63 replies to this topic

[samg2007]

Another nice addition, if possible, would be an extra flick of the wipers after the screen washers have been used - to clear that annoying screen wash dribble

 

Yep, great idea and super easy to implement - just two lines of code. I'll get some demo videos up when all the bits arrive.

[Ethel]

Another nice addition, if possible, would be an extra flick of the wipers after the screen washers have been used - to clear that annoying screen wash dribble

That's the beauty of what Sam's up to. Without any hard wiring you can implement pretty much anything you fancy.

[samg2007]

 

No usb header for programming?

 

Missed that Ethel, decided not to include the USB to serial converter to keep the component count down. Included the icsp header on the board so you could use one of these to program the chip directly https://www.sparkfun.com/products/9825

[Mini Manannán]

Where's the best place to start with all this Arduino shamonkee?  I really fancy the idea of and I've tried to get a grip on it, but I need basics.

[samg2007]

Where's the best place to start with all this Arduino shamonkee?  I really fancy the idea of and I've tried to get a grip on it, but I need basics.

Get yourself a cheap arduino uno and download the IDE. There are a bunch of examples to play with, and go from there.

http://www.ebay.co.u...3D1164876675451

Edited by samg2007, 07 February 2017 - 09:47 PM.

[Ethel]

Ditto, there's a massive Arduino community online. Even if you struggle to get to grips with programming, which is very user friendly, you'll be able to find something useful just to download that makes the outlay worthwhile.

 

I was just wondering where you're going with this Sam - if it's a completed plug & play box of magic, the usb isn't important. I guess the same applies if you want to offer bare boards to soldering iron brandishing geeks. There might be a middle group who want to personalise bits of code, but having a breadboard mock up on your desk first might be better than experimenting on your car I guess.

[Mini Manannán]

 

Where's the best place to start with all this Arduino shamonkee?  I really fancy the idea of and I've tried to get a grip on it, but I need basics.

Get yourself a cheap arduino uno and download the IDE. There are a bunch of examples to play with, and go from there.

http://www.ebay.co.u...3D1164876675451

 

 

 

I'm not sure what the IDE is (yet) but I've bought that one Sam, here goes!

[samg2007]

I was just wondering where you're going with this Sam - if it's a completed plug & play box of magic, the usb isn't important. I guess the same applies if you want to offer bare boards to soldering iron brandishing geeks. There might be a middle group who want to personalise bits of code, but having a breadboard mock up on your desk first might be better than experimenting on your car I guess.

 

I hadn't considered providing boards to anyone, it's just something i've wanted on my car for a while. If theres interest I could always supply the pcb files and arduino code to allow you to build your own.

 

 

 

I'm not sure what the IDE is (yet) but I've bought that one Sam, here goes!

 

 

Here you go: https://www.arduino....n/main/software let me know how you get on! PM me if you need any pointers.

[paulrockliffe]

This is a really cool project.  I wish I had the time to sit down and learn all this stuff.  Maybe a job for next winter...

 

When you had the rev counter from the Bini working, have you managed to get an input from the coil working with that via the Arduino?  I'm thinking that if you have it should be pretty trivial to code in some shift lights?

 

Is it possible to add a proper screen to run as a dashboard, rather than the little LCD you've got?  It would be really neat to drop all your data to a screen.  Or perhaps there's already an android app or similar that you could link the data to to run a digital dash that could incorporate Google Maps and your choice of media player.  Like Android Auto, but with data from the car over-laid.  Thinking about it the best bet would be finding a way to output the data, then using an Android tablet to access (and record?) the data as it must be as easy to build an app as it would be to build the display.  Maybe.

 

I was planning to build a custom dash for my Mini soon, but I'm wondering if I should put it off for a while until I have time to look at this properly.  I built and installed a HTPC in my first Minus that ran a 7" monitor for navigation and music, but that was very clunky, didn't work very well and didn't interface with the car at all.

[samg2007]

This is a really cool project.  I wish I had the time to sit down and learn all this stuff.  Maybe a job for next winter...

 

When you had the rev counter from the Bini working, have you managed to get an input from the coil working with that via the Arduino?  I'm thinking that if you have it should be pretty trivial to code in some shift lights?

 

Is it possible to add a proper screen to run as a dashboard, rather than the little LCD you've got?  It would be really neat to drop all your data to a screen.  Or perhaps there's already an android app or similar that you could link the data to to run a digital dash that could incorporate Google Maps and your choice of media player.  Like Android Auto, but with data from the car over-laid.  Thinking about it the best bet would be finding a way to output the data, then using an Android tablet to access (and record?) the data as it must be as easy to build an app as it would be to build the display.  Maybe.

 

I was planning to build a custom dash for my Mini soon, but I'm wondering if I should put it off for a while until I have time to look at this properly.  I built and installed a HTPC in my first Minus that ran a 7" monitor for navigation and music, but that was very clunky, didn't work very well and didn't interface with the car at all.

Yeah, I've got the rev counter working off the coil, so shift lights wouldnt be too hard, although not something I'm not sure I'll do.

 

The little LCD is completely optional, it could be possible to use an android tablet or UDOO board and the sdk to connect to the arduino, building an all in one. I've explored this option, but am focused on the basics atm. 

 

UDOO is a pretty cool bit of kit and merges android and arduino on one board. I recommend having a look at that if you're interested.

[samg2007]

And the boards are here! Yay.

[Mini Manannán]

have you built them up yet Sam?

[tiger99]

For goodness sake DO NOT control the headlights in this way. They are already unreliable on a standard Mini. If the headlights fail on a dark night at speed, what is going to happen? And what will your insurance company do? You are required to have sufficient independence of headlights that a single common failure will not take them both out.

To undertake a project like this in safety you really NEED an experienced development team which provides independent verification and validation of the software, plus a full Failure Modes, Effects and Criticality Analysis and lots more.

The ramifications are so great that it is just something that should never be attempted at home. You could usefully implement a reduced system with only side and tail lights, fogs and spots, and reverse included, which might usefully include failed bulb detection.

In general, safety-critical software can not be designed at home, and only a very minimal amount of safety-critical hardware should be attempted, where the hardware functionality is simple and straightforward. I have been involved in design of safety-critical systems for most of my working life, which is why I know that it must not be attempted.

Sorry, but that is how it is, and the possible loss of human life is involved.

[lawrence]

And the current mini electrics are better than ardunio controlled ones? All it takes is some old crusty wiring to short out and you loose everything nt just your lights! Your comment is valid however but the ardunio is no less likely to break compared to a mini wires

[samg2007]

For goodness sake DO NOT control the headlights in this way. They are already unreliable on a standard Mini. If the headlights fail on a dark night at speed, what is going to happen? And what will your insurance company do? You are required to have sufficient independence of headlights that a single common failure will not take them both out.

To undertake a project like this in safety you really NEED an experienced development team which provides independent verification and validation of the software, plus a full Failure Modes, Effects and Criticality Analysis and lots more.

The ramifications are so great that it is just something that should never be attempted at home. You could usefully implement a reduced system with only side and tail lights, fogs and spots, and reverse included, which might usefully include failed bulb detection.

In general, safety-critical software can not be designed at home, and only a very minimal amount of safety-critical hardware should be attempted, where the hardware functionality is simple and straightforward. I have been involved in design of safety-critical systems for most of my working life, which is why I know that it must not be attempted.

Sorry, but that is how it is, and the possible loss of human life is involved.

 

Thanks for your thoughts. I've had safety and redundancy at the top of my list while designing. The I2C chip I have controlling the relays, if triggered will stay outputting a high value in the event of the ATMega chip burning out - meaning the lights won't go out. 

 

I actually started looking into this idea when my headlight switch burnt out. With some testing, I don't see any reason this system won't be just as reliable if not more so, than the original.

 

I really like the idea of failed bulb detection, I'll take a look at that for the next version. I'm still testing the fully built up board at the minute, although life keeps getting in the way!