Comprehensive push-button start
 

So During labor day weekend I started a big project to upgrade the electronics in my honda. It's a 95 white 4 liter manual transmission sedan so I didn't feel like putting hundreds under the hood. (Just my opinion, but I have major respect for those who do the work under the hood). What I have been doing is moving all my standard feature, manual windows, locks, etc... with the electronic versions.

Over the weekend, I disassembled the consoles and door trims to begin work on redying the interior and with that, I decided to begin the electronic upgrades.

The main upgrade is going to be the button start. Because the car is a stick, I had to be very weary about how to do it and a whole bunch of problems: shutting off while driving, steering lock, security, reliability...

Here's what I've come up with:

Security: I bought a set of transponder systems and set them up inside the center console. To turn on the accessories, ignition, or start the car, the keys with the transponder must be place in the center console. When this is done, the car will power on as though it was turned to ignition. Once the keys are removed, the car powers back off. For added security, I placed a kill switch in the car that disables keyed start so the transponder is the only source. I also added a simple alarm system.

Steering lock I simply clamped down the steering lock pin on the steering column so the car can drive without the key and never lock. Though this defeats the security feature for it, I believe the transponder will work fine.

Shutting off while driving: My solution to this required a lot of digital logic and electrical tricks. The intended result is this: The transponder is only needed to turn on the car. Once the button is pressed, control of car power changes to a touchscreen button mounted on the console. When this button is pressed, the car will turn completely off as though the key was turned and removed. The caviat here is, the button will not become active unless the handbrake is raised and the brake is depressed. This way, you are sure to have the intention to turn off the car. The point of this whole thing is that if the transponder happens to fail while driving, the car will not shut-off. Also, if a curios friend of yours tries to press the button, it will not shut-off unless you are braking with your handbrake up. <- so know drifting with monkeys

Reliability.... we'll find out soon.


well, if you have any questions please ask away. Also, If you have any comments about some aspect I may have overlooked, please let me know. Thanks for reading!

cool idea. i like your project!

you can reference install manuals for remote-start on manual cars - that will help flush out anything you may have missed. most companies (Autopage, DEI) have install manuals available online.

it's hard for me to really comment on concepts without schematics of what the car has and what you did. did you develop schematics? as an EE, i always start with schematics so I have a reference.

i assume the 95 didn't include an immobilizer system?

so you have to enter the car, put your keys in the center console, then you can start the car?

is the range on the transponder so short you can't leave the keys in your pocket?

That's a good idea, thanks. I attached the schematic I made for it. I have a wiring diagram too, but that's just for pin layouts so if I make a tutorial for it then I'll probably include that. All the logic gates and digital components in the diagram are running off a 12v supply. Rather than voltage dividing the power and inputs to a 5v rail, I decided to go with the 4000 series cmos family.

The schematic is pretty barebone. I don't think I need to include any bypass caps since I won't be dealing with any high frequency signals. what do you think?

As for the transponder, it's a passive rfid system so the range is only a couple inches at best and only reliable within one or two. I ordered a stronger transponder unit but that one is good for 20ft so I need to find a way to weaken the signal so the car isn't on when I'm in the room next to the garage lol. Even with the other transponder, I wouldn't feel at ease having it control the power to the car while driving since the signal could disconnect for just a second and cause the car to power off.
Attachment 17988

So I got all the parts for the circuit. This week I will be wiring and testing everything and installing it in the car on the weekend. One thing to mention, because the 12v source is coming from the car, the currents on the lien are going to be on the order of amps while the cmos chips can only accept up to a couple mA. Because of this, my 12v input and all the signal inputs have a 2.2k resistor in series to limit the current to roughly 6mA. This is a crucial part for anyone attempting this.

you could run through a voltage regulator with 5-9V output also, because the voltage source will vary between 11VDC - 14.4VDC. I've found with logic circuits (esp. 555 timers) that you want a regulator to keep the voltage constant despite varying supply voltage or you can get false logic triggers.

also, during cranking the ignition circuit loses power and battery voltage drops. a few stout caps on the power supply (diode isolated) to "ride through" the cranking may help with circuit stability.

Completely overlooked that. I'll throw a 7810 or 7809 in the power circuit with a 100uf cap. I guess I should throw a bypass cap on the gates too.... Ive gotta figure out how my amp is gonna affect the circuit since it hits pretty hard and probably drops the voltage some. It may drop below the 11.5 volts needed for the 7809... I guess an array of caps could help smooth it out with a long enough RC time constant lol. I could throw in a 9v battery and use a transistor to switch power to it if I drop below operating voltage for the regulators...

Note any diode to prevent cap discharge towards the car (which will negate them completely) will also introduce a voltage drop. Caps on the regulator output should be fine and sized for the combination of current draw by the logics circuit and the duration of cranking.

I don't think I would put a diode in since the back current will be on the order of mA. I also don't think I'll be plugging it in backwards either (hopefully). After work today I'm going to drop the chips into the board I laid out and power it through a battery for testing. I'll let you know how it turns out.

If it turns out nice and safe, I may send it to a board house and offer them on the forum. We'll see...

I was referring to the diode isolating the stiffening caps from being drawn down by system/car demands. Any caps used for "ride-through" should be isolated from the vehicle.

Prototype is the next step. Then testing. Breadboards are handy for this step.

I see what you mean. I'll have to grab a diode for this. I was originally going to use a 7809 to regulate the voltage down to 9volts but now I'm considering the 7808... Since all the inputs are all 12 volts and up to 7 amps, I need to bring them to 9volts and a maximum of 50mA.

Will using a 7809 or 7808 for each input work? I'd rather use on than a voltage divider since the regulators will be safer for current draw.

Btw, I prototyped everything and it all works fine outside the car with test signals. As soon as I can get the power distribution and input signals solved, I'll drop it in the car.