Subaru Impreza GC8 & RS Forum & Community banner

2461 - 2480 of 2558 Posts

·
Impreza Actual
2001 RS-T
Joined
·
1,270 Posts
Whelp apparently the car had a dead alternator. After loosing gauges and having spark blow out on my way to work, I was able to grab a new alternator and swap it in the parking lot. Runs like a charm.
That's so weird. I raced last weekend at BIR and my alternator failed too, ending my day. The post for the cable came apart internally. What was wrong with yours?
 

·
They call me Garrett
98 RBP V4 STi Swapped RS
Joined
·
4,090 Posts
Discussion Starter #2,466
That's so weird. I raced last weekend at BIR and my alternator failed too, ending my day. The post for the cable came apart internally. What was wrong with yours?
No idea, didn't look at it any further than it was ouputting 0V. Into the trash it went.

Are you storing tissues in your window molding?
Old receipts. Gotta get rid of that damn air leak on the highway!

nice work! i like the temporariness of the graphics
I would of at least rocked it for the summer or something lol then remove.
Event based liveries are a thing!
 

·
They call me Garrett
98 RBP V4 STi Swapped RS
Joined
·
4,090 Posts
Discussion Starter #2,467
This is sort of a mod... added some launch control to the car. Copying a post here that I also typed up for a local group. I don't know how much sense any of this will make, but there's videos at the end!




So I took an old ROM patch and added in launch control to my ROM today. I also decided to dive into the ROM patch to look at what it's actually doing and it's really fun.


I made this little picture, but it's about as easy to follow as a child on cocaine... but if you can get through to my ADD you may be able to see how computers work.
Anyways, I'll type it out in English.. well.. software-ish English:


Basically inside of the ROM, there is a function/routine that checks the RPM against the rev limit set value. Makes sense. If RPM >= fuel cut value, apply fuel cut. This function is used every 10 or 20ms, or whatever time the computer queries it at, so it's got it's own set spot in memory so that it can be reffered to, instead of being written out tons of times (saves memory space). That address never changes, so different functions can call to that same location and know that the RPM is going to be checked against the fuel cut limit RPM.


This patch replaces that address value, with the address value of it's little function/routine. Neat! Makes sense! Maybe? So when the software jumps to compared the RPM of the engine, it'll run through the launch control function first.


The launch control function works as such:


First, the vehicle speed value (which is in RAM, if you don't get why, don't worry it doesn't really matter) is loaded into a register.. which is like a box where math can be done. The actual vehicle speed is then compared against the value we set the LC (launch control) rev limit to be disabled at (in my case 3kph). If the value of the actual vehicle speed is GREATER than the value of the
LC disable speed, the software jumps back to normal operation.


If the vehicle speed is LESS than the value of the LC disable speed, then it keeps running the program.



What happens next is the neat part.


In the software, with ROMraider or whatever, you define an RPM offset from your fuel cut RPM. In my case my fuel cut is 8k RPM, and I wanted my LC to be at 5k RPM, so i set the offset to 3K RPM. 8-3 = 5. Cool!


This offset that I defined (in a special place in memory) is then added to the register (remember, a register is a special box where math can happen) where that function from above, the one that checks the RPM against the fuel cut RPM, is going to look for our engine RPM. Then the function jumps to the address we talked about above where that function resides.



This is neat because that function will run exactly how it would during normal operation, however (in my case) it's got 3000 RPM added to the value. This TRICKS the software into thinking the engine RPM is higher than it is, and thus applies fuel cut to create the rev limit we need for fuel cut.



The software then continues running like it normally would, jumping to the vehicle speed check function to disable the addition of the RPM offset.
Make sense? maybe? no? I dunno, i thought this was absolutely brilliant and super cool.


What's also cool is because it uses the stock rev limit function, we can "tune" the LC fuel cut hysteresis just like how we would with the rev limit. From the factory there is a 200 RPM debounce programed in. In my case, the function looks like this:
If the RPM= 8000, then stop giving the engine fuel, and don't let it have fuel back until the RPM hits 7800.


By changing that last value to 7950, we then make there only be a 50 RPM debounce... and this will also apply to our LC.
With the 200 RPM debounce I saw a 250ms time from fuel cut to fuel back on.
With the 50 RPM debounce I saw a 190ms time from fuel cut to fuel back on. Much nicer for keeping a constant RPM and building boost.


Well Garrett, just set it to be a 0 RPM debounce and you can ping pong POW out your exhaust and have a very even rev limiter!


Well, it doesn't really work that way. Life is a feedback system, and feedback systems take time to work. It takes the computer time to figure out what RPM is the engine is at. If that time window is too small for the computer to both figure out what the RPM is apply fuel cut, you'll start to runaway in RPM because the computer can't stop giving it fuel in time. This means a proper debounce is needed for the computer to figure out what to do.


I did see the RPM getting to ~ 5100 with the factory debounce, and about 5250 with the lowered one. So finding the happy medium is why tuning is tuning!


OEM Debounce:

50RPM Debounce:
 

·
Premium Member
96 Impreza Coupe
Joined
·
1,583 Posts
Awesome! LC is really nice! What determines how much boost you will build?

I’m not sure is the carberry rom goes about launch control differently than the tinywrex patch but on my car I see about 12-15psi for my car with LC(this is also on e85 so that may be a factor as well, I’ll see how much boost I make next time I have 91 in the tank).
 

·
They call me Garrett
98 RBP V4 STi Swapped RS
Joined
·
4,090 Posts
Discussion Starter #2,470
Awesome! LC is really nice! What determines how much boost you will build?

I’m not sure is the carberry rom goes about launch control differently than the tinywrex patch but on my car I see about 12-15psi for my car with LC(this is also on e85 so that may be a factor as well, I’ll see how much boost I make next time I have 91 in the tank).
Yeah Carberry basically uses Group N stuff, which is spark cut. Fuel is still flowing and burning a bit so it can create a quite bit more exhaust velocity to spool the turbo up.

This uses fuel cut, so exhaust velocity is not rising.
 

·
Premium Member
96 Impreza Coupe
Joined
·
1,583 Posts
Yeah Carberry basically uses Group N stuff, which is spark cut. Fuel is still flowing and burning a bit so it can create a quite bit more exhaust velocity to spool the turbo up.

This uses fuel cut, so exhaust velocity is not rising.
Hmmm, mine is for sure fuel cut. Carberry has the option for either.
 

·
They call me Garrett
98 RBP V4 STi Swapped RS
Joined
·
4,090 Posts
Discussion Starter #2,473
04 STi Rear Diff

Do we call this a mod, or do we call this maintenance?? I'll call it both.

So yeah, as mentioned before the previous R180 diff was getting REAL clunky, and I believe a hypoid bearing/spacer is trash or something inside of it. Don't know, need to tear it down. In the mean time, I decided I wanted to upgrade the R180 a bit with one with the larger flange and larger driveshaft.

Got myself an 04 STi rear R180 with the mech plate clutch LSD which supposedly had 56k miles on it. I dunno it looked nice (and wet) when I got it in the mail.. so.. sure. I'll take the guys word for it.






Also bought myself a fatter driveshaft to mate to it without issue from an importer. The center section ended up having trashed u-joints but luckily the importer sent me another section without much hassle. God I hate those stupid importers..


Anyways. dropped the old diff, and popped it open to just have a look. Yep, that's a closed unit Suretrac alright.





Swapped over the normal fill plug instead of the broken temp probe on to the new diff, and popped her in.

75w90 Motul LSD oil went in next.




...and a comparison on the driveshafts... fatter + multipiece on the bottom = goes in car.



And it's in! That's a new diff! It feels like the old one with less clunks!





Sending it at an AutoX this weekend so I'll see if there's a performance difference. Not expecting anything life changing.

I'll rebuild the other diff at my convenience and update on what broke. I expect damage due to high miles and not hard use.
 

·
They call me Garrett
98 RBP V4 STi Swapped RS
Joined
·
4,090 Posts
Discussion Starter #2,475 (Edited)
Detroit SCCA Summer Heat Solo AutoX - Schoolcraft College - June 30, 2019

Whoaaaaa I went out for an AutoX for the first time in a few years. I had a few coworkers who wanted to go and I wanted to take my girlfriend for a ride.

Pretty standard AutoX, waiting in the heat for 6 hours for 4 minutes of driving.. but still a good time. Car did great, dampers were turned up, and the ambient temp was in the 90s. It was a hot one.

Unfortunately I still have the clunk in the rear, although much less loud, and the new diff has a low speed shudder (thanks seller :curse:), but it stayed together.

The course was pretty neat for sake of the start which was a 90 degree bend and THEN the timing gates... I couldn't use my launch control at all. Tires felt a bit cold for 3/5 sessions, and they seemed to screw up timing. Results still aren't in, but I had thought I got 2nd in SMS (Street Mod Street Tire), but apparently I didn't get the trophy when they announced it, so I have no idea where I ended up. Lame.

Either way, here are some lame pictures.











I also attached my GoPro to my girlfriend's helmet, and the results were awesome. I totally lost my line of sight on the first run, but you can't even tell because she was looking right on course and I was not! Either way the sense of speed is much more appreciated in AutoX with a helmet cam.



I also took datalogs with my new logger of the 5 sessions found here:

datazap.me | equinox92 | DSCCA 9-30-2019 Runs 1-3
datazap.me | equinox92 | DSCCA 6-30-2019 Runs 4-5


The logger was PERFECT. I am super happy with the performance of it, and here is a video of the data layed on top of the fastest session.


 

·
They call me Garrett
98 RBP V4 STi Swapped RS
Joined
·
4,090 Posts
Discussion Starter #2,478
Fun times at least!
Beats a lazy Sunday for sure!

It's frustrating getting knocked out of a trophy position, but solid videos!
Yeah kinda upset at the event's handling of the timing, I still don't see results posted and the timing for like the entire 3rd heat was scrubbed because it got messed up.. grrr. I'd care more if I was trying to be competitive, but I really just want some hardware for once haha!
 

·
They call me Garrett
98 RBP V4 STi Swapped RS
Joined
·
4,090 Posts
Discussion Starter #2,479
Oil Cooler AN Fitting Conversion + Phase 1 AC Condenser



Holy hell this should not have been as big of a deal as I made it out to be. Ever want to spend double what you did on the factory kit to make it just that much more custom? Do you like designing things you never thought you'd be designing? Well boy do I have a guide for you. The plus side is that you can at least use your AC after it's all said and done.

So yeah, basically that's what facilitated all of this, me being able to use AC.

A little while ago I popped in the Phase 1 condenser I had to be able to use with my radiator, which I had to use because of my fans, which I had to use because of the wiring, which I had to use because I used to have a V4 STi engine in the car. Got it? No? perfect.

That condenser fit everything perfectly, except for the fact that the oil cooler lines (were installed without a condenser in) were blocking the inlet of the AC completely. The remedy this, an AN line setup was going to be used, and I was going to make sure it was done waaaaay too proper.

Alright, so to start this off I bought a Spec C -> AN adapter from Fuji Racing in the UK. Not a hard piece to make, but I didn't feel like doing it. Really good idea for adapting the OEM lines to AN fittings.




Now, I decided I wanted to use -6 AN lines, and I know that sounds small.. but for fitment it was required, and it's really actually equal to the factory lines anyways. I don't think flow is an issue here so I went with the -6 AN adapters.




Making the lines was pretty simple, but took me a while to do. They were going to go out the same way the factory lines did, tie to the chassis, and then come up to the cooler and attach to the factory rubber lines with barbs. Easy.




Tying these to the chassis was the fun part here. I needed a bracket that would clear them away from the AC lines, and also remove relative motion from the engine to the chassis where the core was mounted. To do this I just drew up a design I thought would work in SolidWorks, and then because I've spent WAAAAAY too much time setting up a solid 3D Printing infrastructure, I was able to print of iterations in 20 minutes using ABS plastic, which should work with the heat in the engine bay.





The first iteration worked pretty dang good. Clearance was perfect, however the grip on the lines wasn't as good as I wanted it to be. So I added a bit more interference into the design to really grip the line. That worked... until it didn't get the clip deflected too much. So I made the grippers meatier.




That did the trick.


Here's all of the few iterations in various forms. Some design errors, some printing errors, all very quickly created. The one on the bottom is the one that was used.








And if it breaks... well.. I can just print out some more!


In the last picture there you can also see that the new AC hardline also fits with the lines in place. The rest did too, and with the GD intercooler the N/A AC line doesn't hit the IC...for now. That's great news!







And finally... all the lines bolted up!









Gave the AC a vacuum, and I had some leaks.. a few O rings swapped later and the car has AC... and less ghetto oil cooler lines!

Woo woooooooo




The only downsides are that some areas may still have some abrasion to them, and I think I want to put some better camps on the barbs to the oil cooler.. they aren't perfect. So far no leaks and my bracket hasn't melted even after REALLY heatsoaking the thing hard in the mid 90s ambient temp today. Time will tell!
 
2461 - 2480 of 2558 Posts
Top