1 sec ON when High (3V), 1 sec OFF when low (0V) - no relay!

[alec_t]

these caps, why they are connected that way and not in paralell with the power?

To provide the correct timing, such that the switching action occurs immediately there is an Ethernet pulse edge.

what exactly the diodes are doing?

They provide a rapid discharge path for the caps.

 

[Ibob]

Alec_t, THANKS A LOT!
Here you can check how it works:
https://youtu.be/W4kBuB4gDKM
If I need to change the timing what would you suggest? Changing the R or the C?
Awesome!

 

[alec_t]

Nice to see it working.
You can change R or C, whichever is easier. If you have some spare Rs, for example, you can use them in series or parallel to change the effective R value.

 

[Ibob]

Nice to see it working.
You can change R or C, whichever is easier. If you have some spare Rs, for example, you can use them in series or parallel to change the effective R value.

Actually if I decide to change R, should I keep the 1meg resistors which are in parallel with the diodes or I need to change their values as well? Thanks!

 

[alec_t]

You can keep the 1meg values.

Edit: Sorry, I was confusing the various resistors. You can keep the 1meg values for R3 and R4, but you would change R1 and R2 to adjust the timing.

 

[Ibob]

You can keep the 1meg values.

Edit: Sorry, I was confusing the various resistors. You can keep the 1meg values for R3 and R4, but you would change R1 and R2 to adjust the timing.

Great, thanks.

Now I have a PROBLEM.
The ON and OFF buttons of a channel on the remote have a common Vcc (+4.6V) and there are GND for the ON signal and GND for the OFF signal:


Our circuit has a common GND. I believe I cannot use it that way, right? Maybe I need to have a common power and 2 GNDs? PNP transistors or what then should be used?
Another issue is that, when the pulse overs, if the remote is still transmitting the signal it gives back the voltage to the circuit and the pulse actually is prolonged a bit, at the same time, the other transistor activates, i.e. can i connect some diodes in a way that no current goes from the remote buttons back to the circuit?

Many thanks in advance!

 

[alec_t]

I can't believe the remote's switches are wired like you show. Closing a switch would short the 4.6V supply directly to ground and let out the magic smoke.

 

[Ibob]

I can't believe the remote's switches are wired like you show. Closing a switch would short the 4.6V supply directly to ground and let out the magic smoke.

Yep, it is exactly like that, but of course, there is somewhere in the middle.
Imagine a kind of resistance or something:


Then why not to be possible.
What if your circuit is modified in a way that the Vcc is the common wire, and PNP transistors instead of NPN are used?
As far as I know PNP would allow switching the GND to Vcc? But I am not sure exactly what I am going to do with the HEX triggers, RC, etc...

p.s. When I tested the way it is.... I realized that the remote couldn't switch the receivers... I thought it is dead. But this morning it was working again... with the buttons, i never tested again with the transistors as the polarity is different.

Cheers!

 

[alec_t]

I have been reading your pic as showing a 4.6V supply. Are you saying this 4.6V is 'somewhere in the middle', and that it is the result of a voltage drop (from the remote's 12V supply?) in 'a kind of resistance'?

 

[Ibob]

I have been reading your pic as showing a 4.6V supply. Are you saying this 4.6V is 'somewhere in the middle', and that it is the result of a voltage drop (from the remote's 12V supply?) in 'a kind of resistance'?

If I got you correctly, then the answer is as follows.
The remote operates with a 12V battery. Inside I can recognize a voltage regulator with capacitor before and after, producing 5V supply. All the rest is some ICs which I cannot really say what they are doing, also some smd elements... and eventually I think this 4.6V is a voltage drop, yes.

I really would like to use the transistors as switches, because going with optocouplers/relays would take too much space and it will be more expensive. Also it will not be so original and smooth.

 

[alec_t]

Useful info re the 5V voltage regulator. Can we use that 5V to power my circuit (i.e. can you make a connection to it)?
I'm sure we can adapt my circuit to do what you want, but first we need to determine exactly how your remote's switches are wired. If you check your latest pic you will see that both 'ON' and 'OFF' are permanently connected to GND. That can't be right

 

[Ibob]

OK I will try to follow as much as I can the pcb layout (when back home) and i will post here more details.
No problem to take the 5V out of the regulator output. Great idea to use this to power your circuit.

And for the ON and OFF being permanently connected to GND, that's maybe because I cannot depict the situation properly.
What I know for sure is that the ON and OFF "lines" are low, compared with the common wire of each pair of on/off buttons.

I will see what I can do to give more info on the remote circuit.
Thanks!

 

[Ibob]

I am away from home and still didn't look at the remote buttons in details, but I was googling around..
Maybe we need to go with a kind of such pnp switching:

Then, I guess I would need 3 instead of 5 inverters and thus one IC would drive 2 channels?
Am I in the right direction?

 

[alec_t]

PNPs are part of the solution, but we need to know more about how the switches/buttons are configured before we can interface the PNPs to them safely.

Am I in the right direction?

There's a bit more to it than that. One factor is that the Ethernet signal is 3.3V whereas the remote has a 5V supply. We don't want 5V getting back via the PNP to the Ethernet device.

 

[Ibob]

Hi alect_t,

OK, it is not that easy to follow the circuit of this remote. Everything is connected to a IC (unknown for me).
On top of this, there are buttons on the remote which switch ON/OFF all the 4 channels, and these buttons have the oposite polarity, e.g. the common wire is (-) and the ON and OFF wires are (+), while the other 4 channels are as described before - common (+), ON/OFF wires - (-).

So I have decided to use a low current reed IC relays (very fast, almost totaly silent, etc.). Thus the module I am building here will be more universal and I will be able to use it for other purposes/drives, etc. For instance I am going to use another remote controlling not RF Sockets, but RF wall switches.

So I've connected the two relays to the NPN transistors and your circiut works perfectly.
BUT, when I press the button (this button is only to test), the ON line and ON relay are producing maybe 10 very fast switching when the time of the signal overs. The OFF relay works perfect - it never produces more than 1 switching. This behaviour is the same also if no relays (the LED is kind of bliking very fast for a short period at the end of the signal).

How to solve this? Is it something related with the hysteresis effect? Do we have hysteresis effect with HEX TRIGGER at all?
Maybe I didn't connect something properly, or how to check this? BTW, the issue is only seen after 1-2 cycles, i.e. after several seconds of idle, the 1st trigger is always good.

Thanks for your time and apologies for the delay but I was away from home these days.

 

[Ibob]

The above issue has been solved when I've connected the unused trigger input directly to GND.
Continuing with the remote pcb analysis...

 

[ucok]

Hello Ibob,
Check this blog [MODNOTE]Deleted Non English Text[/MODNOTE]

 

[Ibob]

SOLVED !!!

Big thanks to alec_t !

The project:
1. The hand held remote control is powered with 12V from the Ethernet controller.
2. The Ethernet controller has 16 logical outputs: 0V - 3.3V (constant signals for ON/OFF).
3. The 'Constant to Momentary' circuit proposed by alec_t is powered with 5V from the remote control, as there is a 5V regulator anyway.
4. I've modified the circuit a bit so that there will be a LED to show the current ethernet status (light = ON), and there are led for ON momentary and OFF momentary time.
5. I've decided to use reed relays in IC (2 relays per channel). Silent, low current. I've decided this because if I go with PNP or NPN transistors it will not be so universal, e.g. I am planning also to use another remote control and also, the 'ALL ON/OFF' buttons are wired the opposite way, i.e. for NPN and not for PNP transistors. The relays however are powered from NPN transistors and the 12V rail.
6. Finnaly I am able to switch these RF sockets coming with the remote via internet and the iPhone..

I am now close to finalize another part of the project, that is to use not only RF sockets, but also RF receivers, built in the wall bulb switch. If you are interested, check the progress here:
https://www.electro-tech-online.com/threads/transformerless-power-supply-with-only-live-wire.134152/

 

[alec_t]

Glad you've got it working.

 

[Ibob]

And that's the finished module - 5 channels, controlling one remote: