GeneralDisorder

Interesting. The drive gear on the 5MT's is part of the pinion and cannot be changed. No provision for different driven gears was provided by Subaru because the final drive is part of the pinion shaft of the front diff. Unlike a Datsun where there is no final drive in the transmission but you still need a way to change the speedo ratio correspondingly when different final drive rear diffs are used. So this doesn't really apply to the Subaru MT's. GD

My purpose was not to be rude. My purpose was to call into question your tools and your ability to use them. Think it rude if you wish but to discuss the results of a compression test over the internet without first determining if the test was conducted properly and with the right equipment would be a silly waste of time would it not? I do not know you or your qualifications. Thus I must assume that they are a source of potential error. Whether you like that I do this or not is of little concern to me - be as thin skinned as you like about it. I am only here to solve the problem. I just don't assume that the problem isn't human error. If I did then I wouldn't be the guy people ask to solve their problems . Yep. You and everyone else on here :cool: You assert that the gauge is inaccurate yet you rely on it's results to determine variation.... that is not compatible logic. If the gauge were off by a percentage then this logic would hold - but you have NO EVIDENCE that the error space of the gauge is linear with respect to the correct readings from a working gauge. Thus you cannot assert that variation percentages will be the same in it's current, damaged state. Carbon build-up *where* in the engine will cause poor idle? Generally speaking - carbon build-up will cause a compression problem (if it's around the compression rings or around valve stems causing the valves to stick) - which in turn can cause poor idle. But you have not performed an accurate compression test so you don't really know anything at all. So could I and everyone else in the world. Non sequitur. Low compression only really shows up at idle. Even on engines with exhaust valves burned all to hell and 150 psi lower compression in a single cylinder you can't even tell when the engine is above 3k RPM. No I don't agree with you at all. Valve *guides* have nothing whatsoever to do with compression. The valve seat and face are what do the sealing. Bad guides can contribute to oil burning but not generally to compression problems. In my shop it is never reasonable to "assume". You theorize - then you test - and then you KNOW. I will agree that it's a possibility. But I will not assume that it's the only one. Seafoam has worked for me in the past but not usually a single can in the oil as you have done. As Stubies Subie posted above - it can help under certain circumstance and when used WITH science. You have not tested the compression to determine if anything *really* changed. That's not science so YES there is plenty of arguing with it. Which is what I'm doing. You see rude. I see ignorance. Just words. You are here in this forum and so am I. I'm here to learn and so are you - you will have to answer the skeptic's challenges though or no one will take your posts seriously. I have questioned your logic and if it can't stand up to my questions then it's not very good logic and it needed to be questioned didn't it? GD

No problem. Glad to help. I know how it feels when things go sideways and I want to see that beautiful hatch of your's on the road. GD

I have had the EA81 flywheel's resurfaced to the EA82 step spec and used the EA82 PP (with an EA82 transmission) - it works great. I have *seen* the Nissan TOB installed on the EA81 bearing holder and I have looked at the specs of both bearings - they have an identical ID and both fit on the same cast TOB holder. I know someone that used this setup for a 4 speed behind a Hi-Po EA81 with an XT6 clutch (same as the EA82 PP) and it worked great. The Nissan unit is an identical design to the EA81 TOB - it's got the same ID and fits the same holder. It just has a larger OD and contacts the PP fingers more securely. You could use the EA81 TOB you have now but wouldn't it be nice if it were a bit larger? That's what the Nissan unit will give you. You can view pictures and specs of both bearings on www.rockauto.com if that would make you more comfortable. Also - I wouldn't tell you to do it this way if it wasn't going to work. GD

That's enough evidence for me - you DO have an EA82 pressure plate. Someone at Sachs has screwed the pooch. A proper EA81 PP should have fingers that describe a circle 1/4" smaller diameter than that of the TOB. The bearing should contact SOLIDLY away from the tips of the fingers. Beck Arnley should be good. Exedy is THE best for all Subaru applications but like I said I can't find a part number for a 225mm EA81 kit anymore. Sachs might be using Exedy as an OEM and that could be part of the problem. B/A might have the same issue.... I don't know. This stuff is getting all F'd up by the parts suppliers. If it were me I would find an EA82 flywheel, resurface it, and change to the Nissan TOB. The EA82 PP is stronger anyway and you already have it. For the cost of a used flywheel, resurface and a $35 TOB you could be down the road. Could also have the EA81 flywheel cut to the .815" step spec.... and use the Nissan TOB. Then it will simply work. Less than .1" and a new TOB away from clutching perfection.... GD

I call it like I see it. I do my very best to be honest in everything I do and to save you, me, and the rest of the members of this fine community valuable time. If I come across as a "wongleflute" it's because it was neccesary to sharpen my point. You are most welcome. As for examples - well you posted your compression results and seemed concerned - you should be happy that I'm giving you hope by telling you to have the test done with properly calibrated tools and someone that is familar with their use and what the results can/cannot used to determine. Again - you're welcome . Objection - calls for speculation. It's "imbalance".... again though - you have what evidence of this? In my experience that is not likely the cause of a misfire at cruise RPM. Improved due to what exactly? If you can't trust the tool then how do you know they "vary" and besides that how do you know how much variation is "bad"? GD

Yeah that sounds fine on the flywheel - I would say you have a bad pressure plate. If you can source an Exedy that's how I would go. But last time I checked I wasn't able to come up with a proper Exedy clutch kit for the 225mm EA81 application. Another way you can go is to switch to the EA82 flywheel, use an EA82 pressure plate with the EA81 disc you already have and a Nissan 720 pickup throw-out bearing (fits the EA81 holder). How does the diameter of the hole in the PP fingers compare to the diameter of your EA81 T/O bearing? If the bearing barely touches the fingers then they put an EA82 PP in your kit. GD

First off - the numbers themselves are completely meaningless. They are only useful in relation to each other. 70/95 is ~74% so your highest and lowest cylinders are still within about 24% of each other. It would be nice if that were more like 10% to 15% but 24% on an engine that's nearly 30 years old is not that bad. That said I would guess that your testing method is probably not ideal, your gauge could be wildly innacurate, and both of these are contributing to low numbers and probably to the variation in your results. With the right methods and equipment it's entirely possible that the readings will jump up to normal levels and even might be within the magic 10-15% range. So basically - if the engine runs fine and you did the compression check just for kicks..... put down the cheap-o tools and leave this kind of diagnostic work to the pro's. GD

That's not a coolant port - that's the EGR exhaust port - hose will not work here. Would have to be steel tubing. Remember that it's going to be about 600* F in that tube.... thus it's heat sheild... I would drill and tap the port and plug the open one. GD

I would get a set from over here on the west cost and weld in new one's. Tubing is not going to seal on your door correctly, etc. Not the way I would be doing it. GD

Sounds like HG's to me. Cold bottom hose while the top is hot indicates the water pump is cavitating. You may have just done them - but did you have the heads surfaced or checked for cracks? HG replacement takes patience and attention to detail and can easily go sideways if you don't know what you are doing. GD

Yikes! A 4 door no less. That's not worth beans. Certainly not what they are asking. $800 tops. The valuable (relatively speaking) RX's are the 87 to 89 three door coupes. That's just a sedan with a flapper MAF turbo engine that's going to leave you stranded half-way home. No cool body style, no AWD. Just a big pile of poo. Bad. Deal. GD

Terrible engine. Untouched, with good cosmetics and low mileage - maybe $2k. Worth almost as much without the engine or non-running if the cosmetics are really good. Perfect candidate for an EJ swap. The stock motor is a boat anchor and people routinely get burned buying these and not understanding what they are getting into. You are looking at around $2k to swap the engine over to an EJ and have a really decent ride with a neat retro feel to it. $4800 is a big stretch. NADA is off their rocker. Hope you didn't pay that much. GD

The fuel pump will only run for a couple seconds when you first turn the key. Then it waits for you to crank the engine to continue running the pump. Under the dash on the driver's side there is a set of green and a set of black connectors that should be currently disconnected. Those are the self-diagnostic plugs. If you plug the green connectors together the fuel pump and the radiator fan will cycle on and off when you turn on the ignition. This is the easiest way to check the operation of the pump. Plug in the green connectors and turn the ignition to on (do not crank it) - you should hear the fuel pump relay clicking under the dash and you should be able to hear the fuel pump from the tank fill cap. Once you have verified you have fuel at the tank - check for spark. Pull a plug wire off the coil pack and hold it close to the where it plugged in while someone cranks the engine. You should see a hot blueish white spark jump the gap from the plug to the coil pack. Check all four wires - the coil pack has two coils in it. Another thought - check the plugs at the passenger kick panel near where the blower motor plug is located. One of those plugs contains the power wire to the fuel pump and it could have a poor connection, etc. You were just in that location so maybe moving it caused an existing problem to get worse.... GD

I'm sure I could build a special tool to do this. It would involve a lash adjuster deflator that could apply constant force via a lead-screw to slowly deflate the HLA. Then another seperate tool to hold the valve open. Once the tool was built it would take only a short time to use it. Maybe 30 minutes to deflate the lifter. Of course - desiging and building such a specialized tool would probably involve quite a few hours of work. I just built a press adaptor for dissasembling outer CV's from the shaft in a correct and controlled manner. No "pipe trick" as has been popularized by the youtube that will likely cause spalling of the races inside the joint and doesn't work with 100% repeatability. One part of the tool required a machining step with great accuracy that took several hours and another peice required an hour or so of machine work. And the rest of the bulk peices of the press adaptor probably ate up another four hours of time in sourcing materials, cutting, drilling, tapping, etc. There's probably 8 man hours into that tool - and it's a wonderful tool that I will get a lot of satisfaction from using and laughing and the fools using the pipe trick, chipping their concrete driveways, and generally looking like a circus act and swearing at inanimate objects. There is probably just as many hours in the creation of a special tool for rocker installation without pulling the cam tower.... likely not worth it for this old of a vehicle. But I could do it. I'm sure of that. GD

Fork could be bent, aftermarket cable, wrong cable, poor fitment on the pedal end, etc. Lots of reasons for strange adjustment issues. GD

EA82 PP would bolt up but takes a larger T/O bearing. It would work at first though. The EA82 disc has the wrong spline count so it's unlikely they gave you the wrong "kit". I've had the wrong parts in the "right" kit before though.... GD

The step measurement is from the PP mounting surface to the friction surface. So if the step is too large it will slip - larger number means step is too deep. The EA81 step should be (IIRC) .906". EA82 step height is shorter at .815" so a pressure plate for an EA82 would grip poorly on a stock cut EA81 flywheel. Could you possibly have put an EA82 PP in it? Could just be a weak PP also. Wouldn't be the first one I've seen. Either way you are pulling the engine to find out. Typically when I tighten the clutch cable there is about 1.5" to 2" of thread past the lock nut. But that's a general rule. If it's slipping and you have slack in your cable (you can push it toward cabin with your hand) then you don't have a cable problem. The problem is on the inside. GD

Sounds like a bad pressure plate or incorrectly stepped flywheel. Did you have the flywheel resurfaced? Improper surfacing of stepped flywheels does happen occasionally if the machinist doesn't understand that they have to cut both the friction surface as well as the PP mounting surface. The step height needs to be .905" IIRC on the EA81's. If they made the flywheel too deep it will not grab properly. Having to let the cable out all the way to the tip of the threads is an indication that the PP is not clamping the disc properly. The PP fingers should be almost flat when it's torqued down to the flywheel. If the flywheel is too deep then the fingers will be raised and that will push the clutch fork forward and cause the cable adjustment to be way out at the end - and of course it will slip also because the PP isn't under enough tension to grab the disc. GD

The actual ECU and associated sensor grounds are always on the manifold itself. But you still need a good path from there to the body and from the body to the battery negative. With swapped cars it's pretty hard to troubleshoot over the internet. Most people that aren't prepared to drag out the o-scope if something isn't right with a swap probably shouldn't be doing one. GD

Custom trannsmission x-member and linkage plus a cobbled together clutch assortment that will make your head spin. Not worth it - the 4 speed's are junk. That's a huge step backwards. GD

14 psi ?!?! Possibly with E85, alcohol injection, monster IC, and a LOT of hopes & dreams to keep it together. I doubt the pistons would take that much abuse for very long. The tune is going to be the key and quite frankly you will need a LOT of insturmentation to monitor something that's on the razor's edge with cylinder pressure's, and exhaust gas temps. Doing it right would probably cost more than just buying an STi front clip and calling it good at 300 HP . Check out the threads on NASIOC about that type of build. There's been a lot of folks that have done high compression turbo builds. It's quite easy to end up detonating them to death. It's not something I would sugest for someone that isn't intimately familair with turbo charged tuning. GD

Actually for the engine to *just run* - once the harness is stripped to only the ECU and associated wireing - you only need to make 4 connections. Constant power, switched power, start signal (only hot while cranking - to the starter solenoid), and fuel pump power. The ECU would like to also have a few more inputs - it would like to know if the vehicle is in gear or neutral (a switch that grounds when in-gear or when the clutch is released), and it would like a speed sensor signal. This is a square wave generated either by the speedo head or by a wheel or driveline sensor similar to what is used for cruise control systems. Otherwise all the sensors for the engine are mounted on the engine intake manifold with the exception of the O2 sensor which is mounted in the exhaust, the MAF which is right after the air-filter in your intake duct, and - depending on the year - possible some MAP sensors that are on the fenderwall of the donor car and connected to the engine with a vacuum line. When the harness is stripped you keep all the connectors for the engine and it's sensors as well as the ECU, fuel pump, and ignition relays under the dash. There are a couple board members including myself that offer harness stripping services to remove all the uneccesary portions of the harness and label all the connections that you will use. I charge $200 plus shipping for the OBD-II harnesses like the '96 you found. GD

Slip the clutch and get it moving and you should be able to pull it out. Happens on the 4's from time to time. Tell her to always go to neutral when coming to a stop. GD