Remote Vehicle Starter Help

[Art]

I'll have to look at the ADCON1 register. I think there is some insight for me there.

The initial value for bvoltage is set higher than any 10 bit reading can be so the flag is not triggered on the first cycle.
It could just as easily be 1024, but it's a 10 bit value stored in a 16 bit word var, so I can use a higher number.

bvoltage is the last reading stored for X number of cycles to be compared to the current value which is avoltage.
If no branch condition is met, then the current value becomes the last value for comparison the next time.

[b1arrk5]

I'll try to see if I can find some code. I've been doing Arm7 C programming the last two years, so I have to dig out my old basic files. My code worked well in two Ford vehicles, and I drove them both through a full winter using the starter every cold winter morning! I built the first one on a Lab-x board from melabs, it has an LCD to show whats happening, but the output to the laptop was safer for testing because I had history to look at after I drove the car.If the initial battery voltage was about 11 vdc, the processor would refuse to try to start the engine. I typically see 12.4 to 12.8 before the starter engages, and 13.8 or so as soon as the starter is disengaged. Happy New Year!Jerry

[mister_e]

Happy new year y'all!

I've installed remote car starter for 12 years and there's quite a few method you can use if you don't have any Tach signal generated by the Alternator, available on the Cluster connector or ECM.

The oldest method in early 90's
1) use a vacuum switch...
Con:
not always easy and doable to find a proper vacuum source/hose,
need to cut the hose, and take chance the switch could freeze

The next most popular in those years
2) read the Battery voltage, should be 'round 13.8ish V when the vehicle is started
Con:
Starter often stay engaged when not properly implemented or with "Soon to die" alternator

The most valuable and still in use solutions are
1) use a Field effect sensor attached on the top of the alternator.
Easy to implement, but some alternator do not generate enough magnetic field, OR have their own
"Strong Sweet spot", so you may have to monitor it to find the sweet spot. And once you find it, sometime it's just not simple to sit your detector there. You also need to route more wires.

2) read the Alternator output, the wire that goes "High" when the vehicle is started
The easiest solution, easy to implement with an averaged ADCIN routine (plus some extra pause and double/triple check) and a really strong power supply, board and noise filtering design. The safest and easiest in the installation field. Most alternator have this specific output.

3) grab the junk generated by the alternator on the 12V line and use it as "Tach" signal
Really popular available and patented by many car starter brand. If properly designed it works a treat for all type of car, truck, everything and do not require any extra wiring.

Enjoy, back to my batcave.

[Art]

Thnaks for the ideas mr_e

mackrackit,
I set the pins to analogue after setting them digital, so I don't think that's the problem using ADC.

[Art]

Now, I can measure any number of signals simultaneously - on any pin

How, if there's only one interrupt pin (RB0), do the newer pics have more, or is there a trick?

I'm up to this:


& would like to do a tachometer if it could be done on the same chip.
Maybe a way to cause the input for every pin pulse RB0, and then check the status of other pins..

[mackrackit]

Comparator Interrupt maybe

Darrel has several if not all setup for ease of use.
http://darreltaylor.com/DT_INTS-14/intro.html

[Art]

I'm thinking you could connect a signal diode between each pin and RB0 so that a pulse to any pin would also trigger RBO interrupt.
Then in the interrupt service routine go and check which pin has the pulse on it... if it can be done fast enough.