Transformerless power supply with only Live wire

[MrAl]

Hello,

I gave you a simple circuit with a single diode, cap, and that was it. What was the result of the test?

 

[alec_t]

What is the current draw of the RF receiver and relay?

 

[Ibob]

What is the current draw of the RF receiver and relay?

less than 6mA standby, plus operating the relay
(Check above posts for more info and link)

 

[MrAl]

Hi,

When do you plan on doing that test?

BTW what alec says about the voltage going high is valid so again be careful. That is if the triac goes open which it very well could. So use a cap with a significant voltage rating.

 

[Ibob]

Hi,

When do you plan on doing that test?

BTW what alec says about the voltage going high is valid so again be careful. That is if the triac goes open which it very well could. So use a cap with a significant voltage rating.

This weekend.
Which cap you are talking about? The DC one?
Is it ok to use a switch for the test instead of a triac?
Thanks!

 

[MrAl]

Hello,

The cap that connects to the diode in my drawing.

Absolutely not, you must use the circuit as is because that is what the test is for: to determine if the real circuit can put out any small but usable voltage when the triac is on, but see the triac being on is not the same as a switch being on. The circuit that they have designed may be able to put out some voltage even when it is turned on because it isnt turned on all the time but there may be a pause which lets some voltage get to the circuit. This is what we need to test as this could save the day

 

[Ibob]

Hello,

The cap that connects to the diode in my drawing.

Absolutely not, you must use the circuit as is because that is what the test is for: to determine if the real circuit can put out any small but usable voltage when the triac is on, but see the triac being on is not the same as a switch being on. The circuit that they have designed may be able to put out some voltage even when it is turned on because it isnt turned on all the time but there may be a pause which lets some voltage get to the circuit. This is what we need to test as this could save the day

Ok MrAl,
I assume that cap is the blue one? The DC one? Is a DC one right?
Because the diode is connected to two caps and I want to be sure which one you are referring to.
Also, in order to trigger the triac I need to connect somewhere the Gate of the triac to. Could you please advise me how to finish the circuit that you gave me, as there are some dots (not finished) I am not sure how to finish it. Thanks for your patience.

 

[MrAl]

Hello,

Well no the triac is supposed to be the triac in the circuit you have already, not one you are building. That's the whole purpose here, to measure the actual circuit. You need to use the circuit you purchased for this test, that's the only way.

The cap could be a DC cap but it should have a good voltage rating like 500 volts so you may want to use a smaller cap just for this test. All we have to do (assuming that's the actual circuit schematic and not something you just made up) is measure this voltage so we dont need any current yet.

We need to get on the same page here...

You only do tests on the original circuit you already purchased. You do that because we are trying to tap energy from that circuit to drive your transmitter.

If you are designing your own circuit that's entirely different. Are you saying perhaps that you are designing your own switch circuit and using just the RF receiver?

 

[Ibob]

Hello MrAl,
Now I see.
So I want to build an entirely new (own) circuit and not to use the circuit I've purchased and shown above.
I've opened it to see how they did it. When I measure the biggest DC cap on their circuit it shows 19V, when I measure the smaller DC cap - 6V. That is exactly what I need because the RF receiver will need 12V and the triac control elements - 5-6V. Their circuit provides these values no matter the triac is on or off.
So if I want to use it, I am ready to do so. But then I need to purchase these sensors just for the supply which will be too expensive at the end.

Yes, I am saying that I want to design own switching circuit which will power the RF receiver (I am waiting to arrive soon I hope - from China).

My final idea is to build a mock up that works and then to step in to "production" as I am planning to change all wall switches with this solution.
There will be a button on the wall - push ON, push again OFF - for that purpose I will use 555 IC to go from momentary to constant. This constant signal will drive the TRIAC via optocoupler. The same button will be drived by the relay of the RF receiver (momentary as well). All of these I will put together at the junction box, replacing the regular hard switches. Thus, I will be able to control the lights manually (button) and remotely.

Actually when I get the RF receiver I will check if I can get rid of the relay at all and just pull out the signal directly to drive the triac. But thats easy and not mandatory.

So if you look back at my last circuit posted... (prepared in Word what do you think?

 

[MrAl]

Hello,

Ok well that will require a lot more circuitry as you must know.

Using your own triac circuit, you'd have to design a phase control circuit for the gate. For example, a 555 timer that triggers the gate at a time that is somewhat after the zero crossing but not at the zero crossing. That way we always see at least some voltage across the triac, even if it is smaller than the line voltage.
The circuit has to be powered from the line voltage, and then later when the triac comes on it has to get it's power from the tap off circuit with the diode and capacitor as i drew previously. In this way your circuit will always have power and the 'bulb' will see most of the waveform cycle but not all, but the part it does not see should be small enough so it doesnt affect the brightness too much.

Does this make sense so far, and do you have any ideas yet?

 

[Ibob]

Hi MrAl,

Then I guess I need to use a non-crossing zero optocoupler like MOC 3023 instead of MOC 3041 I have. 3041 is a crossing zero one.
As far as I got it this zero crossing is problem? Sorry I cannot understand exactly what problem. It did powered the TRIAC when I've tested the button solution, powering from a 12V battery.
This 555 IC is only to convert the button press to constant and vise versa, nothing else.

Maybe I should put the whole circuit diagram here so we discuss. Ohhh, so much trouble for just lighting the light..

Actually what you are saying.. the bulb will be off every half period because of this crossing zero or I cannot understand.

 

[Mr RB]

Hello Ibob.

Exactly what is the project meant to do?

You have said it needs to connect to just the two wires in the light switch wiring, but can you tell more about the project?

How many lights will it switch (volts, watts) and what controls the switching? Those things will make a lot of difference to the design.

 

[Ibob]

Hello Ibob.

Exactly what is the project meant to do?

You have said it needs to connect to just the two wires in the light switch wiring, but can you tell more about the project?

How many lights will it switch (volts, watts) and what controls the switching? Those things will make a lot of difference to the design.

Hi RB,
OK, I will start over.
The project is to have the posibility to remotely control the bulbs at the ceiling at my apartment.
At the same time, one should be able to manualy switch on/off the bulbs from the light switch on the wall.
DIMMER FUNCTIONS ARE NOT REQUIRED.
The requirement is that no cables should be layed. When I say remotely I mean not with the remote control comming with the receivers, but from Internet via SNMP (or/and WEB). I already have the Ethernet controller - it drives relay board with up to 16 relays. I can control them with SNMP commands or via web.
Then I will use a RF transmitter/receiver to send signal to the wall ligth switch (junction box). The receiver is small enough to fit in there and to control the bulb.
The receiver triggers momentary signal (the relay of the receiver is on while the remote control button is pressed), so I will need to build a transition circuit "momentary to constant" - that's clear how to do with 555 IC. I will need to replace the standard 2 position wall switches with a button-switch, as this will be the manual ON/OFF button. The signal from 555 momentary to constant switch will switch a TRIAC via MOC3041 optocoupler and the light will go ON/OFF.
Button press - reverse the state of the bulb. Reveiver pulse - reverse the state of the bulb. That's it. BUT...
The receiver needs 12V power, 6mA, I think the MOC3041 needs 15mA or so... so I need to have 12V DC inside the wall ... in the junction box - where the light switch will be. Fine... no problem.... if I live somewhere else, but here in Bulgaria we usually have the Neutral going in the ceiling and only 1 wire - the Live one goes in the light switch junction box. So my problem is to supply the reveiver and the other elements from that wire, and to use the wire to switch the light on/off. Look at the posts above... I already have some ideas and also with the help of others I think I will be able to achieve it.
For sure it is possible as I have sensors already working that way - these are just PIR sensor and put the light on for a short time... see pictures of these in previous posts.

I am available for questions and to receive ideas

 

[Ibob]

OK, based on the alec_t considerations, I've modified the circuit by adding a zenner diode and R1 (fat one).
Instead of the zenner diode I could use regulator, but I guess for the RF receiver driving a relay and the 555 IC + optocoupler elements driving the triac I don't need such, right?

View attachment 73103

I have tested the above circuit.
When the switch is On, bulb goes On but no voltage at the end.
When the switch is Off, I have 20 or 5 volts, depending on the zenner I put (i have such zenners).
But if I connect the MOC3041 in order to drive the triac instead of the switch the voltage drops to about 3 volts or 1,5 if I use the 5v zenner.

Thus, the optocoupler cannot drive the triac.
What can I change? Resistors change/remove, or use regulator instead of/after the zenner?
The DC cap I used is 50v 100uF.

I hope if triac is connected instead of the switch i will have voltage?

From MOC304x datasheet:
"All devices are guaranteed to trigger at an IF value less than or equal to max IFT. Therefore, recommended operating IF lies between IFT2.(15 mA for MOC3041, 10 mA for MOC3042, 5 mA for MOC3043) and absolute max IF (60 mA)"

 

[MrAl]

Hi MrAl,

Then I guess I need to use a non-crossing zero optocoupler like MOC 3023 instead of MOC 3041 I have. 3041 is a crossing zero one.
As far as I got it this zero crossing is problem? Sorry I cannot understand exactly what problem. It did powered the TRIAC when I've tested the button solution, powering from a 12V battery.
This 555 IC is only to convert the button press to constant and vise versa, nothing else.

Maybe I should put the whole circuit diagram here so we discuss. Ohhh, so much trouble for just lighting the light..

Actually what you are saying.. the bulb will be off every half period because of this crossing zero or I cannot understand.

Hi,

yes post the circuit we need to start working with some concrete ideas or we will be here for the next two years

The triac turns on EVERY half cycle, it does not skip half cycles. Let me explain.

The triac normally turns on at the zero crossing. That means the entire half cycle gets to the bulb. But if we do that, then we dont get any voltage for the RF circuit when the triac is turned on. So we need to modify that so that we can get some voltage from across the triac even when it is being turned 'on'.

The way the triac works is it turns on at the zero crossing and stays on, but that means very very low voltage across the triac (like 2v max). By delaying the turn on time to a place just AFTER the zero crossing, that means the sine voltage rises up and stays up across the triac for that small delay time. We then tap off of this voltage with a diode and capacitor and that gives us some voltage to power the RF circuit with.

Now using a full wave bridge is a nice idea, but we probably need a common ground for the triac and the RF circuit so that the RF circuit can energize our triac circuit properly, so we probably have to stick with half wave. But we still have to prove that this works without delaying the triac turn on too long or else the bulb will start to dim too much.

 

[Ibob]

That's the plan and what I believe should work with some maybe adjustments?

 

[MrAl]

Hi,

Here are some corrections:

 

[Ibob]

Thanks MrAl,
No need of fuse?
Looks my MOC3041 is not working anymore. Maybe I burned it.
Should I buy non-zero crossing otocoupler or zero crossing?

 

[Mr RB]

...
* The project is to have the posibility to remotely control the bulbs at the ceiling at my apartment.

* Then I will use a RF transmitter/receiver to send signal to the wall ligth switch

* The receiver needs 12V power, 6mA, I think the MOC3041 needs 15mA or so... so I need to have 12V DC inside the wall ...

Thank you for that. So you need a circuit that delivers 12v at 30mA or so, whether the lights are ON or OFF, powered entirely in series with the light bulbs.

It's not going to be that easy. When the lights are OFF you need to extract 12v DC 30mA from a 230v AC source.

When the lights are ON you need to extract the 12v DC 30mA from the lighting current itself (using a 12v AC power zener which must be capable of handling and *dissipating* the full lighting current at 12v)!

OR you need to extract 12v from the beginning of the mains halfcycle, and charge a large cap, then the lights can be ON for the remainder of the mains halfcycle. If you choose that way, the problem is that the same circuit must be able to work safely etc when the lights are OFF and it gets fully exposed to 230v AC.

 

[MrAl]

Hello again,

MrRB:
Thanks for participating in this thread. You might spot something that helps with this task too like the two levels of power being very different.

Ibob:
Why ask about the fuse? It's still in the drawing isnt it? So it's still there Break the connection at the red "X".
Yes as i said several times now we can not switch at the zero crossing if we are going to test this method of extracting power from 'across' the triac when the triac is what we are calling 'on'.
And as MrRB points out, we will see two different levels of power available which are wildly different: one comes from 230vac directly, and the other comes from maybe a 30v spike of power which will be much much less. Thus, any power supply we build must be able to handle both extremes and still have enough power to drive the RF device.
Since you indicated that the RF device has a (and i quote) "relay", that means we can in fact use a full wave bridge rectifier which does indeed help when the power come from our power spike.
So a little more detail about the power supply is in order, and also the 555 should be a CMOS type.
The power supply looks drawn correctly for the most part, with a capacitor being in series with the input so that reduces the power during the 230v input mode of operation, but we'll have to see if we can use that same circuit for the time when the triac is 'on'. We may have to modify this.