GTSBoy

Fix the stuff that is obviously broken, obviously. All the engine bay rubber stuff, timing belt, water pump. Plan to do that immediately.

The wires literally START at the 12v rail and run to the earth rail. I have NEVER seen arrows on an automotive wiring diagram. This isn't kindergarten. Just follow each wire back from the plug (on the diagram). One will go earthwise and the other will go to a power source.

What do you mean lol? I had one. It did that. Except the death part. I f**ked it off for a Toyota before it failed.

Nah. Subies are famous for head gaskets from the first until the last. ~2010 Forester and anything else with that woeful EJ253 or whatever the hell they call it. 130000 km + or - and they will start to weep at the gasket. Don't fix it within 15000km and it will shit the bed.

More boost.

Many cars will push a little bit more coolant out to the overflow bottle every now and then, and will lose it if the overflow overfills. Then when it cools it will pull in some of the original overflow volume, leaving you with less. If this happens a few times, it will look like it is being lost in the engine, when it's not really. Some engines will lose some through seal leaks, drips, etc. Subarus will usually lose it because they are shitboxes in which a blown headgasket is a matter of when, not if.

You can always work out which wire is which and where the power comes from from the wiring diagram.

Firstly, it is ~30 years old, so it will not have 85k miles on it. It will be lucky if it is less than 185k miles. Is it still the original RB20DET? If so, budget on an engine upgrade to something that can make it move. RB25DET Neo, VQ30DET, VH56DE, LS3. Anything. If it is the original engine then everything rubber is suspect. Vacuum hoses, coolant hoses, intercooler connectors. Budget on replacing all those, plus the timing belt, water pump, front and rear seals. The gearbox should be fine if it is not crunchy, but it will not tolerate significant upgrades in power for very long without wearing out the input shaft bearing. So if following advice to embiggen the motor, budget on adding a big box too. Ditto the brakes. Bigger motor = bigger brakes. The suspension could be in any state at all. If all original, then most rubber bushes will be crook and need poly replacements or entire arm replacements (to obtain adjustability). The front upper control arms are shit on R32s and the only workable aftermarket substitute is by GK-Tech. Those stays on the rear wing are stupid. Otherwise, go for it. It is very similar to the car that I have owned for the last 20 years and have done the entire list of upgrades above, and more, to.

Are you doing this without the freely available wiring diagrams in the workshop manual?

R34 looks awesome while 33 looks like a bloated whale.

Well, I did say the only CU in that spot was the TCU.

Porsche Coxter in one of the hot hot specs. Lotus Exige if not for the sudden addition of MCM tax. Or find a well looked after FC RX7?

What are you talking about? "Order" vacuum lines? No-one buys OEM vacuum lines. Just go look at the car. Work out about how much length you need of each size, go to the shop, buy at least that much in black silicone in each size, and come back and cut it up and fit it.

Block has been horribly compromised. Significant stress in the casting where that force was applied. Probably deformation. The only reason to use an RB30 block is to make bulk power. Who's going to want to trust that block to stay together?

Well, it clearly threw the rod. It's fatal. Why isn't it already at the scrap metal merchant's?

Latest topic updates and latest topics panes on the RHS don't indicate unread/new content with bold any more. It was a bit unreliable before the update, but not doing anything at all now.

The AC signal from the gearbox sender goes up to the cluster on one of two wires (I don't know which, looks like it will be 15 or 16). The cluster VSS does its interpretation and puts out the speed signal to the ECU on another wire (the opposite of whichever of 15 or 16 is the incoming) and it looks like the speedo gauge hangs off of pin 17, although I'm not sure why it has to go via a plug if it is all internal to the cluster. The incoming signal will go positive and negative wrt body earth (ie, pin 22) on that wire (15 or 16, whichever is incoming). The VSS output is 0-5V square wave, on a single wire, wrt body/ECU earth. That could be what they are describing on the above drawing as 2P / VSS. The rest of the wiring diagram is required to see what wires is connected to what pin. That will answer the question definitively.

They're both the signal. It's AC. Sawtooth. +/-1V or thereabouts.

Actually, not so much. Meters can't pump enough current to demonstrate anything beyond mere "continuity". Even using them to measure ohms doesn't tell you what happens when you try to put actual (serious) current through the contacts. A switch can measure up fine and collapse when presented with real load, because it gets hot or the carbon/varnish moves, etc etc. Those binnacle switches have nice wide contacts but there's really only one narrow path that the sliders run on, and the contacts pivot on fixed brass parts that give a really small contact area. There's plenty of ways for them to give trouble. Dismantling and cleaning with a little CO spray and a wipe with a clean cloth is simple enough and had brought back mine from the brink about 10 times over the last 20 years.

No. Just in case you're talking about a stock ECU. No. You cannot take the AC signal from the gearbox's speed sensor and send it straight to the ECU. The stock ECU expects a square wave PWM signal.

Whilst I don't disagree with a thing you said..... he did say that 3 out of 4 headlights are working. That would suggest that the main switch on the binnacle is working and probably the dipper also. Because there is no left-right switching IIRC. But, I haven't got the wiring diagram close by and couldn't be arsed looking if I did! More to the point, I regularly disassemble the main binnacle switch to clean the contacts whenever my headlights start playing silly buggers**. (That's the price of adding relays into the circuit. The relays don't pull enough current to properly whet clean the main contact when switching.) So, I know it is very possible to dismantle and visually inspect that switch and see that it should be working. Apart from carbon build up or melting, there's very little that could go wrong in it. Anyway, with 3 out of 4 headlights working, it's much more likely to be a dud wire/loom connector somewhere out in the field, rather than in the control room.

Good luck. I reckon there would be somewhere between zero and none anywhere outside of Japan. Yahoo auctions, or a service like Jesse Streeter or Import Monster will likely be the only way to find any.

Doesn't it look like the boot goes to a penetration through the side/pillar that is below floor level? Maybe poke a bit of coathanger through it from the door end of it and listen for scratching noises.

Get the wiring diagram out. Follow from the power source to destination and work out where it is stopping.

Actually, it's not that at all. The turbo business is bankrupt. The turbo business was spun off by itself, but Honeywell somehow managed to put all those Bendix asbestos liabilities onto Garrett on its way out the door. https://www.autonews.com/suppliers/garrett-motion-suing-honeywell-over-ongoing-asbestos-liabilities Garrett is actually in some peril as a result. They launched the lawsuit, have since had to file for Ch 11, and will either disappear or go through period of restructuring/pain, same as any company that goes through Ch 11. The bankruptcy is at least partially caused by COVID-19 because turbo sales are down by something huge, like 40%. Some of that is aftermarket, but I imagine that OEM consumption went right down. For a company that was struggling for other reasons, loss of a significant fraction of turnover can be a death knell.