Light Dimmers

Figures 1 and 2 show a schematic diagram of the ML4831EVAL-220V ballast and control interface circuit. An additional secondary winding of 25 turns is wound on T1 (see figure 2) to obtain the bias voltage for circuit operation and the required safety isolation. Components D17, C26, R37, C29 and D18 rectify, filter and regulate this voltage to 18VDC. Components R36, R35, R34, C27 and the Programmable Unijunction Transistor, Q4, form an astable sawtooth oscillator.

The electric power which can be controlled is decided by the permission value of the electric current which can pour into the TRIAC. I used the TRIAC which can apply the 12-A electric current to the circuit this time. In the calculation, in case of AC 100V, a maximum of 1200 W can be controlled but in the actual use, about 700 W or 800 W are safe.

Three power levels (7 / 10 / 20 watts) *Single pushbutton control for on/off and power level The PIC16F84 chip is re-programmable, allowing unlimited changes to the basic code.

This reference design is a high efficiency, high power factor, digital dimming electronic ballast designed to drive rapid start fluorescent lamp types. The design contains an active power factor correction circuit for universal voltage input as well as a ballast control circuit using the IR21592. The design also includes a PIC16F628 microcontroller and an isolation circuit for connecting to a Digitally Addressable Lighting Interface (DALI). Other features include EMI filtering, transient protection and lamp fault protection.

The 555 timer "light dimmer" schematic circuit is shown in figure 1 below. For the light dimmer to work the 555 timer is configured as a "variable cycle", astable oscillator running some where around 300 Hz. The power mosfet used here would be a TO-220 type such as MTP3055E or similar. Note the need for a TO-220 type heat sink for full rated loads.

Figure 1 shows the classic way to use a backlight-driver IC to provide dimming. By modulating an external PWM signal, the circuit controls the white-LED current. By adjusting the on/off-time ratio, or duty cycle, the circuit can provide drive ranging anywhere from full-on to full-off. This circuit relies on the fact that the baseband or application processor has a PWM timer available. A second method of obtaining dimming is to use an adjustable analog-input interface (Figure 2). The drawback of this approach is that it requires a DAC block in the digital-baseband or application processor.

The dimmer circuit in Figure 1 can change the intensity of the light from zero to maximum. The dimmer operates at approximately 12V, unlike the usual ones that function by adjusting the firing angle of the 110 or 220V mains supply. The dimmer works to inject a constant current into the halogen lamp and to regulate that current using pulse-width modulation (PWM) according to a potentiometer-controlled input, or a 0 to 5V signal, or even an analog output from a µC. 12V ac from the transformer, converted to 16.8V dc, powers the SG3524 PWM circuit (IC1). An RC circuit sets the approximately 10-kHz operating frequency. The output of the PWM IC drives the power transistor (Q1), a pnp Darlington.

Virtually all domestic and professional dimming systems are based on triacs. These devices will conduct once they have been fired, only while current flows in excess of the holding current of the device. These dimmers work very well with a resistive load such as an ordinary Tungsten filament light bulb as the triac can be fired at any point during the mains half-cycle and will continue to conduct until very close to the end of the half-cycle as current is drawn continuously over this period. In this way the lamp current can be adjusted from maximum to zero.

Figure 1 shows a simplified schematic of a typical low-voltage halogen-lamp transformer without the protection circuits and EMI filter. Q1 and Q2 with C2 and C3 comprise a classic, half-bridge topology that works in self-oscillating mode. The circuit provides positive feedback by placing the primary windings of transformer T1 in series with the bridge output. To achieve a high power-factor value, a rectified but unfiltered mains voltage supplies power to the circuit. The working frequency is approximately 30 to 40 kHz.

As you can see, this is a tried-and-true diac & triac combination that offers both simplicity and high performance. With a load up to 3.5A, the power rating is 750W at 230V. A heat sink on the triac is a must! Note: for 110V operation, add a 220k resistor in parallel with the 470k pot. No other modification is necessary. But since the maximum allowed current is still 3.5A, the highest possible power output will drop to 380W.

The IR remote control for the system is also based on a PIC10F200. It uses the wake-up on change function of the I/O to wake it whenever a button is pressed. It then generates one of two different modulated outputs, depending on which button was pressed. The carrier frequency is 38 kHz to match the receiver modules in the dimmer circuit.

A 555 timer generates a PWM drive signal to a power FET. Bootstrapping the entire control circuit across the FET using D1 allows an n-channel FET to make two-wire operation possible. Operating the PWM at a low switching frequency of about 120 Hz and shaping the rise and fall time minimize EMI without using any filter components. The optional components in the dotted box add short-circuit protection and may be unnecessary in many applications. You should size the FET's on-resistance to the load (potentially saving some cost), but make sure the FET can handle the lamp's start-up current, which is typically 10 times the steady-state current.

At first triac TR1 is not conducting; C1 is charged through R1/P1 until the trigger level of diac D1 is reached. When the diac trigger level is reached (about 30 V), D1 fires and Triac TR1 is switched on. The triac will remain in conductive mode until the mains current is lower than the triac hold current at the end of the half mains period.

This circuit is a real core of the dimmer system. This circuit generates ramp 100 Hz signal which is syncronized to the incoming mains voltage. The ramp signal which is generated will start form 10V and go linearly down to 0V in 10 milliseconds. At the next mains voltage zero crossing the ramp signal will again immediatly start from 10V and go down to 0V. This same ramp signal is fed to all of the 4 comparators in the dimmer.

The controlled is based on the PIC18F252. This sets the delay before the lamps are switched on after the mains zero crossing. Included are various I/O line to switch the lights on or off, variable resistors to set the brightness, and a RS232 link to enable computer control at a later stage.

The ballast control section is built around the IR21592 Dimming Ballast Control IC (IC2). The IR21592 is used for preheating and igniting the lamp, controlling the lamp power and detecting fault condition (over temperature, over current, VCC fault and DC Bus/ AC line fault). The IR21592 contains a voltage-controlled oscillator (VCO) controlling the half-bridge frequency while maintaining a 50% duty cycle, a high voltage half-bridge driver, an amplitude control, fault circuit and an analog dimming interface.

The PIC16F628 has control of this line and determines if lamp should be on or off based on fault conditions and user requested settings from the DALI. There are two signals used for fault detection, lamp -out (RB6), and fault (RB5). The lamp-out signal indicates the presence of a lamp or lamp fault.

The following circuit is a lamp dimmer circuit that is capable of controlling up to 1200 Watts.

The way that this circuit works is as follows. The AC line voltage is rectified by D1 and D2 which connects to a voltage doubler circuit made up of the two 22uf capacitors. The Flash Freq. Pot and the 10uf capacitor charge up which triggers the Diac and causes the triac to turn on.

Pretty nice design for a cheap consumer grade flashlight. The power switch was not exactly as shown since there was an option to have the lamp on solid.