log1call

I might have a manual... you could read up about transmissions. There is an 02 Impreza transmission manual here... http://cid-4ca3c3459aaa7f7f.skydrive.live.com/browse.aspx/.Public Some of those other manuals have trans sections in them too.

Have a browse through here... http://endwrench.com/main.php?smPID=PHP::search_articles_action.php&

Yeah but RD I did say the first sign.

Hard to get into first and reverse is the first sign of a worn out clutch!

It will work but it's a really old system and there are simpler solutions now. If you really want to connect to the serial port, because your laptop doesn't have usb say, then you would be better off using a rs323 chip rather than all those reststors. If you had a usb port then you can buy a lead with a rs323 chip built into the usb plug. They will run straight off as a serial connection to almost anything. You will need software to go with that plug and most of the software out there is for use with OBD2 interperating chips, so they won't work with a serial connection. There is a limited amount of serial connected OBD2 softawae and most of it is fairly limited in the information it will gather. If you have an OBD2 car it's better to buy a chipped lead made to interface between the software and car. Have a read up on ELM 327 chipped code readers.

Yeah... ultrusound I think it is.... clever fellows those Kiwis!

See on that white spot there is a little dark bit on the right side high up, and another dark bit to the left of the white patch? Well those marks look like the rings are not touching there... perhaps. It's really hard to tell from a photo and not knowing whether you have wiped it with a rag, how long it sat before being pulled apart etc. Also, just to the left of centre on the white patch there is a mark/scratch thet runs up the bore... That could be a crack. Best thing is to take it to someone that has had experiance or get someone around. Photos of shiny metal are really deceptive.

It's a pretty simple circuit Kosto, the wire going from the start position of the ignition switch to the starter branches off to the ecu. It's pin ten on the ecu for a 1990 legacy, you'd need to check for the rs though. When you crank the car pin ten should be getting cranking volts which is about ten volts. I'm not sure where the branch actualy happens but if it's not getting the power I'd just run a new wire from the key switch to the ecu.

You should change the rings. First off, how do you think you are going to get all the metal out of the motor with the pistons still in there? And, if you do the rings, bearings, valve grind, seals and gaskets, it should all wear out together. Doing half the job is just looking for trouble.

Might be something here... http://www.northursalia.com/modifications/misc/wiring/wiring.html

If you want accurate temperature readings you are going to need something a good deal more than that i'd imagine. It's easy to get a relative value using a temperature dependent resistor but if you want accurate figures you have us a proper temperature sensor which requires special wire and formulas applied to it's readings to compensate for the lenght of wire. The simplist thing would be to have two temperature sensitive resistors with one in the air cleaner and one at the throttle body then comapre the two and get an idea of the difference. It would still need formula applying if you want accurate temperatur readings but it would be simple to get some reletive figure that would be reasonably accurate and that could be scaled up or down in a fairly linear way..

It's best to take the fuse box out if you want to get to the relay, but you should try and confirm the fuel pump is good(or at fault) first.

Temporarily hot wire the pump in place and see if it cures the problem. If it does fix it don't leave that set up like that but fix the problem. The relay, if it's in the drivers door pillar, is way up above the fuse box and hard to see.

It should be a simple job for a mechanic and the seals are just common seals, they don't have to be in a subaru packet as long as they are the same design and size. If the mechanic is reasonably competent even then I'd suspect there is some other problem, like play in the cam bearings perhaps or the leak is between the cam retainer and the head.

If you disconnect the temp sensor the ecu will default to a typical warm running temperature which should get it running better. Try that when it's playing up and if it runs better then it will probably be the temp sensor..

As far as I know, if you are using the coils on the manifold you don't need the ignitor. I thought those manifold coils had built in ignitors? I suppose I had better have a look under the bonnet and see whether my car has an ignitor. Nah, on second thoughts, and being a mechanic, I think I'll leave the bonnet down... like I like it.

Screw the leather seats. Up here in the hot weather leather seats fry your arse! I think it's ugly too if that matters.

The circuit diagrams I have seen only show a simple transistor switch arrangement in the ignitor, so I'd have to say the ecu controls the dwell. (Edit).. From a pdf manual called "Ignition diagnostic and operation", The crankshaft position sensor is an inductive pickup type which uses a six tooth reluctor ring located on the crankshaft sprocket. As the reluctor teeth pass the crankshaft position sensor, they cause the sensor to generate a weak inductive voltage signal. This signal is forwarded to the control unit, which uses this information to determine the proper dwell and timing advance for changing engine operating conditions.

So if you put the clutch in and try and keep the revs up with the throttle does it still die? Or is it just that it won't idle with the clutch in? If it's not idling with the clutch in I'd suspect the heutral/clutch switch is playing up.

Agreed! If the marks don't line up you probably need to move the crank a little, well one tooth really, which equals half a tooth on all the other sprockets, then wind them all forwards or backwards and they will line up until the belt tightens when, generally, only the left inlet moves a little.

Ha, that sounds like a mate of mine that couldn't start his commercial fishing boat. He reckoned he had a crook starter. When I went to look at it the batteries seemed flat so I suggested we jump start it, which got it going. Them I looked at the charging gauge on the dash and there was no charging, so I pointed that out to my mate and he reckoned "Oh no, the charging is ok, it's just that gauge, it stopped working about two weeks ago". He was wrong though. Why do people assume the gauges aren't working right when the gauge doesn't show what they want to see? Why do people assume the car has been made wrong when it doesn't do what they want it to do?

The starter signal trouble code is because the ecu is not receiving a signal telling it the starter is being used. It needs the starter signal so it can forget about trying to set ignition timing and fuel mixtures off the usual sensors because, at cranking speed the signals from the air flow meter and the crank sensor are too irregular to be reliable, so, it sees the start signal and sets a fixed amount of fuel and ignition timing. Without the starter signal telling it to inject plenty of fuel and set the timing they CAN be hard to start. The crank sensor trouble code can be set if you crank the motor over too long without it starting, or, if it backfires and kicks backwards. See the connection now? Fix the start signal that's not getting to the ecu and both codes will probably go away.

White smoke out the exhaust? That won't be unburnt fuel surely... it would have to be gallons of the stuff. Are you sure it isn't oil burning?

Oh Ok. I think I see. So it's not strictly a load limit but just when the MAP is too high for the fuel map and the fuel map is related to the load calculation. So it's a defacto load cut... makes sense. I have read all the diagnostic charts and the trouble code setting criteia for heaps of models but I had never clicked to that connection(load to fuel maps). There is always something new to learn! Cheers.

That's an intersting bit of information Sanchez, about the load cut I mean. Where did you get that information and do you have it in print that I could get off you? I'd like to have a read if possible. Cheers.