Need Help Comrades! - Dark sensitive led flasher circuit

Thread Starter

stn_93

Joined Aug 24, 2020
2
Hi Guys, hope all's well.

I was designing a dark-sensitive LED blinking circuit for myself and need some help. Following is all about my little project.

Scenario
I want to switch on my LED bank automatically in the dark. There are 150 LEDS in my LED bank. The attachment includes my design. However, while simulating on proteus, the circuit doesn't work and an error occurs. Can anyone help me fix this?
Note:-
  1. I am not using a ready-made LED lamp or light.
  2. There is only one LED (D1) showed on the schematic for simplicity instead of the LED bank. I will feed the output from the 555 timer to my LED bank via a transistor.
  3. The 1 megaohm resistor (R24) will be replaced with an LDR in the actual scenario for automatic switching purpose.
  4. I am designing this circuit without using a micrcontroller.


What I need to know
  1. Whether my design is correct or not.
  2. Why the proteus simulation displays that error.

Thanks ;)
 

Attachments

KeithWalker

Joined Jul 10, 2017
3,063
I don't know about proteus, but LTspice would complain because you have two separate circuits with no link between them. Try connecting the 0V of the two circuits together.
The simulator should be used on each circuit separately as there is complete isolation between them.
The sense circuit will probably not work reliably because the output on the LM358 will not go below the base to emitter turn-off voltage of the transistor. A pull down resistor is needed.
R22 and R23 could be increased to at least 10K to reduce battery drain.
A pre-set potentiometer could be connected between R22 and R23 with the slider connected to the op-amp input, to adjust the turn on point.
Regards,
Keith
 

djsfantasi

Joined Apr 11, 2010
9,156
I don't know about proteus, but LTspice would complain because you have two separate circuits with no link between them. Try connecting the 0V of the two circuits together.
Just add a temporary connection between each circuit ground to make the simulation work. It’s not needed in the final product.

Warning! You’ll likely find other errors.
 

ci139

Joined Jul 11, 2016
1,898
i don't see anything dark sensitive in that circuit

other than that you can set isolation in LTSpice by large R small C or the (series/parallel/mixed) combination of prev.
you could try large R first ? 4GΩ if errors presist try dividing down by powers 10
// for mains ~AC decoupling i use series RC 1...10 MΩ 0.1 ... 10nF in LT Spice -- the used inductor (if it is such) in the circuit - the coil of SPDT tranformer may cause you to fine tune such "RC link"
 

Thread Starter

stn_93

Joined Aug 24, 2020
2
The simulator should be used on each circuit separately as there is complete isolation between them.
The sense circuit will probably not work reliably because the output on the LM358 will not go below the base to emitter turn-off voltage of the transistor. A pull down resistor is needed.
R22 and R23 could be increased to at least 10K to reduce battery drain.
A pre-set potentiometer could be connected between R22 and R23 with the slider connected to the op-amp input, to adjust the turn on point.
Regards,
Keith
Hi Keith,

Thanks for your support.

I simulated the two circuits separately (the op-amp side and the timer side) and it works fine. But when I join the two circuits using the transistor and relay, this error occurs.
You mentioned there's complete isolation between the circuits. Can you please point out which part is that?
I am driving the second part of the circuit(the timer one, on the right) through the transistor and relay circuit. When the op-amp outputs a voltage, it is fed to the transistor Q1 which switches the relay and then the circuit for the 12V battery is complete and the timer should work. But that doesn't happen of course and I cant understand why.
 

AlbertHall

Joined Jun 4, 2014
12,345
The two parts of the circuit are connected only by the magnetic field in the relay. This division will work fine in the real world but it confuses simulators. They have no way to know the voltage between say the op=amp supply and the '555 supply. To make them happy to provide a way for them to know that voltage, for instance in this case just connect two 0v lines together. You only need this link in the simulation and you can remove it in the real world circuit.
Clipboard01.jpg
 
Top