CovertRussian's 05 LGT Build Thread

Elbow installed straight into the EVAP hard line

http://legacygt.com/forums/attachment.php?attachmentid=259609&stc=1&d=1513487974

 

The other end goes to the electronic solenoid, correct?

Yup, into the outside facing port.

 

I've been hearing a creaking after driving for a while, doing the fender push test put the noise closer to the center of the car. Which means either the swaybar endlink or lower control arm forward bushing was going out. Since the endlinks are fairly new I decided to switch to poly forward facing bushings. I've bought the bushings a few weeks ago, but utc_pyro actually motivated me to install so I could say if they make a difference or not

 

Part Number: W0506

 

 

Before replacing the bushings I wanted to see how much play the stock ones had. While this is hard to notice in pictures, stock bushings have a good amount of in and out and directional play. Subaru went with an interesting design on these, between the crush tube and outside shell there are two metal halfs, that seem to mainly limit how far the center will move (my Infiniti wasn't this fancy, but still had similar amount of movement)

 

Left picture is unloaded, right is loaded with screw driver providing diagonal load on one side.

 

Removing the Stock Bushings

I took the lazy way out and replaced them without removing the arms, which probably cost me more time, but since I can't use my press on these small bushings, I would still be heating and hammering them just off the car.

 

Heated the bushing, mainly the center, to separate the rubber from the metal housing.

 

Afterwards I was able to rotate the center tube with a screw driver and push it out

 

Once the center tube is out, there are two more metal half that need to be removed.

 

Finally can start digging out the outer sleeve. I had to use a screw driver and my die grinder to weaken the shell then nudge the screw driver under it to break the circle's strength.

 

I marred up the LCA arm a little with the screwdriver, ground it down and repainted the surface.

 

 

Installing Poly Bushings

Removing takes the longest amount of time, installation is generally fairly quick.

 

Pressed in the two halves without the crush tube. Notice the "Knurling" inside the bushing, this is what makes good poly bushings not suck on the street.

 

Whiteline and SuperPro both use the knurling, which also helps retain grease so they don't go dry as quickly. Years ago I installed ES bushings in my Infiniti, hated them, I could count every brick and crack in the road. Last year I installed same design Nolathane's (rebrand Whiteline's) in the Infiniti and was expecting the same harshness as before, but I was amazed at the comfort, felt pretty much the same as stock bushings.

 

Crush tube installed, this one was a pain, I would go to push it in, but it would push the other half out, this part took a lot of trial and error to get in.

 

Since these bushings have a lot less give, they are a lot less forgiving to the control arm not being perfectly aligned. Used a jack and screw drivers to get the holes to align.

 

Secured in place and torqued to 70.1lbs as FSM calls. Beauty of poly bushings is not having to torque everything at unladen height. You can torque them with the struts at full droop and it's all good. This is why I bought a set for the '12 outback too, that subframe makes it impossible to fit my torque wrench.

 

Bushing fits the U channel well without any slop.

 

Last thing I noticed was the crush tube on the whitelines had less play in it too, so I measured it all out.

 

Bolt:

 

Crush tube ID:

 

Even though the crush tube ID only .2mm smaller on the whiteline, it greatly reduced the side to side movement of the bolt. Really hard to spot on camera, but once again the play noticeable in person.

 

Results

Overall NVH didn't seem to increase, if it did it's very small and honestly my tires at 40psi probably make the ride quality worse more.

 

As for handling difference, I can't honestly say it's night and day, but there is a slight improvement rear does seem to go out before the front, maybe after I drive more then 4 miles I'll know for sure.

Edited June 7, 2018 by covertrussian

Some more evidence that these bushings are more centered then stock, a couple weeks ago I measured my front toe as 1/32" toe IN, this morning I measured the toe at 1/32" toe OUT. These bushings caused the toe to go toe out by 1/16".

 

Which means the stock bushings were probably slightly off center, which caused the control arm to be off center, which affects camber and in turn affects toe. A better test would be a before and after camber measurement, but I don't have the time or money to pay for two trips to the alignment machine.

 

Also driving around more, car feels great and pretty solid in turns, best part is I still haven't noticed an NVH increase.

Edited June 6, 2018 by covertrussian

Just want to say thanks for all your great post and pictures. Keep it up.

 

Last time I removed/reinstalled the rear shocks I noticed that my rear trailing arm bushings have seen better days. Also since all rubber bushings are torqued at ride height, aligning the rear shock's lower mount can be a real pita, requiring a prybar to push knuckle down, or to unbolt and retighten (at ride height) the trailing arm bushing every time.

 

For this reason I decided to go with poly bushings, which rotate freely and can be torqued at any height. Plus it would be real hard to install OEM style, or metal sleeved poly bushings, since you need to press them in with a press, which requires removal of the knuckle. Since I'm on my bushing replacement kick, decided to finally replace the rear trailing arm bushings especially since these Whiteline bushings have been sitting in my garage since last year .

 

For detailed How to Steps see: How To: Install Rear Trailing Arm Bushings Without Removing Knuckles

 

On the left side the bushings were cracked and rusted.

 

On the right side the bushing started separating from the outer sleeve, I saw it separating while trying to unscrew the bolt!

 

Normally people seem to remove the whole knuckle and just press them out. I didn't want to have to unbolt all of the control arms, hand brake cable, ABS sensor, the axle and shocks. Thus I decided to do it in place.

 

At first I tried to remove the bushing bolt without removing the trailing arm bracket, but the bolt was too long and would hit the body. Thus I had to unbolt and remove the trailing arm bracket.

 

The bushing was still sitting a little too close to the body to safely burn it out or have enough room to use a hammer and flat head. I found that I could rotate the trailing arms down by jacking up the toe arm side of the knuckle.

 

This time I wasn't trying to remove the center crush tube like with other bushings, I actually wanted to remove the whole bushing and then remove the sleeve after. I heated the outside of the bushing enough to knock all of the glue loose.

 

That allowed me to easily rotate it out with just a screw driver

 

Next I removed the metal sleeve with a hammer and a flat head. My left side was so rusted the sleeve came out with a few hits, right side required a little more effort.

 

Both sides had to be cleaned up with a file, rust caused the metal sleeve to bind to the trailing arm. I then painted them and then hand pressed in the new Whiteline W63398 bushings.

 

All done and everything attached:

 

Conclusion

Rear is now NVH did go up a little, it's now on part with front (which is 100% poly now), but it's still very minor and my higher tire pressures (40psi front, 38psi rear) probably make it stand out more then it would with lower pressures.

 

Handling wise, it made a slight improvement, but not quite as big as I would have anticipated from how bad it was.

 

Update: After 200 miles, it does look like the bushing walked out a little:

Edited December 10, 2018 by covertrussian

I'm going to replace my struts this summer and this is one of the "while I'm in there" items I might add on. Thanks for the awesome pics.

 

I've been wanting folding seats ever since I bought the car in late 2013, it almost made me not buy the 2005 to have folding seats, until I saw the 2007+ prices... Anyway I acquired rear folding seats from an 07-09 Legacy Sedan in 2015, but needed more parts and have been putting it off until now (lets just say some of my gun cases are too big to fit in the trunk).

 

More detailed How to Guide: How To: Retrofit Rear Folding Seats into 05-06 Sedans

 

Stock Seats:

 

2007-09 Seats:

 

Stock seats and brackets removed:

 

Lower hinges bolt right up

 

The hardest part is installing the rear hooks which require drilling then using a nut from the back to hold the bolts holding the hooks in place.

 

Which of course makes it a fun time tightening the bolts with nuts spinning easily. Luckily you can kind of fit a wrench from the back.

 

Hooks are all done:

 

The next modification was for the side cushions. They clip in from the top and bolt from the bottom. There needs to be a slit cut to get them to seat right.

 

Test fitting

 

Final touch is the plastic trim and bolting everything up.

 

All done

 

Unfortunately folding seats do come at a price, they are 27.8lbs heavier overall then their non folding seat counterparts, but it's well worth it to be able to fit things.

Edited June 18, 2018 by covertrussian

Nice! I never knew that was possible

Nice! I never knew that was possible

 

With enough time, money and stupidity anything is possible

After a year and 5 months and 12.6k miles the GoodRidge stainless steel braided clutch line (GOO 24216-CLUCL) failed today. Luckily I was able to limp it home before all the fluid leaked out.

 

Failed at the braised to metal connection on the firewall

 

Now I'm not faulting SS line, not at all because my two Nissan's have SS lines that are about a decade old, this is all GoodRidge (My friend's WRX line, that I bought for him at same time, failed too but way earlier).

 

I was half tempted to steel the line from my Nissan, but decided to keep it simple and go back to stock (plus it was starting to rain).

Lower ball joints were one of the first things I replaced on this car in 2013. I had a lot of fun digging out the stock ball joints out of my rusted knuckles.

 

After 47k miles, Mevotech ball joint boots were gone. Now I'm not sure when exactly the boots ripped, but ball joints still had some grease and no felt movement. I actually didn't know they were ripped until I was under the car changing the oil, from the side the boots look fine.

 

To replace them I used the super lazy and quick removal method, where you don't have to remove the LCA. This took me 1.5hrs for both ball joints (with grabbing pics for ya'll).

 

Since I drilled out the knuckle and used longer bolts, loosening the typically painful ball joint bolt was a breeze.

 

I used this method (raise the knuckle up, stick an item between knuckle and LCA, lower the jack and it should pop it out) to pop the ball joint out of the lower control arm, then I removed the ball joint bolt, and used the same method to pop it out of the knuckle.

 

Now I did get a little too lazy and didn't unbolt the swaybar end link first, which caused it to have a little too much power and popped the knuckle a bit too far out, which popped the axle out of it's socket. Don't be lazy and unbolt the the swaybar at very least.

 

I'm surprised it came out that easy, didn't seem like the ball joint really rusted in place like OEM one did.

 

This time I went with OEM everything, I got the ball joint kit off ebay for $33 shipped. For searchability, oem ball joint part number: 20206AJ000

 

Here is where the Mevotech failed:

 

Did you notice that Mevotech is taller? It basically has more back end to it, maybe that's why it didn't rust in place, there is not as much room above the ball joint for water to build up?

 

Installed in the knuckle easily. I then secured it with the side holding bolt, then pivoted the lower control arm down while aligning the ball joint stud & knuckle. Took a few tries but is doable without unbolting the lower control arm.

Finally getting around to posting the 2005 stock spark plugs (SILFR6A) vs the 2010 stock spark plugs (SILFR6B8) data, which I've been working on since May.

 

The biggest difference is the B8 spark plugs have a rectangle platinum welded to the ground tip. The stock ones should have platinum coating on ground strap too, but this provides are finer point.

 

The biggest benefit of this is reduces the surface area under the center electrode which reduces spark quenching. This is basically a factory side gapped spark plug.

 

I won't bore you with the install details again, but I will mention that I did still index the spark plugs (which is a pain on a DOHC Subaru). Indexing the spark plug means you face the open end of the spark plug towards the intake valves where the fresh mix is rolling in. It's a tiny benefit, but it's a free mod (sans time). I mark the location of the ground strap (where it's welded to the plug). Then just make sure that it's not right under the intake valves.

 

Notice that I made some marks on the head, that's the exhaust side, it helped me figure out the picture orientation. Taking pictures of the spark plugs is the only way to really see what's going on on DOHC Subarus...

 

Now onto data and testing...

 

Power Gains

Red line - Stock plugs indexed and gapped to 0.028"

Blue line - SILFR6B8 indexed and gapped to 0.028", Colder day

Red line - SILFR6B8, on slightly warmer day, running -1* less of timing

 

(I'm only doing fuel/occupant weight and temp corrections now, and not barometric pressure (Virtual Dyno forces a default value when you try to temp correct). Temp is probably not required since the runs are pretty close to each other in temp).

 

Power gains were minimal, withing margin of error. Blue line, which is running the same timing, had a midrange bump but IAT's were cooler too. Green line shows that these plugs liked about 1* less timing and thus made more peak power.

 

I wish I had better comparison with less variations, but these plugs required me to retune the car to get it to run similar timing and AFR's, which was a pain and caused more variables, plus the ever changing weather.

 

 

Engine Load

This is why I had to do a retune, old plugs 2.60g/rev, new plugs at 2.45g/rev at similar temps. This meant my timing and AFR tables were now off. Even though my AFR's don't change much between 2.40 and 2.60, the car was dangerously lean (12:1 AFR), thus I had to redo the whole fuel map.

 

MAF Voltage

As expected MAF voltage is lower too.

 

Injector Duty Cycle

Post retune IDC is about the same now, jiving with similar power levels.

 

Fuel Economy

This is the part that took the longest, city MPG tests take a week for me to do, and you can't base data on just one run either, so couple before, couple after and a month has gone by. This can be really seen with temp ranges going up.

 

Highway MPG Testing

 

SILFR6A:

Base IGN: 29.56mpg - Temps: 55-62F

Cruise Timing -2*: 29.68mpg - Temps: 74-88F

 

SILFR6B8:

Base IGN: Didn't test since -2ign map made slightly more power, but after seeing city results I might be testing it...

Cruise Timing -2*: 30.01mpg - Temps: 70-91F

 

Highway Gains: 0.33mpg

 

 

City MPG Testing

SILFR6A:

19.94mpg - Temps: 46-89F

21.04mpg - Temps: 55-82F

 

SILFR6B8:

Base IGN: 21.76mpg - Temps: 64-83F

Base IGN: 21.43mpg - Temps: 67-83F

Cruise Timing -2*: 21.02mpg - Temps: 69-84F

Cruise Timing -2*: 20.30mpg - Temps: 60-80F

 

Peak City MPG Gains: 0.72mpg

Average City MPG Gains: 1.11mpg

 

I didn't expect average MPG gain to be higher then peak, I guess that's why averaging is good, fishes out the outliers. One thing I will say is, weather did get much warmer, but I don't see 55F vs 65F making that big of a difference around town.

 

Conclusion

Why haven't you installed these yet? Slight power increase plus MPG improvements for the same price as stock plugs.

Edited July 9, 2018 by covertrussian

The fact that you had to retune AFR when changing plugs is still odd to me. When I put the SILFR6B8s and cosmos coils on my car AFR was within 1% of what it was before.

I have a question on your statement that AFR of 12:1 is dangerously lean. I thought the higher the number the leaner it is. I'm new to this so please forgive my ignorance. I recently got a COBB stage 2 done on my 06 GT, the up/down pipe and airbox w/accessport. My AFR monitor on the AP shows 14.5 or so when cruising and progressively goes to 11.5 or so at full boost (16.5 psi). The speed shop that installed all this stuff and loaded the tune from the AP says that's right where it's suppose to be. What do you think? I'm a little confused on what the numbers mean. Thanks.

The fact that you had to retune AFR when changing plugs is still odd to me. When I put the SILFR6B8s and cosmos coils on my car AFR was within 1% of what it was before.

 

Right this is weird indeed, maybe my sparkplugs were shot? I got them used from a friend as a stop gap, but they looked fine and power output says they are good. When my other plugs were acting up I lost 40whp and I could really tell .

 

 

I have a question on your statement that AFR of 12:1 is dangerously lean. I thought the higher the number the leaner it is. I'm new to this so please forgive my ignorance. I recently got a COBB stage 2 done on my 06 GT, the up/down pipe and airbox w/accessport. My AFR monitor on the AP shows 14.5 or so when cruising and progressively goes to 11.5 or so at full boost (16.5 psi). The speed shop that installed all this stuff and loaded the tune from the AP says that's right where it's suppose to be. What do you think? I'm a little confused on what the numbers mean. Thanks.

 

While cruising around town your AFR's will be closer to 14.7:1 (stoichiometric value of gasoline, aka how much fuel needed to chemically bind with all the air). Anything above 14.7:1 AFR is considered lean, anything below is considered rich. When you go WOT (wide open throttle) you start needing more fuel due to combustion inefficiencies, basically not all air has a chance to bind with fuel, so you throw extra fuel in in hopes that extra fuel will bind with all air molecules.

 

12:1 AFR is dangerously lean for most turbo cars at wide open throttle, at full boost you need extra fuel to slow the combustion down to avoid detonation and secondary benefit is, extra droplets of fuel soak up the combustion chamber heat which once again helps control detonation. Since Subaru TMIC's heat soak easily, it's recommended to run about 11.1:1 AFR on them.

 

Next your AP is showing you the stock sensor values, which is inaccurate below about 12:1 ratios. Sometimes it will say your running 11.1:1 when your really running 10.5:1, other times it will say 11.1:1 when your really running 11.5:1. From my personal experience though it generally reads leaner then you really are, thus if it doesn't say 11.1:1 I would worry because there is a good chance your too lean.

So, the general conclusion (edit: for the spark plug analysis) is: go ahead and upgrade - can't really hurt and might be a bit better?

So, the general conclusion (edit: for the spark plug analysis) is: go ahead and upgrade - can't really hurt and might be a bit better?

 

That's my conclusion, I've been running for 2 months without any issues. They really helped my city down low torque, which is probably why my MPG gains were so good in the city.

 

There are some anecdotal theories that fine point center electrodes spark plugs can cause pre-ignition due to the fine point glowing easier, which can be extrapolated to fine point ground electrode adding to the issue. But Subaru is using these in turbo OEM cars no problem, thus even at 300whp I don't think it's an issue.

Right this is weird indeed, maybe my sparkplugs were shot? I got them used from a friend as a stop gap, but they looked fine and power output says they are good. When my other plugs were acting up I lost 40whp and I could really tell .

 

 

 

 

While cruising around town your AFR's will be closer to 14.7:1 (stoichiometric value of gasoline, aka how much fuel needed to chemically bind with all the air). Anything above 14.7:1 AFR is considered lean, anything below is considered rich. When you go WOT (wide open throttle) you start needing more fuel due to combustion inefficiencies, basically not all air has a chance to bind with fuel, so you throw extra fuel in in hopes that extra fuel will bind with all air molecules.

 

12:1 AFR is dangerously lean for most turbo cars at wide open throttle, at full boost you need extra fuel to slow the combustion down to avoid detonation and secondary benefit is, extra droplets of fuel soak up the combustion chamber heat which once again helps control detonation. Since Subaru TMIC's heat soak easily, it's recommended to run about 11.1:1 AFR on them.

 

Next your AP is showing you the stock sensor values, which is inaccurate below about 12:1 ratios. Sometimes it will say your running 11.1:1 when your really running 10.5:1, other times it will say 11.1:1 when your really running 11.5:1. From my personal experience though it generally reads leaner then you really are, thus if it doesn't say 11.1:1 I would worry because there is a good chance your too lean.

 

Thank you for replying. It is probably 11.1:1. I have to look next time I take it on the highway. Right now I'm dealing with a transmission problem (5-MT), which the garage tells me is coming from the rear extension housing. It's a chattering noise which shows itself at partial throttle and goes away off-throttle. Full throttle the noise gets less, but the frequency gets higher. It only shows itself under load. If you or anyone has an idea what it may be, please chime in. Thanks.

Haven't heard or seen that issue myself, but maybe your driveshaft U joints are shot? Otherwise might be the rear diff it self.

My bushing refresh is still going, I'm purposefully replacing only one part at a time to see which bushings decrease comfort the most or improve handling the most, thus I'll be providing my Virginia road based NVH rating along with handling rating going forward. I'll have a dedicated thread for all bushings eventually, meanwhile you can get a taste for it here first.

 

 

Part Number: Whiteline W63393

Design:: 1pc Chemically Bound to Metal Sleeve, with Knurling

NVH Rating: 2 - Noticeable on Rough Roads Only

Handling Rating: 1 - Barely Noticeable

 

I bought these used, pre-installed from a forum member. While I would love to have a proper removal/installation guide saving time and money is always nice too.

 

LCA Removal

First need to unbolt the arm from the knuckle and subframe. This is can be rather painful since you don't have enough room for a proper length breaker bar, especially on subframe side.

 

Picture from the top of the two LCA bolts to remove circled, notice I sprayed them with a lubricant.

 

Front bolt is easy to get to, I had a hard time with the rear bolt, had to resort to using my torque wrench to knock it loose since it's long and ratcheting.

 

The subframe portion of the arm slid out easily but the knuckle portion was stuck. Since the knuckle is cast iron it rusted and reduced the diameter, thus I had to clean the rust a little and then use a flat head to pry it out.

 

 

LCA & Bushing Comparison

The bushings I bought came pre-pressed into STI arms, which are identical to LGT arms. Here is a comparison from all angles

 

Here is an up close comparison of the Whiteline vs Stock bushings. Note I didn't apply any heat to remove the bolts, to avoid damaging the bushing more then the 145k miles did. The stock bushing was definitely showing signs of deterioration.

 

Side shot shows how rusty the stock sleeves are and how far the Whiteline sleeves should be pressed in.

 

 

Installation

Just like removal was not easy from the knuckle side, installation also was difficult. I had to tap it into place with a hammer. The hollow part of the LCA should face rearward.

 

The subframe side was really easy to install, this is why I say you should install the knuckle side first.

 

Torque both of the bolts to 89lbs as per FSM, FSM does recommend replacing the two nuts, but you probably can get away with reusing the stock ones through one or two more removal's.

 

All done done with some STI stickers that increase HP of course.

My dedicated auto cross car sprung a coolant leak that I've been too busy and lazy to fix so I took the LGT out for it's first run around the park.

 

A friend of mine (who used to co drive my auto-x Nissan's) brought his 2015 WRX too, so it was a friendly competition. He's bone stock, I have a few mods, but a crap ton more miles.

 

Yes that's a Tesla model 3 hiding, no it wasn't all that fast (but that's probably the driver and not the car).

 

Line up of other cars that are pretty heavily modded (yes that Festiva is in SMF and it's fast).

 

Straight away accelerations, the car is squatting quite a bit on aged stock coilovers

 

Slight right hand turn lean

 

This was a 90* left hand turn, to my surprise the car is actually not leaning all that much.

 

About 90* right hand turn here:

 

Another 90* right hand turn from the back:

 

Now I was on all seasons, I almost installed my old set of tires, but these Falkens are almost gone at ~13k miles so figured I would finish them off.

 

Results & Analysis

The car was very unsettling in slaloms and rapid direction changes, I had to slow to down a lot in the opposite direction turn to prevent it from washing out. With that said though, the car rotated really easily and at one point rear end started walking out which was easy to countersteer out of. This was very enlightening because it further told me that my G20's rear end still grips too much for the tight turns of autocross.

 

Time/place wise I did 39th out of 54, 9.139 seconds behind the fastest time and 7.978 seconds ahead the slowest time. Since those two can very pretty majorly depending on who showed up, I started using percentages. I placed 72% (top 28%), which is about 10% slower then what I usually do on my fully built G20 with much bigger/better summer tires and coilovers. I say overall that's not bad considering that I'm on stock ~140k suspension that's now shot and on 500 tread ware all seasons.

 

This autocross event completely killed my shocks and this is what what I felt while racing and it's even felt on the street now. Time to fit BC coilovers or buy new KYB's...

 

And of course the million dollar question who won: the WRX or the LGT? Well sadly the WRX with 1.88 sec lead, he scored 28th out of 54. I think non shot shocks would have gave him a better run for his money.

Edited July 27, 2018 by covertrussian

Thanks for posting and sharing your day. Looks like you had fun any ways. That's what's its all about.

Indeed it was a fun day, though the next day my tires were really scary in the rain. It would kick out on me real easily, after driving a little on dry ground looks like the tread wore off enough to be fine again the day after.

 

The car is pretty rough around town now, just feels like riding on 2x4's. I guess I wont be replacing any other bushings until new struts/shocks go in.

Took the car twisty mountain driving yesterday, besides being stuck behind some slow cars overall it was much better then last week. Guess the struts/shocks overheated autocrossing but now seem to be better (doesn't feel like riding 2x4's and less twitchy over bumps), but still fairly loose. This brings to the next point though: With all the recent bushing work, with no power applied the rear end will walk out first.

 

Now I'm thinking, why does the front have more grip then the rear now. The only thing I did up front is the forward facing LCA bushing, which I don't think helped the front THAT much. Which brings us to the back three bushings that I replaced (Trailing Arm bushings and two rear LCA bushings). I don't see the rear LCA bushings reducing traction due to their design & location, they should only increase the traction. Which leaves us with the rear trailing arm bushings...

 

On my minivan beam G20, replacing the rear trailing arm bushings also helps reduce understeer. Which leads me to this hypothesis: I think the trailing arm bushings function like secondary springs, by installing stiff poly bushings I effectively increased the spring rate on that axle, which in turn reduces the mechanical grip on that axle.

 

Maybe instead of rear swaybars we should suggest people install trailing arm bushings as the first mod to make the car understeer less. I'm probably gonna set the rear swaybar to softest point, or go back to stock .