presslab

I wish I had a spare one, if I did I would crack it open and start hacking. So please do! Take some pics of the boards inside, the bigger the better so it is easier to read part numbers off the chips. It would be very helpful to have the memory addresses for various things in the ECU. Anyone have access to a Subaru Select Monitor with a module for the EA82T? This could be sniffed to find the addresses.

The case gasket is pn 31337AA120, about $5. If you want to, you could also get qty two pn 31954AA071; this is for the transfer valve "Duty C" gasket and pn 31946AA022; this is the filter for the transfer valve. The dealer here in Santa Rosa had the gaskets in stock when I bought mine. The case gasket is 0.4mm thick and is part of the end play calculation, so don't leave it out. Because you are replacing the extension housing, you should definitely check your end play. You might also need to purchase a different thrust bearing. If you don't, it could ruin the thrust bearing and break off the transfer clutch hub. It has happened to others. http://www.subaru-svx.net/forum/showthread.php?t=48505 Download this: http://www.mshoup.us/docs/fsm/EA82/88_XT_Section_2&3.zip Read section 3-2, page 135. It describes how to measure the end play.

The intercooler by itself will add a lot of torque. It will also allow more boost for more horsepower. If you want to keep from cracking heads and blowing head gaskets, your money would be better spent replacing all the coolant hoses and making sure your whole cooling system is in top condition.

Ah yes the new 5MT drives the front through the VC rather than the rear, right? There is a diff in there too?! The '02+ autos had VTD, STi 6MT has DCCD, which is a planetary center diff with electronic clutch, yes?

I'd guess you could pull the rear diff, rear axles, and the rear part of the propshaft. This would save a bit of weight, about 75 lbs. You might save 1 MPG if you are lucky from the weight and friction savings. The newer drivetrains are not designed to be used in FWD mode for extended periods of time and it will wear out prematurely. This is what Subaru says although I can't remember where I read it at the moment.

I have that Ford pump in my EA82T pushing 15 PSI boost with no problems; I set my static fuel pressure to 60 PSI, up from stock of 36 PSI. Mine came with a short elbow hose that I used to redirect the output 90 degrees like the stock pump.

http://www.main.experiencetherave.com/subaru_manual_scans/FSM_Scans/at_transfer_ext._blowup.jpg http://www.main.experiencetherave.com/subaru_manual_scans/FSM_Scans/at_transfer_ext._partlist1.jpg http://www.main.experiencetherave.com/subaru_manual_scans/FSM_Scans/at_transfer_ext._partlist2.jpg 31942AA031 is superseded by 31942AA070

That's Will Mahan's car, WJM on here, right? First time I've seen that site, very cool!

Well the rear is much stiffer (152%) than the old one, and the front is not a bad improvement either. I did notice an immediate difference. The car feels much tighter around town, the steering input has a more immediate response. However on the bumpy back roads the car now tends to get bumped side to side when going over uneven roads; before the new bars the suspension would allow the car to track flat rather than the car following the side-to-side undulations of the road like it does now. In terms of overall speed I think I did not gain that much. I could go almost as fast before but I did not feel like I was in as much control, especially during transitions in S-curves.

Here is the latest; this contains the IDA disassembly along with my new code and a simple assembler. http://presslab.us/705005_modded.zip If anyone is seriously considering this project and needs help, don't hesitate to ask. I won't do it for you but I can "teach you to fish."

A few days ago I installed XT6 front & rear bars, both 20mm; thanks John. The old bars were 19mm front, 18mm rear. The new bars reduced roll considerably. They also reduced the understeer, now it only pushes just a little bit. It's good now, I'm happy with it. I suppose I could put the 19mm back on the front but I'm going to try like it is for a while. The difference in bar diameter might seem small but the stiffness is the 4th power of the diameter. My rear is (20/18)^4 = 152% and front (20/19)^4 = 123%. Thanks for your input edrach!

Ahha, I wasn't just imagining it. In section 4-1 of the Subaru XT service manual: [ATTACH]6052[/ATTACH] Yet another reason to keep the air suspension: Position sensitive damping!

It would be at least as good as any other shock. The synthetic shock fluid I used by Silkolene has an excellent viscosity index of 303; it's one of the best in terms of temperature stability. http://www.peterverdonedesigns.com/lowspeed.htm

After a week of driving on the modified shocks, I am quite happy with it. It's not too firm at all, I've done some very bumpy back roads and it's still pretty smooth. Perfect for a daily driver. The corner turn-in is much more stable, and control is much better especially at speeds over a buck. For someone who doesn't have too many bumpy roads the 15wt would work well, but with the roads we have here I wouldn't want it any stiffer. Something I forgot to mention was how the shocks seemed to have two distinctly different damping modes. I noticed this while bleeding the shocks. When the shock is mostly extended, it is much stiffer in rebound and compression damping. As the shock is compressed, about half way in the travel it becomes softer. There must be some sort of bypass circuit. It makes perfect sense, as you would want two different types of damping depending on if the ride height was set to 'high' or 'normal.' I believe the XT4T shocks have stiffer valving in them; my car is an 88 Wagon with stock rear shocks. I am considering what struts to install on the front. The XT4T shocks are about $40 more each; for that price difference it would be more cost effective to use Wagon shocks and change the fluid. I would also gain the ability to tune the damping.

Here is the header I made for a TD04L; the rear hole is slotted to allow either turbo to fit. The downpipe is stock WRX with a custom 3" catback.

It would be good to know if the problem is fuel or spark when the engine dies. You can pull a plug wire and use an extra spark plug to see if there is spark. I would not recommend testing for spark without a spark plug as the ignition coil can be damaged. My guess is that something in the ignition system is failing, or possibly something loose. The ignition coils can have problems; I replaced mine with a MSD #8222 and I also installed nice MSD 8mm wires. The 20 meters the car moves, can you tell if it has normal power?

I bought some very nice Goodyear fuel injection hose at Kragen of all places. It even has a flouroelastomer (like Viton) liner. http://www.goodyearep.com/ProductsDetail.aspx?id=5144 The EA82T motors can't stand to be overheated, it cracks the heads and blows the gaskets. Other than that they aren't bad, I've been running 15 PSI through mine (with other mods too, stock long block) for a while now.

There's nothing wrong with the parallel resistors. 10W is big overkill but it will work fine of course. Digi-Key is great, if you decide to order from them pick "first class mail" and it'll be like $3 for shipping.

Without the tank your inside vents will go all wonky when you go up hills; the engine vacuum is reduced at large throttle openings. It's annoying to say the least. There is a check valve mounted inside the tank on the engine side. Yours is probably stuck closed. Mine failed stuck open. You can probably just drill right down the nipple and open it up. Then go to your FLAPS and buy a check valve for a couple dollars and put it in the engine side hose.

Go to Radio Shack and get this: http://www.radioshack.com/product/index.jsp?productId=2062322 Take two of them and and tie them together so they are in parallel. As in tie one lead of the 1st one to one lead of 2nd one. Tie the other lead of the 1st one to other lead of the 2nd one. Cut off the connector on the solenoid and solder each tied end to a wire. Tape it all up and make sure it won't short out, zip tie it to something out of the way.

I had previously added some grease into the bladder of my rear air shocks to stiffen up the spring rate. This worked fine enough but it didn't really address my complaints of too-soft damping, particularly the rebound damping. The shocks were only a few years old, so they were hardly worn out. The OEM air shocks are a twin tube design made by Tokico. Because of this twin tube design I was more confident I could replace the fluid without drilling into something important. I drilled a hole in the bottom of the shock - upon pressure release it was pretty strong and on the second one I broke the drill bit. I then tapped the hole for 1/8" NPT - the steel is pretty thick here and seemed fine to tap. I paid special attention to not get metal chips inside the shock by keeping some fluid in the shock to flush it out and also putting some grease on the drill bit and the tap. I then drained the original fluid, just about 6 oz. To determine the viscosity of the original fluid I set up an orifice with a reservoir; I used a stopwatch to time how long it took the fluid to drain out. The original fluid took 9.6 seconds to pass through the little orifice I had set up. I had on hand some shock fluid. The Silkolene RSF 5wt (26.7 cSt@40C) took 11.1 seconds and RSF 10wt (47.36 cSt@40C) took 18.5 seconds. I bought Schrader valves from the hardware store and installed them into the tapped holes. http://www.acehardwareoutlet.com/productDetails.aspx?SKU=40899 I used a large syringe to inject 6 oz of RSF 10wt back into the shock, and topped it off with 130 PSI of dry SCUBA air. I know some people insist on nitrogen but that's hogwash; as long as the air is very dry it is good. To bleed the shock, I compressed the shock while upside down, held it compressed, turned it right side up and extended it. I did this about 20 times for each shock to get all the air out. It's much, much better now! The damping is now firm, a very noticeable difference. Total cost was about $25. I'll tackle the fronts too some day, but they aren't near as soft as the rears used to be. [ATTACH]6046[/ATTACH][ATTACH]6047[/ATTACH][ATTACH]6048[/ATTACH]

You're probably right, the connectors do look the same. I know for sure however that the circuit board of the Legacy TCU vs the SVX are very different. It looks like the microprocessor is a different variant as well, although they have the same instruction set. I have found the gear ratio map: seg006:D74D GearRatioMap: fdb $B24 ; DATA XREF: CalcVSS1+4oseg006:D74D ; 1st gearseg006:D74F fdb $62E ; 2nd gearseg006:D751 fdb $400 ; 3rd gearseg006:D753 fdb $2C6 ; 4th gear So it shouldn't be too hard to fix that problem, too. The TCU uses the gear ratio map to calculate the input shaft speed after the torque converter. This in turn is used for TC lockup and Duty C calcs. Redline shift points are adjustable as well: seg006:CDFF engineRedlineMap:fcb $5F ; DATA XREF: LookupShift+4oseg006:CDFF ; Gear 1seg006:CE00 fcb $61 ; Gear 2seg006:CE01 fcb $62 ; Gear 3seg006:CE02 fcb $FF ; Gear 4seg006:CE03 fcb $61 ; Gear 1, manual modeseg006:CE04 fcb $62 ; Gear 2, manual modeseg006:CE05 fcb $64 ; Gear 3, manual modeseg006:CE06 fcb $FF ; Gear 4, manual mode For example, 1st gear $5F = 95 ($5F in decimal) * 64 = 6080 RPM Another issue is with the differences in VSS signals; on the Legacy TCU the VSS2 was no problem as both the Legacy and the GL10 have 4000 pulse per mile. However it expects a 4.11 rear diff with larger tires so that ratio was off. I found the constant for VSS1 and corrected it. There is also a constant for the AWD slip detection that might need to be changed. Mine was only 6% off so it was no problem and I didn't change it.

Thanks! I've been trying your suggestion about not using as much throttle and it seems to be working. It's funny, 'cause normally more power = more oversteer. Also, I find that when the traction is not as good (wet vs dry pavement) it doesn't push as much. I'm a little concerned about slapping on a rear sway cause I don't want it to be too tail happy when off the throttle. Like if I do something dumb like lift the throttle in a turn and spin out like a 911...

The two plugs sound like a great idea! Did you mean EG33 (SVX) TCU plugs? I've been looking into the ROM on how the TCU gets the engine RPM. I'm nearly certain I found it. On the Legacy $C00B = 4. On the SVX $C00B = 6. This number is used to determine how many times the RPM interrupt service routine is called before calculating the engine RPM; essentially a divisor. So it looks like it's as easy as changing one byte in the Legacy ROM. With a 6 cyl engine on the EJ TCU the most noticable thing would be the forced shift at "redline" which would be too low. Also the engine RPM is used for TC lockup and Duty Solenoid A & C calcs.

I recently put a rear LSD (friction type) in my 88 wagon. I was having problems before with both inside wheels spinning on tight and fast corners; however the handling was still pretty neutral. With the LSD installed the spinning is gone, however I have a pretty good understeer going on when I apply power on corner exit. I guess this makes sense as both rear wheels are trying to drive the car straight. Anyone else notice this? My 88 came stock with rear anti sway. What can I do to get it neutral again, go with an XT6 rear sway? Those things are like hen's teeth...