Hi everyone, (first post!)

I am creating a (very amateur) on-board display for cars that reads its data from the ECU via the OBDII port (As an A-Level project). It has the potential for a virtual dashboard, error code diagnosis, and a few other features depending on the car. I have been able to test some of the basic functions already on my Polo, so I know at least some of it works!

Part of my project requirement is that I find out what people would want to be present in something like this.

It would be awesome if any of you could share your feelings on the requirements you would want prioritised in something like this on my (terrible, sorry!) survey: http://goo.gl/forms/bpkcSkH3Bv

Thanks for reading everyone, any feedback at all appreciated!

(Sorry if I have posted in the wrong place, please correct me if I am wrong).

First Welcome to the forum.

Sounds like an interesting project, I know one thing that is a useful function from the VCDS data that I can pull up on my Ibiza, is the actual litres left in the tank rather than the vague markings on the gauge

Real time logging or recording feature, with user selectable parameters.

First Welcome to the forum.

Sounds like an interesting project, I know one thing that is a useful function from the VCDS data that I can pull up on my Ibiza, is the actual litres left in the tank rather than the vague markings on the gauge

Thanks for the reply, and sorry for the delay. I will try and include some general 'vehicle status' things like this. Thanks again!

Real time logging or recording feature, with user selectable parameters.

Thanks for the suggestion. So you would want to select a set of a few things to log at the same time as time went by?

On one of my other OBDll generic readers ELM327 & ProScan 5.9 you can do a "drag strip" timer 0~60 1/4 mile etc. a rolling road BHP calculation & average/instant fuel consuption

Yeah I am using a WiFi ELM327. That's sounds like a really cool feature to include. Do you have any idea how I would be able to calculate the BHP from what I can measure?

Good luck with the project - have fun!

I think the challenge is to identify the end user and then their requirements. My gf would be happy with a display that tells her the colour of the car (and if it matches her nails) neverminding however many noises are coming from under the bonnet!

I use Torque a lot and VCDS a bit - methinks the problem is how much data is out there and how to make it useful. (And unfortunately how much/little data is made available via OBD and how significantly that varies between car models.) My 6p suggestion is a very simple main display (maybe car colour, RPM, litres left in tank, manifold vacuum or whatever floats your boat) but with an engine behind that is collecting and analysing as much of the data as possible. With a simple "drill down" option from a traffic light which is the only other widget on the display. Click on the red light and see current fault codes and any other readings collected that are seriously dodgy (your best 0-60 time is 4 hours). Click on green and see all kind of reports on stuff that seems 'good'.

Impossible to complete within the scope of your project but e.g. my vacuum is around 22 inHg at idle - I actually have no idea whether that is red, orange or green on the traffic light. Similarly my LTFT is always -ve, mostly around -5. Maybe it should be orange on the traffic light because together with the vacuum it indicates a possible inlet leak. (Fabricated example - you probably realise by now I am a pisspoor mechanic!) But my point is accessing the data is easy, making it useful is the trick. There is a limit to how much the driver can digest while driving (ergo a simple main display) but the app working its socks off in the background analysing whatever it can intelligently for later reporting would be cool.

Good luck with the project - have fun!

I think the challenge is to identify the end user and then their requirements. My gf would be happy with a display that tells her the colour of the car (and if it matches her nails) neverminding however many noises are coming from under the bonnet!

I use Torque a lot and VCDS a bit - methinks the problem is how much data is out there and how to make it useful. (And unfortunately how much/little data is made available via OBD and how significantly that varies between car models.) My 6p suggestion is a very simple main display (maybe car colour, RPM, litres left in tank, manifold vacuum or whatever floats your boat) but with an engine behind that is collecting and analysing as much of the data as possible. With a simple "drill down" option from a traffic light which is the only other widget on the display. Click on the red light and see current fault codes and any other readings collected that are seriously dodgy (your best 0-60 time is 4 hours). Click on green and see all kind of reports on stuff that seems 'good'.

Impossible to complete within the scope of your project but e.g. my vacuum is around 22 inHg at idle - I actually have no idea whether that is red, orange or green on the traffic light. Similarly my LTFT is always -ve, mostly around -5. Maybe it should be orange on the traffic light because together with the vacuum it indicates a possible inlet leak. (Fabricated example - you probably realise by now I am a pisspoor mechanic!) But my point is accessing the data is easy, making it useful is the trick. There is a limit to how much the driver can digest while driving (ergo a simple main display) but the app working its socks off in the background analysing whatever it can intelligently for later reporting would be cool.

Thanks so much for such a thorough response!

My current specified user is quite broad, but is basically motoring enthusiasts, though people of all skill levels. For this reason the interface hopefully will be quite intuitive, and produce interesting information with entertaining graphics. It is still important though for it to be functional (with the Error Code Diagnosis and Check Engine light deactivation etc).

Currently my prototype has an underlying fault code and PID database, I could perhaps add fields for the scopes of Red, Green, and Amber ranges so that in the graphic dials can change colour? You have given me a load of things to think about with these ideas. Grading or flagging seems like a great plan.

I was considering putting profiles for seperate cars, so I could include this for separate drivers too, so there could be a leaderboard of times 0-60 perhaps.

Thanks again for helping me out!

Also, the true user is kind of me, as I chose this project pretty much as an excuse to make it, but I have to specify another user group in the coursework.

A blank canvas screen, with widgets.

Me, i'd select a big round rev counter in the middle, with smaller digi speed readout to the side, small fuel and temp to the other side - but the next person might not want a rev counter at all, they might want a clock and economy meter instead. So that flexibility would be good. And various graphic themes. Chronograph. Vintage. Basic. Carved wood. Windows 98. Aviation etc.

Unless you're thinking of programming an ipad mini, this probably isn't what you wanna hear..

A blank canvas screen, with widgets.

Me, i'd select a big round rev counter in the middle, with smaller digi speed readout to the side, small fuel and temp to the other side - but the next person might not want a rev counter at all, they might want a clock and economy meter instead. So that flexibility would be good. And various graphic themes. Chronograph. Vintage. Basic. Carved wood. Windows 98. Aviation etc.

Unless you're thinking of programming an ipad mini, this probably isn't what you wanna hear..

Hi, and thanks for posting!

Flexability is a good sugguestion, so I will try to develop a system where the user can create a basic custom screen with preset widgets. For theme I was planning on being able to change the colours to match the car's interior dash lighting. To have other themes is also a nice idea, as my application so far is based on textures and skins that should be able to be easily swapped in and out.

At the moment I am going to be running this on a laptop on the passanger seat, but in the future it could be build into the car with a monitor and a low-powered computer like a Raspberrry-Pi that could run Java. It could even have a touchscreen eventually!

Thanks again for posting.