First thing to do is add filter capacitors to all circuits. Look at your Latch circuit with the NE555. Do the same for all the other boards. Add the 100 ohm resistor and 100uf cap.
Should also connect a 10uf cap from ground in the LM3914 circuit to the anode side of the LEDs.

 

Is the 5V dropping when the relay overloads it when it is turned on? Then the relay turns off because the voltage is too low and the unloaded voltage rises and goes on and off over and over?

Does the relay vibrate at the 50Hz or rectified 100Hz of an overloaded 5V power supply?

 

Is the 5V dropping when the relay overloads it when it is turned on? Then the relay turns off because the voltage is too low and the unloaded voltage rises and goes on and off over and over?
Does the relay vibrate at the 50Hz or rectified 100Hz of an overloaded 5V power supply?

I have no idea of the cause !
I am feeding 5V. That's it. I think is some sort of LC circuit here. But at a larger scale.
I tested it on my working bench and is doing the same. So, no mains interference at all. It is not influenced by the live wires this time.
If I disconnect the 3'd relay from circuit, everything works fine.
If I put 5V on the 3'd relay, is working fine.
My PSU is showing me 1A , sign that something is in short. But Nothing is in short. It behaves as is in short. Only when I connect that 3'd relay. It is weird as hell. I checked for short with my multimeter and nothing is in short. But once I power it up, it goes to 1A straight.
Actually I tested again right now and is not rising to 1A but to 1.44A ! Way over my custom limit that I set on my PSU . Very-very strange.

 

It was a logical error. Im too tired of it probably.
I solved the problem. It was a short after all, but a logical one. Eh... stuff happens.

 

@sghioto Can you make [Signal Amplifier Module] keep the relay1 close ?
Right now, as you had observed as well, is keeping it open.
Thank you.

 

Can you make [Signal Amplifier Module] keep the relay1 close ?

Depending on what you mean by Open and Closed.
When a relay is activated it's contact is Closed.
So I'm assuming you want Relay I not activated, Open contact, until #10 LED is ON, is that correct?

 

yes... I didnt thought on other meanings.
Indeed I mean the relay power to be off when nothing is activated.
When #10 LED is on, relay 1 goes ON . Basically is in sync with led 10. And is healthy for the relay, that's why I care.

 

Change the NPN to a PNP.

 

Bravo. It is working as I wanted.
Thank you !

 

New update:
The movie is very short (3m40s)
Please watch it, I explain in it what I did and how it works.

 

You need to read what I suggested in post #241.

 

another update:
I made a test for the LM3914, then I made a new test on the workbench and on the wall.
I made a manual switching module with a pot and is working fine.
I was afraid of further interference on the wall, but is normal now with the pot.
I am curious what other sensing modules you can think I should try.
Please enumerate all that you have in mind. I have already mentioned myself a few, in the movie. I don't care for originality but I care for quantity this time. So, go nuts.
Warning, this is a lengthy movie, 23m.

 

This is a hard project. But we don't give up until we find a solution.
- I want another sensing circuit !
Until now I had this one built with 2N7002. It worked fine on the table, but near the live mains, it gets strong interference.
So it is not good anymore. I showed in my last post here a video where I changed the sensing circuit with a simple 10k potentiometer and it worked, on the wall, exactly as on my testing bench, very good, very optimal, all leds full lit, just excellent. I tested if the mains 220V were affecting the entire circuit board, and is only affecting the sensing circuit alone. So that was a relief.
Now I can think on alternatives for another sensing circuit.
Mister @dl324 , @Audioguru again ,@sghioto you are my best friends here, you did a great job and I thank you. My wish is to push it further. Obstacles like these happens and is hard for everyone, especially me. But me alone... is impossible to make too much of it. But with your help i have greater chance of success. And don't get mad on me if I'm not trying something you just said, because I work slow and alone and I have also to decide what to do next. My goal, as yours as well, is to make it work, it's all that matters for us, correct? In the end, when all the best ideas fails in practice, I will come back to the older ideas that i skipped. I am methodical as I can.
I want to try next, the theremin circuit. It is a capacitive sensing circuit and it should not interfere like the FET was. The downside of it, it is a big circuit to make, way outside my original plan. But I dont care of my original plan anymore, I only care to make it work.
The next idea is to use IR. It is probably the best idea, but I am also using IR for the remote control (it is an option for that as well).
And that's it. I only have these 2 other options to try. I cant think on something else to use for the sensing circuit.
It might be another way with a coil of some sort, but I didnt get too much into it and for now is just a vague idea for me. And I think it will be pretty much like the FET , in functionality.
Thank you very much and I hope to hear from you guys.
---
Someone else on another forum had a very interesting idea to add to the existing FET circuit, a "low pass filter" and a "peak detector". They sound very interesting but I never use them before. So what you think about them and how to actually implement them? But this FET idea was already eating my time and my nerves, way beyond my plan. I think I should skip it and concentrate on other sensing circuit options. What do you think?

 

Capacitance sensor might work.
How sensitive do you need?
Generally they can sense a finger or hand like the theremin

 

So you're not interested in revisiting the light sensor. I got excellent response using a 2 inch diameter solar cell as the sensor.

 

Someone else on another forum had a very interesting idea to add to the existing FET circuit, a "low pass filter" and a "peak detector".

As I understand it, the problem is line voltage pickup when you don't want it. So a low pass filter wouldn't help.

I don't see how a peak detector would help because you need your hand to couple signal to your antenna.

I think your best bet is to work on the sensitivity of the amplifier and shield the antenna and other sensitive circuitry from picking up line frequency.

 

If you intend for your circuit to work in the dark or at varying light levels, detecting light level will be a non starter.

 

You need to reduce the length of the unshielded input wire that is the antenna that picks up interference so that only your body near it gives it the interference. You need a rectifier and lowpass filter (or peak detector) to convert the interference picked up by your body to DC for the LM3914 input.

 

First of all, thank you all for responding !

Capacitance sensor might work. How sensitive do you need?

Well, as sensitive as I can make it, at maximum, from 10meters if is possible. Then I will tune it down to what distance I need it. At least that is my plan.

So you're not interested in revisiting the light sensor. I got excellent response using a 2 inch diameter solar cell as the sensor.

I am not saying im "not" doing something. Is that Im balancing what is the best to use. And cool. Since im an artist, i like cool things. And also what I have at my disposal.
I am expecting some solar cells as you told me long time ago. See? I listen to you, even if I dont tell you. This is what I got: Mini Solar Panel But they didnt arrive yet.
The light detectors in general are influenced by daylight that is full super bright especially on that wall for 2 hours , around 10 AM in the morning. I think there is a way to modulate that signal somehow and respond only to a coded light source/reflection. When those solar panels will come, I will announce you. This is another option, nevertheless. And again, if other ideas will fail, this will be next to try. I probably will try it anyway, as an experiment. We will see when I will get the things.

As I understand it, the problem is line voltage pickup when you don't want it. So a low pass filter wouldn't help.
I don't see how a peak detector would help because you need your hand to couple signal to your antenna.
I think your best bet is to work on the sensitivity of the amplifier and shield the antenna and other sensitive circuitry from picking up line frequency.

and

You need to reduce the length of the unshielded input wire that is the antenna that picks up interference so that only your body near it gives it the interference. You need a rectifier and lowpass filter (or peak detector) to convert the interference picked up by your body to DC for the LM3914 input.

Audioguru and dl324, you said it more nicely than me. This is how I formulate it in my head (its the same idea as yours):
I dont see how to separate these 2 signals either. 1 strong signal from 220, and 1 weak hand signal. Somehow, to filter the 220 signal and exclude it and then leave only my hand signal "visible" to the FET. But how to do it, is way out of my knowledge and experience. You guys are more fit to make it work than me, if possible, thats why I mention it. But you find it weird as well. I think some radio - amplifier stuff will be involved here, my wild guess. It's probably possible with some state of the art measuring tools, but I dont have those.
And you both are saying that I should shield the sensitive circuitry from picking up the line frequency. And WE did a step towards it by grounding that metal shield. How to shield it further? It is a mistery. Someone from another forum give me a suggestion of using a thin lead sheet. It's a very interesting idea. I will try other sensing circuits for now, but I will keep in mind this option as well. Its too much money for me, thats the number one reason.

If you intend for your circuit to work in the dark or at varying light levels, detecting light level will be a non starter.

It will be concentrated very hard light on that wall area, in the morning around 10AM, especially summer. Direct sun expose and very intense. That's why I am reluctant to use even the IR. But.. I will give it a try with IR, especially a modulation of its signal. I'll have to experiment it when all the components will arrive. I am also having to insert the remote control receiver module there, and that is another pain in my but, what kind of signal to use. My original plan was to use this FET for sending and IR for remote. But... now I have multiple versions for remote as well, so I start to not be so picky anymore.

I received already this 433Mhz and this dtmf . I didnt do anything with them yet since I will let this remote module to the very end. But I am mentioning it because is influencing what I choose for the sensing circuit.
I am also thinking to use 1 dtmf board to use in both situations, 1 as sensing circuit, and 1 as remote. Both transmitting a specific modulation. This way it can be built with a simple LDR and a led light reflected by my hand, (and also from a remote) all modulated to a specific rithm that dtmf will decode from both sources. Its a bit complex but I can do it. I am also expecting some PIC microcontrollers to arrive as well, so I cant start anything yet. But is an option.

 

Why not look into an ultrasonic position sensor and circuitry.