Hello….I need some help.

I'm developing a DC circuit which will sit behind a wall mounted light switch and perform a function not related to my call for help.

The DC circuit will derive its power from the available AC conductors which exist in the back box at a generic light switch location; these conductors are part of a lighting circuit. The available conductors will be an AC unswitched, an AC switched and a ground - I am assuming that a neutral conductor will not always be available (at least it is not common in the UK) and I do not want my circuit to require a rewire to provide a neutral conductor where one is not available.

I therefore need to generate power from the AC unswitched conductor which I will do so by using a PCB mounted isolating transformer, like the Block VB 1.5/1/6, which I will connect the primary coil in series with the AC unswitched conductor. This will generate a 6Vac supply which I will rectify to DC to power the DC circuit.

In other words I’m aiming to ‘harvest’ energy from the AC lighting circuit to power a DC sub-circuit which operates independently of the status of the AC load.

I have a few sticking points I am seeking a bit of inspiration to move my design forward:

1. The AC supply is intermittent such that power is only available to harvest when the light switch is closed. When the AC supply is available (light switched closed) the DC circuit will store energy in a battery however when periods when the AC circuit is not available (light switch open) the DC circuit will consume more power than can be held in the battery. I have not been able to come up with a way to enable permanent AC supply irrespective of the status of the light switch therefore have I missed something blindingly obvious as to how I can derive a permanent supply without wiring in a neutral? In my mind this must be possible as this product by LightwaveRF (http://lightwaverf.com/product/400-dimmer-lw400/) powers an electronic lightswitch, dimming electronics and a wireless transceiver from the AC unswitched conductor without a neutral.
2. If the isolation transformer approach is a feasible one then as the primary coil is in series with the AC load the full load current will pass through the primary coil. I am concerned that the primary coil will burn out when under full load, is there a way I can calculate the current rating of the primary coil?
3. An alternative to subjecting the primary coil to the full load current would be for the load circuit to bypass the primary coil therefore the bypass would carry the full load and the primary coil circuit would only draw the power it requires. However as there is no neutral present would bypassing the primary coil just in effect short both sides of the coil together with a voltage waveform of the same potential and phase?

I may be doing something completely unfeasible however I also know that the development of my circuit is a progression through consideration of many good and bad ideas therefore I would really appreciate your thoughts on the above.

I’m not a newbie to electronics however I have little practical experience of what works in the real world so I’m looking to you guys for some guidance to help me move one step closer towards my proof of concept. These are honest queries and I really would appreciate open and honest responses.

Many thanks,

Neil

 

If you are in the UK, the light switch will have only a permanent live and switched live,earth no Neutral, so you wont be able to get your circuit to be powered up.

Your only option is to use the earth as a neutral, which I dont recommend.

The dimmer circuit uses the bulb as a neutral feed to reduce the voltage across it using a triac.

 

Dodgydave - thanks for your quick response.

Yes - I'm based in the UK, could you expand on your comment about the dimmer circuit? I don't fully understand what you mean.

Otherwise I've been looking around and found a wireless switch from Lutron, schematic attached, which derives its power for its wireless receiver in the same way...

 

Dimmers work on the principal of phase delay triggering, they use a triac to pulse the mains on by delaying the trigger to the triac, the later the trigger the lower the voltage across the bulb, the earlier the trigger the largest voltage is across the bulb, it takes chunks out of the mains supply, which causes a lot of spikes to be produced.

 

What if ... mind you I'm speculating... a 2uf connected the primary of the isolation transformer to ground?
Would that be possible?
Most likely against a thousand rules as well.
Has anyone ever tried something like this?
I don't even know if it could work. Or would you put the cap in series with AC hot and tie the primary between it and ground?

 

It would have to be a capacitor rated for AC. In the UK you use 220 - 240 so 350 or higher VAC rating. Motor run cap maybe?

Oh and you need a fuse in line with the primary if you can do it thus way at all.

 

Dodgydave, thinking on your point a bit more - would it be possible to 'restrict' the current to the downstream lighting load while the lights are not required to be on?

Could I consume the mA I require for my circuit direct from the source and restrict the series current passing through to the lighting load? Rather like a flow restrictor in a pipework design?

If my circuit consumed say 100mA then could I restrict the maximum current the circuit can draw to just a bit over the 100mA? I could therefore design the circuit to always be live and therefore have a permanent connection to neutral but such that there is not enough power being delivered to the load to enable the lights to illuminate, rather like the dimmer triac. When the lights are physically needed to be on I could bypass the restrictor and permit the full load current to circulate.

 

Kermit2 - I am not sure I follow your thinking.

If I connect the primary to ground via a 2uF capacitor then this would provide a return path for current to flow, and therefore allow my circuit to be permanently live but in the UK it is not recommended to use the ground conductors as current carrying conductors outside of fault conditions.

Or are you talking about generating a phase shift by using the motor run capacitor, which could be used as a reference supply to the hot supply?

 

You still need a neutral feed in the light switch, you cant power your circuit without it.

 

If you can't use a ground connection the you will need the neutral as Dave explained.

You need two connections for power to become available.

It is why batteries have two terminals. Either use ground or run a neutral wire.

 

The neutral is available at the lamp, not at my location.

I have access to a complete circuit, and therefore a neutral, when the light switch is closed (assuming that a lamp is fitted in the light).

Does that make any sense?

 

I feel like we're in a loophole here, unless you run a neutral feed to the light switch, OR use the earth as a neutral you cant make a 230V circuit with one wire.

 

The TS attached the circuit in another thread and I attached it at here.
Current flow restrictor in an AC circuit.

 

Why are you putting your transformer in series?

 

 

BR-549 - would my initial circuit diagram not work?

I assumed the primary coil would act as a resistive load therefore the primary coil and lamp would act as resistors in series, the issue as far as I'd see it is that the circuit would act as a voltage divider therefore voltage would be split between the transformer & lamp.

 

If my circuit consumed say 100mA then could I restrict the maximum current the circuit can draw to just a bit over the 100mA? I could therefore design the circuit to always be live

That would be hazardous. Anyone assuming the switch is 'off' and handling the switched-live wire could receive a fatal shock.
BTW, I would guess the Lutron device keeps its radio alive by passing a few mA through the ground connection. Anything more than 30mA or so would trip the ELCB (you do have one, don't you?).

 

lesneypark - What is it you're trying to achieve on the 6V side of things?

With the switch Off, If the load ( Lamp ) of the lighting circuit is a incandescent lamp then there will be some difference in potential between the Line and the Switch Line as the lamp will act as a path to neutral.

With the switch On, both Line and Switch Line are at the same potential so there will be no
difference.

You shouldn't use the Earth or as a " return" path. This will have complications if you have any high resistance on any other circuits in the house and can potentially be lethal. It'l also cause the RCD to trip if you didn't limit the current to under 30mA.

 

That would be hazardous. Anyone assuming the switch is 'off' and handling the switched-live wire could receive a fatal shock.
BTW, I would guess the Lutron device keeps its radio alive by passing a few mA through the ground connection. Anything more than 30mA or so would trip the ELCB (you do have one, don't you?).

I'm with you on the always live scenario being a hazardous circuit design and I am sceptical of its feasibility however it would be good practice when working on the circuit to isolate it at the MCB not at the switch would it not? Although I do appreciate that people do things differently.

Yes, I have an ELCB and I guess that if the earth leakage from the Lutron device is within permitted limits whatever they might be then I can understand that approach.

While I really do not want to use the ground connection, in my scenario I will be consuming approx 100mA @ 3.3Vdc = 33mW. If I then complete this circuit using the ground connection then the current in the ground conductor would be 33mW / 230V = 1.4mA.

Would this be more than a typical domestic appliance like a fridge/freezer?

 

lesneypark - What is it you're trying to achieve on the 6V side of things?

roady_carl - I will be rectifying the AC into a DC voltage then stepping it down to 3.3V to power a wireless MCU which amongst other things will switch on/off the lighting circuit actioned by a wireless switch, sensor and/or laptop/smart device.