BoostedBalls

Yup, after checking the plugs (NGK BKR6E) and finding them at .045", I decided to put them at .030". They were kinda crusty, so I'll bet the cat is clogged. I'll pull it off and check.

OK, the problem continues with the O2 sensor disconnected. The problem continues with the knock sensor disconnected. The PCV was replaced with a OEM subaru part. I can hear the turbo whistling from the front of the car when I'm under boost but I don't know how normal that is for the ej22t. The intake is all stock except for K&N filter. I'm thinking that I must have a leak somewhere between the MAF and the throttlebody. any ideas?

Well, I disconnected the O2 sensor and I didn't notice a difference in performance. The erratic excelleration is still there. It's weird though, sometimes it makes crazy power real smooth; then later it will be erratic again. I'm gunna check out the knock sensor. I have a spare from a non-turbo model that I can test in there. I'm getting a CEL, with the O2 sonsor disconnected.

1986 Loyal turbo (ea82t)- 255,000 miles and still ran 1/4 mile in 14.2 seconds at 20psi. just sold 93 legacy with 215,000 and ran beautifully. (besides the torque bind)

That's wonderful news. I thought the 91 only had one O2 as well, but I never crawled under the car to find out. It's been too wet and cold here! The j*ck*ss at autozone probably looked up an 01 legacy or some crazy thing like that.

Wow, I just priced the o2 sensores at Autozone. $114 for the front and $99 for the rear! That sucks! I assume the front one fails more often? Should I just replace both? dang, that's expensive!

Doesn't the ecm stop sampling the O2 during WOT? I wonder if the cats are clogged up... I have a stainless downpipe, I may be installing that ahead of schedule. It's setup for a Garrett so I would also have to upgrade the header at the same time. Maybe I'll just cut the downpipe and modify it for a v-band clamp so I can swap over to the garrett a bit easier.

I don't really know that it was ignition but I believe it was because it was instantanious. It didn't feel like a sudden leaning or over rich condition unless it was really sudden, like the injectors lost their ground. That's possible. Either way, it's working again once I plugged the solenoid back in. Maybe the ecu is cutting the injectors or ignition trigger when it commands the solenoid to open and it doesn't get the feedback? I know it wasn't really overboosting though, I was watching the gauge. I may have to revisit this when I setup a home-built boost controller. That will be after the intercooler and water/alky injection. thanks guys!

UPDATE- Ok, I found one of the 3/8" or so hoses that connects to the turbo inlet to be not hooked up right. I know it was leaking and probably really bad when the turbo was at full song. I took the waste gate solenoid out anyway and connected the two larger lines together and left the small line open. Then I hooked up a boost gauge. Then I drove the car. Just as boost would start to come on, the ignition would cut out??? Weird, the engine actually stalled out while I was loading the torque converter to see if I could show boost in gear with my foot on the brake. So I tested the solenoid and it was responsive and sealed tight and seemed very good to me. I shot some carb cleaner in it and blew it out anyway. It was clean inside. I reinstalled it and went for a test drive. The boost would come on and stay around 9psi but the power was still eratic. It didn't seem as bad though. The boost didn't seem to fluctuate in time with the surging. I'm gunna replace the PCV valve like mentioned earlier and I'll scout around for some more vacuum leaks. in addition to the tune-up parts

yeah, good thought. Maybe I'll suck some seafoam through it at idle. I think this car has been babied and it probably has all kinds of carbon buildup inside anyway. Time for water injection

might also look into retrofitting the 2.2, I have the 2.2 and the 2.5. I like the torque of the 2.2 over the HP of the 2.5 anyway

It does sound like HG but I would first top off the coolant at the plug located on the passenger side of the radiator. If the coolant system isn't full, it will sometimes show an overheat condition even if the temp is fine. If it still does this after topping it off, I would have someone test the coolant for hydrocarbons. That engine likes to have new head gaskets every 75k or so. Just the nature of the beast.

I just bought a 1991 Turbo Legacy:clap: that is in really good condition at 158k except for this boosting issue. When I give it 1/2 throttle or more, it seems to come on really strong then fades for a moment and then pulls hard again, over and over. It feels like the wastegate is opening and closing. 2/3 throttle and WOT have about the same feel. It's funny, even with this issue it's still quicker than my N/A Legacy! I haven't installed a boost gauge yet to see what is actually happening so I'll do that in the next couple of days. I'll also replace the plugs, wires, air and fuel filters. I guess some people replace the wastegate solenoid with a GM part for a similar issue??? It may come to that; I can also wire up an LED real quick into the cab to let me know when the solenoid is powered. anyone care to place bets on the cause?

I lived in Montana for 5 years and drove my car in -45F, and I have used both. Go with the pan heater! if not both The block heater is nice but the main issue here is OIL! Your cold, thick molasses oil can push the filter element right out of the filter and into your bearings. The block heater does NOTHING for the oil that is sitting way down in the pan. The convection from the warm oil will help to warm the rest of the engine also. You must make sure your antifreeze is up to par, obviously. PS: stay away from FRAM filters, especially in the cold weather.

I read the whole thing... At first, I thought, what!? That 'mechanic' installed the belt a tooth off. Then I remembered the comment about "foil hats"; I found that they are very easy to make and- the one I made actually works! As I put the hat on, I instantly began thinking more clearly. But seriously, the explination of everything being motionless is only valid if the parts all have tight tolerances. The torque converter does see engine load at idle and it transmits this load through the axles to the wheels. This is not a constant, smooth torque because we are talking about H4 engine, not Buggatti V16's. Our little engines like to shudder a little bit at idle with a little load on them. It's normal. But if you open up the clearances on one of those axles, you change the stability of that 'stationary' part and this may be enough to get the whole thing shakin'. Harmonics can do some pretty amazing things. Read up on that little harmonic device that Tesla made that caused a multi-level building to shudder like an earthquake. The device fit in his pocket. This is just one more reason I do all of my work myself. If it was my car with this issue, I would have thrown a fit over the 'mechanic' goofing with the idle screw. Then I would distrust his abilities. If I had changed the axle and seen this vibration first-hand, I would scratch my head and I would be sure I must have knocked something else loose in the process of changine the axle.

I'm thinking timing belt. Maybe you lost a seal and the belt got oily, that happened to my 93 legacy once. I usually leave the covers off after doing a belt replacement, helps a lot in troubleshooting later on.

I have heard about using WRX struts and brakes on a Legacy. What else works? I want to lower the car a bit and do some autocrossing with it.

Ah, found a table! the outbacks have the same backspacing as the 1990-2004 legacys!

I want to put some 15" wheels and rally tires on my 93 legacy. I have the option to but some alloys from a 97 outback. I know the offset is different for the outbacks but will they work on my legacy?

I am MS'ing my 87 RX and I need to know what kind of pickup is used in the distributor. I think it's opticle but I'm not sure. I need to know because I have to choose which components to solder.

Well, I tested the compression and it was just over 90 psi in 3 cylinders and 60 in another. I didn't have the throttle open. I'm wondering if one of my HLA's are not pumped up enough to open the valve. The rings and cylinders were good when I put the heads on. I did a liquid leakdown on the cylinders and they looked great before I put the heads on. The valves were lapped and were sealing great. I can't imagine the head gaskets are toast. This engine only has about 2 minutes running time. I plugged the line that comes off the the T-stat housing like it was before, no change. I just need a sunny afternoon to finger out what the issue is.

I'm thinking that it would have been easier and faster to saw the dog, plus you wouldn't ruin the wheel. Just rinse it off when yer done. Remember the scene from mad max?: Max: The chain in those handcuffs is high-tensile steel. It'd take you ten minutes to hack through it with this. Now, if you're lucky, you could hack through your ankle in five minutes. Go. [the hacksaw is dropped next to The Kid, and Max limps off] classic, but seriously, I'm a dog-lover. Now, if that were a cat...

Well, I messed around with it briefly yesturday. It seems to like o start but it doesn't stay running. Sorry, I have not checked the compression or the connection ofn the thermostat. I did notice an unuasually large amount of crankcase vapors escaping from places like the dipstick tube. I can't imagine that the pistons/rings or cylinders are bad. But stranger things have happened. I didn't open up the block because the cylinders looked so good. I just did head work and put it all back together. There is only a few thousand miles on this engine (from what I'm told). I picked it up with a bad idler pully. The guy had just replaced the engine and he wasn't a mechanic, so I got a smoken deal on it. I hope the block is goo because...are you ready for this?...I FINALLY BOUGHT THE MEGASQUIRT II !!! You heard me right. I'll start with the parameters listed on DIYautotune.com's 87 RX buildup. http://www.diyautotune.com/tech_articles/how_to_megasquirt_your_subaru_rx_turbo.htm I'm running a bigger turbo, bigger cams, better header and down pipe, no muffler, etc. It should scream. If not, I have an ej22 that I'll put in place of it and part the thing out to you greedy-subie-part-vultures. Cross your fingers that the compression is good!

You don't really need a box to surround the carb. I've done it many times without on different engines. The dual throat webbers and delortos are best IMO but a nice edelbrock performer 500cfm 4 barrel would work great also. They are extremely easy to blow through. I suppose if I were doing a blowthough on an ea series engine, I would go with dual downdraft delortos, like the ones you see on VW's. The jetting would already be in the ballpark too if you followed a buildup from a 1600-1776cc VW to copy. If you were looking for just cheap, easy and still fast- I would go with a single side draft webber or del attached to a garrett turbo from a mid 80's 2.2l Chrysler product and use it in a draw-through setup. They have the carbon seal that you need for the vacuum on the inlet of the compressor, they are all over ebay really cheap too. You could expect 150-180hp easily with that kind of setup. The drawthrough is a lot easier to tune also. And you don't have to worry about a special fuel pressure regulator, intercooler, BOV, etc. It isn't the ultimate in power but it will be enough to eat a stock WRX in the 1/4 mile.