°C steps. package. The thermopile IR sensor is 3 Zero voltage switching for low interference. 3 Triac gate drive: 68mA pulse for 300µs after each zero voltage crossing. connected between pins 2 and 3 while an q q q q q q q q Celebrating 30 Years April 2018  15 WARNING! The Heater Controller operates directly from the 230VAC mains supply and contact with live components is potentially lethal. Fig.1: circuit diagram of the Heater Controller. IC1 monitors the mains zero crossing at pin 5, potentiometer setting at pin 6, the internal temperature at pin 3 and external temperature difference at pin 7. It uses this information to decide when to deliver a gate pulse from pin 2, to switch on Triac Q2 which applies one cycle of mains power to the heater at a time. NTC thermistor is connected between pins 1 and 4. This is important since the thermopile senses the difference in temperature between the room and the sensor itself. The thermistor allows us to determine the sensor temperature. By adding the two temperatures, we can determine the absolute temperature of the room. The microcontroller monitors several signals or voltages, which are internally converted to numbers using its inbuilt 10-bit analog-to-digital converter (ADC). The voltage across the NTC thermistor in TS1 is monitored by input AN3 (pin 3), while the voltage at the output from instrumentation amplifier IC2 is monitored at input AN0 (pin 7). The setting of the control potentiometer, VR1, is monitored at input AN1 (pin 6) while the mains voltage is monitored at digital input GP2 (pin 5), via a 330kΩ 1W resistor. IC1’s GP5 output (pin 2) drives the base of NPN transistor Q1. Q1, in turn, sinks current from the gate of Triac Q2, switching it on. Its gate current flows via the 47Ω resistor connected between the 5.1V supply and the A1 terminal, through the gate and then to circuit ground via Q1. This is a slightly unusual configuration. The gate resistor (in our case, 47Ω) is normally placed between the Triac gate and transistor collector, with the A1 terminal connected directly to the supply (in our case 5.1V). However, with that arrangement, noise on the mains Active conductor could be injected into the microcontroller supply and cause it to latch up. In our circuit, the 47Ω resistor between the mains Active and 5.1V sup- ply provides isolation to avoid this problem while still limiting the gate current in exactly the same manner. Q2’s A1 terminal connects to the incoming mains Active supply via a 10A fuse. Q2 is used as a switch for making power connection from mains Active to the heater, via the A2 terminal. The DC supply for the microcontroller is derived directly from the 230VAC mains supply via a 470nF 275VAC X2 rated capacitor in series with a 1kΩ 1W resistor. The capacitor’s impedance limits the average current drawn from the mains while the 1kΩ resistor limits the surge current when power is first applied. It works in the following way. When the Neutral line is positive with respect to the Active line, current flows via the 470nF capacitor and diode D1 to the 470µF capacitor to charge it The first in this series of scope grabs shows the zerovoltage switching action of the Triac. The remainder show how the number of cycles is increased to increase the power. 16 Silicon Chip Celebrating 30 Years siliconchip.com.au What is a Thermopile? up. On negative half-cycles, the current through the 470nF capacitor is reversed via diode D2. Zener diode ZD1 limits the voltage across the 470µF capacitor to 12V and that supply then feeds a second 470µF capacitor via a 47Ω resistor and its voltage is limited to 5.1V by zener diode ZD2. This is the supply for the microcontroller, IC1 and it is decoupled with a 100nF capacitor close to the micro. IC1 monitors the GP3 digital input at pin 4 and this can be tied to the 5.1V supply or to 0V using a jumper shunt at JP1. When this pin is pulled high, the reading from VR1 is used to control the percentage of full power delivered to the heater. When this pin is pulled low, the heater is temperature controlled instead, using TS1 for feedback. VR1 is connected across the 5.1V siliconchip.com.au A thermopile is a “pile” of thermocouples that are exposed to the outside via an infrared window. A thermo-couple is a junction between two dissimilar conductors, bonded together as shown. One set of junctions is exposed to infrared energy while the other set is shielded from the infrared and thermally connected to a reference thermal mass. This is indicated as a heatsink in the diagram. The thermopile we are using has 60 thermocouples connected in series, so their voltages are added together. The output voltage from the thermopile varies with the difference between the temperature of the infrared-exposed junctions and the heatsink-connected junctions. If the infrared radiation heats the first set of junctions up to the same temperature as the heatsink, then their output will be at 0V. The output voltage can be positive or negative, depending on whether the heatsink connected junctions are colder (positive) or hotter (negative) than the infrared exposed junctions. The fact that the temperature measurement is based on received infrared energy means that this is a non-contact method of temperature measurement. All objects which are above absolute zero give off some infrared energy and the hotter they are, the more they emit, hence we can use this energy as a way to remotely sense temperature. The sensor we are using also includes an NTC thermistor which is joined to the heatsink. This allows the heatsink temperature to be measured. The temperature of the monitored object (ie, the source of infrared energy) is the heatsink temperature plus the temperature measured by the thermopile. So if the thermopile package temperature is 25°C, as measured by the thermistor, and the thermopile infrared temperature measurement is registering 4°C then the actual temperature measurement is 29°C (25°C + 4°C). If the thermopile registers -3°C then the result is 22°C (25°C - 3°C). The thermopile includes an infrared-transparent window that allows the infrared energy into the thermopile. We are also using an external lens, to protect against accidental contact with the circuitry on the inside of the box. One more thing to note about thermopiles: while we said above that the infrared emissions from an object depend on its temperature and that is true, they also depend on the colour of the object’s surface. A “black body” is an ideal emitter with an emissivity of 1.0 and the thermopile is calibrated to accurately measure the temperature of such objects. All other objects have an emissivity between zero and one and as a result, the thermopile will pick up less infrared energy at any given temperature and so will measure less than their actual temperature. Examples of objects with low emissivity includes most shiny objects, for example, with polished metal surfaces. To accurately measure the temperature of an object based on infrared energy, you need to know the emissivity and divide the measurement by this value. In the case of our Heater Controller, the emissivity of a typical room should be high enough, and the difference between internal and external temperature low enough, that such compensation should not be necessary. We do, however, suggest that you avoid pointing the IR window at very shiny objects. Celebrating 30 Years April 2018  17 Fig.2: follow this overlay and wiring diagram to build the full version of the Heater Controller, which is capable of operating in power control or temperature control modes, as determined by the position of jumper JP1. supply with the wiper connected to the AN1 input, as described earlier. The 100kΩ resistor from the wiper to ground holds the AN1 input at 0V, setting the control to minimum, should VR1’s wiper go open circuit. Temperature measurement The resistance of the thermistor in TS1 decreases with increasing temperature and therefore it has a negative temperature coefficient (NTC). Its value at 25°C is close to 100kΩ and will be reduced at higher temperatures. Its resistance is monitored indirectly at the AN3 input of IC1, by connecting the thermistor as part of a voltage divider, ie, with a 100kΩ resistor to the 5.1V supply. The resulting voltage is converted to a digital value in IC1 and that value is used to compute the sensor temperature in °C using a table that lists the expected voltage against temperature. With the thermistor at 25°C, given that its resistance of 100kΩ matches that of the fixed resistor, the voltage between NTC- and NTC+ should be half of the 5.1V supply (ie, around 2.55V). The thermistor resistance changes in a non-linear manner with respect to temperature. This online calculator can be used to determine how the thermistor resistance varies with temperature, by plugging in a Beta value of 3960 and a resistance at 25°C of 100kΩ: siliconchip. com.au/link/aaj1 For example, we can determine that if the sensor is at 15°C, the thermistor resistance will be around 158556Ω, giv18 Silicon Chip ing a voltage of 3.13V (5.1V x 158556Ω ÷ [158556Ω + 100kΩ]) across the thermistor (ie, at Vout1), assuming the supply voltage is exactly 5.1V. So that determines the temperature of the sensor itself. The thermopile output voltage allows us to determine the difference between the sensor temperature and the room temperature, but it is Celebrating 30 Years a very small voltage and needs amplification before it can be measured by IC1. To achieve this, we use an instrumentation amplifier (IC2). The amplifier gain is set at about 211 by the value of resistor Rg, 470Ω. This amount of gain gives IC2’s output a slope of 10mV/°C. The gain takes into account the losses in infrared heat through the lens used siliconchip.com.au Fig.3: this diagram shows which components can be omitted, to build the unit only for heater power control only. JP1 is replaced with a wire link and two additional wire links are fitted where shown. The photo below is of the full version and shows all wiring completed and secured, as in the diagram opposite. to cover the sensor. The output voltage of IC2 is referenced against a non-zero voltage so that we can measure the room temperature even if it is colder than sensor TS1. So if the output from IC2 is at this reference voltage, the thermopile measurement is zero degrees (ie, ambient equals sensor temperature). The refersiliconchip.com.au ence voltage is set by trimpot VR2 to half-supply, ie, 2.55V. If the output of IC2 is 20mV above the reference voltage then the temperature difference is +2°C. The micro adds this differential temperature to the sensor temperature, computed as explained above, to gauge the room temperature. Celebrating 30 Years The negative end of the thermopile (pin 3 of TS1), which connects to the inverting of IC2 (pin 2) is connected to ground. Thus, the positive output of the thermopile, at pin 2 of TS1, can vary above or below ground, depending on whether the outside temperature is above or below the sensor temperature. The thermopile is a voltage source and it can generate a negative voltage, despite the circuit not having a negative supply. However, instrumentation amplifier IC2 is capable of handling input voltages down to 150mV below its negative supply. The thermopile output voltage is typically within ±1mV so this is not a problem. The 100nF capacitors at the AN0, AN1 and AN3 inputs of IC1 provide a low impedance source for the analogto-digital converter’s sample-and-hold circuitry. Zero voltage crossing detection Pin 5 of IC1 (GP2) is a Schmitt trigger digital input. This monitors the mains Neutral via a 330kΩ resistor and is filtered with a 4.7nF capacitor. An interrupt in IC1 occurs whenever the voltage changes from a high (around 4V) to a low level (around 1V) and also from a low to a high. That interrupt occurs when the mains voltage swings through zero volts in either direction. The interrupt tells IC1 that the voltage of the mains has just passed through 0V. This allows IC1 to synchronise gate triggering with the mains waveform. Note that the 4.7nF capacitor at pin 2 introduces a phase lag (delay), but this is compensated for within IC1’s software. The voltage at pin 5 is clamped by IC1’s internal protection diodes. They clamp at +5.4V and -0.3V. Since the 5.1V supply for IC1 is essentially connected to the mains Active via the 47Ω resistor, the sensed Neutral voltage is relative to the 5.1V supply. Controlling power level only If you only want to be able to control the heater power and don’t need temperature regulation, jumper JP1 is set to pull pin 4 of IC1 high. In this case, there is no need to install TS1 or IC2, trimpot VR2 nor any of the associated resistors and capacitors (see Fig.3). This would reduce the cost of building the unit. April 2018  19 Parts list – Heater Controller 1 double-sided PCB coded 10104181, 103 x 81mm 1 diecast aluminium box 119 x 94 x 34mm [Jaycar HB-5067] 1 Fresnel lens for IR sensor (Murata IML0688) [RS components Cat 124-5980] † 1 M205 10A safety panel mount fuse holder with 10A M205 fuse (F1) [Altronics S5992] 1 4-way PC mount terminal barrier (CON1) [Jaycar HM-3162] 1 3-way PC mount terminal block with 5.08mm pin spacings (CON2) 2 cable glands for 5-10mm cable 1 DIL-8 IC socket 1 3-pin header with 2.54mm spacing and shorting block † 1 2m long 10A mains extension lead 1 knob to suit VR1 4 4mm eyelet connectors 4 8mm long M3 tapped Nylon spacers 3 M4 x 10mm screws 2 4mm ID star washers 3 M4 hex nuts 8 M3 x 5mm machine screws 4 stick-on rubber feet 10 PC stakes † 1 100mm length of 3mm diameter heatshrink tubing 1 25mm length of 6mm diameter green heatshrink tubing, if required for eyelet lugs 3 100mm lengths of 250VAC 7.5A mains wire (for VR1) 4 100mm long cable ties Semiconductors 1 PIC12F675-I/P microcontroller programmed with 1010418A.hex (IC1) 1 AD623AN instrumentation amplifier (IC2) † 1 ZTP135SR thermopile sensor (TS1)‡ [element14 Cat 2506255] 1 BTA41-600BRG insulated tab 40A 600V Triac (Q2) [element14 Cat 1057288, RS Components Cat 687-1007] 1 BC337 NPN transistor (Q1) 1 12V 1W (1N4742) zener diode (ZD1) 1 5.1V 1W (1N4733) zener diode (ZD2) 2 1N4004 1A diodes (D1,D2) Super glue, heatsink compound, solder Capacitors 2 470µF 16V PC electrolytic 1 10µF 16V PC electrolytic † 1 470nF 275VAC X2 class 7 100nF 63V or 100V MKT polyester ‡ 1 4.7nF 63V or 100V MKT polyester Resistors (0.25W, 1%) 1 330kΩ 1W 2 100kΩ § 1 1kΩ † 1 1kΩ 1W 2 470Ω § 1 20kΩ multi-turn top adjust trimpot (code 203, 3296W style) (VR2)† 1 10kΩ linear 24mm potentiometer (VR1) 2 47Ω († not required for power control only version) (§ 1 required for power control only version) (‡ 2 required for power control only version) * depends on version o o o o o 20 Qty. 1 1/2* 1/2* 1/2* 1/2* Value 330kΩ 100kΩ 1kΩ 470Ω 47Ω Silicon Chip Resistor Colour Codes 4-Band Code (1%) orange orange yellow brown brown black yellow brown brown black red brown yellow violet brown brown yellow violet black brown 5-Band Code (1%) orange orange black orange brown brown black black orange brown brown black black brown brown yellow violet black black brown yellow violet black gold brown Celebrating 30 Years Note that it would be possible to build the unit so that it could be used in either mode, by fitting a 250VAC-rated single-pole double-throw toggle switch to the box and wiring it in place of JP1. In percentage control mode, the temperature sensed by TS1 would be ignored. Construction The Heater Controller is built on a double-sided, plated-through PCB (printed circuit board) coded 10102181 and measuring 103 x 81mm. This is mounted inside a diecast box of 119 x 94 x 34mm. Fig.2 shows where the components are fitted for the full version, which can regulate the temperature, while Fig.3 shows just the components fitted which are required for controlling the heater power level (percentage). Follow the overlay diagram appropriate to your version. Start by installing the resistors. Table 1 shows the resistor colour codes but you should also check each resistor using a digital multimeter. Three additional wire links are fitted for the heater power control only version. These are shown in Fig.3; one to set the mode, in place of JP1, and two to hold the AN0 and AN3 inputs of IC1 at 0V. Use resistor lead off-cuts to make these links now. Following this, install the diodes which must be orientated as shown. Note that there are several different diode types: standard 1N4004 diodes for D1 and D2, a 12V 1W zener (1N4742) for ZD1 and a 5.1V 1W zener (1N4733) for ZD2. IC1 is mounted on an 8-pin DIL socket so install its socket now, taking care to orientate it correctly. Leave IC1 out for the time being, though. IC2 is installed for the full version, soldered directly on the PCB. Transistor Q1 can also be installed now. Fit the capacitors next. The X2 class capacitors and the polyester types usually are usually printed with a code to indicate their value; see the small capacitor codes table. Small Capacitor Codes Qty. Value F Code o 1 470nF 0.47F o 2/7* 100nF 0.1F o 1 4.7nF .0047F EIA Code IEC Code 474 104 472 470n 100n 4n7 siliconchip.com.au Subscribe to SILICON CHIP and you’ll not only save money . . . but we GUARANTEE you’ll get your copy! When you subscribe to SILICON CHIP (printed edition) in Australia, we GUARANTEE that you will never miss an issue. Subscription copies are despatched in bulk at the beginning of the on-sale week (due on sale the last THURSDAY of the previous month). It is unusual for copies to go astray in the post but when we’re mailing many thousands of copies, it is inevitable that Murphy may strike once or twice (and occasionally three and four times!). So we make this promise to you: if you haven’t received your SILICON CHIP (anywhere in Australia) by the end of the first week of the month of issue (ie, issue datelined “June” by, say, 7th June), send us an email and we’ll post you a replacment copy in our next mailing (we mail out twice each week on Tuesday and Friday). Send your email to: missing_copy<at>siliconchip.com.au 4 4 4 4 4 Remember, it’s cheaper to subscribe anyway . . . do the maths and see the saving! Remember, we pick up the postage charge – so you $ave even more! Remember, you don’t have to remember! It’s there every month in your letter box! Remember, your newsagent might have sold out – and you’ll miss out! Remember, there’s also an on-line version you can subscribe to if you’re travelling. Convinced? We hope so. And we make it particularly easy to take out a subscription - for a trial 6-month, a standard 12-month or even a giant 24-month sub with extra savings. Here’s how: simply go to our website (siliconchip.com.au/subs) – enter your details and pay via Paypal or EFT/Direct Deposit. You can order by mail with a cheque/money order, or we can accept either Visa or Mastercard (sorry, no Amex nor Diners’). If mailing, send to SILICON CHIP, PO Box 139, Collaroy NSW 2097, with your full details (don’t forget your address and all credit card details including expiry!). We’re waiting siliconchip.com.au to welcome you into the SILICON HIP subscriber family! Celebrating 30C Years April 2018  21 Fig.4: this diagram and the photos at right and below show how the sensor is mounted, with its lens emerging through an 11mm hole drilled through the side of the case. The electrolytic types are marked with their F (microfarad) value and must be oriented with the polarity shown. The longer lead is positive while the negative end is usually marked on the can with a stripe. The screw terminals and trimpot can be installed now. The 3-way terminal block for CON2 is fitted with the lead entry toward the lower edge of the PCB. VR2 has its adjustment screw near the top edge of the board, as shown in Figs.2 & 3. If fitting a pin header for JP1, solder this in place now. TS1 is fitted for the full version and is mounted along the edge of the board. This arrangement is shown in Fig.4. The sensor is located centrally within the cut-out on the side of the board, with the leads to pins 1 & 2 along the top of the PCB and the leads to pins 3 & 4 along the underside. You can either use PC stakes to connect the lead of TS1 to the PCB terminals or bend the sensor leads to fit into the PCB holes directly. Make sure the leads for pins 1 and 2 do not short to the pads on the PCB for pins 3 and 4; a small piece of electrical tape or heathsrink tubing could be used to insulate them from the board. Any pigtails on the underside of the PCB, including the ends of PC stakes, must be cut short to prevent contact with the base of the sensor. Leave Triac Q2 to be installed later. Preparing the case Now you need to drill some holes in the diecast enclosure. Drilling templates are provided in Fig.5. The lid requires a 9.5mm diameter hole for potentiometer VR1 and a 4mm Earth screw hole so drill them now. 22 Silicon Chip ORIENTATION TAG ZTP135SR HEATSHRINK TUBING If building the full version, follow the instructions below. Otherwise, go to the heading titled “PCB location”. The first hole that needs drilling in the box base is that for the lens. This is on the side of the box and is 9mm in diameter. Once drilled, fit the lens into the hole with the two protruding locating prongs on the rear rim of the lens housing both facing downward. Attach three of the four 8mm spacers to the PCB with the screws from the top of the PCB. The omitted spacer can be the one nearest VR2 or near fuse F1. Now place the PCB into the box with the thermopile sensor entering the lens. While holding the PCB in place, press the PCB toward the lens so that it is held tightly in place. Align the PCB so it is squarely positioned in the box and mark the mounting hole position for the missing spacer on the base of the box, then drill this hole to 3mm and check that the hole is correctly positioned. If adjustment is needed, file the hole with a small needle file so it is correctly positioned. Once correct, remove the spacers and lens and place the PCB in the box. Align it with the hole already drilled. When the PCB is squarely positioned, mark out and drill the remaining three holes. PCB location Note that when the PCB is in the box, the CON1 screw terminal end of the PCB sits further away from the end of the box compared to the other end. This allows space for the cable gland nuts. Holes for the fuse, cable glands and Earth screw can also be drilled now, making sure these are drilled at Celebrating 30 Years the CON1 end of the box. Re-attach the 8mm long spacers to the PCB. Then bend the Triac leads up at 90°, 4mm from the body of the Triac. Insert the leads into the PCB from the underside. The PCB can now be secured to the case with the screws from the underside into the tapped spacers. Mark out the Triac mounting hole position on the base of the case. Remove the PCB again and drill to 4mm. Clean away any metal swarf and slightly chamfer the hole edge. Re-attach the PCB and adjust the Triac lead height so the metal tab sits flush onto the flat surface, then secure the Triac with the M4 screw and nut. Note that the metal tab is internally isolated from the leads and so does not require any further insulation between its tab and case. Solder the Triac leads at the top of the PCB, then trim them short. Now remove the screws to gain access to the underside of the PCB and solder the Triac leads from the underside. The four rubber feet can be attached to the base of the case now. The case lid should be fitted with a panel label. We have designed three labels. One is for the power-control only version, one is for a temperature-control only version and the third option is for when JP1 can be used to select either control mode. These label files can be downloaded from our website (www.siliconchip.com.au). Once the label has been attached, cut out the potentiometer hole and Earth screw hole with a hobby knife. Wiring Cut the 10A extension lead in two, siliconchip.com.au to provide one lead with a plug on the end and another with a socket. Where the lead is cut depends on how long you prefer each lead. You may prefer a long plug cord and short socket lead, so the heater is located near to the controller. Alternatively, the lead can be cut into two equal lengths. Before cutting though, make sure you have sufficient length to strip back the insulation as detailed in the next two paragraphs. Strip back the outer insulation sheath by about 200mm on the socket lead so you can get a suitable 100mm length of Earth wire (green/yellow stripe) for the connection between the chassis and lid. Then cut the blue Neutral wire and brown Active wires to 50mm. Some of the spare 150mm brown wire length can be used later to connect from the fuse to CON1. The plug lead outer sheath insulation should be stripped back to expose 100mm of wire. This leaves sufficiently long Earth and Active leads. Cut the Neutral wire to 50mm. Pass these wires through the cable glands and connect as shown in Fig.2, stripping back the insulation before terminating to the fuse and CON1. Make sure the plug lead and socket lead are placed in the correct cable gland and wired as shown. Note that when wiring the fuse holder, heatshrink tubing should be placed over the wire terminals. Heatshrink tubing 3mm in diameter is suitable. Pass the wires through the tubing before soldering to the terminals. Cut the shaft of VR1 to 12mm long and file the edges smooth. Then attach the three 100mm lengths of 7.5A mains rated wire to its three terminals and cover with 3mm heatshrink tubing. The other ends connect to CON2. These wires are held in place using a cable tie that feeds through holes in the PCB. Attach VR1 to the lid and note that the potentiometer must be oriented as shown, so it fits beside the mains rated capacitor on the PCB. Fit the knob; you may need to lift out the knob cap with a hobby knife and re-orient the cap so its pointer position matches the rotation marks on the panel. Apply a smear of heatsink compound to the underside of the Triac before installing the PCB inside the case. Connect the Earth wires to M4 crimp or solder eyelets and cover with siliconchip.com.au green heatshrink tubing. The eyelets attach to the side of the case and the lid using an M4 screw, star washer and nut. IC1 can now be plugged in, taking care it is oriented correctly. Insert the 10A fuse into its holder. Press the cover onto the terminal barrier (CON1) to prevent accidental contact. Check your construction carefully and especially check that the Earth wires (green/yellow striped) actually are connected to the case and Earth pins on the mains plug and socket. Check this with a multimeter set to read low ohms. The cable glands need to be tightened to hold the mains cords in place. Because these are easily undone, the thread of the glands should have a drop of Super Glue (cyanoacrylate) applied to the threads before tightening. This way, the glands cannot be easily undone. Attach the lid using the four screws supplied with the case. If you fitted the pin header for JP1, now plug the shorting block into the percentage (%) position. Connect a heater to the controller and with the lid in place, apply power and check that the power can be varied using VR1. The following text only applies to the temperature-control version, so if you built this, unplug the unit, open up the lid and shift the shorting block into the alternative position. Setting up the thermopile The reference voltage needs to be adjusted now and this procedure also compensates for any offset voltages present in TS1 and IC2. Do not plug the unit into the mains directly during this procedure! You will need a supply that is between 5 and 9V DC, (eg, a 9V battery) connected between the 0V PC stake and the V+ PC stake on the PCB. Monitor the thermopile voltage using a multimeter connected to the 0V (black) and Thp (red) terminals, then adjust VR2 so that the voltage at Thp is half the voltage measured at the terminal that’s labelled 5.1V. So for example, if the supply is 5.1V, Thp should be set to 2.55V. This adjustment must be done when the thermopile is measuring the same temperature as its own body. To ensure this, place a matte black object larger than the diameter of the lens directly in front of it. This object needs to be the same temperature as the sensor, so place it nearby and leave it for an hour or so, to ensure that they are very close in temperature. This black body can be a block of wood painted matte black, the side of a black plugpack, a piece of matte acrylic, black-handled kitchen utensil, etc. Final calibration Final calibration is done with the unit powered from the mains, so make sure the lid is in place. Firstly, measure the ambient temperature with a thermometer and set VR1 to that setting, noting that calibration can only be done if the ambient temperature is within the range of 1531°C. Switch on the power and after about five seconds, check if the load is on or off. An incandescent lamp Close-up view of the mains input and output section of the Heater Controller. Don’t forget to place the protective shroud over the mains terminal block. Celebrating 30 Years April 2018  23 makes a suitable load (and you can see if it’s on!). Calibration is made by adjusting VR2 but this adjustment must be done only after power is switched off. So the adjustment will be a trial and error procedure. VR2 needs to be adjusted clockwise if the load does not switch off when VR1 is set to just under ambient temperature. Adjust VR2 anticlockwise if the load is still off with VR1 set to ambient temperature. During this procedure, ensure that the lens is not blocked and has a good view of the room and is not pointed at a large window, oven, fridge, air conditioner or another source of heat/cold. Temperature control mode When using the Heater Controller as a thermostat, the thermopile will need to be placed so that the room temperature can be monitored. This room temperature is best detected by using a non-shiny black object placed near the thermopile lens. The object should absorb the surrounding heat from the room air, allowing the thermopile to take the temperature reading. In practice, you may just need to place the controller box and lens near to any object of any colour to have satisfactory temperature control. The acceptance angle to the thermopile via the lens is about 84°. This can be visualised as a cone projecting out 24 Silicon Chip Fig.5: drilling and cutting templates for the diecast aluminium box and lid. The “D”-shaped fuseholder hole should be drilled to 11mm and then filed to the shape shown here, to prevent the fuseholder from rotating. from the lens with an 84° angle at the base. This is close enough to 90° that 100mm from the lens, the area that is observed by the thermopile is a circle of 200mm diameter. In other words, it is a 1:2 ratio of the distance from the lens to the spot size. Celebrating 30 Years Take care not to have the thermopile facing the heater itself. The high infrared level from the heater will cause the controller to switch off the heater. Bar radiator elements can reach 380°C, while convection heaters will be significantly above ambient. SC siliconchip.com.au