DE4441733A1 - Light dimmer for conventionally wound transformers supplying LV halogen lamps and/or filament bulbs working on phase-control principle - Google Patents

Abstract

The dimmer has a multistage filter (1) between the line and neutral conductors to reject the audio frequency components in the mains and in addition, through a control circuit comprising a filter member (r1,c1), a multistage filter (the same one as before?) and a window discriminator(2), is in a position to reject DC components in the load current, generated by the magnetic properties of the load or by the control circuit itself, by suitable shifting of the firing angle. With unacceptably high phase-shifting in fault situations with unloaded transformers, the phase section is prolonged and the end stage is controlled for the whole time, thus preventing asymmetries.

Description

The invention relates to a brightness control for con conventionally wound transformers with secondary sides Low-voltage halogen lamps and incandescent lamps after the phase cutting principle according to the preamble of claim 1.

For the stepless control of conventionally wound Transformers based on the phase gating principle are prin zipiell circuits known, usually in the main Circuit a semiconductor switching element, for example in the form of a triac. These circuits must compulsorily are usually equipped with a fuse, which in the Short circuit must be replaced. There are also circuits known, the one after detection of the zero current crossing Determine the ignition timing and, according to the set phase angle, ignite a connected triac. These Circuit arrangements can, however, ripple control Insufficiently negate impulses in the network.

From DE 38 36 128 A1 it is known that in the main circuit a MOS-FET or IGBT power amplifier in series with a load which, in contrast to the triac output stage, passes through at any time Apply the appropriate control voltage to the gate in the sine half-wave can be switched off. This type switching off in the event of a short circuit by sensing the voltage  used via a current measuring resistor. Furthermore, the Ver using a MOS-FET or IGBT power amplifier known after the leading edge principle works. This will be the end stage controlled so that according to the set phases angle an ignition pulse is generated, which the final stage in the conductive state and thus the current flow through the load ensures. Now there is a self-retention of the Control in place that maintains this conductive state receives. If the current runs through the load through zero, this is detected by another circuit arrangement, which in turn now turns off the power amplifier again. This corresponds to the function of a triac in switching behavior. This is necessary to protect the power amplifier, since the Switching off the current outside the current zero crossing in the connected inductive consumer (conventional wound transformer) would generate a counter voltage, which in the continuous occurrence to destroy the end level could lead. But occurs outside of normal loading caused a very excessive current or a short circuit, you are immediately able to switch off the power amplifier and to protect them like that. However, this is necessary for more Occurrence of this overcurrent, a permanent Ab circuit, otherwise the power amplifier through the occurring overvoltage of the inductive consumer zer could be disturbed.

The object of the invention is now a scarf to arrange arrangement in such a way that convents tionally wound transformers and incandescent lamps los can be controlled so that when a Short circuit of the load the arrangement is not destroyed and thus protects itself without a fuse to have to change. Furthermore, the trouble-free operation  of inductive loads, that is, for example not from loaded transformers of any kind up to the empty run and of mixed loads of incandescent and conventional wound transformers. Furthermore ripple control pulses should be negated.

The aim of this task is achieved by the Claim 1 means reached. A further education the invention is the subject of dependent claim 2.

An embodiment of the invention will follow hand of the drawing figures explained in more detail. It shows:

Fig. 1 is a block diagram of the brightness control;

Fig. 2 is a block diagram to prevent the asymmetries of the brightness control.

A fundamental problem when dimming transformers is the creation of DC components. This can to destroy the transformer and the dimmer to lead. DC components can be different Types arise:

• a) When switching on, depending on on results switching point within the network half-wave one more or less high direct current, which with the by the Trans formator certain time constants decays and the the alternating current is superimposed. For example the time constant for a transformer 500VA round 13 periods, that is approximately 0.27 seconds.  
• b) Is switched off in a transformer there is a high level of residual magnetism Switch on to saturate the magnetization curve (asymmetrical)
• c) Is at a transformer at the time of the maximum Induction switched off the secondary load, this can be too lead to a permanent DC component (failure of the Halogen lamp).

The invention recognizes this DC component, as can be seen in FIG. 1, by averaging the output stage voltage 4 (combination R1, C1) and uses this as a control voltage for the following controllable window comparator 2 , which thus determines its window center. Furthermore, this window comparator, the filtered and 180 ° ge rotated network signal from filter 1 is supplied. If the network signal passes through the comparator window, the capacitor C2 is reset. This creates a ramp voltage. This signal is fed to a further comparator OP2. If the ramp voltage intersects a DC voltage value proportional to the set dimming angle, an ignition pulse is generated. If, for example, a positive control voltage occurs due to the above-mentioned developments, the threshold of the window comparator 2 is shifted in the positive direction. Since the network signal in the three-stage filter 1 is now rotated by 180 °, the ramp in the negative half-wave is reset later and the ignition pulse is also generated later. The current flow in the negative half-wave is thus shorter and the positive DC component is regulated. Experiments with conventional dimmers and the use of this circuit arrangement have shown that permanent saturation peaks are reduced down to 20A after the load has been removed by the asymmetry correction to 0.4A.

The requirement remains to negate ripple control signals. This is implemented in the filter 1 already mentioned in the asymmetry control.

The suppression of ripple control pulses in leading edge dimmers is a problem that has been discussed for a very long time. A higher-frequency signal is superimposed on the network of the 50 Hz fundamental wave by the power supply company. This leads to shifts in the zero points of the mains voltage half-waves. Since the ignition angle is derived from these zero points, this also fluctuates, which is noticeable by fluctuations in the illuminance of the illuminant. If you take control in two-wire operation (series connection phase - dimmer - load - zero) between dimmer and load, you can achieve improvements with a sieving, but they have their limit due to the time of the sieving. A successful example is shown by the conventional so-called rotary dimmer with its changing sieve factor depending on the current flow angle. Fluctuations in brightness caused by ripple control impulses are noticeable to the human eye, especially when dimmed down. However, since the rotary dimmer has its largest sieve factor here, it dampens the ripple control influences the most at this point. With fully electronic dimmers, so-called touch dimmers, this sieve chain can only be fixed and based on the maximum brightness, and thus only have a small sieve factor, which has an unfavorable effect on the damping of ripple control pulses. For this reason, with ripple control pulses, a zero point detection for fixing the ignition angle by detecting the voltage between phase and zero is advantageous, as a result of which results can be achieved which exceed those of the mechanically adjustable dimmer. Since in this arrangement the time constant and thus the phase shift through the sieve chain remain independent of the current flow angle, it is possible to select a very high sieving with a phase shift of 180 ° relative to the zero points of the mains voltage. With, for example, three-stage RC elements in filter 1 , the effects of ripple control pulses are very much reduced and almost unrecognizable to the human eye due to fluctuations in the brightness of the light.

Another new circuit concept is required despite this regulation ( Fig. 2): If you control an inductive load (for example an idling transformer), there is a phase shift of up to 90 °. Since the ignition pulse is always related from N to L and thus has no phase shift, you can only set phase angles from 90 ° to 180 °. Every phase angle <90 ° results in a current flow of 180 °. Since a current flow angle other than 180 ° cannot be reached from 0 ° to 90 °, care must be taken that phase angles in this area are switched to continuous current flow. This can be achieved as in Fig. 2. Ue1 is the high signal that begins when the average is cut by the ramp voltage and ends with the discharge pulse. Ue2 is a high pulse that is obtained from the termination of the current flow through the output stage 4 . The voltage Ue1 is differentiated via C3, R4. Two voltage criteria can now be distinguished at the input of the OP1. During a time to be selected, the voltage at R5 is so high with a positive C3 pulse that Ue2 is unable to set the output of OP1 to low. After this time, the latching of the OP1 via R7 and R5 can be reset at any time by a pulse Ue2. The time for the "forced posture" is sensibly chosen with 0.5 half-wave, ie 5 ms.

Claims (2)

1. Brightness control for conventionally wound transformers with secondary-side low-voltage halogen lamps and / or incandescent lamps with a load connected in series to the MOS-FET or IGBT output stage, which works according to the phase-gating principle so that the output stage operates in normal operation when the load is at zero current is switched off, however, that when an excessive current or short circuit occurs, the output stage can be switched off and thus protects itself, characterized in that a multi-stage filter ( 1 ) is connected between the N and L conductors in order to negate the ripple control effects in the network , and in addition by a control circuit, consisting of a filter element (R1, C1), a multi-stage filter ( 1 ) and a window discriminator ( 2 ), is able to produce DC components in the load current, caused by the magnetic properties of the load or but negated by the control circuit itself, by suitable shifting of the ignition angle ( Fig. 1). 2. Brightness control according to claim 1, characterized in that in the event of impermissibly high phase shifts in the event of a fault due to idle transformers, the leading edge is extended in such a way that the output stage is controlled for the entire time and asymmetries are prevented ( FIG. 2).