greatest invention i am tyrying to build

Guys as you know i am highly skilled in the art of madness for electronics.I have build remote starting kits to building light and rain sensors(proof of concept and beta/alpha testing)build circuits that make cars talk to you via negative inputs,build voice activated features that worked 60% of the time. I have done many things that only new cars have...my motto is what you have in your new car i can build it in a old car:D. Now the time has come for a new project which i need so many feedbacks from this excellent forum. My next project from scratch is *drum roll*BUilding a circuit that will monitor your battery in your car and when the battery gets flat(enough amperage to start the car) the car will start itself and when full charged it will switch off gain...and this will cycle all the time until the kill switch get applied . This application will be great for those guys that play there audio system when the car is off. So far i have the circuit for the low voltage .But getting a buzzers output to trigger a relay is a bit tough.What you guys think.I still need help on the full charge partthe only interesting answers are the one that destroys the question.

Instead of starting the car when battery voltage is low why not disconnect the battery?  That way, the car won't run for days on end (until it is out of gas) if you leave the lights on.  I would much rather come back to a dead battery then a dead battery AND a car with no gas!

I think there is a relatively inexpensive part out there called a "Battery Buddy" that does this already.  If the charge on the battery drops past a preset point the battery is physically disconnected from the car.  The driver can then reset the battery buddy by pushing a button on the battery. 

If you want to get creative, build your module and have it send you a text message when the battery gets low.  That would be much more useful then just starting the car and letting it run (imo). 

I'm not a battery charge expert but I would question how reliable a circuit would work just monitoring voltage (with no load on the battery).  Hopefully OldSpark will chime in on this!

What do you mean by "getting a buzzers output to drive a relay is a bit tough"?  What has a buzzer on it?  What is hard about getting it to trip a relay?  What is the relay going to do?  It seems to me that you could do pretty much the entire thing with one chip - monitor battery voltage and start the car.  The only relays you should need are for the ignition, accessory, and starter.

Kevin Pierson

thanks for the reply.What i mean is that the cheapest circuit i can build has a buzzer or led.the thing is that the voltage and amps might be to low for a relay coil for triggering.I could use a very small 5amp smd relay to drive a standard relay,This could work...and like you said pulse the starter relay and ignition(pulse to latched)and when the battery is fully charged the latched relay will engage again to switch off the car. In South Africa we dont really have those gadgets and if we wanted them the price would be staggering.....hence the trying to build this for us poor folks ;D the only interesting answers are the one that destroys the question.

I'd suggest forget it. You need a battery condition monitor (BCM) to be of practical use, and they are not cheap. (It involves impedance testing, history tracking, load monitoring etc.) It is not merely a case of voltage monitoring. The best is a voltmeter that tells you the battery or charging system is low, and the brain to interpret that.    The brain can be human with a bit of knowledge, or a chip with LOTS of programing.... (see BCM above). There are various voltage alerters (with inside/outside thermometers, clocks etc) from ~$20-$50. PS - If you know enough about batteries, you'll start to understand. EG - how do you determine when the battery is full? Plus, normally you would be talking about a faulty battery else alternator else system, and the solution for that is a new battery or fix the problem.   

If it has a buzzer or LED output a simple transistor will fix that. However, I would think the LED output would be a usable signal to trigger the remote start. Are you designing this system from scratch or just piecing things together to get it to work? How are you planning on starting the car?Kevin Pierson

In retrospect, I was going to suggest a solar panel - with regulator!

KPierson wrote: If it has a buzzer or LED output a simple transistor will fix that. However, I would think the LED output would be a usable signal to trigger the remote start. Are you designing this system from scratch or just piecing things together to get it to work? How are you planning on starting the car?

I am designing the system from scratch with the use of many kits and making them unison.The way i figured to start the car would be by using a output from the led(latched to pulse operation)To stop the starter swing i would have a circuit built in series with the d+ wire to tell the starter to stop engaging(this circuit will have to be very intelligent hey)we dont want over swing. This sets me up for a headache..lol. I am good in what i do and and for years i have pleased many a customer at custom car shows.This will be a difficult feat from scratch but in the end all will end well :Dthe only interesting answers are the one that destroys the question.

Peter, email me on this. We don't know if you're highly skilled, we've never heard of you, your use of English unless it's a second language makes me doubt that. Low battery monitoring and remote start has been available for many years with remote starters. Any way working 60% of the time = total fail. Any further comments, Kevin or Peter?

I don't think it is a terrible idea, it just needs to work perfectly and have numerous safeties built in (like only two or three starts of the car before the key is used to start the car, tach monitoring, etc.) I don't know enough about battery monitoring but I would think if you used a 10 ohm resistor and apply voltage to it for a second every 15 minutes you could possibly accurately judge the state of the battery without draining the battery too much. It would be a 1A load for 4s every 60 minutes. I think once the car has started you let it run for 25 minutes and then shut it down. You then lock the system out for an hour or two before checking for voltage again. The thing I would NOT want is for the module to consistently try to start the car over and over and over again. Of course I think it is a terrible idea to hack modules together to try to do something this advanced. Do it right or don't do it at all. I don't consider myself to be "highly skilled in the art of madness for electronics" but I am confident I could design one module to do all of this (from the firmware programming to the circuit board layout and everything in between) and get it working in a car!Kevin Pierson

Yes - my first reply.... I didn't even mention the complications and hazards for self-starting... I think of all the interlocks - timings to ensure non-repetitive starts; exhaust fumes killing eveything; starting in the dead of night; suitability for EFI only; anti-theft; etc etc. Thee are many safety if not legal issues... And using a flattening battery to start a vehicle to recharge itself.... If it is a faulty battery, that ruins any chance of a "normal" start later.... Then there is the complication of determining a "low voltage" battery - ie, accounting for different wiring and normal loads; battery resistances; temperature effects. That is something that has to be user configurable, but with some internal smarts for battery capacity etc. I am well known for doing the impossible. This can be done - with a lot of work (especially for non-specific or public use!) - but it will not be worth the effort. It won't succeed as a product because it will be too expensive and complicated (for the near future).    I consider my "smart EFI & Ignition" simpler than this project. But nevertheless, simple voltage alerts or voltmeters (which should be considered "mandatory" for anyone worried about charging and batteries) will probably win hands-down for price, simplicity & outcomes. And maybe a solar panel with regulator to keep the battery topped up - hence also extending battery life by avoiding sulfation. I suspect many still do not understand how difficult it is to measure battery condition and remaining capacity. [ It seems the "experts" still replace batteries ahead of time rather than try condition monitoring - eg, replace 10-year UPS batteries after 5 years unless used several times before then.... ]