car junkies paradise!

[gsr20det]

so i don't know about many of you, but i am a bit of a automotive enthusiast. i actually purchased the raspberry pi for the soul purpose of automotive integration. i have read online of a few people wanting to use them as a stereo, or using them to read vital engine data and outputting to a little LCD screen, but i want to take it a few steps further. now my knowledge of the ARM processor is quite limited, as well as my knowledge of electronics, but the purpose of this adventure was to learn! so i am now on my way to building a mini car computer (not a full blown car computer) i am hoping to gear it towards high performance and primarily turbocharged applications. so i am starting a thread here of what you can do with a raspberry pi, and your car. this is going to be a long project. i am currently enrolled in a few programming classes (level 2 of c++) but i know i have a long way to go.

idea 1:
Basic Electronic Boost Controller
many companies charge 300-500$ for a effective boost controller, but i believe that i can be done with something like a raspberry pi. all that these systems involve is a little screen, a pressure sensor and a solenoid to control something called the wastegate (a wastegate is what a turbocharger uses to bypass hot exhaust gasses from the turbocharger so it doesn't over spin or over pressurize your car resulting in EXTREME MECHANICAL FAILURE, for more info look here....http://en.wikipedia.org/wiki/Wastegate) and the idea of the program is relatively simple. the boost controllers you see for 500$ range of course have tons of toys, bells and whistles like boost by gear (pressure changes according to the gear you are in), or even boost by RPM (higher rpm = lower boost)
something that the honda community and tuning has taken a step further though, is now boost by traction. where it will take in the data of all 4 wheels ABS (antilock braking system) and compute how much boost by how much faster your tires are spining by comparison to all the others. so for example, 20% traction, would automatically make the front tires spin only 20% faster than the rear, by adjusting turbocharger pressure.


update: 5/12/1013
with the boost controller, i now have a few circuits on the pi to sense manifold pressure, and i bought a universal solenoid from the local auto parts store, and have a basic program running to control pressure at 10PSI. it is using a exponential graph to compute duty cycle, so the solenoid will start to pulse at 5psi, and anything over 10 psi it stays wide open. i haven't had the time yet to put it in my car, but on the table it seams to be working great!

5/17/2013
got it put on the car, having a little bit of a boost "spike" issue, but its working and my car is now running at 7-8 psi stable with i!!! so it is working! i need to fine tune my exponential graph for duty cycle though. i need to also find a better way to power the pi though, and also how to run the program automatically at start-up. and most likely a auto shutdown??

5/22/2013
i'm now creating a BLOG to hold all of the information, specifications of the car, and all raspberry pi projects. including the code. i will post a link asap.
this will have one root blog for the car as a project, and "children" blogs for the mini projects

http://gsr20det.blogspot.com/
http://boostcontrol.blogspot.com/

[gsr20det]

i'm kinda surprised, i thought more people would be interested in something like this?

[mishrask]

Hey,I am interested in automotive application of RB Pi.Will keep an eye here ..your project is interesting ..gimme your mail ID so that I can contact you directly .

[ghans]

@gsr20det

Perhaps publishing you code on github.com and adding info to
a Wiki there could attract more people. Or you could browse the
forum yourself and PM others which did similiar projects -
there are enough so that a fruitful collaboration is just a thing
of initiative.

ghans

[gsr20det]

@gsr20det

Perhaps publishing you code on github.com and adding info to
a Wiki there could attract more people. Or you could browse the
forum yourself and PM others which did similiar projects -
there are enough so that a fruitful collaboration is just a thing
of initiative.

ghans

newb question.... what is github? lol, i have a couple of videos, but they are on my Facebook page. whats a good hosting site for everything? do you recommend i get a blog?

[ghans]

By github i mean github.com.
You can share your code there (if you've got anything
you would like to), and even have a Wiki for your project.

You can also setup a web page on it , but github isn't really a
webhoster , so you'll be limited to static pages.

ghans

[SirLagz]

I love the idea !
Look forward to how this project pans out

[tzj]

For a power supply to run off your car battery, how about this? http://pihw.wordpress.com/2013/01/07/sw ... ation-kit/

[RunningWithScissors]

I'm really interested in how your getting and handling the pressure readings with the RasberryPi and MAP sensor. Great idea though. Would of been helpful if this and the Pi came out a couple years ago when I was working on a 7mgte supra swap. Please post some pictures, code for lowly engineering students doing research, when your going to boot to 30 psi(:^), etc.

As for auto run/stop start with: http://www.raspberrypi.org/phpBB3/viewt ... 26&t=35404

For power, I would just wire in a disassembled cigarette to usb adapter(http://i.imgur.com/te2M7.jpg), but you can't just wire it into the accessory/power buss because the RPi will start to boot and then get hard reset when you engage the starter. Probably directly wire (fused obviously) to the battery. So if your car is running and then is shut off... have a script shutdown the RPi if it has received atmospheric pressure for a given amount of time. Then the RPi will be in a shut down state but receiving power(hopefully negligible). Then have a relay (or wire in to the buss protection relay(forgot what it's called)) to reset the power when you engage the starter on startup?

Please excuse any missed etiquettes, as this is a first time ever post to a discussion.

[clifgray]

This looks awesome, I'll definitely be following the rest of your progress. Do you know of any other automotive projects with the Pi? I would love to learn about more.

[gsr20det]

i haven't had much time to play with it since i am also studying, i had some major errors with my code the other day when i was testing and spiked the boost to 22 lbs (i don't want my car to pass 18 without a good turbo) so i have found quite a few errors in my code, but when i get a few free minutes, i will post it up! i definitely need some help with this project...

hopefully i can also get a few temperature data loggers built into it too, i don't imagine that would be too hard, i have found a few projects of temp data loggers on here.

[gsr20det]

For a power supply to run off your car battery, how about this? http://pihw.wordpress.com/2013/01/07/sw ... ation-kit/

this looks like it would work great to give me stable power, i read somewhere that car signals are
"dirty" and i definitely don't want to destroy the pi, which now could result in major car desctruction!!

[gsr20det]

For power, I would just wire in a disassembled cigarette to usb adapter(http://i.imgur.com/te2M7.jpg), but you can't just wire it into the accessory/power buss because the RPi will start to boot and then get hard reset when you engage the starter. Probably directly wire (fused obviously) to the battery. So if your car is running and then is shut off... have a script shutdown the RPi if it has received atmospheric pressure for a given amount of time. Then the RPi will be in a shut down state but receiving power(hopefully negligible). Then have a relay (or wire in to the buss protection relay(forgot what it's called)) to reset the power when you engage the starter on startup?

.

welcome to the forum! (as i am new too!!! lolz)

this is one of the big issues that i was facing, and was talking about with one of my friends on a local Nissan site. i think it would be best to power it down, but he was thinking that it would be fine to hard restart whenever the starter started. considering that my car is at stake here under high performance conditions. i would rather play it safe and send it the shutdown signal.

one thing i would like to know, is there a way to program a SD card with ONLY this program on it, so it starts immediately when the pi receives power? no networking, no LXDE, just go straight to the boost control?

[tzj]

For power, I would just wire in a disassembled cigarette to usb adapter(http://i.imgur.com/te2M7.jpg), but you can't just wire it into the accessory/power buss because the RPi will start to boot and then get hard reset when you engage the starter. Probably directly wire (fused obviously) to the battery. So if your car is running and then is shut off... have a script shutdown the RPi if it has received atmospheric pressure for a given amount of time. Then the RPi will be in a shut down state but receiving power(hopefully negligible). Then have a relay (or wire in to the buss protection relay(forgot what it's called)) to reset the power when you engage the starter on startup?

.

welcome to the forum! (as i am new too!!! lolz)

this is one of the big issues that i was facing, and was talking about with one of my friends on a local Nissan site. i think it would be best to power it down, but he was thinking that it would be fine to hard restart whenever the starter started. considering that my car is at stake here under high performance conditions. i would rather play it safe and send it the shutdown signal.

one thing i would like to know, is there a way to program a SD card with ONLY this program on it, so it starts immediately when the pi receives power? no networking, no LXDE, just go straight to the boost control?

To reduce the startup time use arch Linux, it will boot to command line within 10s.

[JumpZero]

Hello,
This kind of application needs a real time Linux, I think, uch as Xenomai
--
Jmp0

[LetHopeItsSnowing]

To run a command a startup and stop at close, check out the following link

http://www.stuffaboutcode.com/2012/06/r ... rt-up.html

It describes how to create a init.d script which does exactly that.

[LGOPs82ndAbn]

I'm rather interested as well but I plan on going a step further than you are/wanting to take this. I was hoping I could take a RB Pi and use it in place of my ECM for my Neon (or any vehicle I have really...)

[tzj]

I'm rather interested as well but I plan on going a step further than you are/wanting to take this. I was hoping I could take a RB Pi and use it in place of my ECM for my Neon (or any vehicle I have really...)

Might I suggest using a guzunty board connected to the pi, that way you have a more reliable controller while the pi can log data and change the controllers 'cores' while operating.

[SirLagz]

I'm rather interested as well but I plan on going a step further than you are/wanting to take this. I was hoping I could take a RB Pi and use it in place of my ECM for my Neon (or any vehicle I have really...)

I wouldn't trust the Pi to do something as critical as running an engine...
Generally the Pi doesn't run a Real Time operating system, and in an environment where timing is important, milliseconds of latency can make a real difference.

[TXBDan]

I'm a car guy myself and this sounds like a really interesting project. A DIY megasquirt.

I agree with the above, though, this isn't the right hardware for the job. There is far too much going on inside the Pi that could crash and cause catastrophic trouble for your car. You're very ballsy for testing what you have so far! I'd create a simulator of sorts.

What you need is a microcontroller. You should start with an Arduino Uno, that'd have all you need to do this job and it would be much more reliable and predictable.

My next project will be a digital dashboard/datalogger/video recorder for autox and track days. If you're interested I'll start posting about it on www.projectsofdan.com.

[LGOPs82ndAbn]

I'm a car guy myself and this sounds like a really interesting project. A DIY megasquirt.

I agree with the above, though, this isn't the right hardware for the job. There is far too much going on inside the Pi that could crash and cause catastrophic trouble for your car. You're very ballsy for testing what you have so far! I'd create a simulator of sorts.

What you need is a microcontroller. You should start with an Arduino Uno, that'd have all you need to do this job and it would be much more reliable and predictable.

My next project will be a digital dashboard/datalogger/video recorder for autox and track days. If you're interested I'll start posting about it on http://www.projectsofdan.com.

I do have an arduino uno board. I've thought about it as well.

[XEON]

I really feel the Pi is not fast and stable enough under linux or android for this unless you drop allot from it.
I suggest running this with riscOS. "hard yes"

I'm running a pi in my car with a 4"lcd that has a 5V reg that i'm pulling power from for the pi "so far very stable but not enough oomf for a usb hard drive too"

I'm planning in extending the pi's gpio and controls using a teensy 3.0 @96MHz


Good luck with your project.. i'm dying to see how this turns out.