Follow along with the video below to see how to install our site as a web app on your home screen.
Note: This feature may not be available in some browsers.
Which automobile bulb are you talking about? Is it a headlamp, taillight,or panel lamp?swedluap said:Forgive me for being a complete noob.
I want to build a circuit to delay an automotive light bulb by merely 1/4 to 1/2 second. I want to keep it small and simple. I have no idea where to start.
Sceadwian said:Have the switch connect a simple low pass filter (a big resistor in line with a modest capacitor connected to ground) to a Schmitt triggered buffer. When the switch is triggered the capacitor will take time to charge up to the Schmitt buffer's trigger voltage. If you use an SPST switch you'll also get the delay during turn off. If you want the on delay with instant off you'd have to use a DPDT switch so that when the switch was off the Schmitt buffer input was shorted directly to ground after the resistor. The Schmitt buffer itself would drive a relay or transistor to actually turn the light on. If this description isn't enough let me know and I'll draw out a schematic for you. All of these terms are searchable on Wikipedia if you're really confused.
k7elp60 said:Which automobile bulb are you talking about? Is it a headlamp, taillight,or panel lamp?
With the values shown, that is the 10k fixed and the 1M you can adjust the time from no noticable delay to about 1/2 second. The actual time will depend on the actual value of the capacitor and the applied voltage. In reality how ever long it takes the capacitor to charge to about 5 volts to turn on the transistor.swedluap said:Wow, thanks!! I think I actually understand that!! What size pot would I use in place of the 1M 10K to get 1/4 to 1/2 sec variable?
I think you need a few more parts. As drawn, it might work well for a single closure, but for a repetitive circuit like a turn signal, the cap never has a chance to discharge fully if you make it large enough to give you an appreciable time delay. The diode solves this problem.swedluap said:Ok, I'm just about to build it. One more question. It looks like most of the load is handled through the transistor, so would 1/4 watt components be sufficient for the cap, 10k and 1m, or should I beef that up? If so, 1/2 watt? 1 watt?
The slightly higher voltage should be no problem. The problem is probably noise on the supply when the engine is running. Cars are notorious for this. Unfortunately, you can't add a filter or a regulator without making your lamp dimmer. I tried to simulate the effect, but could not duplicate it. I suspect that it might not happen if you get rid of the NPN positive feedback. Try it and let us know if it works.swedluap said:Update:
It works well with the engine off (12v) but the bulbs flash simultaneously with the engine running (14v). A voltage regulator should fix that, right? Or is there something else I should think about?
The MOSFET only conducts when the gate is below the source by about 5V.swedluap said:Wow, thanks!! I think I actually understand that!! What size pot would I use in place of the 1M 10K to get 1/4 to 1/2 sec variable?
I think you need a few more parts. As drawn, it might work well for a single closure, but for a repetitive circuit like a turn signal, the cap never has a chance to discharge fully if you make it large enough to give you an appreciable time delay. The diode solves this problem.
The NPN (any small-signal general purpose NPN) gives positive feedback, causing the MOSFET to switch rapidly and much more efficiently.
1/4 watt resistors are fine.