Subaru Impreza GC8 & RS Forum & Community banner

2561 - 2580 of 2587 Posts

·
Registered
'00 RS / V5 STI
Joined
·
590 Posts
Nice to see some progress being made!

Valve cover gaskets are terrible. I've learned to drain the oil, thoroughly clean all surfaces, and apply hondabond (or toyotabond or equivalent) to the gasket on both sides, then let it sit and cure under hand tightening for at least 1 hour, then torque to spec. Wait till the next day to add in oil. Never had this problem with the EJ251, only with the EJ207.

-TC
 

·
Registered
2001, Impreza 2.5 RS
Joined
·
36 Posts
Great update! Sucks about the valve cover but love that the turbo is ready to rock. My VF37 has some buggered up blades so I'm also looking for an upgrade so curious to see how things go once you get this all dialed in!
 

·
They call me Garrett
98 RBP V4 STi Swapped RS
Joined
·
4,109 Posts
Discussion Starter #2,564
Nice to see some progress being made!

Valve cover gaskets are terrible. I've learned to drain the oil, thoroughly clean all surfaces, and apply hondabond (or toyotabond or equivalent) to the gasket on both sides, then let it sit and cure under hand tightening for at least 1 hour, then torque to spec. Wait till the next day to add in oil. Never had this problem with the EJ251, only with the EJ207.

-TC
Yeah I don't know why I had this problem. Ain't my first valve cover rodeo and usually they leak at the half moons. Weird as hell.

Great update! Sucks about the valve cover but love that the turbo is ready to rock. My VF37 has some buggered up blades so I'm also looking for an upgrade so curious to see how things go once you get this all dialed in!
Well if this guy gets dialed in nicely and is happy, I may have a spare VF37 for ya hah!

Very nice! Why'd you choose a stock inlet over an aftermarket silicone one?
I knew someone was going to ask this eventually haha.

Never been a huge fan of the aftermarket inlets. Most fit like garbage and don't offer anything over a factory one except ability to leak a bit more.

Flow characteristics are more than likely unchanged, if not worse.. but they will last longer as far as heat cycling goes. I'll give em that.

Basically, it was just more expensive than beneficial to me, especially on a stock turbo.
 

·
#32
95 Coupe, 95 Sedan & 95 Wagon
Joined
·
12,326 Posts
Well yea but did you do a 5 why? Was their FOD during install? You just rush? Was the gasket already defective and not caught from the supplier? I'm assuming you approved for source inspection and did not inspect too much in house?

Aerospace got me drinking the juice now LOL
 

·
Registered
Cheap Kitchens
Joined
·
1,898 Posts


Went to start the car after putting the IC and airbox back in place and it was not a happy camper. I think it was flooded from all the gas leaking into the cylinders during my injector debug. Finally started and it ran an idled smoother than ever! I was letting everything all warm up and was excited to hear the VF42 spool up... and then I heard something leaking..
Hah, this picture said sensitive content and was hidden so I looked for a solid minute trying to find the surprise penis. If it is there you have hidden it really well and my jimmy hat is off to you
 

·
They call me Garrett
98 RBP V4 STi Swapped RS
Joined
·
4,109 Posts
Discussion Starter #2,573
Yayyyyyy I got the car tuned up and out! Drove it with the previous cal and boy was it ready to party! I had to tone it down a little because I was hitting 21psi in 3rd gear at 4k RPM.. which I mean.. I guess if it's happy to do...

Either way, it was clear low RPM, high load driving wasn't gonna work with the previous stuff I had on there so I ahead and did this:

-Smoothed out my timing (this was done last year and I just never uploaded it)
-S204 turbo dynamics
-S204 target boost table (little bit modified for more low end)
-S204 WGDC Init/Max (modified a little to interpolate)
-S204 Boost cut (just bumps up absolute pressure like 3psi)

What I noticed:

WAAAAAY more response than the VF37. Usually when I swap between my Mini (1.6L supercharged) and my Impreza I notice a very distinct lack of response in the Impreza below 3k RPM compared to the Mini. Makes sense, it's just the nature of both engines. With the new turbo that lack of response in the Impreza is completely gone. Well sorta.. I made it worse with the NEW cal. More on that in a second.

Either way, the response is MASSIVELY improved and I love it. I tried to show it with some logged data.. but both cases aren't quite the same so I wouldn't call it a perfect apples to apples comparison.. either way, the data backs up exactly how it feels. I found a 1-2 pull at WOT log with the VF37. At 5300 RPM in 2nd gear, full WOT and had been after a shift for a while, I was at 9.82 psi. Same (ish) 1-2 pull with the VF42, not at 100% throttle (66% max throttle, this was as I was still increasing throttle postion) after a 1 to 2 shift, at 5300 RPM i was at 14.66 psi.

And that is all with a WGDC table that is neutered compared to the stock tables with the VF37. Amazing. I'd dig out the old log, but it's a pain and not exactly easy to directly compare to.

Anyways.. back to that neutered WGDC. tl;dr new map isn't getting to full target boost (~21psi with compensations added). This is because the WGDC is hitting the max allowed that I programmed... so in order for the turbo to be able to give more boost, it needs to keep the WG closed a bit more. Easy enough. It's important to note that the WGDC IS hitting the max value, because that means our turbo dynamics are still okay for the time being. We are adding enough WGDC in with the proportional for it to hit max. If we saw that the WGDC never reached the max, was still getting added to, and we still didn't hit target boost, we would know the proportional would need to be modifed a little in order to give it more beans when it is spooling up. Integral control is a bit more complicated and I don't know how to explain that yet, because I don't understand the "error of time" caluculation. Anyways.. not the point... I will dial the car in, but for now I can drive the car safely, and even still have fun with it!

Another cool thing to note about the turbo dynamics PI controller is the difference between the VF37 and VF42 in the respective stock ROMs. Comparing the VF42 values to the VF37 values the VF42 allows for way LESS correction upon error, positive or negative. Why would you want less correction when you could have more?? Simple.. you're not moving around as large of a mass (think turbine CRHA) with the VF42. So the inertia is lower and you need to give it less of a kick in the ass to get it going, and less correction when it deviates from it's target. It's like trying to move a ball full of air, vs a medicine ball. When you flail the ball full of air around, you don't need to fight it as much to move it back the other way. If you apply too much correction to the light ball you'll start oscillating because you're moving it back and forth too damn fast from where you actually want it. The medicine ball actually probably will move YOU when you move it, so you have to apply more force than the ball of air to get it where you want it as you move it.

I love when the physics of things makes sense to the practical application of it!

Anyways, now that I've lost 90% of anyone reading this, here's where I give up my goods:

Current ROM: https://github.com/equinox311/V8-EJ207-tunes/blob/master/LC_added_Smoothed_Timing+S204 TD + S204 Boost + S204 WGDC + boost limit up- 4-25-2020.bin

Full initial log of the run (there's a lot there):datazap.me | equinox92 | Impreza VF42 - New tune
View of a decent 3rd gear pull: datazap.me | equinox92 | Impreza VF42 - New tune

Aaaaand pictures!



 

·
They call me Garrett
98 RBP V4 STi Swapped RS
Joined
·
4,109 Posts
Discussion Starter #2,576
More tuning stuffs!

So I noticed on my drive home from work, I had a pretty gnarly knock event. Went to look at the data and here's what I saw:

tl;dr



Looks like a super lean spot on tip in... tip in enrichment not doing its job.. which is very odd.

Dug into things a little more and found this table..


Basically (it's already correct in my boost table in this picture) it looks like there's a tip in enrichment compensation based on boost error... I'm not sure why.. but it looks like at 0-2.48 psi of positive boost error, it will pull out tip in enrichment. When I modified my boost table, I think I just fudged the numbers a bit during highway cruise such that it put me into this category of pulling out tip in enrichment.. so I changed the tables back.

I think what I may end up doing is taking out the tip in decrease... but since I don't understand why you'd do that in the first place.. it stays.

After that, I flashed on this new map: https://github.com/equinox311/V8-EJ207-tunes/blob/master/LC_added_VF42+more WGDC after initial+cruise boost tables 5-1-2020.bin

Along with the modified boost target, I added about 15% more WGDC... haven't been able to do a proper log... my logger decided it didn't want to record data on my drive today.. but initial impressions are good. Hitting 19lbs in 2nd gear, response is better, and no knock when stabbing the throttle like in that event.. but again, I need more testing to call it solved.

Yay!
 

·
They call me Garrett
98 RBP V4 STi Swapped RS
Joined
·
4,109 Posts
Discussion Starter #2,578
Back with more tuning stories because I finally flashed the car again! I'm just ever so slightly bumping up WGDC to fine tune that turbo response. This time I bumped initial up ever so slightly and also bulk raised the max a TINY bit.

After going for a drive, I really should have done more, but I guess a little a time is better than a lot and then boost cut.

Either way, car feels like it spools a little faster, as it should.. but I really still need to add in a LOT more initial because it is lacking the response at lower RPM i desire. The stock cal really gets the turbo moving at 90% DC initial under 3800RPM.. and I think I need to take that shotgun approach because everything over 4K RPM is working exactly as I expect it to given the calibration I have.

I also got rid of that tip in reduction in fueling... I think I was able to convince myself it was for emissions reasoning they did it (in reality STILL have no idea) so I decided because of that to disable it.

Here's my changes compared to my last map, and the ROM: https://github.com/equinox311/V8-EJ207-tunes/blob/master/LC_added_VF42+more WGDC initial+ no tip in pull out 6-6-2020.bin

83683


And here's my really sketchy log going on an onramp to 75 north during construction: datazap.me | equinox92 | 6/6/2020 Revision

83684


Yellow marker denotes where WOT was achieved. Spool is still kind of bad, but it was a hot day out. I also forgot to log IAT -_-

Dunno! Gonna keep sending it!
 

·
They call me Garrett
98 RBP V4 STi Swapped RS
Joined
·
4,109 Posts
Discussion Starter #2,579
Oh also, totally had an Optima red top blow up on me. The sweet smell of sodium hydroxide killing me won't be missed.

Also, good luck trying to warranty an optima. The process is terrible and you'll end up just shelling out more sheckles for another. Been in the warranty process for over a month now still.
 
2561 - 2580 of 2587 Posts
Top