The LED's Wings Project and sensing circuit.

Thread Starter

q12x

Joined Sep 25, 2015
1,657
Open this image first:
(this image is HiDef and you can read the smallest text in it, but you must open it properly, either download it, or in browser 1click to enlarge it and then rightclick on it and choose "view image")
dBenzin100_ziud_22 circuit 3914 and LDR copy 1.jpg
I have 3 important problems to resolve:
probl 1: (!!!!!Very important!!!!!)
The area of the light is too small and the LM3914 circuit will create an On/OFF state for all leds, when interupting the beam. It will interupt suddenly.
Its not the circuit fault but exclusivly the light and LDR fault. Their dimentions are too narow and my finger is too wide for the circuit to sense it slowly. I need a way of enlarging the LDR surface that will allow penumbra to play its role on the lm3914 circuit and the leds to behave corectly. I have to evectively get into the beam light and interupt it right now, so it is no room for penumbra to leave the sensor to sense anything. Its too fast for the ldr to be able to cath other little diferences in shadow, other penumbras diferences. I need to enlarge the shadow area of the LDR. I dont know how to say it better than this. The shadow area right now is 0.5cm (or 5mm) (or 1/5 of 1 inch)
I wish I can enlarge the shadow area to about 5cm (or 2 inch) around LDR sensing area. When this area is so large, the ldr have the time and space to dim the leds 1 by 1 more elegant. Right now is too abrupt, too brutal, too fast, too narrow. You get me? If not, ask me, and ill add more details on this matter and drawings if needed.

probl 2: (not very important at the moment, is working as it is)
The new Laser diode, I inserted for the very large distance i can separate the triger elements. But, the Laser is too dim at 3V and im afraid to push it over 3V ( i can, i test it already and is brighter) but it will overheat and burn in a short period of functionality. Im not very sure how stable will be as it is at it's normal parameters.
My thought is to amplify the LDR signal to pin 5 trough a transistor because of this small laser signal. It is good light right now and working ok but not very bright and leaves other external lights to interfere. Of course ill have to make a new case for the ldr to defend it from external lights and concentrate only my beam on it.

probl 3: (important, but it can wait until the end)
All this trigger part, (laser diode + ldr module to pin 5) will have to be integrated with another pre-existing circuit. I'll have to change ldr module to that one, and adapt this one to do a double function: to action on pin 5 as it is now, and also to triger a relay from the other circuit. What a complication. I hope i can split the signal in 2, one for the pin4 and other for a transistor that will drive the relay. Ill have to test it and keep you in touch.

Here are the 2 circuits i must combine:
Light switch: https://www.deviantart.com/q12a/art/dBenzin100-ziuc-12a-Mains-Beam-Light-Switch-Latch-848347670
Vumeter with 3914 and LDR : https://www.deviantart.com/q12a/art/dBenzin100-ziud-20-Vumeter-with-3914-and-LDR-852598799
 
Last edited:

Audioguru again

Joined Oct 21, 2019
6,673
It looks like the LEDs are 3V white.
The very high value of R2 is limiting the total LED current to (5V - 3V)/100k= 20uA. Therefore each LED gets 2uA when they are all turned on or 20uA when one LED is turned on. An LED is bright with 20mA which is 20000 times more current.

Why use R2 when a feature of the LM3914 is an adjustable regulated current for each LED? The LED current is about 10 times the current from pin 7 which is usually done with a resistor. Your circuit only has the 12k ohms resistance of the pin 6 ladder.
Your pin 7 voltage is set to 1.28V then the current in the 12k ladder is 1.28V/12k= 107uA. Then the current in each LED will be only 1mA without R2. Add a 680 ohms to 2k ohms resistor from pin 7 to ground for more regulated LED brightness.
 

MrChips

Joined Oct 2, 2009
30,708
The response time of LDR is too slow for your application.
Use a photo-diode or photo-transistor with a shorter response time.
 

Thread Starter

q12x

Joined Sep 25, 2015
1,657
I know you have the LDR working now with the LM3914 but what if you redesigned that circuit using a small solar cell for detection.
SG
Yes, you are on the subject and is as intriguing for you as is to me i bet, I thought of solar panel as well but i dont have one, but I go even further, to a mini coil to create a 10~20cm(4~8inch) little bubble/sphere of radiation around it (electromagnetic in this case or even radio?) but im not that good in this particular sector of electronics, I only remember some bits and ideas of circuits from old magazines i read long time ago, when i was a child/adolescent. This circuit is more a memory wish to fulfill than a practical COMERCIAL thing. So im open to anything out of the ordinary as long as it WORKS as I wanted it! Thank you so far, you did great, even if you didnt win the first level. :) haha haha Welcome to the 2nd level. :)
 

Thread Starter

q12x

Joined Sep 25, 2015
1,657
It looks like the LEDs are 3V white. The very high value of R2 is limiting the total LED current to (5V - 3V)/100k= 20uA. Therefore each LED gets 2uA when they are all turned on or 20uA when one LED is turned on. An LED is bright with 20mA which is 20000 times more current.
Yes they are 3mm white bright leds . Indeed pin1 led gets a lot brighter when is alone lit, and lower when all are lit.
I see your point but for some reason my IC doesnt like it. I try it in the first place, before talking on the forum here with you, and I tried diferent ways and the only way i find to dim the leds was with my simple solution.
This is what i did following your advice:
V+ common to all Leds = 5V
A] pin6 and 7 are linked. I put 1k ,then 10k to pin 7 and the leds get to full brightness, indifferent of the rezistor i put there.
B] I unlinked pin6 and pin 7. I put 1k pin7 to gnd. Nothing happens.
Absolutely no dimming in brightness, only maximizing them up.
Aparently, with only 1 resistor on pin7 to gnd is not working (for me). Is what I tried before writing in the forum as well.
What I didn't do yet (and is my mistake not doing it in the first rounds, because i looked for simple solution on internet), but I will in the near future, is this (again) voltage divider from datasheet:
1598186060557.png
Thats why my solution that involves 1 resistor is more appealing than this hay here. But i completely agree with you because what you say is corectly regulating, by adding the same current to each led (i hope) and not letting them in the breeze with very high current through them as it is now. And is very good that you mentioned that first led higher current, because I observe it as well but didnt give it much thought, until you bring the curent calculation explanation illumination. Ill do it properly, in the end. But understand that now, in the heat of creation, I must kill some of them. :) haha. (im joking).
 

Audioguru again

Joined Oct 21, 2019
6,673
The datasheet shows that when pin 8 (ref adjust) of the :M3914 id grounded then pin 7 is regulated at 1.28V. Then a resistor to ground from pin 7 sets a regulated current in all LEDs.
BUT you added a resistor to ground at pin 8 causing pin 7 to try to be higher than it can go and be unregulated.

If pin 8 is grounded then pin 7 will be regulated at 1.28V and if pin 7 is connected to the 12k ladder of pin 6 and they have a 1k resistor to ground then each LED will have a current of 13.8mA (fairly bright). If the resistor to ground from pin 7 is 20k then the current in each LED will be about 1.71mA (a little dim).
 

Thread Starter

q12x

Joined Sep 25, 2015
1,657
BUT you added a resistor to ground at pin 8
No, I never did. Pin 8 was always on gnd. Look more carefully. Also look in the video.
Here is what I did so far:

dBenzin100_ziud_23 circuit 3914 copy 1.jpg
I suspect Maybe... is something wrong with my IC.
If not, is a minor problem that i will repair with the datasheet circuit in near future.
 

Audioguru again

Joined Oct 21, 2019
6,673
Pin 6 is not supposed to be connected to nothing like in your "B".
Pin 4 is supposed to be at a lower voltage than pin 6.

The datasheet circuit uses its R1 and R2 in series so that the pin 8 Vref sets a pin 7 voltage of 5.3V and the supply (minimum 6.8V) is at least 1.5V higher than pin 8 so that pin 7 has good voltage regulation.
 

Thread Starter

q12x

Joined Sep 25, 2015
1,657
Those are tests I made to check what is going on and how to make it work. Inevitably I made the weird test like in B. I make mistakes, until i get the good combination.
I see your point, so you are saying that the regulator needs a minimum of 6.8V to properly work, and pin 7 to have good voltage regulation. Hmmm, i didnt think of this at all. I mean, i see it, read it, but not paying attention too much to it as you did.
.... Im doing the tests now because i postpone it too much.
Ok, I did all the tests I could think of.
The result was the same all the time:
All leds at maximum and the same behaviour as it was conected to pin5, turning OFF Led by Led until the last one, but with some errors. I used POT to divide the voltage as prescribed in datasheet.
at 3V, 5V, 6.5V , 7V - same result.
I changed 5K pot to 10K - same result.
I put and take out my V+ Rezistor to see its efects in diferent permutations. It is efectivly protecting the leds from overburning, especially on higher voltages. I dont want to risk it, though it is mentioned in datasheet that it have an internal regulator, diferent from this external one that we are playing now.
I used this new circuit, with the pot conected to pin7, to pin8 and gnd.
I changed the IC with a brand NEW one and is having the same results !!!!!
I put the exact values from the datasheet for the 2 resistors (instead of the pot), and only first 2 leds were lit brightly.
I recheck all the links i made and all the wiring, all is ok (it wouldnt work with the older circuit anyway) but I check it just to be sure.
What else? I did more tests than I just enumerate here, and i cant remember all, i am putting down what i remember.
I am thorough and i dont give up easy. We will find its problem.
And thank you for taking time for this.
dBenzin100_ziud_23 circuit 3914 copy 2.jpg
 

Audioguru again

Joined Oct 21, 2019
6,673
A pot at pin 4 messes up the voltage sensing.
A pot at pin 8 messes up the LED current regulation when the pin 7 voltage becomes within 1.5V from the supply voltage.

1) Connect pin 4 and pin 8 to ground.
2) Connect a 10k pot to the supply voltage and its slider to pin 5.
3) Connect pin 6 to pin 7 and connect a 1k resistor from them to ground for 13.8mA in each LED. The pot at pin 5 will be able to light no LEDs or the 1st LED or any number of LEDs. All LEDs will light when its slider is turned to 1.28V or higher.
4) Then try a 10k resistor to ground at pins 6 and 7 to ground to see the LEDs dimmed a little at 2.3mA each.
 

Thread Starter

q12x

Joined Sep 25, 2015
1,657
ok i did your test step by step.
All leds stay ON as I turn from end to other end the POT.
But, lucky for you, I did this circuit in the beginning and you only omitted the ground of the 3rd leg of the pot. Now when I rotate the pot, every led is turning OFF, one by one, or ON, if i reverse the direction. BUT... no matter if i put 1k or 10k, the intensity of the leds jumps up, they lit brighter a tiny bit more, indifferent of the value.
But i see my mistake with your test here. I did kept my 100k on the V+ of all leds, and no intensity change visible when switching 1k and 10k. When I put direct wire on V+, their intensity got up a tiny bit but more obvious this time. Also I see dark spots now. I also cover them but the cover get off sometimes and the bright light directly into my retina, and im looking from an angle at them, not from the front, and still attack me like a welding flame.
I also put a POT on pin 7, then8 then gnd, as in specs. And is does something but very unstable, litting leds 1 by 1, and unpredictable, jumping around and not the same intensities. Very strange.
oh, i also changed the IC with a new one and repeated the tests and same result.
 
Last edited:

Audioguru again

Joined Oct 21, 2019
6,673
Your 100k resistor in series with all the LEDs reduces their current and brightness much lower than the design of the LM3914 regulated LED currents. Your resistor also provides no current regulation.
 

Thread Starter

q12x

Joined Sep 25, 2015
1,657
- I understand it now, also thanks to you, but is very efficient while other methods simply dont work. I wish they do and Ive tried them before coming to the forum also, so a couple of times. The circuit you put me to do last night, i did it a couple of times in the past, but i will not complain because is easy to make and if is telling you something, then ill do everything i can to give you as many hints as possible.
- I thought of it in the start and choose only 1 resistor 100k for V+ , but also I could add 1 for each led (1k), this way dimming them in the same amount each. But that is for the final circuit board. Because from datasheet, it can not be dimmed. So I make my own solution, especially when i was so infuriated of that blasting brightness into my eyes.
Im sure you understand my reasons and also keep trying to do it the right way, and i respect that, but the right way, does not work for me at least, and we must move on. There are alternative solutions. If you really insist on finding the right solution, im very happy , since I dont get much help anyway, so im happy when i receive help, no matter how small or large.
It should have worked from those pins they described in the DSheet. Why it doesnt? I have no clue. It's the SF or twilight zone as usual.
Can you help me with the original problems I made this post ?
Im also on halt right now, thinking hard for a solution, and as i mentioned already:
"
... I thought of solar panel as well but i dont have one, but I go even further, to a mini coil to create a 10~20cm(4~8inch) little bubble/sphere of radiation around it (electromagnetic in this case or even radio?) but im not that good in this particular sector of electronics, I only remember some bits and ideas of circuits from old magazines i read long time ago, when i was a child/adolescent. This circuit is more a memory wish to fulfill than a practical COMERCIAL thing. So im open to anything out of the ordinary as long as it WORKS as I wanted it! ...
"
 
Last edited:

Thread Starter

q12x

Joined Sep 25, 2015
1,657
Ive made a video following your directions and i talk a bit, explaining everything we discussed here.
I hope is informative enough. Seeing is believing.
 

Audioguru again

Joined Oct 21, 2019
6,673
I have never seen a white LED like you have. It is performing correctly.

You have the pin 7 and pin 8 set for a regulated reference voltage of 1.28V then into the 12k ohms of the resistor ladder at pin 6 the current is 1.28V/12k= 0.11mA. Then the IC regulates the current into each LED at 10 times which is 1.1mA each. Most LEDs are fairly dim at 1.1mA.

When you add a 1k resistor from pin 6 and pin 7 to ground then its current is 1.39mA and the current in each LED is regulated at 13.9mA which is fairly bright.

You can try adding an opamp as a follower with its very low input current at pin 7 and its output feeding pin 6 to make the LEDs extremely dim, a resistor (try 100k or 220k) from pin 7 to ground might then set the low brightness that you want.
 

Thread Starter

q12x

Joined Sep 25, 2015
1,657
My leds i have are called [super bright] and are from good old ebay. Here is a similar link to what i have because I have them from long time ago and the original link is no more. My key words for search in ebay are: [100pcs 3mm LED white superbright] Hope it helps. Unfortunatly is Cold white. I like Worm white, and i have some smd leds like that. Very retro looking and i like it.
https://www.ebay.com/itm/100PCS-LED...sh=item35fe881b8b:g:~HwAAOSwpdpVdbVN&LH_BIN=1
All I have is UA741 - General-purpose op amp.
So you say to add an op-amp to those pins. Ill do it since I am not that experienced with op-amps anyway. It's an occasion for me to learn something new. I'll announce you to my progress very soon.
 

Thread Starter

q12x

Joined Sep 25, 2015
1,657
1 thing I always forget to mention, this LM3914 had an internal regulator. Different from pin6,7,8 which can form an external one with external components.
What this internal regulator is, from what i remember from pdf, is for protecting the leds from over voltages.
Now, back to your "fear" for the first led on pin1 (when 100k is on V+), that is getting an over voltage, or too much current through it, and its life might be in danger.
My impression is that the internal regulator of the IC, will intervene and take care of it not to get in any way damaged.
Because now is very dim lit, when is remaining alone, it is more obvious that rise in current and ilumination through it. Also from your calculations the rise is real. BUT... again, my impression, is that it is protected internally.
This is not my faith, it is my impression and my deduction, after reading some of its specs.
This is an idea i had but flew away a couple of times before mentioning to you.
 

Audioguru again

Joined Oct 21, 2019
6,673
An LM3914 has a voltage regulator with its pin7 its Output and pin8 its Adj similar to the LM317 adjustable voltage regulator. Usually its pin7 output feeds an LED brightness resistor to ground and the pin6 resistor ladder. When the voltage regulator is working properly (an output voltage no higher than the supply minus about 1.5V) then it also regulates the current in each LED. You can wrongly set too much current from pin7 (its max allowed output current is 10mA) to blow up each LED with a massive current of 100mA and the IC will over-heat. It does not protect itself or protect the LEDs if you set it wrong.

An LED sets its own voltage and it is fed a certain current, not a voltage. The maximum supply for the LEDs on an LM3914 is 25V and 2V or 3V LEDs will be fine because the IC regulates their current.

Your LEDs are rated to be very bright blue-white.

A 741 opamp was one of the first opamps designed 52 years ago and it has many problems that modern opamps do not have.
One opamp in an LM358 dual opamp can be a follower and provide a very small load on pin7 of the LM3914 to produce a very low current in the LEDs reducing their brightness and the output of the opamp can feed pin6.
 
Top