Wow, this looks amazing man. You provided so many pictures I've been looking for!

 

more power to you for having the patience to deal with this. it looks good.

 

Swapping the harness is not hard. The harder part is removing the items the harness is around.

 

Intercooler piping is hard to remove?

 

Ha! That's funny. Obviously for everything you have done ot has been more than just removing ur ic piping.

 

my engine bay looks like a mess

 

Everything I have done is more than installing a harness. To install the tucked milspec all you would really have to do is remove the UICP and spark plug cover.

 

Also forgetting to mention you would have to remove the plastics for the windshield wiper motor as this is powered via this harness too. It would be easiest to also remove the intake manifold to remove the stock harness. I know with have the F1-i that I have I had to remove to get the stock harness out. Much easier to do this than to suffer trying to work around it.

 

wire tuck looks awsome, props to ChaseBays!!

 

Any further progress? I think we all want an update. BTW looking hella good.

 

Hey guys, I want to point something out. There could be benefits from having a rats nest. When wires are bundled together, they become electromagnetically coupled. Back in college, I was doing an electronics lab, and although me and my lab partner checked our work a bazillion times, the finished circuit just wouldn't work. Finally we gave up and decided to take it apart and return the parts to the bin. But first, we unbundled all the wires, and tried turning on the circuit again... and voila, it worked. It turned out that my lab partner was a neat freak and had bundled all the wires together for a cleaner look. We surmised that some of the clock signals must have gotten unwanted noise from being coupled to another wire, and unbundling the wires removed the coupling. The circuits we were working with were capable of going to several MHz, but frequency has nothing to do with the problem. bottom line is, if Mitsu designed the wire harnesses a certain way, I personally would not rearrange and rebundle them with something else. You might create an electrical gremlin that would be impossible to hunt down...

 

The stock wires are bundled together as well, just criss-crossed horribly. The milspec setup has worked on hundreds of other cars out there, and I'd dare say nothing in our car is going to clock near the kHz.

No offense, but it sounds like you didn't properly shield or electrically isolate wires and it induced crosstalk. I deal with T1's at work all day that run 1.544 mHz, tightly bundled, and shielded properly and don't have that issue at all.

Did you ever toss an O'scope on there to verify your hypothesis?

 

Now that is a nice engine bay it must be nice to have so much room to work on anything you need to

 

This was a lab experiment put together on a solderless breadboard, using 22AWG solid wires. It was a digital circuit, not analog. We had hundreds fo wires, and the normal procedure was to just wire them up, in point-to-point rats nest style. With this style of wiring, no two wires would ever run together in parallel for any appreciable distance. The lab experiment was not supposed to be put together in a neat-freak fashion. I think we are talking about two different things, no offense taken. But to either hunt down the gremlin or to shield all the wires would have taken the rest of the semester, and definitely not worth the trouble (and at that point in my education, i would not have the ability). I only brought this up because I've seen posted pictures and saw a spaghetti mess of wires in the engine bay. But come to think of it, my engine bay looks nothing like that, so the poster must have unbundled some wires. I would just be nervous about bundling together wires that were not originally bundled.

 

ahhh I just thought of another thing. You said the stock wires were criss crossed horribly. I have not looked, but they may have been done that way on purpose, to prevent crosstalk. I my current work, which involves chip layout, sometimes we criss-cross adjacent signals on purpose to reduce cross talk when we dont have the luxury of shielding. This mainly works for complementary signals. There is a technique called bit-swizzling, which, if you added up all of the length of the negative version of a wire with the positive version, u find that it got equal amounts of exposure to the positive and negative version of the neighboring signal, and theoretically, any crosstalk induced by the neighbor would have been self-canceled. Could the horrible criss-crossing on the stock bundle be going after this effect?