Motion triggered battery powered dimmable COB LED light

[NiculaeGeorge]

My view on life, in any field, is not to produce more but to consume less. Everybody told me this is a very eccentric weird mentality. Perhaps they are right. Nevertheless I have to build my project to make a highly efficient, extremely low drain, motion triggered 12V battery powered dimmable COB LED light for a room. That’s a mouthful I know. Sorry. I have STPD and paranoid fear of not being understood. My battery is charged by a solar panel. But that’s not important for now. I don’t want to use switches, potentiometers or relays. Let’s eliminate those from the start in any design. I also want to consume minimal power. When I enter the room I want the light to automatically turn on. When I leave the room, after 1-2 seconds the light turns off by itself. I don’t want any switches or potentiometers, If light too bright or too dim, I want a touch sensitive surface in rare occasions to control the dimming. The entire assembly will be mounted in a box on the ceiling of the room (except the battery of course).
In my mind there’s this logic diagram with 4 basic minimal components:

​

• 12V battery
• PIR motion sensor HC-SR501 (with BIS0001 IC and a HT7133-1 3.3V 30mA voltage regulator onboard) . The module has the jumper set to repeat trigger mode. Has 3 connecting pins: ground pin, input voltage(4-24V DC) and Output pin with a 3.3V TTL positive logic output, LOW=no motion, HIGH=motion has been detected). The sensor will have to be continuously powered forever. I’m ok with that, since it draws under 50uA quiescent current. Pot1 CW direction increases ONTIME when motion has been detected. Pot2 CW direction increases the area of detection (from 3 up to 7meters) Datasheet here> https://www.mpja.com/download/31227sc.pdf
• FC-106 Touch LED dimming PWM light regulation module with onboard SGL8022W IC. The onboard SMD LED is removed. The onboard S8050 Y3J NPN transistor (https://www.sunrom.com/get/448000) rated for 40V 0.5A might be small, if so, it can be replaced with a bigger one. No problem there. The FC-106 jumpers are extremely and very carefully set (T1=1 and T2=0) to regulate lightness continuously with lightness-MEMORY meaning the light level when LED being turned off is stored in memory,which will be set as the initial lightness when next touch operation turning on the LED. This is crucial. If power to the module is completely cut off, the stored light level will be lost, therefore the SGL8022W IC will have to be continuously powered forever. I’m ok with that, since It only draws 12uA on standby. Datasheet here> https://datasheet.lcsc.com/szlcsc/Sigma-Micro-SGL8022W_C81650.pdf
• And finally 120mmx36mm 10W 12V COB White LED

​

I don’t want to ditch the FC-106 module to design a different PWM block dimmer controlled by a potentiometer as a normal individual would think of doing. I’m not normal. I have issues. I have an allergy to potentiometers. They are prone to failure after a few years due to dust, moisture, grease or other mechanical failures. I want something that is maintenance free and lasts thousands of years. I’m crazy, I know. We live in the 21 century. Let’s keep up the pace and use touch sensors instead of switches. Call me a germophobe weirdo if you want. I don’t care.

​

The dimming of the 12V COB LED works flawlessly. Why it’s sold as “undimmable” is way beyond my comprehension. The maximum current I was able to measure was 0.8A at 12.9V meaning 10.32W. Math checks out. The onboard S8050 Y3J transistor gets warm but didn’t got upset since the SGL8022W when first powered automatically starts from half level light and not full power. That transistor is the least of my problems with this project. The problem I’m facing is interfacing the PIR sensor with the dimmer. I’ve managed to solve half of the problem by sandwiching the touch sensor module between 2 capacitor plates and a thin dielectric film not seen in the attached picture. Applying at least 1.5V on this makeshift plate capacitor seem to trigger the SGL8022W chip. (I took care of not touching the wires going to my makeshift plates)
This only solve half of the problem because I still need some sort of “buffer” between the PIR sensor and the dimmer. They don’t speak the same language, so I need a “translator”. Excuse my weird terminology. English is my 2nd language.
Somehow interfacing directly to the touch capacitive sensitive pin 5 of the SGL8022W IC ideea did cross my mind but I have to keep in mind that I don’t want to discard the touch sensitive area of FC-106 module, because I still want to occasionally manually control the dimming. And besides the signals are not compatible (see picture of signals required). According to SGL8022W IC datasheet a short pulse less than 550ms is required to turn on/off the light. More than that will dim the light. Thus the interface circuit will have to put out a pulse with the width less than 550ms (RC time constant).
I’m thinking along the sides of a both rising & falling edge retriggerable one-shot. I’m ok with building complex circuits, however, since there might be an easier alternative, I have to explore it. I sense there’s a simpler solution for my project, which is why I’m posting on this forum. I know I’m missing something.

Thus I was thinking, why can’t I use this type of circuit to act as my interface between PIR sensor and touch dimmer?​

​

I need some help. It took me 6 hours just to compose, write, draw and raise this question on this forum. Please show me the courtesy for all my work and reply more than a few words, a sentence or a question. Thank you.

 

[SparkyChick]

If you really want to use lots of modules, you can get 3.3v to 5v level converters ready made. Here is one from SparkFun:-

SparkFun Logic Level Converter - Bi-Directional - BOB-12009 - SparkFun Electronics

The SparkFun bi-directional logic level converter is a small device that safely steps down 5V signals to 3.3V AND steps up 3.3V to 5V at the same time.

www.sparkfun.com

But personally, I might be inclined to go down the line of using a PIC microcontroller such as this:-

Dynamic Product Page | Microchip Technology

That chip has built in capacitive touch capabilities along with PWM drivers to provide the dimming (all you would need is a suitably size drive transistor). It's operating voltage is 1.8 to 5v so it will readily interface to the 3.3v of the PIR module. Whilst you would need to write some code, it's likely to be a significantly less complex solution in the long run than what you are currently proposing. Of course it would also require you to design a PCB and to get that manufactured (no longer a problem as there are a wealth of PCB companies who do great pricing). You may even consider checking out the capabilities of the Arduino SBCs as they may have hat's that can provide the functionality you require.

 

[NiculaeGeorge]

If you really want to use lots of modules [...]

You are only fuelling my paranoid fear of not being understood. I'm clearly not being understood. I don't want to really use lots of modules. I expressed my desire to light a led by motion and dim it by touch. Period. End of story. It's that simple. The means towards this goal are many. I want the simplest with the fewest modules possible. Using a microcontroller doesn't simplify the project, does it? It still uses 4 logic blocks: battery, PIR sensor, microcontroller and LED. And besides FC-106 SGL8022W module is way cheaper and simpler than a microcontroller.
Thank you for nothing.

 

[DPG]

I'd also go with something like a low power PIC. But you obviously don't want to go down this route. Not sure why you don't though - I would still class it as one module.

 

[westward10]

The end of your post in #3 is disrespectful and will result in this thread disappearing should it continue.

 

[SparkyChick]

Thank you for nothing.

You're more than welcome.

What I didn't include in my original reply was a suggestion that you lose the attitude the closing paragraph of your original post made me think you have.

Hide behind your claimed mental health issues if you like, but these are no excuse for being rude.

Your response is almost exactly the kind of attitude I felt you would display if someone replied with something you didn't want to hear. As @westward10 has already said... if you continue to display such an attitude with other members, the thread will be closed.

I'm out and you're now missing out on the knowledge of at least one person who has the skills, tools and experience to help you. Have a nice day.

 

[NiculaeGeorge]

Ok. I've managed to make it work. It's similar to the schematic I've originally though about. I'll post it if someone is actually interested which I doubt.

 

[westward10]

What is going to keep the 12v battery charged.

 

[NiculaeGeorge]

What is going to keep the 12v battery charged.

I have some overunity vortex coils which extract heat from environment and converts it into electricity. They are basically two interlocked tori in quadrature. Each tori has multiple short circuited coils acting as resonators and harmonic amplifiers of a fundamental frequency. Technobabble when in operation the more I load the system the more the tori are getting colder.

 

[westward10]

So the charging of the battery is a system flaw.

 

[NiculaeGeorge]

So the charging of the battery is a system flaw.

I don't see it that way. If you read my original post, I've already stated how the battery is charged.

 

[westward10]

I don't see it that way. If you read my original post, I've already stated how the battery is charged.

So you did my error.

 

[NiculaeGeorge]

So you did my error.

I did and I thank you for that. You really opened my eyes in how this actually works. Thank you.