OT: Fake LED car alarm improvement

[ejaf]

Hi all...

Coming here since I am not an electronics genius, and I've received tons of help here before (even enough to build my own Lipo charger!).

There's a device I can get on EBay, which is a flashing LED with 3 wires. One goes to +12V (always on), one goes to ground, and one goes to an accessory +12V, that's only on when the car is started. The LED flashes once you remove the ignition key, and stops when the car is started.

What I want to do is improve on this circuit, in that when the car isn't on, and the door lock switch is hit (I suspect some king of +12V pulse), the LED light flashes, and when I hit the unlock switch (a pulse on another wire), the LED would stop.

My guess is that it would be some kind of relay between the constant +12V and Ground, but I have been searching for a circuit example that would point me in the right direction. Been looking into SCRs (silicon controlled rectifiers), which I don't think fit the bill, and even SCSs (silicon controlled switches), but I'm not sure if that's what I really should use, rather than some type of relay.

If I indeed should be using a relay, I guess it would need two inputs, one to close, and one to open, but I am getting lost on digikey trying to find something that fits the bill.

I am also curious as to what is being used to turn of the LED once the 2 wires into the device are hot (don't have it here, so I can't rip it apart).

Any pointers, circuit example, etc. would be a big help.

TIA...Eric

[Almaz]

I really don't think you can do that because the car doesn't know when doors are locked or unlocked unless the door is physically open. I guess you can make it work with mechanical switch or something like that. The other way you can tap into lock/unlock button. When the switch goes up it triggers your circuit relay ON, when switch goes down it triggers your circuit relay OFF. I think any standard Single pole, double throw (SPDT) will work if wired correctly to the lock/unlock button

[ejaf]

Forgot to mention, perhaps, that the doors are controlled electronically, so I'm assuming that when I hit the key remote, there is some kind of pulse sent to all doors to unlock/lock. Makes sense?

[Almaz]

.,,

[ejaf]

OK...but what if I know the right wires already?

[Almaz]

OK...but what if I know the right wires already?

As a matter of fact it might not work as well without an extra circuit because relay works while the wire is hot. Using lock/unlock switch/wires won't work as well because the wire goes high only while pressing the switch. Simple Pic12F675 IC, transistor/fet, couple of resistors to divide 13v to 3,4v (Pic doesn't allow anything higher than 5v) and relay would work. The other question can you program them? :) You can also use discrete components to run relay but someone else should comment on this with a complete circuit.

[Almaz]

Now after thinking about it and to add to my last post you don't even need a relay. Replace fet/transistor, relay with LED and you have a complete product :) Just the way you want it to be.

[ejaf]

Using lock/unlock switch/wires won't work as well because the wire goes high only while pressing the switch.

That's why I started looking into something called an SCS (silicon contolled switch), which acts like a momentary pushbutton, causing hysterisis (latching) when a pulse is sent. Not sure that would fit the bill, though.

Didn't want to get into PIC programming for this project.

E

[Almaz]

I see what you are saying. I guess in your case cheaper to go with discrete circuit. I don't have time right now to build it but take a look on google for keywords flip flop relay.

[ejaf]

will do...thanks! That's one keyword combo I haven't tried yet ;)

[darkith]

I built something very vaguely similar, an adapter from a remote control garage opener to unlock (and relock ) my car doors.

I basically used the output pulse from the garage opener receiver to trigger a 555 timer setup as a one-shot (to lengthen the pulse to a stable length). The 555 output was used toggle a D flip-flop back and forth (which chose the unlock or lock relay), and also to fire the chosen relay. It could get out of sync, as it didn't know the current state of the locks, and had only one button.

One issue you may want to keep in mind is the current drain of your solution. I've heard it referred to as "dark current". You don't want to be drawing much current, or you'll flatten your battery if the car sits (i.e. you leave it at the airport for a week while you go on vacation and come back to a flat battery). In an ideal world, you'd be able to set it up so that you could use a relay that was *non-energized* when the car was off...so it was only consuming current to energize the relay while you were running the car.

You'd have to see what sort of signal(s) you're going to use. If it's too separate signals, it could be as simple as two relays.
One relay would control power to the LED, it would allow power to flow when *de-energized*. When the unlock pulse is sent to the locks, it also energizes the coil, which would switch the power from the LED to the coil, latching it energized.
The second relay would be spliced into this latching loop, when it's energized by the lock pulse, it trips briefly, breaking the latch loop and allowing the first relay to de-energize and return power to the LED.
(You'd probably need to isolate the inputs from the lock signals with diodes so that you don't feed power back from the latched coil).

If it's a single lock/unlock pulse to the lock motors, it could be more difficult and require something like my above flip-flop description. But doable.

Your best bet is to tear into the car and find the signals you could possibly use, then design for those.

D.

[ejaf]

thanks much...I'll definitely have to breadboard it first, and check power consumption...pretty cold here in the Northeast, and I'd hate to have no power to start the car. Thanks all, I think I have enough to go on.

Eric

[Dare_Devil]

i would go with a thyristor and then have a relay to shot in mometaroly when the doors are un locked this will latch then stay on till shorted

[everydayflyer]

Here is a great site for automotive most everything electronic

http://www.the12volt.com/doorlocks/doorlocks.asp

Charles

[ejaf]

Thanks a ton for that link!

[Dan Baldwin]

I would use a 12 volt single coil latching relay. You hook 12 volts to the coil in one direction, and the relay will latch on, in the other direction, it will latch off. The door lock actuators typically work exactly the same way, so you should be able to just hook the coil of the latching relay directly across the door lock actuator.

Dan

[ejaf]

Thanks! Looked at the datasheet for one of those yesterday. Although one other issue. If I hook it into the lock/unlock circuit, as mentioned, it will work, but I also don't want it on when the car is started. That means that my relay needs (what I've run into in my reading as a "gated" SR latch) an "enable" (or more specifically, "NOT enable"), that will reset the relay regardless of whether the doors are locked or unlocked. I do have a working circuit that uses a PNP transistor to pull the current away from the LEDs when given +12v from the accessory power, but with the latching relay idea, could I slave the "unlock" +12v in parallel with the +12v into the reset prong on the relay? I couldn't make out from the datasheet whether the relay could handle the constant 12v instead of just the pulsed 12v.

Sorry, but I haven't had time to put a circuit diagram together to explain easier. Will try to do it, after pulling double duty due to my wife's broken ankle :(

Eric

[Dan Baldwin]

I'm not sure if the latching relays will handle a 100% duty cycle signal or not. Generally they just get a very short pulse to set them. I gather that the problem is that your car locks the doors as soon as you put it into drive, which would turn the LED on. You might be able to use the switched 12 volts to inhibit the set on the relay.

Dan

[ejaf]

Thanks...will look into that idea.

[ejaf]

One other dumb question about the relay...how would I do this:

"You hook 12 volts to the coil in one direction, and the relay will latch on, in the other direction, it will latch off."

If I'm reading my wiring diagram right, I have 2 different +12v into the lock, and one ground. So I'm not sure jst how to flip flop the current direction...I did, however, eyeball the 2-coil relay, which has two + and one -, so that may be what I need, if I understand correctly.

Eric

[Dan Baldwin]

One other dumb question about the relay...how would I do this:

"You hook 12 volts to the coil in one direction, and the relay will latch on, in the other direction, it will latch off."

If I'm reading my wiring diagram right, I have 2 different +12v into the lock, and one ground. So I'm not sure jst how to flip flop the current direction...I did, however, eyeball the 2-coil relay, which has two + and one -, so that may be what I need, if I understand correctly.

Eric

The lock actuators I have seen are 2 wire motor driven devices. You lock the door by applying 12 volts in one direction, you unlock the door by applying 12 volts in the other, exactly the same as the 1 coil latching relay. If you do indeed have a separate 12 volt signal for lock and unlock, then a 2 coil latching relay should work fine. A 2 coil latching relay would probably be a better bet anyway to cover all your bases. The 2 coil latching relays I've used can also be used as a 1 coil latching relay by using just 1 coil.

Dan

[ejaf]

Awesome! Thanks much...think I'm going this route...probably with a transistor up front, to pull down the voltage when accessory power is applied.

I'll try to find more specs on the actual lock actuator, to determine either single or dual for sure.

E

[ejaf]

Dan..looks like you nailed it, if I'm reading the diagram right (see attached).

Thanks, all.