The X-Prize Sonata

[mpgmike]

Most impressive! Having dealt with patents in the past, you can use any US Patent for your own use and are not infringing upon the patent. The moment you try to sell it, that's when they have jurisdiction.

[mpgmike]

I thought I'd share some of the Roush Testing process with you.

Testing takes 2 days per test. When we arrive, they drain the fuel tank of pump gas. They then pump about 5 gallons of specially brewed test fuel into the tank. We wasted quite a bit of money, as we filled it up as we rolled into Detroit. They then push the car into the dyno booth and strap it down.




They run a full FTP-75 test to get the ECU happy with their test fuel. The engine has to then cool over-night. The actual test is conducted the next day.

Emissions are sampled with very expensive Horiba test equipment.




The test is broken into 4 parts; Cold Start, City, Hot Start, Highway. A Driver sits in the car and works the throttle pedal to keep the speed between the Max/Min parameters.




The Operator is in a booth monitoring the entire process on 3 different screens.




Once the test is complete, they shut it off, disconnect everything, then push it out of the dyno cell into another room. Total time is around 2 hours per test segment (including all 4 Cycles).

[mpgmike]

Back in 2008 when I was recruited to help out on the X-Prize project, I didn't know a whole lot about electronics. I stumbled to learn about LM555 Timers, LM339 Comparators, LM358 Op Amps, transistors, and so forth. The controller that made it to Roush testing used very simplistic analog components.




Soon I had grasped the basics and moved on to microcontrollers. We then created a series of Controllers that used board-mounted pots to adjust parameters.



As my skill sets improved, so did our Controllers. Towards the end we were interfacing a laptop for USB tuning, and tapping into the OBD II to see how the ECU liked our "tweaks".

[pgfpro]

and I thought I had it tough back in 2005 tunning on my first car with a Haltech ES6 DOS based standalone management system

[mpgmike]

I should add that before we started using our own controllers, we first purchased a MOTEC Controller (top of the line, and very expensive). I'd be driving along tuning it and all the sudden the screen would show 19,000 RPM and the engine would freak out. I was on the horn with East Coast and West Coast Support to no avail. We eventually pulled the MOTEC and installed a Link PC G2 I had laying around. It gave us the same issue. Out of desperation, I went with a MegaSquirt MS3.1, but alas, got the exact same issue.

It was around 8:12 AM on the last day of the event (July, 2010), I was sitting at the outdoor table sipping on morning coffee when it hit me...


The freak-out was caused by the Cam Position Sensor! It monitored the cam that twists (Variable Valve Timing stuff). When the factory ECU would change cam timing, the aftermarket ECU wouldn't know what to do with it. If I eliminated the CMS and controlled the engine as Batch Fire, all of those controllers would have worked.

By then we had already committed to using our own controllers.

[mpgmike]

The earliest controllers were wired direct. If I had to make a change, this made for way too much work. I had the wiring harnesses apart many, many times.







Later, I tried to use a somewhat standardized connector that could be reused with later versions of controllers. I'd get at least 2 if not 4 versions before having to wire in new connectors. I landed on Cinch enclosures & connectors. They are water-tight, and the connectors are top-notch.