stuckagain

UPDATE - no oil pressure mystery solved. I've been down in the lower 48 for the last 4 weeks but am home now. Today I decided to put in a new oil pump. When I pulled off the timing belt covers the toothed pulley for oil pump fell to the floor. I'll file this one in the "lessons learned" folder. No mention in the Subaru EA82 manual of loose or detached pulley on oil pump in the Engine Lubrication trouble shooting section under "no oil pressure." I have no idea how the pulley managed to come loose from the pump, but it did - hence, no oil pressure. I've decided to leave the timing belt covers off for a while; never liked 'em much anyway.

Sorry to be so chatty, but if this helps someone then it's worth it. Is it possible that the oil pump became "vapor locked," perhaps trying to pull up thick oil? I seem to recall that one of the manuals I read suggested packing the oil pump with vasoline before installing it so that it would create a vacuum when first used. When I removed both the "oil pressure sensor" and the "pressure relief valve" oil flowed from the sump and out of the open holes. Perhaps enough oil flowed into the line to create enough of a vacuum for the pump to begin working again???? Just a thought.

Problem solved - solution cloudy. I'm happy I stopped driving the car when I did. I was correct - there was NO oil pressure. It was a balmy 18 F. in the parking lot so working in my shorts didn't present any problems (unlike those days when the tools stick to your skin because it's sooo cold). Here's what I did. 1) Double checked oil level - OK. Checked for obvious leaks - found none. 2) Checked dash gauge. Grounded the wire that connects to the oil pressure sensor and turned on ignition. Gauge took off and slowly moved up the scale. Gauge - OK. 3) Removed oil pressure sensor from the oil pump and installed an oil pressure gauge (Matco) to check the pressure. Started engine. Oil pressure moved up to 14 psi (very low) at cold idle and after about 45 seconds dropped to zero. Turned off engine - pronto. 4) Removed end plug from "oil pressure relief valve" and extracted spring and plunger (also spent time trying to find the small spacer that fits between spring and plug after it fell on the ground). Put the valve back together and reinstalled oil pressure sensor. 5) Started engine. Oil pressure gauge rose ever so slightly so I did not shut off the engine but rather, on a hunch revved it to about 3000 rpm several times. Something opened or dislodged itself because the oil pressure started coming up after about 45 seconds. The sound of the engine changed the instant the pressure gauge started to rise; the engine becoming much, much more quite. I can only speculate that perhaps the relief valve had become stuck in a return flow position or, more likely I think, something was blocking the flow of oil into the pump. Perhaps both. Dunno Thanks to all for the suggestions.

Well like I said, I don't think it will be fun working on the car on a snow covered parking lot, in my shorts. Oil level is good - oil is clean and clear. Just did an oil change about 150 miles back.

UGH! 10W-30, but it's not that cold here. Yup, will be the first thing I test with meter tomorrow. I have manual - shows relief valve to be opposite oil filter on oil pump assembly. Should be able to get at it. I'll check it before I pull the pump. Thanks again!

Ross, thanks! That was my first thought - but timing belt didn't break and there didn't seem to be any odd noises coming from the pump and no leaks around it. Is it possible for the pump to fail that quickly? The car has 115,000 miles on it and I think the pump is original equip so it's due for a change.

Does anyone have any idea what would cause a rapid and immediate loss of oil pressure? There do not seem to be any leaks. Oil is fresh and clean (about 150 miles on it) though it does have a slight odor of gasoline to it. I am almost certain it is NOT the sending unit or gauge that is at fault. I took my 1990 Loyale 4WD SPFI out for a drive today. It's cold here now, last night the temperature was 10 deg. F. After warming up the engine I drove about 2 miles when I noticed that oil pressure gauge was reading ZERO. I listened for lifter noise and decided that it was louder than usual. I pulled over opened hood and based on the noise coming from the engine shut off engine immediately. I have not yet tested the sensor or gauge, but based on the engine noise I am certain that it is working. The car is parked about 2 miles from my home and I don't relish the thought of working on it outside at 20 F parked on a snow covered gravel parking lot so I'll probably have to tow it. Any suggestions will be greatly appreciated!

Thanks for post - I ended up pulling the knuckle out after I got carried away with a metal hammer Turns out I gouged the seat (in the knuckle) when I switched, out of frustration, to a metal hammer to bang on my brass punch. I didn't notice that the bearing had gone sideways and being harder than seat in the knuckle the bearing left a mark/ridge in the knuckle. With the knuckle out of the car I carefully sanded down the ridge I made with emory cloth using cutting oil (don't try this without the cutting oil). Since the ridge left by the bearing was near the outside of the seating area I took care not to remove any material below the ridge (inside bearing). Thus, the bearing should still form, hold and ride O.K. The tolerences are pretty tight in there - I noticed that when I was trying to pound the bearing in, and it got cocked, it wouldn't turn in the race; as soon as I pounded it out it rolled fine.

OK - I was just skeptical about the nuts. I installed them yesterday. I match marked them 'cause I'm paranoid I guess. They're not spec (like NAPA said they were) and it cost me a day because it caught me off guard. I just want to make sure it's OK (there are other posts on other lists asking the same thing - so I'm not alone). The mfg makes them this way not for my benifit, but to reduce their costs. Not all parts are the same - ask any 70-80's era VW mechanic who had to install mufflers made in Mexico. I just want to be sure I'm doing it the best way possible.

Wow, you folks have way more miles than me - I'm a pip squeak at 117,000 - I've got a long way to go. You do help me realise why I like fix'n my own car though: $$$$$. I could drop $ 18,000 - $60,000 on a new car or work on a $1000 beater (they used to be $50 -200, but I'm an old guy I guess). I think it was Lee Iochoca in a speach to the SAE who said that no car that Chrysler produced cost more than $4000 to build. In the same vein I have a friend with an Ivy League MBA and who's made tons of money as a venture capitalist who always points out that the cost difference between making 100,000 cars and 100,001 cars is about $1.00. The moral: watch your backs. Car mfgs are coming up with all sorts of ways to make sure you can't drive that old car forever; you'll have to buy a new one, it'll cost too much to fix the old one.

Shouldn't be a problem. Just remove bolt that holds the lower arm to the frame, take off everything else that needs to be taken off and the lower arm and the steering knuckle should come out together. You'll have to disconnect everything attached to the lower arm (except the ball joint of course) and do what you would have had to do to get the knuckle out as well. I just put a knuckle in this afternoon (did balljoints, wheel bearings and 1/2 shaft); removed the lower arm mounting bolt to make things easier. OBTW - did you use cutting oil when you tried to dirll out the bolt? Handy stuff.

TurboSPFI wrote: "If they didnt work why would they sell them?" The fact that something is sold is no guarantee that it is safe or works. I used to be a mechanic for VW-Porsche-Audi. We used to receive reams of "Service Bulletins" advising us to change things that had already been sold (some for many years). I assume that this is true for other makes of vehicles and for parts. Here's a corrected link to the "Service Bulletin" I found on the web: http://www.dmcnews.com/bulletins/SC-01-11.81.pdf I called a friend who has been an automotive engineer for the last 20 years; his field of expertise is military trucks. Here's what he had to say about self locking nuts: A high-quality self-locking metal/metal nut is generally a good torque retentive fastener as long as it is installed correctly and used only once. A castle nut with a cotter pin has some field repair advantages in that you can reuse it. You can reuse the torque retentive nut as well, but it will generally not have the same locking capability on subsequent uses unless it has a torque retentive thread form such as "SPIRALOC" Torque retentive threadforms are more than just squished nuts it is the actual thread form that prevents torque loss. They must however be properly installed, and if they lose preload in service they may lose effectiveness, so the best combination is a locking thread form like SPIRALOC combined with a crushed nut. One bad thing with respect to metal/metal self-locking nuts is the bad habit of nuts, particularly large sizes and stainless steel of welding to the bolt during installation particularly when using an impact wrench. It can also be more of a problem if a self-locking nut is installed on a soft-bolt. It will likely chew up the thread as the nut is harder than the bolt. SPIRLOC nuts are used to fasten the solid rockets to the space shuttle - this part of the space shuttle has not failed yet. Nyloc elastic nuts are generally not used in my business for critical fastening because they start to melt around 150 deg F. There are high-temp elastomeric nuts that have higher temp limits. Overall I think it is the trend, and good practice to install a metal/metal toque retentive nut in this application. It will be harder to remove, but maybe that's a good thing, and will make future replacement more time consuming. But that is only important to the customer, umm, oh that's you, huh....

Are you certain Nylock (or other brands) of self locking nuts are O.K? Can you cite a reference of some sort? Self locking nuts on a 1990 Subaru would be considered a modification. I found at least one example of "self -locking" nuts that failed when used in the front suspension. The sevice bulletin advised drilling out the stud and using a cotter pin. A copy of the bulletin is here: www.dmcnews.com/bulletins/SC-01-11.81.pdf

I thought I would finish the front end work today on a 1990 Subaru Loyale 4WD. I bought new balljoints from NAPA last week. They sell two varieties: cheap and expensive. I opted for the expensive ones. When I pulled one out of the box I noticed that there is a self locking cap nut, but no hole for a cotter pin. Has anybody else used these? Should I take them back? They appear to be well made and identical in size to the ones I pulled off the car. As always - thanks in advance. Stuckagain

1990 Loyale 4WD wagon needed new front axles on both sides. Pulled axle on one side using Edrach's method (worked like a charm) so the knuckle is still on the car. The inside bearing seal was in bad shape so I decided to re-pack the front wheel bearings. I removed the bearings using a brass punch - so far so good. Cleaned bearings - re-packed with hi-temp grease and installed outside bearing using a brass punch - no problem, BUT the inside bearing does not want to go back in. I checked the matching surface in the hub and it is clean and scratch free. Anybody have any "hints," special tool sugestions (the cheaper the better - I don't have $250 to spend on an OTC Hub Tamer just now) for getting that bearing to go back in without having to remove the knuckle from the car? Thanks!!! P.S. I'm a bit worried about installing the new seals - should they pop right in, or will I have to pound them in?

I think the article mentions this, but the flywheel marks for the cams ARE NOT the same marks you use to time the engine (tripped me up at first when I did a timing belt). As NW mentioned use the 3 mark grouping - not the grouping with way more marks that are used to time the engine.

I'm flummoxed! TPS now reads correctly ??? The problem has got to be me, I guess. Did everything the same as yesterday: cold engine, Fluke 87-V M.M., same leads. I triple checked all my readings yesterday so go figure. FWIW I started with your suggestion to read working voltage between B&C with the ignition on / engine off - 5V and dropped as resistance increased [if your new to Subaru repair see HINT below]. Surprised, I read the resistance on the connector lugs and came up with: B&D 4.93 k ohm C&B 0.5 k ohm (closed) to 4 k ohm (open) I guess I saved a wad of cash but I've still got an engine that is running just a tad rough when warm. Air flow and IAC are OK so I'm off to test PCV and then EGR solenoid and if they check OK move on to vacuum hoses. Thanks for the suggestions Cougar! HINT: having never worked on a Subaru before I was using a small piece of phone wire to read signals from plugged in connector until I realized that you can peel back the rubber cover on connectors like the TPS and Airflow and then access the conncetions from the back of the plug (this is described in section 2-7, page 12 of the Service Manual "Air Flow Meter.")

Tailgatewagon - duh sorry, manuals would be of no help. Blush....

Try manuals at www.finleyweb.net/default.asp?id=142

I don't think I'd touch it until I have a spare. Spent some time searching old posts and found nothing about rebuilding one of these. I have a schematic from the Service Manual it labels the posts as follows: A - Idle switch) B - (GND) C - Throttle position signal D - + 5V I assume the ground is for the ECU, perhaps it's isolated.

I continue, with much help from USMB and its' members, to work on a friends 1990 Loyale 4WD. I replaced the timing belts a couple of weeks ago and have been dealing with other small problems since. The idle is a tad rough after the car warms up, rhythmic but rough, so today I thought I'd start checking things. Since I tested and adjusted the "idle switch" when I timed the engine I decided to check the "Throttle Sensor" resistance as outlined in Sec 2-7, of 1989 EA2 Service Manual available from www.finleyweb.net/default.asp?id=142 (PDF part 2, page 25). The TPS in the Loyale has 4 posts referred to as A,B,C,D. The sensor passed one check and failed 2. I did get continuity from A and B as described in the manual (idle switch) but B and D read open rather than giving me the 3.5 to 6.5 k ohm the manual suggests. A check of B and C seemed promissing, but the connection went open after resistance rose a bit rather than slowly rising from 400 ohm closed to the prescribed 2.4+ k ohm at full open throttle (connection went open at same position as idle switch). The manual says that post B is supposed to be ground, but I get no reading between B and the metal throttle body. Aside from the rough idle the engine, when warm, seems to accelerate OK - am I doing something wrong??? The TPC seems to be a very $$$ item unless I can find a working unit at a junk yard. As always thanks for any pointers.

I adjusted the timing on my 1990 Loyale yesterday. Both the Subaru SM (1989) and Chilton say that when I connect the "green" diagnostic plug (on the firewall in my 1990 Loyale) the "check engine light" should be lit. But...my CEL is already on (EGR solenoid problem) and when I connect the "green" connector my CEL turns off???? Dazed and confused. Any advice?

Yup! Didn't put it up on ramps, but squeezed the hose coming off the thermostat housing a bunch to get all of the air out. Dash guage works now, RPM drops once engine is warm and fan kicks in when things get too hot. THANKS to all.

You are correct, the radiator switch ("thermoswitch" in Chilton), when working correctly, is open until the temperature is high enough to cause it to close. It occurred to me after I posted early this morning. When I got up I marched over to the stove with the switch, a thermometer and my multimeter. I connected the meter to each of the 2 leads and turned on the heat. Sure enough at about 208 degrees F. the switch closed.

Any idea why both sensors would fail at once (the radiator and the guage)? I didn't touch anything electrical - at least not on purpose. I checked the sensor in the radiator with a multimeter - it seems to be open: I get no reading either in series or from connector to ground.