"It does not operate properly" he said holding it in his hands.
One day, a friend of mine came to me holding in his hand the dome light from his car. "Its does not work properly" he said. "When i close the door, the light turns off immediately". Obviously, the turn off delay circuit had gone bye bye.
Searching the Internet, i found tens of different dome light off delay circuits. Thus, i decided not to design a new circuit from a scratch. Again, i had to slightly modify it. The difference was the power supply. I had to invert it, as the position that the circuit indicated the positive, was actually a big sheet metal.
This is the circuit that i found:
The circuit on a breadboard for test
I will start explaining the circuit from right to left. The 3-resistors net (R1-R2 and R4) performs a voltage divider that controls the voltage on the base of T4. The T4 will control how fast will the C1 discharge through it's CE, that is parallel to the R3. The more the current through the T4, the faster the discharge.
You should find 3 connectors in the dome light shell
When power is applied to the circuit at the connectors CON1 (positive +12V) and CON2 (0V), the capacitor is fully charged. The transistor T1 will allow all the current to go through. The current is controlled from the transistor T2 that is attached to it's base. The transistor T2 is controlled by the transistor T3.
When the power is switched off (the car door is closed), the capacitor will start slowly to discharge. As the voltage across R3 (base of T3) is falling, the T1 will gradually allow less current to flow through. This is the main idea of this circuit.
This circuit is connected parallel to the switch of the dome light that is controlled by the door switches. This is how to connect the circuit:
In the dome light shell, you should find three connectors. One goes to the ground. The other one goes to the battery directly. The third one is connected to the door switches. These switches are connected in parallel directly to the ground. The circuit that i found had these switches connected in parallel to the positive of the batter (???). Anyway. As shown above, the circuit is connected in parallel to the negative (CON 2) and to the door switches (CON 1). So simple. No other wiring is required.
The circuit was built on a pre-drilled prototype board.
Testing the construction
Actually, it was already getting too late and my friend was in a hurry. I suggested a nice PCB construction, but he insisted on putting the things on a pre-drilled protoboard. What the heck. After all, a PCB would significantly increase the car's current value... its an old Ford for crying out loud.
With no further ado, i start soldering the parts on the PCB. The final size was around 3cm x 3cm. It could be much smaller, but there is plenty of space under the dome light.
Bill Of Materials
Resistor 2.2 KOhm 1/4 Watt 5% Carbon Film
Resistor 33 KOhm 1/4 Watt 5% Carbon Film
47 uF 25 Volts electrolytic Capacitor
1N4002 General Purpose Diode Rectifier
1N4148 Switching Diode
BD243 NPN Silicon Power Transistor
BD140 Plastic Medium Power PNP Transistor
BC547 Switching and Applications NPN Epitaxial Transistor
When a door or boot is closed, the Dome Light circuit takes over and keeps the light illuminated for an extended period of time.
It does this by immediately seeing 12v across terminals CON1 and CON2.
The electrolytic is initially uncharged and it gets charged very quickly via the base-emitter of transistor T3.
This charging turns on T3, T2 and T1 and the voltage across CON1 and CON2 drops to a small voltage. This voltage is just enough to put a small voltage across the capacitor, plus the 0.6v on the base-emitter junction of T3. The capacitor charges a little bit more and T3 turns off a small amount. This turns off T2 and T1. This allows the voltage across CON1 and CON2 to rise.
As the voltage across CON1 and 2 rises, it robs the globe of voltage and current and the globe dims.
Eventually T3 turns OFF and so does T2 and T1.
The capacitor is now fully charged.
When the door switch is closed, it puts a "short" between CON1 and 2 and the capacitor is discharged via diode 2.
T4 "robs current" from T3 to alter the delay time.
Hi, Kammenos. I changed the connections, CON1 to the LED cathode, insted to door's pin, and now everything is OK. The circuit works fine and the side effect disappear. All the same, i'll try to apply circuit to control LEDs, maybe must to change some components values, don't know yet. When i build phisical model, i'll post the result.
I have no time for soldering 'n searching required components, so i tested the circuit in Proteus(virtual simulation) instead. Well, it works, but i found some unexpected and unpleasant side effects. When the dome is OFF and switch over to AUTO, the LEDs lit for a while. Is this bug appears in the real model? And if exists, how to get rid of this? I'll think twice about this problem and how to fix it. Please, write here if you have any ideas. By the way, your site is very useful, regards!
Is this circuit suitable for LED dome light (12pcs SMD3528)? I just now made modification of my dome light - from standard to LED. Because of negative controlled dome light circuit, i can't just put a capacitor before LEDs. Can i use the above circuit to add dimming effect on my dome? Thanks in advance.
Devinz, i am so sorry, i am not good at this. You should search for a transistor with same characteristics, but that would be a great deal of work. There are books that have transistor equivalents, but unfortunately i do not own one.
Indeed, i did some research myself. I just think that, using the ground for doors is more practical, as you save one part of the wiring: that is from + to each door. The switch needs only to be ground, and only 1 wire departs from the 4 switches (in parallel) to arrive to the dome light.
As for the beer thing, except the time that i consume for the site - that i do with pleasure FOR my fun - there are actually some expenses to keep this site. Not that much comparing to other yearly expenses, but is a respectable amount of money. Books, tools, components, and another $180 for the web hosting and name reservation. As i share all my the knowledge with all of you, i thought that sharing some expenses would be quite fair.
P.S. Following the ads on the right side daily, can help a lot and it will cost nothing for you.
There is a reason that one circuit you looked at had the door switches going to power. Older cars are wired different from your drawing and they are wired in one of two ways.
With one version, power will come from the battery, through a common fuse, through a dimmer rheostat then through the door switches that are all wired in parallel. Then the ground side of the door switches will all connect to one side of the interior lights and the other side of the lights will go to ground.
With another version, power will come from the battery, through a common fuse, through a dimmer rheostat then to the interior lights. Ground is provided through the door switches that are all wired in parallel.
There is yet a third version where the dome light has one side tied to ground and power to the light is managed trough a lighting controller. With this version, the lightiong controller determines what needs to be sourced to the interior lights.
The circuit you designed will work great for a "smart" dome light, one that is designed to be controlled externally. Also, I think it's a neat design.
BTW, I like the "buy me a beer" pop-up and I will as soon as I have some extra money.