high torque servo for remote car ignition start

[atbrown]

I am wondering if someone could help me with a simple robot design.  I live in Alaska where it's common to have remote car starters for cold mornings so your car will be nice and warm by the time you enter it.  But these things cost several hundred dollars.  I'm wondering if there's a cheaper alternative where I can simply use a high torque servo to turn the ignition.  Ideally, I would just leave the key in the ignition (I'm not worried about anyone stealing my truck) and then the servo would be mounted in a solid structure that can be installed over the ignition, and be easily removed when needed.  I would use a microcontroller to control the servo to turn the key to a predetermined angle to start the ignition, and return.  And to make it remote, I'd like to use one of the those systems that can send an electrical pulse when it's called from a cellphone (I forgot what they're called).  The problem I see happening is that sometimes it takes a little bit longer to start my truck than others on very cold mornings.  I'm willing to give it a try though.  Any criticism, ideas, questions?  Do they make cheap servos with enough torque to turn the ignition?  Thanks in advance for any help.

[rbtying]

If you're looking to remote-start a car, the easiest/simplest/cheapest method would be to add a relay to the solenoid on the car.  Turning the key in the ignition just closes a switch to the solenoid, so adding another 'switch' in parallel would work well.  It also wouldn't require a key in the ignition =). 

As for the servo, you could probably build a geared system for < $20 or so, but a single servo with that much torque may be much more expensive (you should probably try to measure the torque needed to turn the key).  Also, realize that the cell phone -> electrical pulse device you talk about its also very expensive.   At the point where you are using one, you can just use multiple commands for longer/shorter times and etc. 

Good luck on your project!

[Soeren]

Hi,

[...] I would use a microcontroller to control the servo to turn the key to a predetermined angle to start the ignition, and return. 

Like already mentioned by rbtying, it wouldn't be the best way to do it with a servo, although they do come with much more torque than needed (at a fairly low price).
When you turn the key with thumb and index finger, you probably use less than 1kg-cm force (guessing here, haven't measured it) and the stronger standard servos comes in torque ranges several times that.

Installing a heater (electric or fuel) is how it's usually done (you pop one of the anti-freeze plugs and insert an immersion heater with the electric (mains driven) version and although I haven't seen an "oil burner" up close, I assume it's a similar setup. That way, starting isn't that hard on the engine, as it is already warmed up a bit.

To start the engine by remote (or timer), you need two "switches" (relay contacts or transistors), one to turn on the ignition and one to crank the engine (or rather to activate the starter) - just like the contacts on the ignition key switch.

To make sure it's starting, the controller just need feedback from eg. measuring when the RPM goes above what the starter can manage - when that happens, the starter should be shut off (but the ignition should stay on of course). It is necessary for the engine to be able to be running idle and if it dies down, it can be detected as well and a new start attempt can be made.

To avoid running down your battery (or worse - melting the windings in the starter motor), careful thought should be given to how long it is allowed to run in one go, or thermo sensors could be installed on the starter motors frame (if it can be done in a mechanically safe way), to keep the starter from hurting itself.
There should be a limit to the number of times it attempts to start, reasonably pauses built in and the condition of the battery should be monitored as well.

The "easy" way to make a remote starter is to sacrifice an old cell phone (if you can find one that works during the lowest temperatures you'll likely experience (perhaps a problem in Alaska during the cold season? In your pocket it's held at reasonable temperatures, but in a car shut down for 8 hours or more...).
A PMR or Walkie Talkie or dedicated transmitter and receiver could be used as well, as you're not likely to be that far from the car when you want to start it anyway.

Most cell phones can be set to use different ringtones, depending on the callers phone number, so can be set to eg. a pair (or even a sequence) of DTMF tones which can be decoded directly from the speaker wires (speaker could be disabled if wanted), via a DTMF decoder chip and fed into a controller for further checking and can then give prerecorded response if needed. Several phone numbers can be selected for this action - just give them the Open Sesame ring-tone. The "all (other) users" can just go to a silent ring tone.
That way, no other phones than those you select can be used to start the car - easy security.

Feedback, if needed, could be in voiced messages like: "Attempting start". "Engine running", "cabin temperature is now nn°", "Battery is now nn.nnV", "Scantilly clad blond female leaning on car", "Oh, it's just your wife who locked herself out trying to see who's starting the car", etc.
Your imagination (and the amount of complexity you are willing to add) is more or less the limit.

The problem I see happening is that sometimes it takes a little bit longer to start my truck than others on very cold mornings.

That won't be a problem when a controller is used, as already described.

That said... If you haven't done any electronics or microcontroller work at all previously, you're in for a steep learning curve, so you have to be certain that you're gonna reach the goal, or you'd just waste time and money that could else be directed towards a ready made engine heater.

[Admin]

I assume you wanted a servo because you understand RC control and don't want to get overcomplicated with microcontrollers.

As such, you can use this:
http://www.dimensionengineering.com/BattleSwitch.htm
http://www.dimensionengineering.com/PicoSwitch.htm

Attach it to your ignition wiring and a RC receiver to activate it.

[aerowenn]

I agree that switching connections in the ignition system would be much more reasonable/easy versus actually turning the ignition switch (key) with a servo, BUT, the point of such projects is to do things the way you want them... otherwise you'd just buy a system. Goodluck!

[garrettg84]

I just got a remote starter for one of my cars and installed it myself. Cost less than 75$ shipped. Provided the car doesn't need an ignition bypass module (factory ignition disable/alarm system) you should be good to go at no more than $75-80.

This is the system I got:
http://www.jcwhitney.com/rs160-remote-starter/p2020143.jcwx?filterid=j1

It was on sale shortly before November for about 50$ and I had it over night shipped. Looks like the price came back up a bit, but still reasonable. You may be able to find it cheaper elsewhere but I trust jcwhitney.com so thats where I looked.

If you do go the route of the servo, take a look at this one:
http://www.hobbyking.com/hobbyking/store/uh_viewItem.asp?idProduct=9620

I have a few of them and they are CERTAINLY powerful enough to turn the ignition on any of my vehicles.

[Soeren]

Hi,

[...] BUT, the point of such projects is to do things the way you want them... otherwise you'd just buy a system.

Apart from you waking up a dead tread (a more and more common thing among new  users unfortunately), it's not particular good advice, to give people what they ask for, if it can be done in a cheaper/more secure/etc. way without increased outlay (in this case it's even cheaper to not use the servo) and removing the ignition key from the equation means the steering lock is engaged.

There's a world of difference from a slightly improved, but still DIY solution to a store bought same.

[diskman]

{ Deleted by author }

[Soeren]

Hi,

Unfortunately the only safe and the easiest way

Why do you think the only safe way is buying something (probably Chinese made, with all what this usually entails) and then installing something that you don't really know the inner workings of?

A cellphone that respond only to one or a few select callers is safe and so is a KEELOQ enabled controller with rolling codes.

You will probably have at LEAST $100 into a DIY solution, plus your time.. 

How do you figure an amount of "at least $100", without knowing what it takes?
Knowing what it takes, I'd guess $20 to $30 tops and the time... People usually don't DIY to save, but rather for the experience and to know how their stuff ticks.

[...] this is one instance when a solution is already available for safer and cheaper.

Not safer, not cheaper and certainly not half as fun.

[Soeren]

With the real remote starter, you get features like:  Set the car to run for 10, 15 or 20 minutes, remote keyless entry, your heater controls activate with the remote module, theftproof (engine stops when brake pedal is touched)

Just a few lines of code - takes about 5 minutes to do.

and it will turn the engine over until it starts.  OP mentioned that his engine is hard to start when it's cold.

Not that smart. It may melt down the starter motor and/or drain the battery.

I understand that this is a robotics forum and that the DIY runs very strong.  In this case, it's a matter of ease.

It's an afternoon or two to make and the install won't be different from a ready made unit.

Even the remote starters have rolling codes so security isn't a concern really.

But the "Made in China" may be a major concern - did you ever open a Chinese electronics product?

OP:

...Probably won't see this, it's an old thread.

Sounds like you're selling these gizmos.

[diskman]

Not selling.  I am a senior HP copier tech for a large corp in town.  I have put these starters in all my cars now and I was trying to be helpful.   Since my comments were not viewed as helpful, I have since removed them.


Sorry if I offended you.

Off to work on my bot for now.

Upgrading the system board to a shiny new EPIA Mini ITX..  Hopefully to cut down on my power usage.

[Admin]

It really depends on what your goals of the project are.

If it's to get the job done with the least amount of time, money, and hassle spent, the off-the-shelf method is *always* the best route.

If your goal is to learn about that system, or the off-the-shelf item is overly expensive or lacking important features you need, then the DIY route is better.