asavage

I don't advise substituting a resistor for a solenoid. First, the solenoid I tested remains mechanically open when de-engergized, so you'd also have to plug the feed line. Second, there is absolutely no power loss to having a working charcoal canister purge system, and there's a very good reason to keep it operating: it keeps massive amounts of unburned hydrocarbons out of my breathing air. EGR: if disabled, you need to recalibrate the spark timing curve, or you'll get pinging. Or you'll have to up the octane of your fuel. Cheaper and easier to keep the EGR system working correctly. EGR is what allows you to use relatively high compression with low octane fuel and still keep oxides of nitrogen low -- the stuff that contributes to photochemical smog. On using an ohmmeter to test a PCS or EGR solenoid: Set your ohmmeter to the 1x scale, unless it's an autoranging digital unit, in which case you don't need to set it to a scale, but just to "ohms". This is usually marked with the greek omega symbol Test the ohmmeter by touching the probes to each other: the meter should read near zero, or possibly up to one ohm (on cheaper units). If not, there is a problem: you do not have the meter set to a resistance setting, the probes have an open, or the meter's internal battery is faulty. Very old ohmmeters: you may want to use a "ohms zero" adjustment to set the "probes touching" measurement to zero. If your ohmmeter does not have a thumbwheel marked "zero", skip this step. I haven't seen a modern ohmmeter with this setting, but there are a lot of serviceable older units out there. Next, locate the PCS solenoid connector. When looking over the radiator, it's to the right of the thermostat housing, and to the left of the throttle body. I find that it's much easier to remove the black rubber-like air intake duct from the throttle body, disconnect the two 17mm hoses from it, and move it to the spare tire area. Now you can look right down at the two solenoids. The PCS and EGR solenoids sit next to each other, look identical to each other, are fastened to the engine via a single 10mm head bolt each. The PCS solenoid is the one further toward the passenger fender. Disconnect the PCS solenoid's connector. Sometimes a No. 0 flat screwdriver is helpful to break the two connectors loose from each other. Turn on the ohmmeter (or set it to 1x or autorange as above). Connect one ohmmeter lead to one terminal of the PCS connector, and one lead to the other. I mean, the solenoid side of the connector, not the engine harness side! Probe polarity is not important for this test -- you can put the red and black probes on either terminal. Unless you have clip leads or aligator clip adapters for your ohmmeter, you will find it a bit of a challenge to hold the meter's probes against the connector's blades and still read the meter: everything wants to fall off. The ohmmeter reading should be PCS = "under 100 ohms" EGR = "32-39 ohms" You will probably find that your ohmmeter does not change reading or move at all, indicating an open circuit. This is a bad solenoid (electrically speaking). Wiggle the probes against the solenoid's terminals to ensure you are making good contact. If you still get no change or reading, the solenoid is bad.

http://groups.google.com/groups?hl=en&lr=&ie=UTF-8&oe=UTF-8&selm=397DFB0F.31F6%40mit.edu&prev=/groups%3Fq%3Dsubaru%2Bclock%2Bdim%26btnG%3DGoogle%2BSearch%26hl%3Den%26lr%3D%26ie%3DUTF-8%26oe%3DUTF-8 W. Curtiss Priest, Director, CITS Center for Information, Technology & Society 466 Pleasant St., Melrose, MA 02176 Voice: 781-662-4044 BMSLIB@MIT.EDU Fax: 781-662-6882 WWW: http://www.CyberTrails.org Technical Note July 24, 2000 1980-1992 Subaru Automobile Repairing Intermittently Displaying Car Clock (Flourescent green digital display style) 1988 Model DL 2 Wheel Drive 1800 cc Single Point Fuel Injection (SPFI) Preface: these repair notes are written for those who would rather fix it themselves than purchase car repairs. And, if you've collected enough electrical parts, you can often do these repairs without leaving home! Also, when you take into account the time spent bringing a car in for repair and either waiting, or having to arrange other transportation, you are often ahead in time, doing it yourself. I do get occasional notes from mechanics at Surbaru garagesabout why I would want to do this when the part is "only" $$. They forget that they need only walk over to the parts counter and spend your money on parts that cost many times the original manufacturing cost. *** This particular repair is very quick when you know what to look for. You need not even disconnect the clock, so access is about 10 seconds. Total repair time is less than ten minutes. *** Symptoms: 1. clock display goes dim or out 2. pressing one of the setting buttons may cause display to return Repair: ------ Clock pops out of dash, lightly prying side Remove 4 wires from back clip. Remove cover, prying slightly on clips. Unsnap PC board prying slightly on one of two side clips Inspecting board: ---------------- Solder around power resistors looked "cooked" -- I shook off loose solder by heating and tapping the PC board on the dash. I checked for tinning on resistor leads. These looked good, I resoldered, and there was no change in the problem. (probably a good idea to resolder these anyway) Flexing board caused display to change brightness. Problem and Repair: ------------------ The 150 ohm 1/2 watt resistor to the left of the buttons had a poor solder connection to the circuit trace running to the buttons. I resoldered that end of resistor. Clock got bright, no amount of board flexing changed it. Clock was fixed. Amusing: ------- So, the reason the clock got brighter when an hour, minute, or second button was pressed was because this flexed the board, causing the 150 ohm resistor to "reconnect" due to the mechanical flexing. In Case You Wanted to Know: -------------------------- The reason there are four wires running to the clock is: 1. Ground 2. Battery voltage for when ignition is off 3. Battery voltage for when ignition is on 4. Battery voltage for when you switch on parking or head lights (this causes the clock to be dimmer at night !) ============================= Al's addition: I had to resolder an area on the left side of the board, too, to fix mine.

I watched a driver's side head gasket replacement today, on a '86 sedan turbo. Once you get the PS, A/C, and all the intake and pipes/hoses out of the way, and the whole front of the engine (timing belts, covers, fan clutch) out of the way, oh, and try to move the turbo crossover pipe out, it didn't look too bad. Took a professional mechanic most of a whole day, more or less. In a shop full-o'-tools. I was impressed -- but then, I'm easily impressed. If you don't have to deal with the turbo stuff, it's likely a whole lot easier. If it's the passenger side HG, you don't have to mess with the distributor. BTW: I didn't know before, but I know now. The o-ring for the camcase, if purchased from Subaru, is a special "stiff" o-ring. I think it's reinforced somehow. Wish I'd thought of that before I glued a regular o-ring in when I did mine. The '86 above was being re-gasketed for a severe oil leak from head gasket. You know where. $800 later . . . Yes, $240 is pretty reasonable for this job. But if you've got the tools, the work area, the time and inclination, you're better off doing it yourself. The gotchas are things like stripped holes, broken bolts, and you almost always have things that are worn out or need to be replaced that you didn't expect. The upside is that you have total control over the quality of the work done.

I had similar: my noise started 10 secs after startup, and went away completely when warmed up 15 min. later. I had a bad oil pump seal & gasket, a very common EA82(T) problem. Pics in this thread: http://usmb.ultimatesubaru.org/forum/showthread.php?s=&threadid=3517 But fixing the pump mounting didn't fix my noise. I replaced four hydraulic lash adjusters, and it's been silent for a month. See also Cameron's advice: http://homepage.powerup.com.au/~camncath/ea82_hydraulic_lifter_fix.htm

Pull the connectors, use an ohmmeter to measure their resistance. PCS = "under 100 ohms" EGR = "32-39 ohms" AFAIK, the ECU only checks for current draw on those solenoids, so I think they can be mechanically bad and still not trigger a code, but that's only a guess. If you're getting a code, the solenoids' coils are probably bad, or unplugged.

I have had an opportunity to use Intermittant wipers a few times since I unplugged/replugged the wiper motor connector, and it's still working well, so I'm going to call it fixed. Add that one to the list of possible causes for jerky wipers.

My rig is down at the moment -- I'm chasing a throttle body impossibility, WRT the fuel pressure regulator, and I've got the whole thing apart and on my bench. I faxed a request off to AllData today (our contract allows me to do this -- we pay them enough money!), and they couldn't come up with a decent picture or vacuum diagram either. I do not have manifold vacuum at the regulator at idle or at any other time, and the port to which the regulator is connected does not appear to be designed to ever have vacuum, it looks more like a purge dump port. Or something. Tomorrow at noontime, I pick up a complete TBI assy for testing purposes. It's good to have a friend at a Subaru wrecking yard Anyway After I get it back together, I'll try to localize my rattle -- the last couple of days I drove it (Mon/Tue), I could not get it to rattle very much at all. But I guess I could try harder. I thought that the wye pipe (with the funky junction where the O2 sensor mounts) is a simple wye -- is there catalyst material in it? If so, I can't imagine trying to fab one using a generic cat. I was thinking that it might be the "actual" catalyst rattling, the one on the pipe to the rear of the wye pipe. That one's a catalyst for certain. I have tried to "squeeze" a noisy cat in the past, unsuccessfully. It wasn't a Subaru cat, though. Through-bolting with a long bolt might shut it up, but it definitely won't catalyze nearly as well afterward. I had access to an '86 Sedan Turbo today, MPFI, and his headpipe is bad -- this is not the crossover, but the pipe off the turbo to the rest of the exhaust. I called the dealer, turns out this pipe contains a precatalyst, and cost is over $600! Yow! He's gonna get one from a 'yard. I don't blame him. (This Turbo's fuel pressure regulator has vacuum on it at idle just fine. Of course, the whole setup is completely different from my SPFI, but the principle is the same. Until I can find better info or see a working vacuum routing setup, I can only guess.)

Based on what I've read, and my bad one, you will find that the solenoid's coil is open. Use a ohmmeter, remove the PCS connector, and check the solenoid's resistance. Should be ~35 ohms. It will read infinity (open). Coil is bad. Bad news: dealer wants over $110 for it. Good news: it's the identical unit as the EGR solenoid that sits right next to it, with the exception of the wire connector. You can get one from a junkyard or wherever and cut off your old connector, solder it on to an EGR solenoid, and it fits perfectly . Bolts right in place, hose connections are identical and everything. And, I have a local source for a few new EGR solenoids. Call A&G Import Auto at 360.385.0002 in Port Hadlock, Wash., ask for Tony, and he'll set you up for a lot less than $110, and all you have to do is solder two wires. Mention Al Savage and he'll only jack up the price another $10 (for my cut, you understand :-p ). I did this myself last week, and posted about it here. You're supposed to run the D-check procedure. Warm it up, shut it off, connect both the D-check green connectors and read mem connectors. (edited later: I left this part out: Turn on the ignition, verify that the CEL light comes on, press accelerator to floor, then back off to half throttle for two seconds, then release it completely.) Then drive it constantly above 5 mph for at least 39 seconds (yeah, I know). CEL is supposed to blink, indicating no codes found. If, after about two minutes of driving above 5 mph, you get the CEL light again and it's not blinking, it's done with the diagnostic mode and it found another code, which you can then read. On my rig, I sometimes get a 51, Faulty Transmission Inhibitor Switch. Balls -- switch tests fine. Maybe it's intermittant, but it's sure no fun to get out and work with: I've done it once.

LOL A couple of years ago (3 or 4), I was driving my old dsl Maxima wgn down through SF, running along at the limit, and I heard this loud bang/clang. Sounded like a bolt came off something rotating and hit some sheet metal. I could hear it over the CD I was playing. Hmmm. So, I signalled and got over on the shoulder, what shoulder there is, in CA there are a lot of concrete sound barrier walls right next to the right lane, so there is barely enough room to pull off the road, and getting out of your car is really scary. I popped the hood, then carefully exited the car, walked around the front and began to open it, when I noticed that the RF was lower than the LF. Looked over, the RF tire was shredded. And flat, of course. I never even knew I had a blowout.

If you want my resumé, please post your position offered. Otherwise, you'll have to ask a lot more nicely. I didn't ask for your qualifications to post here. Do you think it's relevant to require a BA or PhD to be able to list your experience with a business, or to post an (informed) opinion? I might as well ask, "what makes you know what you know", but it's not necessary. Heavens, where did you get the notion that I want to convince you? Go do your own research, form your own opinions -- they're like armpits, everyone's got a couple. You are mistaken: 26 PSI is the recommendation for tires on several current model vehicles. Toyotat 4WD vehicles, for one off the top of my head. You don't hear of a lot of RAV4s rolling due to underinflation, do you? Nope, it was a suspension instability that was exacerbated by defective tire blowout that led to rollovers and deaths. The tires were not underinflated. Again, 26 PSI is a spec still in use -- though not by Ford for current Explorers I completely agree. Just for fun, try this: use your favourite search engine, search on the phrases: "les schwab" lousy "les schwab" bad "les schwab" ****ty "les schwab" terrible Read few stories. My point is that Les Schwab is not good to everyone but me. YMMV That's what most of us do, I suspect. Good question, one I'm happy to answer: In the beginning (many, many years ago, before I aquired as much, er, experience as I now have), I was awed by how Schwab trained their folks to interact with us peon customers. Every other tire store I'd used treated us like cattle. "Those pesky customers" is a phrase that comes to mind. Schwab always seems to want my business. I want to encourage businesses to acquire and develop this attitude. It was only after several bad experiences with more than one Schwab that I began to discern the pattern, that being one of friendly but not always knowledgeable Schwabites. Ref. my short list above. I will admit that I am a demanding customer. If I pay somebody for service, I'm going to get that service. It's going to be done right (as far as I can ascertain), and the rig is going to come back to me as clean as I gave it to them. If not, there's always hell to pay. I don't think that it's unreasonable to expect good service, and I don't pay for bad service. And I know the difference. Huh? I don't think I've complained "a lot". I complained that our local Schwab refuses to utilize the recommendations of highly-paid automotive engineers. Then, when you asked for us to not bash Schwab, I listed my valid and relevant experiences with Schwab to justify my panning of the Schwab empire in the PNW. Targetted and specific. I'm not offering any alternatives or suggestions because there are a lot of folks reading this who don't live where I do and couldn't make use of my advice if I recommended Bill's Garage or whatever. Generically, I advise to support local businesses (wherever your "local" is), but that's only Step Zero: beyond that, you have to do your own research for your area. Well, this has nothing to do with either Jiffly Lube or Schwab. You're broadly generalizing about the residents of an entire region, and your conclusion is based upon faulty logic. And it's useless discussing faulty logic. Perhaps your comment is inspired by your particular environment (ie Renton). There's a lot of Puget Sound that is not Renton, I assure you. And not everybody here has moved from California, bringing their driving habits with them, though if you have to commute on I-5, I doubt you'd believe that. I can remember when the Totem Lake area was all blackberries. Sigh. You want to discuss tires or Quick Lube joints or anything Subaru-related, or you just want to yell that Al doesn't have any positive suggestions?

Well, I'm on my way to being stumped. After reading more about how the fuel pressure regulator is supposed to work on my '93 Loyale, I decided to retest it. Test gauge teed after the fuel filter. Key on engine off, while the pump runs: 27 PSI. After pump shuts off: 23 PSI (OK) Cold idle = 28 PSI (spec is 20-24) (bad) So, there's a problem. There is a possibility that my test gauge is miscalibrated, though. So I follow the test sequence and pull the vacuum line off the regulator and hook a vacuum gauge to it. Spec is about 19" Hg. I get nada with engine iding. I used compressed air on the TBI fitting, still no vacuum. Took the top off the TBI, and I can't for the life of me see how you could ever achieve vacuum at idle: the tap is bored to a single hole above the throttle plate. I did a vacuum check with a hand pump, while covering the hole with my finger, and it holds, so I'm assuming that either there is no other passage drilled down to the manifold, or that there is, but it's been plugged or occluded by some other repair. I don't want to pull it off the manifold, because it looks like I'll have to drain the cooling system, and I can't get into that tonight. I'm mystified. Both AllData and Mitchell are explicit: the fuel pressure regulator is connected to manifold vacuum. Mine isn't. Help! Somebody with a SPFI EA82 please pull the vac line off the fuel pressure regulator and verify that, at idle, there is vacuum on that port. If verified, I'll have to tear the TB off the manifold tomorrow and see what the hell is going on under there. The fuel pressure regulator on a SPFI EA82 is horizontally screwed to the driver's side of the throttle body, is black in colour, has one fuel line connected to its top, and one vacuum line off the bottom (which is hard to see). On my rig, that vacuum line is about 2.5" long and is connected to a vacuum tap on the rear of the throttle body -- the port which has no vacuum at idle on my engine. Even aside from this issue, I'm not getting proper regulation from my regulator: with no vacuum on it, I should see somewhere above 30 PSI (simulating WOT). I've applied 20" vacuum to the unit, and it seems to drop too low (15 PSI). Unfortunately, this is another $160 part, so 1st things 1st, I'm going to sort out this vacuum feed problem.

I can't post a pic right now, but I have the CEL/ECU info. The ECU has three wiring connectors. On mine, they're yellow plastic. One of them has 12 pins. One of the two wires nearest the lock latch is Red w/Yellow stripe: that's pin 40, and it's the CEL lamp ground. For the following test, you'll need to drop the ECU down. It's behind the knee bolster under the steering wheel. Remove knee bolster (several screws). The ECU is retained via three 12mm head nuts, and the ECU mountings are slotted. Supposedly, you can just loosen the three nuts, and pull the ECU toward the rear of the car, but I've always had to remove at least the rightmost nut completely. Once down, locate Red/Yellow, straighten a paper clip and insert it into the back of that pin. Connect a voltmeter's red probe to it -- a clip lead works best, but use what you have. Connect the black voltmeter's probe to a good ground. Measure that pin 40 with Key On Engine Off: a) if 12v or close is present, ECU is bad. if 0v or close is present, * CEL lamp is lighted, or * CEL lamp is bad, or * no power to CEL lamp If , I can get wire color for lamp feed, but one step at a time: check that voltage.

Quite by accident, I found another reference to this: http://216.239.53.104/search?q=cache:LdmUiC_gdQcJ:www.smpcorp.com/download/view/tt1q00.pdf+subaru+diagnostic+codes+1993+loyale&hl=en&ie=UTF-8 SUBARU-1989 TO 1993 W/1.8L ENG.- NO START If you are working on an '89 -'93 Subaru L series or Loyale with a 1.8L single or multi-point fuel injection system that's a no start, has spark and fuel injector pulse, but no fuel pressure, here's something to keep in mind. Before investigating the fuel pump circuit, please be aware that most wiring diagrams for these vehicles indicate that the blue and black wire from the fuel pump relay coil is grounded by the ECM. That would be at pin #47 for the single point injection system, and pin #6 for the multi port system. This is true for vehicles that do not have automatic shoulder belts. For models equipped with automatic shoulder belts, the fuel pump relay's winding is grounded by a Revolution Sensor or fuel cut unit. This unit is similar in operation to the fuel pump relay used on many European models (for example, Volkswagon). When this unit receives a tach signal from the ignition coil negative terminal, it grounds the blue and black wire, thus energizing the relay. The Revolution Sensor is usually located just above the hood release cable handle.

Yup, Glen described my situation pretty well. I covered the step-by-step (sorta) here. (Man, it's like pulling teeth, to get a specfic post ID number from this system -- I had to go read the source.) Key On, Engine Off, the CEL should light steady. On my '93 Loyale, it's on the lower right end of all the red lights under the instrument panel. If it does not light, I think that it may be easier to put a voltmeter on the ECU connector to which the CEL is connected, and with the key on, see if that pin is near 12v or near ground. If near 12v, the ECU is defective -- or, most likely, the transitor driver array uPA1478H is bad, like it was on mine. I do not have the wiring diagram for that circuit with me right now, but if you want to do this check, post back here or email me (asavage at iname dot com) before 6am PST and I'll get you the ECU harness pin number and color code of the wire to check for your '90 Loyale Turbo. If you want to try to fix the ECU (and the engine does run), you can replace that uPA1478H with one that Glen will sell you, if you are comforable with desoldering through-hole plated PCBs. If not, you can take the ECU and the new part to pretty much any TV repair shop and have them do it. I didn't go that route: the CEL driver is only one-fourth of that uPA1478H, so I found a single-transistor replacement, snipped three leads off the uPA1478H, and grafted my transistor to them. It's still working (and tonight I got around to paralleling a second (yellow) LED to the red one. Mine's on a coax jack and 6" lead, so I could watch it blink while driving home, though I did have to watch it between the steering wheel spokes, which isn't too safe.). If my crappy digital camera would do any kind of decent focus for shots <4', I'd have taken some nice pics. Unfortunately, it won't (anybody wanna buy an Olympus D380 with orginal box/accys/charger, w/128MB card for cheap?). Lastly, you can send me the ECU and I can do either repair (graft-on driver or replace the uPA1478H with Glen's). No charge, you just pay shipping and buy the part. So, the no-CEL can likely be repaired. Replacing the entire ECU is another idea. The thing (for your car) can be had for as little as $50. Use http://www.car-part.com/ to see where. On mine (93) there were three different ECUs available in that year, and the cheapest I could find was for something like $150.

It's a bit more complex than that, I'm afraid. If it's a model equipped with automatic shoulder belts, then there exists another box that controls the fuel pump relay's ground instead of the ECU. I can't find a wiring diagram of this circuit in AllData, but here's the "alternative" fragment: Also, there's a TSB on checking this Fuel Control Unit: Click on each image for a larger version to print.

What's the normal range, what should I expect? Almost all my driving is what would be considered "highway" (ie few stops).

(Here's the article that Skip posted on checking codes and code meanings.) That's a relay that cycles every two seconds. It's normal in D-check mode. (Warning: some of the below may not be correct; I'm writing this off the top of my head, and it's not straightforward stuff.) There are at least three modes: * U-mode: no connectors connected, this is the condition you drive around in. The Check Engine Light ("CEL") tells you that you need to check for some codes. No CEL = only intermittant codes might be stored. * Read Memory: (Loyale: white connectors connected) On later (some models, about post-85??) ECUs, there is some RAM to store intermittant codes. Check for these by connecting the white Read Memory connectors. * D-mode: (Loyale: green connectors connected) use this for setting ignition timing, and for running the diagnostic procedure, which purposely tests all sensors and switches in the emissions system, over the course of two minutes, but you have to do specific things, and there are a couple of things to avoid doing during this procedure. D-mode will set new codes, which will then have to be tracked down. If all you're going to do is check codes for existing hard codes, you don't need the D-mode (green) connector. (It's important that, if you are going to run the D-mode test procedure, that the engine is warmed up. I am not including the D-mode procedure here.) To check for existing codes, you check in two modes: a) with no connectors (U-mode) with Read Memory connectors (white ones, next to green ones, next to the fuel filter on the left shock tower on Loyales) (Read Memory mode) No, the CEL will not flash codes on the Loyale series. It will come on (to indicate a hard code, that can be checked in U-mode via the O2 LED), and it will only flash in one case -- and you won't see that in any mode except after completion of a D-mode test (and the flashing does not indicate a code value in that case anyway). Loyale: the CEL comes on with key on engine off. If it doesn't, fix that first, because it's absolutely necessary for further code checks. Next: If, with the engine running, the CEL light stays on or comes on, read the code(s) on the ECU's red O2 sensor LED. The codes will repeat, so no hurry, you won't miss any. You don't need have to have the engine running (but it can be) (edited: no, it can't be: if the engine is running closed-loop (warmed up), the O2 LED will monitor O2 transitions in U-mode, not output codes. Must be Key On Engine Off to read codes in U-mode), and the engine does not need to be warmed up for this U-mode check -- NO connectors need to be connected at this point. If you don't have a CEL light with the engine running, you probably have no U-mode codes, and when reading the O2 sensor LED you'll get a sequence of short flashes only, repeating. (with key on engine off). My '93 Loyale, non-Calif, 4WD, gives seven shorts flashes for the Processor ID. Confused yet? The range of possible Processer ID values is around three to eight, I think. Again, you will only get this sequence of short flashes if you have no U-mode codes set. Next, you can read stored intermittant codes, the ones that come and go, and cause the CEL to come on but later go off. Do this by connecting the Read Memory connectors with engine off, then turning on ignition (Loyale: white connectors next to green D-mode connectors, near fuel filter on left shock tower). Read codes the same as in U-mode. All codes start with a "tens" flash, so if you are only getting short flashes, you have no codes. If you get a long flash at any time, then there is a code stored, and that's a code tens digit. Count the long flashes, then count the short ones that follow, and add them up: three long followed by two short is 32, which is an O2 sensor fault. Etc. Use Google to find a code listing chart for your year, there are several. If there is more than one code, there will be a pause between the first and subsequent codes. At the end, they will all repeat, and continue to repeat as long as the key is on. Once you fix the cause for those U-mode codes, you can reset them by connecting BOTH the D-mode connector AND the Read Memory connector, then running the D-mode diagnostic. Forget about even trying this until you've found the cause of the U-mode codes, because they won't reset until you fix them, period. Hope this isn't too confusing. If it is, I can't help it -- it's the early Subaru way, and I don't do my best tech writing B4 breakfast.

Spark Plug Wires! They're often not replaced as part of a tuneup, and they get leaky and can be the cause of misfire into the intake tract. Skip is correct, a lean condition can exacerbate this, but with a modern ignition system in good condition, lean misfire doesn't (isn't supposed to) occur; instead, you just get a power loss. Anecdote: I am (I really am!) trying to get my Loyale wgn whipped into decent shape for my friend without spending as much as is my wont. This car had had a tuneup recently, I have the receipt, but they didn't replace the plug wires, and I vacillated over whether I should. As a test, I grounded a long screwdriver with a clip lead, then with the engine running, I ran the tip over the wires, ends, spark plugs, and all over the dist cap. When I brought it near the coil top, it drew an arc. I shut off the engine, then removed and carefully cleaned the wire end, boot, and terminal. Retested, no arc. Next day, I decided to reset the timing. I plugged in the D-check connector, warmed up the engine, removed the bellhousing plug, loosened the dist and connected the light. I decided to back off the timing a few degrees, so I put a shop rag on the cap, and with my gloved hand, tried to rotate the dist. I got zapped hard! Thirty seconds later, I was on the phone, ordering a new wire set. What galls me is that the day before, I couldn't get an arc to a solid ground, but next day it zapped me through a clean, dry shop rag and a layer of latex! I guess I'm just a magnet for automotive abuse.

It was either a Volvo or Toyota owner's manual I was reading last year, that said to their seat belts were designed to be replaced every 10 years! I've replaced worn/frayed webbing, but if the retractor mechanism gets worn, I'd look hard at the replacement one. If it was a 'yard one, it'd get a good cleaning and lube. A lot of junk gets into them.

The Turner Pick-A-Part has particulary good prices. I bought an item there about three years ago, for half of what I had to pay elsewhere.

Let me add my voice to the "no ARI" list. Aside from the iffy joint rebuild, their boots don't seem to hold up very well, either.

That reminds me . . . Car & Driver commissioned a dual-engine Honda Civic, back in the '80s, and did an article on it. My fantasy from my teen years: a dual V8 VW Squareback. At one time, they were sold with a script "1600" emblem on the tailgate. Dual engine: '76 Cadillac Eldorado drivetrains, one for each end. 500 ci each = ~16,000 cc . Add one more zero to the "1600" script Put one in Drive, one in Reverse, use up tires to your heart's content. Or more likely tear up the chassis. Well, it sounded good when I was seventeen. A few years later, at a Bug-In (at SIR), I saw a quite well done Squareback with a rear Toronado conversion (455 ci). It was a good job. Sano, as we used to say.

For my '93 Loyale. I'm not certain, but at least once I got an odd resistance reading from mine. New is not cheap -- gee, is there anything cheap for Subaru FI? Anybody got a junk EA82 SPFI motor they're parting out? This sensor is mounted on the right (left, if you're bent over the radiator) side of the thermostat housing. It's hard to get out if the upper rad hose is still attached. You have to use a wrench, not a socket, because it is the one that has a six-inch two-wire harness integral (ie this isn't the single-spade-terminal sensor that's next to and behind it, that feeds the dash gauge). I'm very fond of parts-swapping in the case of intermittant and rare FI faults, to eliminate -- or at least reduce -- variables while troubleshooting (If it gives you any trouble, . . . ). I'd be happy to slip a Alexander Hamilton (who was never a President) (for the part) and a Abraham Lincoln (for shipping: ZIP 98368) in the mail in exchange for one of these. (This offer void outside the US -- shipping doesn't make it worthwhile, and supply is plentiful here.)

They're not cheap to buy. I bought Spanish ones at cost, and they were $26 ea. If they weren't making noise when you took it apart, I think I'd disassemble & clean them and reuse them. I just can't see forking over $200-400 for them, when I could spend that money on something else, and take two hours to thoroughly clean them instead. This page shows the details. The only hard part is prising off the cap to get at the guts, and it's not really all that bad. I used a deep socket to put the caps back on the bodies. That said, I'm apparently one of the few folks who really did have a bad one: I replaced a bad oil pump mounting seal, and I took the camcase off and replaced the feed o-ring and cleaned out the relief valve galley, but I still had a warm-up tapping. Two days later I took the camcase off again and replaced four lifters (I could not be certain which one was bad), and it's been silent for a couple of weeks now. So, in my case, I needed one lifter. But from what I've read here, in the majority of cases they don't go bad all by themselves, they have to be helped along (by not keeping the oil clean).

Would you believe that that is a $320 hose? I know, because it's common for them to crack, both at the outline of the cast-in "SPFI" rectangular logo border, and around the PCV and IAC "nipples". I had both cracks. I carefully black-RTV'd mine a couple of weeks ago.