GeneralDisorder

What's wrong with the pistons? GD

The trick is going to be cam position sensors and trigger points. You will probably have to block the exhaust AVCS camshaft oil passages and use the Baja exhaust timing sprockets. Assuming the camshaft position sensors on the intake cams are compatible or can be swapped.... It should work. Ask for pictures to confirm. GD

It makes zero difference in the end result what gasket you use. The ports don't match in other areas on those gaskets. Subaru has used lots of mismatched gasket/head combos in stock configurations. We use the 770 gasket on everything. Including 205 heads on 257's making 400 AWHP. In no circumstance has the coolant flow across the gasket interface been any concern whatsoever. GD

You can't buy a scanner for under $2500 that will do what you are asking. And some in that price range wont either. When you reset the computer, or disconnect the battery, the idle relearn procedure will be required. There is no easy way around this. GD

It's not that big of a deal in practice. I haven't seen a major difference in rod bearing longevity between 48 and 52. And the only thrust bearing issues I've seen were on 22 and 22T engines.... The #5 thrust does seem to be more reliable but thrust bearing failure on the #3 engines isn't common by any means. The best upgrade is to just put the D heads on the 251 bottom end using 770 head gaskets. GD

A 13 year old turbo car doesn't sound like the best choice for her. These are neither simple nor cheap to maintain. Fix it and get something more suited to her budget would be my advice. GD

This will yield an unsuitably high compression ratio. Somewhere in the 11.5 range. It's actually pretty close to using 96 25D pistons which yields a compression so high it will melt pistons - over 12 IIRC. Don't do it. The acceptable combinations are: 22E heads with 97 to 99 25D pistons using 25D thick head gaskets. About 10.6:1 25D heads using 99 to 04 251 pistons, and '06+ STI head gaskets. About 10.2:1 Both run quite well on premium. I know of no other combinations that do not require custom pistons and head gaskets, etc. GD

Having examined both heads side by side I can tell you they have the same basic chamber size. The only way to tell them apart is by the exhaust port design. There is no compression gain, therefore no real performance from using them. You get an oddball exhaust header port - that's about it. GD

You don't need to compression test it. Just pull off the valve covers and check the valve lash after correctly timing it. If the lash is really huge on any of the intake valves then they are bent. GD

The alternator only requires 12v from the indicator lamp to flash the field - which it has. Sounds like it's dead. GD

90/91 have bad solder joints on the circuit board mounted connector of the HVAC control panel. This is almost SURE to be the problem. The Subaru relays don't fail often. GD

The diff from the EA82 will bolt up. No problem there. GD

It's not something that can be explained with a post on a forum. You will need to get a manual. Try eBay for old paper manuals. Best idea would be to throw away the 4 speed and get a 5 speed from a later EA82 car. GD

Any tips he might have (doubtful) wouldn't apply to a 97. He has a 91. They are different generations of the legacy platform. They share the name and some mechanical but that's about it. GD

H6's don't have a timing belt. GD

Check the temp of your heater core hoses. Might be a clogged heater core. I suggest you make a separate thread as this has nothing to do with a the OP's lack of airflow problem. And a 97 is only loosely similar to a 91. GD

Chances are very high that the circuit board in your HVAC panel needs resoldered. The 90/91 panels are notorious for that. Just remove, disassemble, and remelt all the solder joints. It's actually very easy to fix. GD

Denso is OEM. NTK should be fine also. They make spark plugs for Subaru. GD

That's rare enough that I haven't seen it (on a Subaru). The valve broke completely off which is possible but again super rare. I've personally thrown a 251 belt at freeway speed and that didn't happen. GD

Your reader probably isn't CANBUS capable. That's actually an air fuel ratio sensor (wideband). Heaters fail. It happens. GD

The intake valves get bent. That's it. GD

Oil consumption of 1 quart, give or take, in 3,000 miles, is not terribly excessive. The engine was ENGINEERED to use some oil. We have gone over this in other threads. Oil consumption, in and of itself, is not an issue. In fact the loose tolerances yield considerably better fuel economy and the price of the quart of oil is peanuts compared to the savings in gasoline. The desire to have good fuel economy and good performance are at odds with extremely tight tolerances that yield little to no consumption. Remember - loose is fast. The only issue is that consumers both do not understand the intent of the engineers and the engineers should have provided a larger oil capacity and a dipstick that doesn't reflect a "problem" till the engine is multiple quarts low. The whole issue is a huge mess. If consumers would just check their oil and add as needed and stop worrying about the consumption and being "inconvenienced" and "worried" about things they patently do not understand this wouldn't be an issue. Similarly if Subaru had designed the thing to not cause customer concern with warning lights and low dipstick indication.... again it wouldn't be an issue. Look to lose 2-3 mpg with the tighter rings. More friction = less consumption = less efficiency. Period. It's a very difficult situation. I build engines and we build them to use as little oil as possible but its still typical to see them use 1 quart in about 4-5k. This is because customer satisfaction demands low consumption and when building turbo engines they need that ring seal or you end up with severe blow by. For these economy isn't an issue. More than 1.5 quarts in 3k is borderline. These engines all have about a 4.8 quart capacity so as long as you have more than 3 quarts after 3k then it's good to go. To an extent - the closer you are to that happy place of 1 quart in 3k (non turbo), the better the economy will be and the happier your wallet will be with fuel economy. Personally - we prefer not to use the 0w20 oils, etc. The engines were not "engineered" to use this. They tolerate it for the US market. The turbocharged variants and most other countries run 10w30 in the same engines. We routinely run 15w50 in our race engines with only slightly larger than stock clearances. It has much higher film strength than the thinner stuff. We generally run either 5w30 or 5w40 in stock engines depending on their age and oil consumption/leakage level. GD

Sounds like the rust probably precludes welding it also. Yes you can use a spacer to eliminate the P0420 code on pre-2005 models. We don't do that as a shop - it's highly illegal. Check out Amazon for 90 degree spacers these work best. GD

Looks like either the clutch coil is getting weak or there is too much clearance in the clutch. Try pulling out the spacers that sit in down in the splined hub of the pump flywheel. GD

If it's that shot, I would have gone to some shops or searched around craigslist for someone doing a part-out and got a used factory cat for $100 from someone. Legit businesses can't sell them but people do sell them on the side for cash, etc. Rust isn't the only issue. Many of the really cheap one's won't actually get rid of the check engine light so they are worse than useless. Broken right out of the box. AND - as I said YOU HAVE AN O2 CIRCUIT CODE!!! The cat is not the problem and never was. You need a new O2 sensor. GD