Does anyone know a possible way to create a auto engine cranking using arduino?
So the inputs would be a signal to activate a timer.
15 secs ON (cranking),.... ( if Started before 15 sec, then stop the timer, a feed back is available)
if not wait for 15 sec and crank again.
Cranking should be attempted 3 times, if unsuccessful print Engine Fault.
if Successful print engine running and (or turn led on)
Your topic was MOVED to its current forum category as it is more suitable than the original
This is a 100% perfect example of where you should think "state machine!" - why not start to read up on them ( for example here and/or here) and draw the state diagram for your system.
Yes, but you’ll have to pay for it.
I did exactly this back in 2007, with up to four start attempts, variable pre/choke and crank timing.
Selectable for diesel or petrol engines.
Oil, overtemp and speed sensing, fuel level - everything.
RF keyfob, LCD panel, RTC auto start and web remote - a complete system.
Back then it was on PIC, but easily done now on an AVR or newer platform.
Sounds very much like an auto start generator... 
Yep, it was used in a few applications…
Fire pumps, generators, etc.
@asmit123 I had an idle moment yesterday so drew up a very simple state diagram. It covers only what you require, none of the sophistication ala @lastchancename. I'll post it if you like: but you may prefer to look into state machines and their diagrams yourself first.
This, I think, is one of those cases where to get to some kind of simple solution is pretty easy but then (in true Agile fashion) once Pandora's Box is opened you get submerged in "yes but, what about 'this' and hey what about 'that' now" especially when you look more carefully at safety, liability and regulatory aspects.
The code for what you ask is very simple and could be done quickly, with an on/off switch as an input to mimic the feedback from the plant to say it started ok. But after that, a whole different project may emerge.
EXACTLY.
I did that unit years ago…. it started simple, then grew into the available hardware!
More recently I built something very similar on an ATMEGA1284P, it’s comprehensive, but doesn’t have the number of ‘engine specific’ features of that original unit - with the addition of some extra features.
4 x inputs,and 4x relays out… it could be rewritten as a multi-start engine controller, but someone has to pay for it ! 
And …. The cars electrical system is very noisy , especially during cranking when voltages drop .
You need to take precautions that a fault won’t make it crank when the engine is running
Indeed- although I said above that my state diagram (which I'll post if OP desires) does only what the OP asked, it does do that test. The engine might for example have been started correctly and then the Arduino lost power. When the Arduino comes back, it must detect the fact that the engine's already running and not go into the waiting_for_start_state.
I wonder where the OP has got to?
Thank you so much for your reply. This project is about the water shortage in rural parts of our country. We are refurbishing old decommissioned fire pumps and providing them to the farmer in rural parts of our country. We have incorporated few local collage students and they will be designing this as their final year project. I was just looking around if anyone has done it yet. The OEM controllers are know as ECON-f manufactured by PROCOM. They usually sell for around 80 to 100 USD. I will update once a model and a code has been developed. This forum will help those students get an idea about the things that they need to considered. Anything and everything is welcomed, thank you for your time.
P.S we are not selling or making money out of it. There is absolutely no market for this, new pumps comes with OEM ECU Controllers.
Hi,
A nice project, most auto starts, especially on generators, monitor the Oil Pressure Switch or sensor.
This will indicate the engine running, although some engines now can crank up the oil pressure to turn the oil pressure light off before the engine starting.
Monitoring the pressure after the start cycle would however indicate if the engine is running or not.
I have repaired a few industrial auto starters and Oil pressure is a main input, battery voltage after startup is another to prove the alternator is okay.
Will this be a remote unit, or just a press to start for convenience?
Thanks.. Tom... 
haha.! Well thankyou so much for spending your precious time on it.
This is basically going to be a feedback loop with 2-3 conditions. The engine may only crank if all of them are true.
For example the
This two conditions will ensure that the engine doesn't crank if it's already running.
There are couple of other faults like
Over Speed, High Cylinder Temperature, High Water/Coolant Temperature. But we will use them just for indication purpose as of now. or they could be used as a stop command for the engine. ( we could add all this condition to the primary loop where the engine wont crank if either of them is true)
The most important thing is the break condition if the engine fails to start, With three crank attempts with 15 sec delay. So that battery dose not discharges. As we will be providing them with solar chargers which are not very efficient, and pump won't be running for too long to keep them charged all time.
What an idea. Love it..! never thought this could be one way to verify one way if the engine is running (oil pressure can also go low because of leak). voltage sensors are cheap compared to rpm (optical sensors). Might as well use an optocoupler to determine. So remote that they would hardly be serviced, and used by people who don't know anything about it.
Most of this engines are 10-12 year old.
That's fantastic. Make sure they embrace state machine thinking in their work 
One other feature to consider…
A pre-warn BEFORE the crank cycle starts.
Receive the start command - activate a beeper or strobe for 3-5 seconds before the crank starts.
Simply to warn anyone that may jump in surprise, drop a bottle or get tangled in the drive.
Here @asmit123 is the state diagram I drew up and you're welcome to use it as a starting point. The beauty of doing this work in a diagram is that it's easy to see that (say) a certain state might be redundant, and /or two states can be combined, or that a link between 2 states is missing etc etc.
This is very much just a starting point, and your students may find it useful. Or not. YMMD.
In ArduinoLand, those states can easily be handled as the cases in a switch...case.
Hi,
Hope I am not trying to teach a grandmother how to suck eggs.(1)
Along with @octopirate state diagram, you draw a block diagram of your controller and I/O devices.
Like a block for the controller, a block for the go button, a block for the oil pressure sensor/switch, a block for the voltage sensor and what sort of connection they have to the controller block.
Then develop each block separately, getting circuitry and code to work, so you have an group of codes that work and ready to combine into your final project.
Tom...
Thankyou, i will make sure they will refer to this.
Absolutely not, this grandmother has left kitchen long back. The last time i had used Arduino when I was in Uni Melb for Warman designed and build competition. We will definitely draw a block diagram. All the sensors operate on 12-24 vdc. So we will also need a optocoupler and level shifter. Some of the sensor are analogue, we will also have to incorporate adc's for the same. It's a long way to go, or maybe not. Thanks a lot. I will keep you update.
