Maker Pro
Maker Pro

Unusual ultra-simple 2-lamp alternating flasher

gamebox

Sep 14, 2014
7
Joined
Sep 14, 2014
Messages
7
Hello all. :) I have some questions regarding certain ideas I have for some simple circuits.

Here's what it's about. There's a set of christmas lamps - make it 20 in series on 230V (12V each). The first one contains a bimetal switch. I need to flash another set of 20 the opposite way. I intend to use the changing voltage on the flasher lamp for that purpose - it is low (12V) when on, and high (230V) when off. To make things simple, the graetz or diode supplies rectified mains to all the lamps. First I tried using a small transistor whose base I connected to high voltage on a flasher lamp through some high-ohms resistor, and it was driving a TIC106 or some similar thyristor. It worked well for a couple of minutes each time, but from some point the second set would just stay on (I was said sine-wave voltage peaks in transistor's base circuit were to blame). The better idea I was given was to use optocoupler (MOC3020) instead, which worked well but they are somewhat harder to find and more difficult to solder without the PCB (as I need to do). So, I came to an idea of a super simple, ultra cheap circuit using PCR406 thyristors which can be had virtually free as I have many broken modern christmas sets using those in 8-function controllers. So, this is what I don't know about my idea - can I connect the PCR406's gate to the mentioned changing voltage (12-230 V) of a flasher lamp through a high-ohm (how big?) resistor but using *something* (like two 1N4007's in series) as a stable voltage reference between gate and ground if necessary - would that idea work? What is important is a full shut-off for the second set when the first one switches on, and I also don't want the current supplied to the electronics to leave enough voltage+current in the first set to make it "glow" when it's flasher lamp is off (lamps could be rated even at 0.05A, but should be typically around 0.1A). And as a second thing - what would be the simplest/cheapest blinkers (without dedicated chips) I could use to trigger a single PCR406 or a pair of them to flash the lamps? A blink-LED crosses my mind first, but can something simple also be made with a condenser/resistor or so?

Thanks in advance, folks. :)
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
You could replace one of the bulbs with a 12V relay that draws similar current to a bulb.

The problem with using a transistor across the bimetal strip (if that's what you're doing) is that it has 240VAC across it when it is open. You would need to be very careful in order to use this as a method of driving a transistor.

The voltage across the bimetal strip could also be used to trigger a triac directly. You would want to arrange it so that the gate current drops to zero as the triac turns on or you could end up with some glow through the lamps that are "off".
 

gamebox

Sep 14, 2014
7
Joined
Sep 14, 2014
Messages
7
Thanks Steve. :)

The relay in series with bulbs is a problem as I'd like to make a few sets like this, with different bulb voltages and current.

Direct use of 230VAC to drive the SCR is my idea. I want to use PCR406 thyristors I have aplenty, and use rectified mains on the bulbs. Now how could I arrange this - is it enough to supply (pulsing) 230V to SCR's gate through some resistor or do I need some way to stabilize those voltage pulses too (by some component between gate and ground having fixed voltage - like two 1N4007s (1.2V), or some electrolytic cap.)? Would that voltage fixing on the gate (in case it is needed) keep SCR conducting when supply falls to 12V?

PCR406's datasheet is as follows: Gate trigger current - 200 uA, Gate trigger voltage - 0.8V, Holding current - 5mA, Latching current - 6mA, Peak gate current 0.1A, On state threshold voltage - 0.95V. As it seems to me, since it is a low power device, it's gate current is small enough not to cause any glow trough the first group of bulbs.
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
Firstly, it needs to be a triac, not an scr.

Secondly, it needs the gate to "anode" voltage to fall to a small value after the lights are triggered. This essentially means that if the thermistor is connected to the neutral end of the chain of bulbs, the triac needs to be on the active end.

There will also need to be some resistance in series with the gate.

I haven't fully thought this out, and be VERY careful considering you're working with the mains!
 

gamebox

Sep 14, 2014
7
Joined
Sep 14, 2014
Messages
7
Hm, but why is a triac instead of a thyristor necessary on (pulsed) DC operation? The whole circuit (bulbs and electronics) will get rectified mains (through graetz or diode).
Basically, I just need to get the PCR406 conducting by using pulsed 230V DC (310V in peaks) as a gate input and that challenges my knowledge. The resistor in series with the gate is necessary, but I don't know how to select it's value given that voltage input would change. Or if some additional component is necessary here (a stabile voltage reference of sort, for example).
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
Sorry, an SCR will be sufficient then.
 

gamebox

Sep 14, 2014
7
Joined
Sep 14, 2014
Messages
7
Cool. :)
So, how do I make it conducting using (pulsed) 230VDC? Is plain resistor in series with gate enough? Or I need something like zener diode between gate and ground to keep SCR's control voltage stabile?
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
Something like this:

Flash.png

The switch shown is the flashing bulb. I obviously haven't shown all the bulbs. D1 is just there to ensure you get pulsed DC from an AC input. If you already have pulsed DC, you don't need it.
 

gamebox

Sep 14, 2014
7
Joined
Sep 14, 2014
Messages
7
Sorry for the delay, I lost a link to this page when my system previously crashed.
Yes, *steve*, that's the schematics. The "switch" (bimetal bulb) doesn't need to be the last one in the strand, I placed it first to count out the resistance of the whole strand of the bulbs. But, the SCR can not be triggered with full mains voltage like you draw there, that's the problem I originally had. I tried to solve it using the transistor to trigger the SCR with lower voltage, but on a second thought I saw that it might not have been necessary at all to perform the basic full-on/full-off switching.
During these holidays I did a small test. In a set of cheap Chinese Christmas lights with 8-function controller running on mains I used the components supplying the IC (180K resistor charging the low voltage electrolytic (to about 4V) which is then fed to the chip) to trigger the existing SCR controlling one strand of lamps there. I simply put a wire across this board, feeding the SCRs gate with 4V from the IC-supply electrolytic, overriding the built in chip's gate control voltage, and in the end that strand remained steady on while the other was blinking. So, it seems 180K (110K for half-wave current) resistor and an el-cap of couple of mf on 16V (22-47mf) is seemingly all I need to control the SCRs gate (4V gave gate current of about 8mA). I yet have to test it in my original circuit though to see if it will work there with 12V present when it's supposed to keep the lamps off.
Now for that other question - is there a way to use two SCRs fed by either a diode (half-wave rectification) or a graetz (full-wave), to blink two strands of lamps alternatively, but in a cheapest and simplest way possible and with the most affordable components and the minimal part count? I thought about using blink-LEDs to trigger them first, but the problem with blink LEDs is that they have a very short "on" interval. Does someone have some idea how to put this together - the circuit should use simple stuff like resistors, diodes, capacitors, and the like (transistors only if necessary, and no ICs).
Thanks.
 
Last edited:
Top