GeneralDisorder

Bearings are definitely a lot more delicate than people often treat them. I am thankful that I worked in an industrial machinery rebuild and repair shop where bearing installation was a daily, common task. And unlike automotive applications - there was never a pre-packaged, "monkey-safe" option like loaded hubs, etc for the equipment I worked on. I was schooled by the experts and sent to classes and lectures sponsored by bearing companies like F.A.G. on such minutiae as grease compatibility and bearing failure analysis. If you stick to the golden rule at all times you can't go too horribly wrong: Never drop a bearing or cause impact forces to be transmitted through the rolling elements. If you do - discard it immediately. This causes spalling of the races. A bearing is not made of material which is the same hardness all the way through - the outer surfaces that contact and transmit force are very hard. But that is only a few thousandths deep. Under that is a softer material - think of window glass sitting over mud - that is a good analogy for the race surface. Now think of the balls in the bearing as a bowling ball. If you set the bowling ball on the glass (atop the mud) and hit the ball with a sledgehammer.... what do you suppose will happen? Bad things of course - the same things that happen to a bearing that's been dropped or improperly installed. The balls or rollers slam into the race surface and cause the extremely hard race to fracture like a pane of window glass. You can't see this fracture - it is not visible to the eye. But once installed the ball or roller will travel over this area and flex the cracks over and over again till peices of that cracked hard surface begin to pull away and get crushed. Now you have chunks of hard metal IN your grease - grease with hard grit in it has another name - Grinding Compound! Not to mention a pot-hole in the race that just increases in size the more load is put to the bearing. At that point you have complete failure. Allowed to continue it will chew everything to a pulp and eventually just fall to peices and cease to function. There is tons of experience and education that goes into how to select grease, grease compatibility and recognition, setting adjustable preload, bearing ABEC grades, axial and radial load handling ability, monitoring operating temps, vibration analysis, assembly, dissasembly (destructive and non), custom tool making, pre-heating for interferance type fits, using a press on bearings..... and so forth. You could, and many companies have, written enormous volumes on this stuff and as much as I would love to try and pass on what I have learned - I don't have the time or the neccesary memory recall to dredge it up when I'm not "in the moment" of a bearing related job in my shop to do so effectively. I would encourage anyone that wants this type of education to go to a bearing house and ask if they have any extra books laying about published by F.A.G. or possibly Timken, etc. that talk about all this in more detail. GD

First and foremost you have to *inspect* all your mating surfaces. Rolled edges, burrs, and gouges in the metal need to be filed down smooth. The axle nut threads need to be checked - dressed properly with a thread file or die if needed. I have installed hundreds of these EA axles - probably multiple of every brand and reman under the sun. I push the axle into the back of the knuckle and I get the nut started on the end - then I use two 12" pry-bars to pull it through the bearings. When I run out of purchase I thread the nut on farther. When that's not enough I put the cone washer on upside down, followed by the spring washer and then the nut - but the time I finish with that combo it's far enough through to slide on the hub and use that to finish it off. No tools are ever used that can cause an impact between races and balls. I have NEVER had a fitment issue. Without fail - when someone is doing this job in my shop and starts sniveling about how they can't get the axle installed - it takes me about 45 seconds to pull them through. You just have to perfect your technique. GD

The bearing should NOT just slide on. The fit between a 6207 and an EA outer CV joint is called a net-zero fit. These will almost never fit without some amount of resistance. NTN is the OEM manufacturer of all Subaru CV components - both past and present. They (IMO) make the best quality joint money can buy. There are so many things wrong with the methods and comments in this thread by almost every poster that, in my opinion, it should simply be discarded and a proper guide written on how to replace bearings and axles. I'll make this really simple - DONT BEAT ON BEARINGS AND AXLES WITH HAMMERS! That includes slide hammers. If you would like to not be repairing this stuff again in the short term you should really leave it to the pro's that know how to treat bearings or you should learn the ins and outs of bearing fitment and how to PROPERLY deal with them. GD

I would agree - get a wire feed machine and try using solid core wire with sheilding gas. The flux core gets in the way with all the slag, etc. Messy process. GD

I appreciate your input. You can indeed foul things up if you apply uneven pressure, or don't check your progress often and insure you are creating a flat surface. I grip the head only by the sides and let the weight of the head do all the work. I apply my force to move the head as close to the paper surface as possible. It's key not to rock the head in any way while performing this task. I have to say that I fundamentally dissagree with not publishing this type of methodology - the concept is sounds and it's up to each individual to decide if they want to risk doing this themselves or pay a machine shop. I think that our modern society is too far removed from old-world techniques - those of us that can and do practice these tried and true methods and have been given this knowledge by the "old timers" we have worked around should spread it as best we can. If, due to the use of this information, someone should damage a set of heads then they will have at least learned how *not* to do it the next time. People make mistakes when learning. It happens. I don't think they should be insulated from knowledge just because they might hurt themselves though. Sadly that's all too common in our increasingly socialist world - the removal, by the state, of the individuals right to fail. And in a world where you cannot fail - success is meaningless. Failure is the essense of what it means to be human and when you take that away you diminish the humanity of us all. GD

They both have their uses. MIG Stands for Metal Inert Gas welding - which is not what you are doing if you are using flux core wire. That's FCAW (Flux Core Arc Welding). With solid wire and 75/25 sheilding gas I can weld mild steel as well as I can with a TIG machine but 10x faster. TIG is a SLOW process and doesn't lend itself to overhead or welding in tight spaces or on unprepped surfaces. TIG is the aerospace version of welding. You can do it in a white lab coat - but it's very slow and tedious and still requires sheilding gas (Argon - not a mix) as well as tungsten electrodes that are carefully shaped and are a consumable and filler metal that is carefully selected based on what material you are welding. It's more expensive and it's extreemely time consuming and somewhat limited due to the difficulty level, required cleanliness of the base metals, and two-handed nature of the operation. You can't just point and shoot with one hand while you hold something steady with your other. It's not a matter of better or worse - it entirely depends on what you are welding as to the appropriate choice of welding apparatus. As a general rule FCAW or MIG (GMAW) is the most versatile and simplest form of welding today. Everything is integrated into the machine - filler metal, sheilding gas or flux core wire, and no other major consumables. TIG is far, far superior at aluminium and stainless steel welding. But an inverter machine will probably not have great aluminium ability. You really need a proper square wave high frequency machine for aluminium. GD

Thermal Arc is Tweeco's welder division. It's a quality machine that competes with Lincoln and Miller. Not a cheap unit. All welders have a duty cycle (even if it's 100%) and most that aren't full industrial machines have one that's 80% or less at full output. Duty cycle is not a measure of time but of percentage of time - so say that you are welding along on some fairly thick material at full current output of your machine - at some point you will have to take a break and the cooldown time will be 20% of the total between on and off time. So hypothetical numbers would be something like 8 minutes of weld time and 2 minutes of cool-down. The machine will have a thermal overload that will trip and stop you from welding when it needs to cool down. Now - that's all fine and dandy but in the REAL world most fabricators don't over-temp their machine very often. You do so much moving, adjusting, cutting, and preping that the machine has a chance to cool anyway and rarely do you end up having to wait on the machine. Fan cooling just means it's not water cooled. Only BIG machines are water cooled. Inverter machines are fine. They make really excellent quality one's now and they are tiny in comparison to the gigantic transformers they once used for high frequency welding. The machine rectifies the incoming 240v AC (RMS) power source to about 320v DC and then passes it through an inverter that creates a low voltage (~10 to 30v), high current, PWM 20Khz output for welding. The ultra-high frequency is what you want for really smooth welds and the voltage is dropped to bring the current way up for a nice fat arc that will jump across to the work. It's not a bad thing at all and all the major manufacturers are turning to this technology to save cost and weight. I personally own an inverter plasma/TIG machine and I love what it can do for it's teensy tiny footprint. GD

Cool - glad it worked out for you. One thing you can probably do on the clutch cable is get the rubber "snubber" and associated washers from an EJ cable and slip it on before the two nuts. Those are handy for that sort of thing. They give you additional adjustment if the cable is too long. The snubber and washers add up to a good 1/2" of additional stack under the nuts. Is it an OEM cable? Might be able to modify a cable mount from a 5 speed.... Anyway that's pretty trivial. Getting the cable adjusted to where the fork wants to sit is easy once the clutch is doing what it should and grabbing properly. Absolutely on the write-up for the USRM. Anyone that wants an awesome clutch for an EA81 w/4 speed simply has to use the EA82 flywheel, XT6 pressure plate, EA81 disc, and Nissan 720 TOB. Sure you have to peice the kit together to get it all setup right but that's not too hard. The reason I've never written it up is because most people just ditch the 4 speed for the 5 because it's stronger with a lower "Lo" range. GD

The first hydro clutch used by Subaru was on the EJ turbo - which for the US was the '91 Legacy Sport Sedan. But none of that is compatible with the EA tranny because those are pull-clutch setups. An 88 DL would be cable but would not have the Hill Holder so it would have a return spring setup instead.... not hydro in any case. Your best bet is to swap over to a 95/96 Legacy pedal assembly - they were cable clutch. The cable is a better system IMO - easy to adjust for different drivers, no bleeding, and it has better "feel". If you want to keep the hydro system you will have to figure out a fork solution - probably a custom fork will be involved, as well as slave mounting, etc. No question it could be done but it wouldn't be worth it to me. A little more involved because of the 4WD engagement and the Lo range linkage, etc but not bad. If you are going to split the case - almost silly not to include an OBX Helical front limited slip while you are at it. You don't have to split the case though - all the EA82 and EJ axle bar's use the same 25 spline, 27mm interface. You can swap to EA82 DOJ's as well. Same as any other EJ adaptor plate, yeah. GD

The proper key code is located on the passenger side door lock tumbler - remove the door panel and you can read it with a flashlight. Take that to any locksmith and they can cut a perfect key from the code. Then you just have to get the old one out. As mentioned - possibly some needle nose pliers, etc. But you might have to take the lock off the column and remove the key by dissasembly. GD

There was a JDM version at one time. People do swaps all the time. It's nothing new. GD

That would be standard practice durring a HG replacement. I don't even bother checking - I just resurface the heads with 220 grit wet/dry and a sheet of glass. Any warpage will show up instantly as low spots during the surfacing. Usually the heads will be cupped slightly. The block I clean with a very fine synthetic abrasive disc in my die-grinder. VERY carefully. If the engine is being completely rebuilt I have the block resurfaced when I have my machinist do the ring hone. Though I've been considering doing those steps in my shop as well in the near future. GD

YES. You need to replace it. GD

NOT on a DOHC 25D. Bad idea - they last about 15k miles and then you are doing the socket & wobbly dance trying to get them out and back in again. If he's pulling the engine then invest in the CORRECT platinum plugs and get at least 60k out of them before you have to do that nasty job again . GD

Don't get the timing belt and tensioner from Subaru. Get a complete timing belt/wp/idler/tensioner kit from "theimportexperts" on ebay. They are well respected and you will get quality parts. No need to get ripped off at the dealer on that stuff. That kit will also include the cam and crank seals and thermostat seal so take those off the dealer list. Get Denso platinum or iridium plugs. About $6.50 each. Clutch kit - Exedy ONLY. Amazon has the best price and free shipping. Here's your dealer list: HG's Intake/Exhaust gaskets Valve cover gaskets and grommets (take VC in with you for a correct match - there's 6 styles) Thermostat Oil seperator plate and bolts (back of engine) NO HEAD BOLTS. They are reused and are not a wear item. Trust me - I know more than your buddy from the dealer. I do this every day. GD

No the starter will be fine with the EA82 flywheel. Remember that the EA81 flywheel is the taller one so the flywheel just moves away from the starter slightly. And that's also what happens when you take 0.1" off the EA81 PP surface. Still works fine. Forgot to mention the grinding - sorry about that. An angle grinder or a die grinder with a cut-off wheel will do the job easily. You just notch the bottom of the bell housing where you have it marked. That's the only mod needed. You also have to ditch the cover plate on the bottom of the engine if you have one. No big deal there either. I take my cut-off wheel and make a mark at the proper location of zero for the timing. The rest of the marks can be interpolated from the zero - you just have to count the slash marks or subtract 7 as you noted. GD

"Vehicles may only be used for exhibitions, parades, club activities and similar uses" You might have trouble with that. I wouldn't do it - too much risk and likelyhood of being pulled over and questioned about not having a front plate. That's happened to me more than once with normal plates. GD

I have never seen an EA81 leak at the HG interface. It is possible that the HG's were replaced at some point and the surfaces not properly prepped or trued up before assembly. When correctly installed they should not need sealer and they should not leak. GD

Make sure the bottom end is good before you bother with the HG's. Pull the oil pan and inspect for sludge/metal, etc. When these overheat - they often destroy the bottom end (rod bearings) in the process. I've seen engines completely seized at 140k due to overheating. And I've seen others that "ran" but had severe damage to the bottom end making HG replacement an expensive waste of time. If you decide to do HG's - PULL THE ENGINE. And use only Subaru OEM gaskets from the dealer. If you do all your gaskets, timing belt and WP, and have the heads resurfaced it's going to run about $350 in parts and machine work. GD

The earlier ones likely did but the later EA71's I think used the same pilot. .815" is what I have always measured on EA82 flywheels and it's the number always qouted for the EA82T and XT6. Which used the same basic clutch setup as the regular EA82. If you measure an EA82 flywheel step it will be right at .815". I'm sure that .827 would probably work too.... only talking about 12 thou.... GD

Yeah - pilot bearing is the same from the 70's to today. (MOD EDIT-83-84 2wd ea81 5spd has a slightly different sized pilot bearing for some reason, learned that the hard way the other day .5 mm smaller diameter) GD

No. Transfer case and transmission are not seperate. GD

I get the Walker or NTK one's. Amazon is a good source - find the part number on www.rockauto.com and stick it into Amazon. GD

Open deck, 48mm rod journals, piston-slap prone, HG's last about 100k..... this is NOT the engine you want to be turbo charging. You put a turbo on it and it's going to last a week and a half. It is quite likely that you have bad HG's now. The water pump's are not that common of a failure. What are your symptoms? Bottom end failure is very common with this engine - the rod journals were off-center ground from a 2.2 crank that had 52mm journals. This created a stroker effect but made the rod bearings tiny and weak. At 190k with overheating issues it's probably already cooked. You will be lucky if it goes down the road reliably in stock form - let alone with a turbo. GD

No you should be alright on the cable. Having the step height corrected will cause the PP fingers to flatten out and move the fork back maybe 1/4" to 1/2" from where it was. The .1" step reduction will help with that. But you should still be solidly in the threaded area of the cable. The only measurement I know of to differentiate the PP's is the TOB diameter compared to the PP spring diameter. I've looked at all of the combo's in detail and the EA81 TOB doesn't fit the EA82 PP for beans. It barely contacts on just the outer edge. If you compare to the EA82 TOB (which is the same for Legacy, etc all the way to present day models) the diameter difference is huge. Even the Nissan 720 bearing isn't quite as large. Better than the EA81 version for sure though. The fact that you have the EA82/Legacy springs in the PP coupled with the known step height difference and that you experienced slippage is really good evidence that you have an EA82 PP. GD