Power-up/down Sequencer

Whether you’re talking about a home cinema  or a computer system, it’s very often the case  that the various elements of the system have  to be turned on or off in a quite specific order,  or at least, automatically. Constructing this  sort of automation system is well within the  capability of any electronics enthusiast worthy of the name, but in this ‘all-digital’ age,  most of the circuits of this type to be found  in amateur electronics magazines or web-sites use a microcontroller. Even though that  is indeed a logical solution (in  more ways than one!), and you  might even say the easiest one, it  does pose problems for all those  people who don’t (yet) have the  facilities for programming these  types  of  IC.  So  we  decided  to  offer you now an approach that’s  very different, as it only uses a  simple, cheap, commonly-avail-able analogue integrated circuit,  which of course doesn’t have to  be programmed. Our project in  fact uses as it’s ‘brain’ an LM3914,  a familiar IC from National Semiconductors,  usually  used  for  driving  LED  VU  (volume  unit)  meters.

Power-up/down Sequencer Circuit Diagram:

Sequencer Circuit Diagram

Before taking a look at the circuit  for  our  project,  let ’s  just  remind ourselves that the IC has  one analogue input and ten out-puts intended for driving LEDs.  It can operate in ‘point’ mode,  where the LEDs light up in turn,  from first to last, depending on  the input voltage, but only one LED is lit at  any given time. Alternatively  it can operate  in ‘bar’ mode (this is the mode normally used  for VU meters), and in this case, the LEDs light  up one after the other, in such a way as to create a strip of light (bar) that is longer or  shorter according to the input voltage. This is  the mode selected for the LM3914 in the circuit described in some detail below.

So as to be able to control the AC powered equipment  our  sequencer  is  intended  to manage, we are using solid-state relays — four, in our example, though you can reduce or increase this number, up to a maximum of ten. Since the input devices in solid-state relays are LEDs, they can be driven directly by the LM3914 outputs, since that’s exactly what they’re designed for. As only four relays  are available, these are spread across out-puts L2, L4, L6, and L8, but you can choose  any arrangement you like to suit the number  of relays you want to use.

Resistor R7 connected to pin 7 of the LM3914  sets the current fed to the LEDs by the LM3914  outputs. Here, it’s been set to 20 mA, since  that is the value expected by the solid-state  relays chosen. The input voltage applied to  pin 5 of the LM3914 is none other than the  voltage present across capacitor C1 — and  this is where the circuit is ingenious. When  the switch is set to ‘on’, C1 charges slowly  through R5, and the LEDs of the solid-state  relays on the outputs light one after another  as this voltage increases; in this way, the units  being controlled are powered up in the order you’ve chosen. To power-down, all you have  to do is flip the switch so that C1 discharges  through  R5,  and  the  LEDs  go  out  in  the  reverse order to that in which they were lit,  in turn powering down the units connected to the solid-state relays. Easy, isn’t it? If you’re not happy with the sequence speed,  all you need do is increase or reduce the  value of R5 in order to alter the speed one  way or the other.

The circuit needs to be powered from a volt-age of around 9 to 12 V, which doesn’t even  need to be stabilized. A simple ‘plug-top’,  ‘wall wart’ or ‘battery eliminator’ unit will be  perfect, just as long as it is capable of supply-ing enough current to power all the LEDs. As  the LED current is set by R7 to 20 mA per LED,  it’ll be easy for you to work out the current  required, according to the number of solid-state relays you’re using.

In our prototype the type S216S02 relays  from Sharp were used, mainly because they  proved readily available by mail order. They also have the advantage of being compact,  and their switching capacity of 16 A means  you can dispense with a heatsink if you’re  using them for a computer or home cinema  system, where the current drawn by the vari-ous units can be expected to remain under  1 A. These solid-state relays must be protected by a fuse, the rating of which needs to  be selected according to the current drawn  by the devices being powered.

Also note the presence across the relay terminals of a VDR, also known as a GeMOV or  SiOV, intended to protect them from any spurious voltage spikes. You can use any type  that ’s intended for operation on 250 VAC  without any problem. The values of fuses F1  to F4 are of course going to depend on the  load being protected.

Construction of the circuit shouldn’t present any particular difficulty, but as the solid-state relays are connected directly to AC  power, it is essential to install it in a fully-insulated case; the case can also be used to  mount the power outlet sockets controlled  by the circuit. Note that sockets are female  components.
Let’s just end this description with the sole  restriction imposed by our circuit — but it’s  very easy to comply with, given the intended  use. In order to remain triggered, the solid-state relays must carry a minimum holding  current, which is 50 mA in the case of the  devices we’ve selected. In practical terms,  this just means that each of the devices powered by our sequencer must draw at least  50 mA, or in other words roughly 12 VA at  230 VAC, or 25 VA at 120 VAC.

Author :Christian Tavernier - Copyright : Elektor


Post a Comment