Greddy TZ518Z (TD05 8cm 18G) install w/ injectors & upip

[drwrx]

I just figured I'd make this official.

Over the weekend of 10/28/06 Me, Zak and my brother-in-law Matt (some of you have met him, he has the red 05 LGT) decided to install the Greddy turbo that had been sitting in my basement for the last year along with the injectors, uppipe and UTEC. What follows is the arduous tale of that endeavor:

First off let me start out by saying that had I known up-front how difficult this install was going to be, I'd have PASSED!

The removal of the TMIC, turbo, up-pipe and down-pipe currently on the car went quite quickly, I was begining to think we'd finish early (my delusions would dissipate quickly).

Next came the injector install which involved removal of the dreaded "green brackets of death" which also came off remarkably easy.
Now for the new injectors. the Ultimate Racing RC 650cc injectors require different clips, which necessitate the injectors to be rewired. As well as the use of a resistor pack which requires leads to be run from the clips to the resistor and back (the stockers are high impedence injectors the RCs low impedence). Zak spent hours with his soldering iron sorting out the massive spaghetti pile of wire for this aspect of the job. Meanwhile, Matt and I tended to the install of the RC injectors which are nearly 1/2" longer than the stock injectors which required us to use a rather unsettling amount of force to bend the fuel rails out and back to get the new injectors in place. Not to mention the fact we had to get longer bolts and nylon spacers to hold them tightly together. A great deal of measuring and a very successful trip to the hardware store were required to get it all matched up. Needlss to say the GBOD were NOT reinstalled. This is where aftermarket fuel rails probably would have come in handy.

The used, but newly wrapped, up-pipe went in with some effort. But nothing too difficult for us to manage. I used Collins trick of cutting the probe and using it for a plug in the temp bung.

As the turbo was going on, our next issue arose. Where do we route the coolant line, since the turbo is oil cooled only? After some debate it was concluded that a straight hose run from the coolant reservoir directly to the block would be best. Unfortunately, I did not have a section of heater hose long enough and by now it was after 8:30 pm and all the stores were closed. It was also very dark quite cold, and we had not eaten. Thus ended day one.

Early sunday morning saw us all back at it with a trip to the auto parts store for the needed hose and other miscellaneous bits. With the hose rerouted and the tubo in plce came our next obstacle, a project ending one at that! It appeared that the previous owner had stripped the threads on the oil input port on the turbo (very bad) and the banjo bolt was destroyed trying to get it to seal. All was not lost as we were able to find both another banjo bolt at Advance Auto parts, as well as a tapping tool to match the tread pitch at the hardware store! Some gingerly turns of the tap and some careful cleaning of shards later we were back in business!

The rest bolted back together without much difficulty (I'll save the tale of the TMIC as some of you have heard it before) and all that was left was to install the UTEC and fire her up and HOPE everything was copacetic.

After dry cranking some oil into the turbo, we decided it was now or never and proceeded to fire her up. She turned over roughly and the idle was anything but smoothe, but she was running! There were no signs or fumes of gas spewing from unseated injectors, no smells of oil gushing from unsealed oil input lines. the only smoke was from the curing of the header wrap on the up-pipe. Suddenly a steady dripping of coolant indicated something was awry. We found the culprit, my new coolant hose had been clamped too far down the neck and was letting coolant by. Sliding the clamp up and letting the car run soon set that right. She was running albeit a bit lopey at idle.

With a quick 9pm run around the neighborhood it became clear that my HKS EVC IV needed to be reprogramed as it was clearly NOT controlling boost as the turbo would quickly spool to 25lbs, but that would done another day. Everything was in, the turbo was spinning, the injectors were spraying and UTEC was managing the works with no drama.

Tuesday I was able to reprogram the EVC with just a few 3rd gear runs, with my low setting at 18lbs and the high at 20lbs.

Now for my impressions:
1. HOLY CRAP! Even at the 18lb setting this turbo pulls like a beast. My motor now doesn't have a high enough redline. At even moderate throttle input first and second gear disappear in a flash. At full throttle, my hand has to stay on the shifter as the car goes through the rev band so quickly. Third gear, which before was such a great long pull gear, now seems short. I only get a brake once I reach 4th. The charts I have seen indicate that this is a 325-330whp turbo on a ej205 with 93 octane at 21psi (dynojet). Based on how it pulls right now I have no reason to doubt those figures.

2. The idle is not right. At 750 RPMs it's very lopey and even feels uneven, especially during cold starts. Once warm thing seem to settle a little but once a fan comes on or any load is put on the alternator, the rough idle reappears. My hope is the UTEC can control the idle setting and I can just raise it a bit. Anywhere above 800-1,000 RPMs and it's just fine. I'm also getting lean condition CELs and cylinder misfire CELs at idle.

3. It's very smooth just driving around, and at mild throttle the turbo responds well and gets out of vacuum quickly. Although there is a bit more lag compared to my ported and clipped vf22, once your on boost, HOLD ON!

4. I'm going to have to be very careful now, as this set-up could clearly be the demise of both my block and transmission.

[complacent]

When are you going on the dyno for tuning? Going back to ultimate?

And WOW, that does sound like a royal PITA to get everyting in and working.








(and yes, watch out for your trans, it's living on borrowed time now.)

[Mr Kleen]

wow, the wagon wasn't slow before...

and this story illustrates why, if I decide to upgrade the turbo and injectors, I'll be paying somebody to do it.

[drwrx]

When are you going on the dyno for tuning? Going back to ultimate?

The new UTEC was tuned by Phil Grabow of Element Tuning for this very set-up and at 20psi. While I may want to do some minor tweaking, like adding some fuel up top, and setting it up to work with my IC sprayer, I don't think I'll need a tune.

That being said, I do plan to take it to the next MachV dyno day and do some runs with an AFR readout to see just where I'm sitting.

wow, the wagon wasn't slow before...

More is always better!

[drwrx]

this story illustrates why, if I decide to upgrade the turbo and injectors, I'll be paying somebody to do it.

I've been through injector installs four time now. Stockers are a breeze by comparison. 2 people can do a stock set in a couple of hours.

[complacent]

So the rough idle and such are just going to be injector pains and you're going to deal with it? Just curious.


and i wanna ride!

[sirwilliam]

So the rough idle and such are just going to be injector pains and you're going to deal with it? Just curious.


and i wanna ride!

No disrespect to Julian, but I think my injectors are giving me problems as well. I will try to get around it w/ some tuning adjustments from Agile...if that doesn't help, I will try to use StreetTuner on it. If that doesn't help, I may just put the stockers back in.

With this new turbo being only oil cooled is it beneficial to get one of those larger oil coolers to mount on the front?

[drwrx]

So the rough idle and such are just going to be injector pains and you're going to deal with it? Just curious.

I haven't spent a great deal of time trying to find a fix yet, but if the UTEC has adjustment parameters for idle, there should be a solution. And for the record, it really isn't THAT bad, it's just a noticeable change.

and i wanna ride!

Absolutely! We may need to set-up a day, since I'll be missing the next meet and who knows when we'll have free time once the Holidays hit.

With this new turbo being only oil cooled is it beneficial to get one of those larger oil coolers to mount on the front?

I wouldn't think so. There are plenty of "oil cooled only" turbos out there, even on race cars. The only thing that that I will have to do is to let it idle for a bit after running. But no worries there, that's what I have a turbo timer for.

[sirwilliam]

With this new turbo being only oil cooled is it beneficial to get one of those larger oil coolers to mount on the front?

I wouldn't think so. There are plenty of "oil cooled only" turbos out there, even on race cars. The only thing that that I will have to do is to let it idle for a bit after running. But no worries there, that's what I have a turbo timer for.

Does that means that your oil is going to heat up/wear out faster than normal? Meaning, you will have to change it more often, or will it be the same?

[drwrx]

The turbo holds less than half a quart of oil, and it moves through rather quickly. It is not any more likely to be broken down or overheated any more than in a water and oil cooled turbo charged car. It does mean that the turbo will not cool as quickly.

That being said, changing the oil will become more important, though not necessarily more frequent, and if I didn't already run synthetic, now would be a good time to switch as it is far less likely to be coked by a hot turbo than regular dyno oil.

I should also say that most racers I know put new oil in on a race day and change it out soon after.

[Sabre]

Congrats Darrin! Sounds like a heck of an ordeal, but the rewards sound great too! I definitely want a ide next time I see you!

No disrespect to Julian, but I think my injectors are giving me problems as well. I will try to get around it w/ some tuning adjustments from Agile...if that doesn't help, I will try to use StreetTuner on it. If that doesn't help, I may just put the stockers back in.

Hey Will, you REALLY need to get that tune looked at...

[drwrx]

I wanted to up-date the situation following a night of fine tuning with the aid of Zak.

The idle issue was fairly easy to take care of. We simply richened up the 0% load value by 10%. The idle has smoothed out and the Lean Condition Bank 1 and the cyclinder misfire CELs have disappeared. There is still quite a bit of fine tuning that needs to be done. The MAPs that Phil made for the other car don't quite work so well on mine:
1. The other car's boost was clearly dropping off significantly as the revs moved higher, where as mine stays on boost all the way to redline. Phil had a high amount of timing advance up top, where as now I'm having to pull it back.
2. There are crossover problems. Going from full throttle in 3rd, shifting and getting back on full throttle will almost always cause a low count knock event early in 4th. And any downshift into full throttle will likely do the same. As will coming off the throttle and then going immediately back on at high revs.
3. The throttle responce at vac is very, very slow, whereas once boost is detected the timing feels almost too advanced as it can be a bit jumpy. I know this has to do with open and close loop fueling, but it's a bit tough to tune for.

I'm also getting great gas mileage. After a saturday of driving over 120 miles I burned just over 1/4 tank.

Lastly, MachV is having a dyno day on saturday Dec. 9th. I plan on getting an early spot and see what kind of power I'm laying down and see where my AFRs are. Everyone is welcome to come and watch the results.

[cheshirecat]

Congrats on the install.....!

Question:

My Subaru friends in FL (i just moved here) all mod the stock injectors to 800cc and they seem to work quite well, not to mention using the stock boost control with an upgraded GM solenoid. We also use ECUExplorer combined with ECUFlash for tuning.....as well as the EVO guys...

Any experience or comments on those? I'm fairly new to the subaru world (came over from Nissan...) and tuning (at least ECUs) as well. I have noticed a tendency to "over mod" with the Soobs...

[Sabre]

Glad to hear you all worked through the problems with the idle and timing up top! I wonder why the other car's boost fell off on top. Seeing as how it's the same engine, it just seems odd! Maybe you have a better flowing exhaust system or your engine is in better shape!

Are there any updates to the UTEC that you have to help with the close/open loop transition?

[drwrx]

My Subaru friends in FL (i just moved here) all mod the stock injectors to 800cc and they seem to work quite well, not to mention using the stock boost control with an upgraded GM solenoid. We also use ECUExplorer combined with ECUFlash for tuning.....as well as the EVO guys...

Any experience or comments on those? I'm fairly new to the subaru world (came over from Nissan...) and tuning (at least ECUs) as well. I have noticed a tendency to "over mod" with the Soobs...

I've heard and read good things about ECUFlash but I have no experience with it. I have much more experience with the UTEC and some with the Accessport and have local support for those so I tend to stick with what I know. You are correct, the GM solenoid works very well on WRXs with stock boost control. However, I had an HKS EVC IV boost controller left from my Galant VR4 and transfered it to my WRX immediately after it's purchase in 2002. That was many years before the advent of ECUFlash on the WRX or the use of the GM solenoid. There are also some benefits of the EVC IV: 1. it allows me to change boost on the fly. 2. It monitors boost pressure in multiple spots for overboost and lowers boost accordingly as well as beeps when overboost is occuring. 3. It modifies boost based on atmosheric conditions (getting 18psi at sealevel is very different than at 10,000 ft.). 4. The EVC builds boost MUCH quicker than the stock controller with or without the GM solenoid and it holds boost indeffinately. This is both a positive and a negative when it comes to tuning.

The only reason I went with the RC injectors was that the UTEC was already tuned with that turbo and those injectors. Had I known the issues I was going to have with the install I'd have gone some other way. And yes most people use modified stockers without any issues, Julian is a good example. However, there is work that has to be done to make them work properly, and they are by no means foolproof. Will can speak-up as one who is having issues.

If by "over mod" you mean add things that don't necessarily add performance you are correct, and I may be an example. But that isn't much different than lots of other folks who "mod" cars. If you mean add more power than the car can handle, than i deffinately fit that bill as well, as my car's engine and tranny are probably now near their limit, or as Colin put it, "on borrowed time".

Glad to hear you all worked through the problems with the idle and timing up top! I wonder why the other car's boost fell off on top. Seeing as how it's the same engine, it just seems odd! Maybe you have a better flowing exhaust system or your engine is in better shape!

My guess is the other car was using the stock boost controller and they all pull boost up-top. Phil would have known this and added timing to keep the power up.

Are there any updates to the UTEC that you have to help with the close/open loop transition?

My guess is there probably are, but I will always have issues between the interaction between the ECU/UTEC/EVC. I spent some time with Nathan at TurboXS discussing this very issue and his solution was always the same, "remove the EVC from the loop" I'm not ready to do that yet. I may have to spend more time getting everything to work properly, and it may never be "perfect" but I'm willing to live with some inperfections.

[zaxrex]

One option is to lower the crossover point to a very low load site. This would require all load sites be mapped by hand, as you would be forcing it to go closed loop.

A variant on that would be to log the open loop IDC at a high loadsite crossover based map. Then try and tune to replicate those IDCs on a Closed Loop Map set. That way you could leave the Timing values set to ECU to provide better transition feel.

Third would be to place a larger (less restrictive) restrictor pill in the turbo pressure pipe to slow down the ramp up of the boost.

Ultimately, you are just running up through the load sites too quickly for the fuel to keep up with your boost response.

Getcherself a FMIC and all your problems will go away...

[cheshirecat]

If by "over mod" you mean add things that don't necessarily add performance you are correct, and I may be an example. But that isn't much different than lots of other folks who "mod" cars. If you mean add more power than the car can handle, than i deffinately fit that bill as well, as my car's engine and tranny are probably now near their limit, or as Colin put it, "on borrowed time".

No, no, not you. Your mods are well thought out IMO. Even testing the limits of block and tranny are ok in my book (esp when you are married, if you break something it is an easy way to upgrade as opposed to replace ) I'm more or less refering to situations like:

open bov - no regard for rich condition

CAI - no regard for maf structure, obviously there are well structure/thought out intakes out there that take the maf into consideration

huge turbo - running a gt35r without the supporting mods or even the capability to really use that turbo (i think that is a bit of a supra mentality)

high dollar up/downpipe cobb, perrin, etc (while great companies, people forget that it's just a pipe) when an ERZ works just as well if not better

...etc

the wrx is an great platform and like most subarus, kept in context it can become an awesome platform

I bought an uppipe/downpipe for 150.00
Tactrix cable $80
ECUExplorer - FREE
ECUFlash - FREE
Dave's exhaust - ~250.00
Various suspension mods (i spend more money there)
I didn't use the software b/c it was free, I used it b/c i wanted to understand how to tune. However, there is something very nice about having a UTEC and taking it to a local tuner who really really knows what he's doing......i hope to be that guy one day

Another route could've been
Cobb Accessport - $645
Cobb Up/Downpipe - ~$675
Cobb Catback - ~$645
...

I'm not sure if this explains my "overmod" statement or not, if not just my $.02 then....

Thanks for your time, sorry if I highjacked the thread

[Sabre]

^^^ lol, you, Darrin and Zak are going to get along very well. They take pride in their ability to save tons of $$$ by going the non-standard routes!

[drwrx]

One option is to lower the crossover point to a very low load site. This would require all load sites be mapped by hand, as you would be forcing it to go closed loop.

Ideally this would be the best solution, however the second part of the statement is what is keeping me from doing it. How many hours do you think THAT will take?!

A variant on that would be to log the open loop IDC at a high loadsite crossover based map. Then try and tune to replicate those IDCs on a Closed Loop Map set. That way you could leave the Timing values set to ECU to provide better transition feel.

This is pretty much the other side of the coin, but the work load pretty much remains the same.

Third would be to place a larger (less restrictive) restrictor pill in the turbo pressure pipe to slow down the ramp up of the boost.

Actually, I'm not certain it would, as the EVC is constantly adjusting the boost curve it may see the restriction and just modify the boost curve to compensate. With the EVC off all these issues disappear.

Ultimately, you are just running up through the load sites too quickly for the fuel to keep up with your boost response.

Ding! Ding! Ding!! That's it, Zak gets the prize! It all comes back to that simple priciple. I guess I just have to take the bad with the good, or spend a LOT more time with UTEC

Getcherself a FMIC and all your problems will go away...

Just keep pitching it. I'm sure I'm going to buy at some point.

open bov - no regard for rich condition

CAI - no regard for maf structure, obviously there are well structure/thought out intakes out there that take the maf into consideration

huge turbo - running a gt35r without the supporting mods or even the capability to really use that turbo (i think that is a bit of a supra mentality)

high dollar up/downpipe cobb, perrin, etc (while great companies, people forget that it's just a pipe) when an ERZ works just as well if not better

I bought an uppipe/downpipe for 150.00
Tactrix cable $80
ECUExplorer - FREE
ECUFlash - FREE
Dave's exhaust - ~250.00

Your singing my song brother, not to mention preaching to the converted.
There is NOTHING on my car that was purchased new (except the car itself) and I am a firm believer in buying the least expensive stuff I can find. I always call myself the "cheap guy" in the club, although Zak can argue that he has that title (although we may just be under different governing bodies like boxing).

[drwrx]

It's all set!
Saturday, Dec. 9th, 9am at MachV.
I have the first slot.

[sirwilliam]

It's all set!
Saturday, Dec. 9th, 9am at MachV.
I have the first slot.

Hmm, who is going to tune your ride? Or are you just dyno'in' it to see what power you are making?

[drwrx]

I'm just running it to see what numbers it's laying down and (Most Importantly) where my AFRs are.

The UTEC was "tuned" by Phil Grabow of Element Tuning and was running on the previous owner of the turbo and injectors.

His setup and mine are nearly identical (big TMIC, up-pipe, TBE, ect.)
I believe the only differences are that I have headers and I'm using an Electronic Boost Controller (it appears based on the MAPs that he was using the UTEC boost controller)