The Research and Development Thread!!

[TazzI]

There are no words to explain the mind melting, teeth grinding, head smashing couple days Iv had with getting ALDL supported.

The way its implemented by the J2534 protocol, it doesnt actually work well on our commodores. The idea they had for it was ok, but practically it doesnt work out, since the tool will wait until it finds suitable 'idle' time before it tries to communicate... although it never believes there is a suitable time to do so in our commodores!!

But, have managed to come up with a workaround which so far hasnt skipped a beat. I imagine GM probably implemented a similar workaround for their own systems otherwise it simply would not work as advertised.. although they do have alot of problems with communication dropouts, least my workaround doesnt have that problem.

Currently there is a VT BCM, VY LS1 ECU, VZ V6 ECU, VE BCM, VE Radio and VF Cluster all hooked up at once, and can communicate to all of them using the one cable and software!

Before I can transfer this into the new NVS application, will need to do more live testing in multiple cars. I can hook up to almost anything on a bench setup, but its not the same as live car conditions.

Iv had some eager beavers email about helping with testing (Which I am welcoming!), but I do need you to have a J2534 compliant device otherwise you wont be able to do anything. Theres no recommended tools at the moment, Im still sussing out various different suppliers.

 

[Dirbatua]

nice to see your getting somewhere funny how just a new way of looking at something can reenergise motivation

 

[TazzI]

nice to see your getting somewhere funny how just a new way of looking at something can reenergise motivation

Theres no words to describe my motivation for this. The scantool hurdle held me up for too long. Dedicating all R&D time on this now.

 

[TazzI]

Ripped out a white board over the weekend, and brainstormed ideas with some mates on how to implement a vehicle definition scheme that will be 'future' proof for newer additions down the road. I kept comin across limitations in my own design.. so needed some fresh eyes on the problem.

After some of the worse ideas I have ever heard in my life (They really were that bad!), we eventually got some traction and designed a system which I am pretty happy with it. I can safety say I wouldnt have come up with the same kind of solution by my self.

Some of the starting ideas were pretty far fetched. One of the first ideas to come out was to create an application which could self learn and teach itself the entire car so it didnt even need a database.... so.. basically artificial intelligence (AI). My simple response back was "Who do you think I am.. google?". That idea got a bit carried away.. by the end of it I was accused of creating skynet

But overall.. progress! I then managed to stay up until 2am this morning implementing the new design. Theirs no words to describe the relief I had to see it working correctly.

Todays job is to add in some vehicle definitions for the following cars:
- VY V6 ECU/BCM/Cluster
- VT/VX/VY LS1 V8 ECU
- VZ V6 ECU
- VE V6/V8 ECU/Cluster/Radio
- VF V6/V8 ECU/Cluster/Radio

Its basically just going to 'check' those modules are in the car by sending out a simple "Are you there?" message. If this goes smoothly (No reason why not), then its time to migrate it into the NVS project so the visuals are all in place, and I can start putting up some demo videos showing new additions while developing!

 

[TazzI]

Oh, while its on my mind. I have been pounded with the same question over and over... "which J2534 scantool do I need?".

At the moment, Im still working out a list which indicates some of the J2534 devices on the market, and which vehicles/modules they will support. Not all J2534 support all the require protocols (Languages), which is why Im putting together a list of tools.

I wont be able to physically test every tool, so the list is based off of what information I have been provided by the manufactures. As each tool is tested by either myself or someone else, I will update it with a confirmed status.

 

[VS_Pete]

Keep up the good work mate

 

[TazzI]

All good so far!
Tested and working on the modules connected to the bench. And also connection to my VE ute.
This means the core functionality is complete, its now moving onto end user functionality.

Todays (fairly cruisey) day will be to migrate everything into the NVS application.

Next on the list:
- Continue to add vehicle definition files
- Brand new licensing system

Then... finally.. we can start seeing some pictures/videos of the new setup in action.

 

[caaf]

Top stuff Jason. We all do appreciate the time you take to update us on your progress.

 

[TazzI]

Next update.
Migrated to the new application. Was a bit more work then expecting as I had to restructure a few things.

Any programmer will know the feeling, when you look over old coding.. and wonder what the hell you were thinking.. and end up completely redoing it as its way more efficient and to the point. I dont think you ever stop learning with programming, forever growing your knowledge and problem solving ability.

Anyways, all integrated with the NVS application. To ensure everything was logging and working on all the protocols, I did a 2hour logging session on each of them.

The lucky last protocol to test was ALDL, and typically.. it has some weird issue. After logging after for about 26mins (On the dot), my old implementation was freezing up. The actual problem was my 'workaround' solution for ALDL to get it working which I explained on a previous post. It seemed the J2534 protocol didnt like what I was doing, causing it to completely freeze up after so many requests.

So.. went back to the drawing boards and completely re-wrote its implementation over the weekend.. and.. (Thankfully) problem solved!

Rechecked all protocols again (8+ hours total of leaving it logging).. and we are back on track.

Next on the list is to get this live data to be showing on the gauges!

 

[TazzI]

Been a interesting week and a bit!

My VE ute had its instrument cluster fail... then the partners cars ABS module failed (recall covers this) and the new Ford I got for R&D had a hit and run!

*Sigh*

Anyways, just thrown together this basic preview video this afternoon:

Do note, I threw in dummy data to navigate through the screens. Simply because the main database with all the diagnostic information is still being put together. Due to the way that is designed, I cant demonstrate with a half completed section or it will fail.

Will be doing a live data demo for the engine computer in my own VE soon.