Spud's "The Definition of Ghetto-Rigged" M90 Supercharger Build Thread (lotsa pics)

[Skidd]

Twin M90s!
Dude... you is nutz!
That's going to be the most ridiculous Impreza ever!

[Spudchucker]

Originally Posted by Skidd

Twin M90s!
Dude... you is nutz!
That's going to be the most ridiculous Impreza ever!

That's the plan



For now I gotta worry about my fuel getting to the engine though

Buuuuuuttttt, I think I figured it out, finally!!


What's wrong with this picture?




Turns out it's not the "hitting 6 psi" that's giving me problems, that's just when I noticed it feeling a bit different. My car is killing off any tip-in as soon as I hit boost, which easily explains the fuel leaning up a bit as I put the pedal down, despite the G/Rev only hitting 1.20 or so when it happens (this confused me because 1.20 G/Rev airflow is well within the abilities of my stock injectors as they are)

So now with my TB only opening 57% at W1/2T (think about it ), and the Tip-In scaler being fixed, I'm willing to bet I can bring in some good news realllll soon


Back to the car I go!

[Spudchucker]

-Update-


HUGE DIFFERENCE first time out! even with it horribly off from what the final numbers will be on the Tip-In map, it's SCREAMING now!!! wow wow wow wow wow


wow wow wow wow wow wow what have I been missing all this time??? I just safely hit a smidge over 6 psi til 4,000 and it was the fastest I've been in a car for prolly 2 years now (I know, weak haha.. I've driven a few WRX's and a pretty much supercar status Integra a few years ago, but nothing recent. I would say it's slightly, slightly less pull than a WRX, but at friggin 2,000 RPMs, WHO CARES???)

I've never seen a 4 banger pull like this at these low RPMs, ever, which is probably why I'm so ecstatic right now Yellow lights shall never be missed again! It's good to know my car is finally starting to act like it looks To think this was all being messed up by one freaking small "hole in the wall" scale buried under mountains of important looking Tip-In stuff haha. Lesson learned, now back to fine tune my girl some more!

[Spudchucker]

Back again, this thing pulls HARD now friggin sweet!



But I have a problem too, hopefully you guys can give me some input. For the first 10 minutes of driving, it screams, and cruises around like a bat outa hell. No knock, good AFRs (11-12.5 while on the pedal, 13.5 cruising, and stoic at idle), and a very fun drive. Then, you can start to feel it slowing down, slowing down more, then bogging down to the point of being concerned. My coolant is all running fine with no crimps in the line, but it kinda reminded me of how driving with a line pinched felt like when that was an issue for me last month.. When I was (nervously lol) making my way through the 1/2 mile home a few minutes ago, it didn't wanna move, and it vibrated slightly when I pushed the pedal, though there was only one or two knocks which I couldn't hear and the AFR stayed right on target, thank goodness.

The vibration reminded me of the pinched line, though it wasn't nearly as drastic and obvious; It was more like it just didn't feel like moving at all, then kinda shook a bit as it took off (well.. not exactly "took off" ) while the pinched line was more of a hard vibration followed by the not wanting to move. That's what leads me to believe the air is only heating up the cylinder walls themselves due to going through so much stress upon entering the plumbing, not the engine losing the ability to cool itself off (this would mean the other parts of the block weren't getting too hot, not as bad.. I hope). I figure hot air will first kill off the power via not having enough density, then heat up and shake the block as an afterthought, while coolant problems will do more on the line of heating up the entire block (causing the shaking) with cooler air still entering the cylinders and not hurting the power as much.



...So is it the lack of an intercooler, or is it the TB only opening 50% at "WOT"? Perhaps a combination of both?

My theory is that having the TB only opening 50% while making the car run a good 30-45% stronger than stock, there just isn't enough air going through it to whisk away the heat. Judging by the difference in performace simply insulating the headers makes, I could certainly see the engine being a heatsink as a problem with the air being stretched through the small initial opening, then smashed into more air, then blown up. I'm gonna go ahead and bring my Throttle Plate opening back to 100% at WOT (won't be even close to as bad as before, now that I've fixed that little "having no Tip-In at all while boosted" problem) and get the intercooler completely hooked up ASAP. If I'm not able to finish the intercooler tomorrow, at least I know that my car is perfectly safe right now for drives under 10 minutes, which is all I ever really do.



Basically what I'm looking for is reassurance that my theory is right Makes sense to me anyways.

[williaty]

If you look at the thermodynamics of the situation, the airflow is not even close to enough thermal mass to make a significant difference in the block temperature in steady state.

I have nothing explanatory to add, sadly.

[reddevil]

No idea whats going on but..... with regards to your fuel pump, its up to the task of flowing bigger injectors UP TO A POINT. That point is most likely be an RPM range where it just can't keep up with the needed flow. However, that flow is something you are more than likely NOT going to need (BIG flow). Right now you need to be rich to keep the pistons cool until you can safely tune to a leaner condition.

[Spudchucker]

Originally Posted by williaty

If you look at the thermodynamics of the situation, the airflow is not even close to enough thermal mass to make a significant difference in the block temperature in steady state.

I have nothing explanatory to add, sadly.

I think it might just be the SC itself, the half open throttle plate is just the added crunch.. 5-7 psi through an Eaton has to be pretty hot without the intercooler. Granted, the problem doesn't go away when I drive it like a grandpa, but it could still definitely be an intake temp problem.. I'm gonna log that tomorrow

Originally Posted by reddevil

No idea whats going on but..... with regards to your fuel pump, its up to the task of flowing bigger injectors UP TO A POINT. That point is most likely be an RPM range where it just can't keep up with the needed flow. However, that flow is something you are more than likely NOT going to need (BIG flow). Right now you need to be rich to keep the pistons cool until you can safely tune to a leaner condition.

THANK YOU for a good definitive answer! I'll go hack em up first thing tomorrow. Gonna tell the ECU they're 530cc to give me a little headroom

[williaty]

I want to jump on something RedDevil said here. Given his experience, I'm 99% sure he knows this, and this is what he meant, but a lot of people don't know it and will read what he wrote in the wrong way.

There's a very common misconception that adding fuel (going richer) cools down the charge and the combustion chamber because the extra fuel carries away more heat. This is 100% not the reason it works. If you do the thermo, you'll see that taking the cold fuel, vaporizing it, and spraying it into the charge reduces temperatures around 3-4 degrees for volumes the size we're talking about here. Absolutely trivial.

The main effect adding fuel has to to make combustion significantly less efficient. In other words, a really rich charge releases much less heat (and also less power, btw), than a charge at a proper AFR. Because less heat is being released by the combustion, temperatures go down. However, your fuel mileage also went down and so did your power.

Interestingly, several tuning books talk specifically about tapering fueling richer as load and RPM increase for the specific purpose of capping total thermal (and power) output to control cylinder head temperatures.

[Spudchucker]

Yeah i always kinda wondered how just a little liquid could take away that much heat, that sounds more like it. Also good to note that running richer than stoic does produce more power up to a point (yes I know you and RD know this very well). 12.5:1 is give or take the best number for torque, while more gas is weaker but safer from detonation, and less gas is more efficient (till you get too lean and kaboom!)


So I just ground my injector caps off, theoretically making them somewhere between 550-580cc. After the most hardcore/awesome idle sound I've ever heard from a Subie (think: massively overcammed twincharged 8 second musclecar idling ) from tuning the mod completely wrong (oops), I took Williaty's advice to bring the latencies back to stock, as that wouldn't change with the same injector core electronnics. Bringing the scalar to 535.55cc's and keeping latencies stock, the idle was very solid and it ran beautifully.

-It never knocked once
-It didn't search at all
-There were no misfires detected
-The AFR was 14.2:1, +/- only .1-.2 (where guesstimated it should be)
-The temps were all very cool
-It basically acted exactly like stock (just straight idling) EXCEPT:

When I put the windows up or down, the idle went low then caught itself, almost exactly like it does if you turn the wheel hard or turn on the AC. Does this mean there's a voltage problem or something? ECU overloaded? My car is becoming self-aware? Crap

Maybe I just didn't notice this before? It completely surprised me when it happened and I've had this car 3 years, so I don't think it happened before



-edit- oh yeah, I have another question. I looked at all the AFC and AFL percentages, and almost all of them were reading 0.0%, even after warming it up for 15 minutes. Air/Fuel Correction #1-2, AF Learning #1-3, etc, none of them left 0% in either direction. The following were the only ones that weren't 0:

-AF Correction #3 was an unchanging -50.0%
-A/F Learning Airflow Range (Current)* (offset) was an unchanging 1.0

[williaty]

There's a decent chance that you didn't notice it before. Electric motors draw maximum current at 0RPM. So when you just start the motor on the way up or "stall" the motor against the end of the window's travel, it draws a CRAPLOAD of amps.

It's also possible that it's having a little trouble controlling idle. I doubt it, if the scalar is only 535, but it's possible.