IR Remote Light Switch when no neutral is available?

Thread Starter

AlbertHall

Joined Jun 4, 2014
12,345
My light switches have only the two wires to the switch - no neutral and no earth connection. Is it possible construct an IR remote switch to connect to these two wires?
 

Alec_t

Joined Sep 17, 2013
14,280
I believe commercial IR remotes manage it; presumably by drawing some minimal current through the load to store enough energy to power the IR receiver and operate a latching relay?
 

Thread Starter

AlbertHall

Joined Jun 4, 2014
12,345
The circuit will also need power to switch off the relay when it is switched on and the voltage across the switch is zero.
 

Dodgydave

Joined Jun 22, 2012
11,284
All uk switches are like this with no Neutrals, must be and old system without Earths!!!,,
the receiver if you find one will be powered through the lamp load to Neutral...
 

Alec_t

Joined Sep 17, 2013
14,280
The circuit will also need power to switch off the relay when it is switched on and the voltage across the switch is zero.
But is it ever zero? Couldn't the remote receiver present some series impedance which prevents that? Could a supercap provide the energy storage? Small transformer powered by the load current?
 

Thread Starter

AlbertHall

Joined Jun 4, 2014
12,345
My best guess so far. The circuitry at the left powers it when the light is off, while D4, D5 powers it when the light is on.
1577716867460.png
 

djsfantasi

Joined Apr 11, 2010
9,156
I believe commercial IR remotes manage it; presumably by drawing some minimal current through the load to store enough energy to power the IR receiver and operate a latching relay?
The circuit will also need power to switch off the relay when it is switched on and the voltage across the switch is zero.
Is battery power a possibility for the IR receiver? When the switch is open, you could steal a little power to recharge the battery.
 

Thread Starter

AlbertHall

Joined Jun 4, 2014
12,345
Any comments on whether the circuit in #7 would work or reasons why it wouldn't?

It has occurred to me that I could easily add a touch switch function to this circuit.
 

Alec_t

Joined Sep 17, 2013
14,280
Any comments on whether the circuit in #7 would work or reasons why it wouldn't?
I assume the mains supply and load are out-of-view to the right somewhere, and that the load is an off-the-shelf mains-rated LED lamp?
How much current does the LED draw?
Will D4 survive the full load current?
Can C1 pass enough current to keep the voltage across C3 fairly constant while allowing enough gate current through R2?
 

Thread Starter

AlbertHall

Joined Jun 4, 2014
12,345
I assume the mains supply and load are out-of-view to the right somewhere, and that the load is an off-the-shelf mains-rated LED lamp?
How much current does the LED draw?
Correct. The light is max 100W. Mains voltage is 240V so current is <0.5A (ignoring PF - no idea what that is for an LED lamp)
Will D4 survive the full load current?
Planning on a 3W 3.3V zener. Expecting max 1.6W
Can C1 pass enough current to keep the voltage across C3 fairly constant while allowing enough gate current through R2?
I plan on a sequence of short pulses (perhaps 200us period) to keep the triac on and minimise average current. A sensitive triac will trigger with 3mA gate current so that sounds good. It will also minimise the leakage current when the triac is off.
 

Thread Starter

AlbertHall

Joined Jun 4, 2014
12,345
Today I decided to check the effect of the leakage current by simply connecting the capacitor across the light switch contacts. I decided that I would check the contacts with a neon tester first and that I had better first check the tester before relying on it.

I have two neon testers and one LED tester (the type with batteries in it). Not one of them works! I measured the resistors in the neon testers and they are both greater than the maximum 20 MΩ that my meter will measure. I presume they are open circuit. I checked the batteries in the remaining one and they were well flat but I knew I had some replacements. I fitted two of those but it still didn't work. When I measured those batteries they were around 150mV. I checked the date code - July 2006!

I have ordered two new testers, one of each type (a new tester is cheaper than new batteries) and postponed the leakage test.
 

MrAl

Joined Jun 17, 2014
11,389
Hi,

We talked about something like this a long time ago i think right here in this forum.
A possibility is that you can use a current transformer to provide some power when the 'lamp' is turned on. The current to the lamp is turned into a voltage that way and that powers the circuit.
Check out AC current transformers for more information.
 

Thread Starter

AlbertHall

Joined Jun 4, 2014
12,345
Hi,

We talked about something like this a long time ago i think right here in this forum.
A possibility is that you can use a current transformer to provide some power when the 'lamp' is turned on. The current to the lamp is turned into a voltage that way and that powers the circuit.
Check out AC current transformers for more information.
But the circuit needs power when the light is off to be able to respond to the remote control. During the summer the light may be off for many days.
The easiest solution to the potential leakage problem would be to use incandescent bulbs but that is verboten now and in any case they are difficult to get.
 

MrAl

Joined Jun 17, 2014
11,389
But the circuit needs power when the light is off to be able to respond to the remote control. During the summer the light may be off for many days.
The easiest solution to the potential leakage problem would be to use incandescent bulbs but that is verboten now and in any case they are difficult to get.
Hi,

I didnt think that the light being off presents any problem because then some voltage is available at the two terminals. The circuit should be made to draw little current at that point so the current though the 'bulb' is very very low at that point.
The other problem is when the lamp is turned on and then the current transformer can supply power. Again though the circuit should be made to use little power.
 

Thread Starter

AlbertHall

Joined Jun 4, 2014
12,345
The TSOP4838 is specified at 0.9mA max but it's also got to feed the PIC and the switch (which is now a latching relay).
 
Top