[silverton]

Bought it as a 2012 Christmas/New Years gift to myself after my old cars transmission decided to **** out the counter shaft in the transmission. It had 186,000 miles and was bone stock. I'm sure the automatic transmission assisted in its easy life.


As of Aug,2016 it has 236k miles on chassis.
As of Jun,2019 it has 259k miles on chassis.



Engine Mods
-EJ257/251 hybrid
-ARP head studs
-DW 650cc injectors
-Tomei FPR + gauge
-Perrin Up Pipe
-Ported TD04 @ 9psi
-Blitz BOV
-Perrin down pipe
-3" catback to...
-Prodrive Axleback
-Mishimoto GD fitment oil cooler, modified for my GC (currently unmounted)
-Greddy Emanage Ultimate w/ Greddy 3bar MAP and AFR monitoring


Suspension/Brake Mods
-GPSports G-Master coilovers
-Perrin P.S.R.S.
-Cusco Front tower bar
-Whiteline front lower control arm poly bushings
-X brace rear bar
-19mm rear sway bar
-WRX front rotors
-Legacy H6 rear rotors

Interior Mods
-JDM v6 STi steering wheel
-EDM 160mph cluster
-Perrin Gauge pod holder
-AEM Boost and AFR gauges
-Prosport 45mm EGT, Oil Pressure and Oil Temp gauges
-JDM Hazard switch :banana:
-Tinted windows

Exterior Mods
-JDM front bumper beam
-Big Black Wing
-Replica v6 lip
-Prodrive hood vent scoop
-Yellow foglights

 

[silverton]

I've had almost a two year struggle with this car, but I believe it is finally complete(ish)! All of the fabrication and modifcations were done by my good friend David (Mr.510) and majority of the wrenching and all the $$$ spending was me. My brother helped out a lot near the end when all I needed to do was replace pistons. It currently has its fourth engine, which has needed to be pulled twice for cracked ring lands. One the first time, two the second time.

 

[silverton]

October 2014 a head gasket failed

New Years Day 2015 the timing belt tensioner failed; bent 16 valves!

I thought I had a lot of the parts needed to turbo charge the car so I thought why not.

I bought this turbo block, but it ended up having hammered bearings.

But it had this!

I wanted to run these equal length headers, but it required modification I wasn't comfortable with.

The passenger side head needed to be modified to fit the turbo cross over pipe

Had to modify the Passenger Side fuel rail, the regulator was in the way of the turbo.

Here it is coming together. Mind you, I don't know the bearings are bad in this block yet.

Turbo drain was fitted to the PS valve cover

Here we are, middle of June fitting the engine for the first time, stoner problems! I still don't know that that EL header doesn't work

Getting that cross member notched and boxed!

Finished product

Well gosh darnit!

Getting the lines for my Mishimoto Oil Cooler to the proper length

Here it is fitted and done

Outfitted with gauges. the center three pack was done while it was still NA. the Water temp has since been swapped out with Oil Pressure.

Here we are fitting the turbo inlet, had to get a set of phenolic spacers to allow it to run under the intake manifold

Finished product here

The up pipe got wrapped due to its close proximity to the turbo drain, which is also wrapped.

A custom Y pipe was made for the intercooler; Shout out to my buddy David for making this all possible!

Here it is finished

And here is one of the finished products.

Well damn!

I don't remember which engine that was, but that's where I'll end my story. I melted Piston #4 on two seperate occasions, one on the motor with junk bearings, and one when the factory injectors got tired when I decided to fine tune it. It has Engine #4 currently which I've had to pull twice. Cracked ringlands from over boosting and an Emanage Ultimate that couldn't control timing. 4.0 cracked #4, 4.0.1 cracked both 2 and 4. 4.0.2 is running strong!

Carnage Pictures! A bunch of pistons, a turbo, and an oil cooler. Oil cooler is just junk from bearing material from the junk block I installed.

These are the bearings that came out of the CL block I bought that was supposed to be good. Mind you I ran this motor for about 5 miles.

Yeah, that's two large pieces of the ring land broken off completely

 

[silverton]

Had a couple odd misfires happen. once it was 3 and 4. then another time it was all four. Got this weird code too, P2030 "Fuel Fired Heater Performance". Whhhaaaaa???? Weird! A buddy of mine heard me pull away from his house and he described the car as sounding "like an aluminum trash can filled with scrap metal rolling down a hill". Cut the muffler off. Problem solved! But holy was it loud! I kinda liked it. Like we do secretly, but always sneer at the others that do it. hah!

Anyway, I found this Prodrive axle back on ebay and put it on. Music to my ears. Maybe yours too?

Fly-by with Prodrive axle back - YouTube

 

[silverton]

Spoiler alert, i hadn't cracked it at all

I'm fairly certain I cracked the emanage! I've tried and failed at a lot of different tuning methods. Of course the most simple is the one that seems to work.

I got to thinking, what are these people with RRFPR's doing and how do they make it work so well. As in, consistent AFR's in boost and just an all around happy car. So I went back to basics.

I checked and set my fuel pressure without vac to the high end of the FSM spec, 46psi. I had increased it the other day to a idle vacuum of 43.5. Without vacuum that's 60psi! HAH. Needless to say the rear O2 wasn't happy even being in an extension.

In the Emanage I cleared all non-boost cells, and even cleared out the 1-5psi boost cells. And also set my throttle voltage output back to factory. Because I changed the fuel pressure, I have to start from scratch. That's okay, cause that was the plan! Reset the ECU through ecuexplorer and started hunting for the proper latency's. Both the before and after latency's have changed but I found something that keeps the fuel trim happy at idle and 2500 stationary rev. So I went for a drive and started toying with the vacuum cells cause I was seeing some in the 5-10 range so I changed the cells accordingly and kept driving. Now of course the idle trim is off! This is where I've been pulling my hair out this whole time.

So I thought for a second, if it learns like that.... then maybe I need to just LEAVE IT ALONE. I returned all the vacuum cells back to 0 and continued to drive. Sure enough, throughout the rev range they are all in the -5 to 5 range, and trending towards the positive. Which is fine by me cause that will make the final tune just a smidge more rich.

Now that I was happy with that I started building boost. at 1psi it was adding about 15% in the short term. Set those cells to 20. The trims are about -3% at 1psi now. Try for 2psi. It was adding close to 25% in the short term, so my 2psi cells are set to 30. 3 is 40, 4 is 50, 5 is 60. This seems to be keeping my fuel trims happy (under 10 but aiming for 5) at partial throttle in boost, yay! At 6psi is where the auto tune feature starts to take over as I'm deep enough in to the throttle that it's open loop.

 

[silverton]

So the enricher didn't actually work out for me, i think because also emanage

Bought an Enricher module from SplitSec and installed it. The more I work with this car, the more I think Subaru didn't want anything fun done with it.

The Enricher can modify up to four boring O2 sensors, or two wide band sensors. I'm fairly certain these cars have an NTK style wide band O2 sensor so instead of cutting wires I got to tap wires. Initially I only tapped one of the two 0-5v wires on the O2 sensor. With the Enricher in test mode it was obviously doing something, but I wasn't sure if it was working properly as it would send the trim from 0 to -5 before cycling back to 0 and starting over. I don't believe I adjusted the dial and have slight ragrets now.

But! This made me think that the other 0-5 volt wire might be a "double check". So I tapped it too! Now I can either get the trim to swing wildly (30+) rich, or wildly (-27) lean. No amount of micro adjustments seems to be able to get it to hold a steady 0... well any enrichment actually. I've emailed SplitSec to see if I've got it wired correctly or what kind of adjustments I need to make to it so we'll see. I may have to disconnect the second tap and actually adjust a dial to see what happens.

But as it is right now I have it set to engage at equilibrium and go full add fuel. With only the Enricher my AFR's are well below 10 unfortunately. But with the Emanage's auto tune feature it can pull the fuel that the Enricher is adding to give me a proper AFR under boost.

All in all, even without the Enricher working exactly how I want it, the car responds much better to just about everything now. It drives like I would expect it to from the factory, no slight stumble miss from a stop and it also stopped cutting out occasionally once coming to a stop.

Another note, now with this Enricher installed the Emanage is registering cells near 11psi! Last I knew it wasn't getting more than 7 as I've only been running a WRX wastegate.

 

[silverton]

I've rewired the Enricher about four times. I think I finally have a winner but I'll have to do more field testing to be sure. Fuel mileage has suffered slightly but transition in to boost is buttery smooth. It's building 11 pounds with what I'm pretty sure is a WRX wastegate

Decided to see what Craigslist had to offer the other day and sorted by newest. Backstory, my car had been wrecked and the factory beam bent. So! Top result was a JDM bumper beam! Mostly interested in a straight beam but adding lightness is cool too! Emailed right away and was able to meet up with him the next day. Holy crap! Its stupid light. No pictures but it definitely makes me worry. I may switch back to a, straight, USDM beam.

 

[silverton]

I made this to hopefully help somebody out! When I get out to the car today I'll snap some pictures for the visually stimulated. Don't judge me for the mangled wires. Project's a project right? I had my injector scaling way off (don't know how other than emanage wizardry) and it was pulling massive amounts of fuel (almost -40 between the LT and ST) which in turn I believe made it build more boost than it was necessarily supposed to. Now that I have the Enricher connected properly I've got the scaling down to a point that the car is getting better than 13mpg and has also returned to building 7 pounds of boost. Transition to boost is flawless, it's fun! I do have this weird occasional stumble off idle, been a problem since I got the car but since adding the snail it's become more apparent. I'm thinking either cam or crank sensor? The chassis has 245k miles on it so I should start looking in to original sensors and replacing them.

NASIOC - View Single Post - O2 sensor question

This post helped me determine which wire did what. The turbo cars O2 sensors don't use as many wires as ours do, but they still operate on the same idea. Connector B136 Wire 6 is the Vs or Voltage Sense wire. and Connector B136 Wire 19 is the lp or Current Pump wire.

 

[silverton]

Damn! Another motor down. Noticed cranking was sounding a little long in the tooth, so did a compression test to verify. One, Two, and Four had 120psi which I thought seemed kind of low, Cylinder 3 was only pumping 90.

I'm debating getting rid of the emanage and doing a wire merge to open source tune, or just throwing another set of pistons at it and keep going. The bottom end still sounds fine. This will be the third time I've had to pull this motor to replace pistons though. Another part of me wants to put it back to NA, I am bit by the boost bug though.

 

[silverton]

Figured after six or seven uses it was time to buy new head bolts. Bad idea, on the second run of 90° torque angle bolt A pulled the threads at 40°. Threads on B held on til 80°. I did not continue. I have a spare case, going to inspect it tonight and see about swapping internals as I figure all 12 of the head bolt holes will need to be helicoiled on this block.

 

[silverton]

The other block I have is standard bore and I don't feel like buying another set of pistons, or paying for work. So I've bought enough helicoils to do the block, and also opted for a set of ARP head studs.

 

[silverton]

These studs are hot!

 

[silverton]

And we're back! Feels stronger than it ever did, gonna take my time on the tune and not turn up the boost from wastegate pressure. Halfway tempted to leave the WG open! I decided to pull back a lot more timing. instead of only -1 per PSI, I decided to do a baseline of -10 so started at -11 for 1psi. Just casual driving around it went as high at 50* advanced today.

I replaced the engine coolant temp, camshaft, and crank sensors this time around. The from a stop stumble it had appears to be gone, and fuel trims aren't anywhere near as erratic as they used to be. Also replaced the head breather lines I had with some small stainless filters.

 

[silverton]

I've forced my wastegate open to help get a better baseline tune with the large injectors. I'm beginning to think they are too big for what I'm trying to accomplish. I can't get a stable fuel trim for the life of me. I'll see about sourcing some smaller ones if I can't get this figured out in the next few days.

 

[silverton]

lol im dum. the EBCS can't force the waste gate open!

Doing some thinking... And I bought a Perrin EBCS Pro to wire in to the emanage. I don't want to get in to chassis wiring so instead of splitting the TPS signal I've decided to turn the turbo in to nitrous! Anywhere the car is trying to target 14.7(14.4?) in closed loop, I'll have the EBCS keep the wastegate open. Soon as the car hits 3500 or wherever the transition the open loop happens to be, it will close the wastegate to build boost and then open it as target boost is reached.

I think this will keep the general drivability up, and still have the go fast when it's safe. I will need to do a six speed swap to go standalone or be able to have the boost on all the time.

 

[silverton]

The stock injector scalar DOESN'T just work. I'm still dumb.

Based on my research from people using the EMU on these cars, I'd been fiddling with the injector scalar for months and months trying to find the sweet spot since it had been said the scalar doesn't work. Well, it does.

Stock injectors are 280cc with 0.40ms lag time. My current injectors are 650cc with 1.16ms lag time. Entering those numbers in and driving around after a reset would make you think that wasn't right, because the car drove like crud. Patience was a virtue here, as it eventually leveled out and I now have the most consistent​ fuel trims I've ever had.

The Perrin EBCS being delivered today, I plan to use it to limit boost to 2 or 3 pounds when the motor is under 3500 rpm, beyond that I'll target 7 or 8.

 

[silverton]

It really doesn't though. Auto tune doesn't help until you're in boost.

More feedback. I am positive that the injector scalar works. It has perfect trims at idle, and both 35/60 cruise speeds. Before I could get good fuel trims at only one, or two of them, but never all three.

I have some more tuning figured out with these cars and the piggyback. It does require auto tune and forcing the car in to open loop. With the injector settings figured out I used auto-tune set at 14.7 and the airflow adjustment map, along with unplugging the factory O2 sensor to basically fill in a "air-assist map". I only drove around for about 45-60m and plugged my O2 sensor back in and turning auto tune off. There was definitely an improvement. I think more time with the emanage autotuning with the O2 unplugged would make it even better.

 

[silverton]

I don't think this is weird because of Air Assist. Air Assist doesn't actually do much of any thing other than make our lives difficult with injector upgrades.



Just to give an idea of just how weird the adjustment map to make up for lack of air assist injectors is, this is four or five days of driving around targeting 14.4. It still needs some work in the areas where it doesn't spend much time in, like the few moments after a stop.

Here's a short clip, I don't think I'll ever trust this car, I wanted to see if it smoked.

1st gear pull - YouTube

 

[MSRTom]

Auto STM sedan club!

 

[silverton]

Got tired of my Prosport mechanical boost gauge not reading 0 at equilibrium, if I remember right it was open box, so I bought a new one. They sent me the wrong illumination though. d'oh! blue instead of amber, no thanks! So I decided to just match the AEM wideband in the other pod, with their digital boost gauge.