asavage

Yup, mine's got a break on the "1" position coil. I just twisted the two broken ends together, and put it back in, temporarily. Works, but won't last. I live in an area with exactly three junkyards within 50 miles. Two of them have NO Subarus, the third is the one I've been dealing with -- and they don't let you at the stock, it's a "we pull it" yard only. If I was back in Portland, I'd have my choice of U-Pull-It yards, but not around here. Yup, that's what I found. $120 (cost) for me to buy it, and strangely, not in stock. Fortunately, I've got a $60-plus-solder-two-wires solution in hand (see above post).

Well, for the '93 Loyale, it looks like this (175k). The unit on the left is the OEM unit. It's located just behind the water outlet housing/thermostat housing, bolted to the top of the intake manifold with one 10mm head bolt. It has two small vacuum hoses and one two-wire connector (GL = Green w/blue stripe, BW = Black w/white stripe). The BW wire should have near 12v on it with ign ON. [bTW, I ran across a '91 Loyale today, and pulled and measured it -- 36 ohms, and you can blow through it when de-engergized.] The unit on the right (Subaru Part No. 14774AA101) is not for this particular application but I was told it is functionally identical, with the exception of the orientation of one spade in the electrical connector, so I'll snip off my old one and solder it to the new one. [it measures 35 ohms, and I can blow through it.] The dealer quoted me $120 (cost) for the correct one. This new 14774AA101 cost me $60 from a knowledgeable local. A Google search on that number comes up with a reference to an EGR solenoid, so perhaps that's what it's for. At any rate, I'm grateful to just have one. Tomorrow, I'll do the mod and install it, clear the code, drive it, and see if any other codes present themselves.

Engine oil drain plug. 17mm hex, 22mm thread. NAPA numbers: 704-1357 copper washer, 22mm ID 704-1082 fibre washer, 22mm ID I've used both, they work fine, and can be reused without leaking.

I used "Search" but didn't find any good hits. Where do I buy a new Purge Control Solenoid? I've read that the new ones have been redesigned to (perhaps) be more durable (part number ending in 351?). NAPA and Beck-Arnley don't list them. Mine's got an open coil. ------------------ Next, how 'bout the Blower Resistor for the multi-speed interiour fan? Mounts in the air duct, under the glovebox area, two 4mm screws that like to rust into place, and one 10mm head screw on a plastic bracket. Symptom of my bad one: fan position "1" does not run blower. "2"-"4" work fine. When I got it out, the nichrome winding for the "1" position is broken. My local Subaru 'yard says they're "always bad". I can find one source aftermarket, but my cost is over $70 for it. Looking for cheaper -- any ideas?

First, the PC37 bulb/socket combo does not quite fit the dash. I trimmed the socket a bit with an Exacto, but it's not something I'd recommend -- buy the right part. I mention it only to prevent someone else from trying it. Tonight, I got the ECU re-mounted in position, reinstalled the dash panel, and fired it up. CEL worked like it's designed. After about two minutes, I finally got the CEL that I was looking for, so I levered myself down eyeball-to-eyeball with the O2 monitor LED, and read the code: 35: Purge Control Solenoid (PCS) continuously ON or OFF. I see, from a Google Usenet ("groups") search that this is a very common problem. I followed the Alldata diagnostic chart, and found the PCS' coil is open (should be approx. 100 ohms) [correction: should be 35 ohms, checked on one used and one new unit]. It's located just behind the water outlet housing/thermostat housing, bolted to the top of the intake manifold with one 10mm head bolt. It has two small vacuum hoses and one two-wire connector (GL = Green w/blue stripe, BW = Black w/white stripe). The BW wire should have near 12v on it with ign ON. NAPA does not list this part, neither does Beck-Arnley, so I removed mine, put it in a Ziplock baggie with a note, and hung it on my local Subaru junkyard's door, so they'll find it tomorrow. Hopefully, they'll call me with good news, but it sure looks like I'll be better off buying a new one. Thanks to all for your help and support. Now I'm off to start another thread or two.

It does! Plugged the repaired ECU into the harness this morning, connected a test lamp to the dash connector (the lower dash is out, waiting for NAPA to get me a PC37 bulb/socket combo, as one of mine was missing), and KOEO the lamp lights. Start engine, lamp goes out. Almost as if it was designed to work that way You must not have the GL licensed in Portland. I lived there for years, you gotta pass the smog test to get your tags every two years. Whatja do, register it in Newberg? One nice thing about Portland, you can take the smog test for free, and as many times as you want (you still have to wait in long lines, of course). Lets the home DIYer do his own emissions repairs for cheap. Some tough cars I worked with, I used that resource a lot.

One problem/advantage of network-based interaction is that you get to make up all kinds of stuff . The job I've had that I liked best was an electronics bench tech for two years, computer monitors & terminals. Until about five years ago, you didn't need to know much . I own a TDS-210 'scope that I bought new, a Pace desolder station, Fluke 23 & 79s, and the usual complement of assorted junk, all of which mostly sits in storage these days -- I am not really educated enough to do commercial-level electronics work, and you can't make a living fixing the consumer crap anymore -- six years ago, my average repair cost for a monitor or terminal was $110; nowadays, it's like a broken VCR: you don't repair it, you toss it and buy a new one. Being able to find an open junction on a part that drives the CEL lamp is a non-paying skill. I can only call it a hobby and write off my time as such; nobody in the North Olympic Peninusula is going to pay me to do this kind of work, and I'm not moving to Portland or Seattle to chase a job that has probably already been outsourced to India, Pakistan, or Korea. ========== A kind soul on sci.electronics.repair decided that an NTE2338 would be a perfect fit: http://www.nteinc.com/specs/2300to2399/pdf/nte2338.pdf and it looks good to me. Not only that, but a cursory look through my junk bin yielded two (2S)D2018 parts, which cross to NTE2338, so I'm all set. The NTE part is under $2, but nobody within 70 miles of me stocks NTE parts; I'd have to drive to Bremerton to get it, or have it shipped. I'm off to the storage unit to get out the Weller station and stereoscopic helmet and dust them off. If this doesn't work, I'll take you up on your kind offer. Undoubtedly, replacing the uPA1478H with a like part would be the most professional solution, but the 2338/2SD2018 is an acceptable alternative for my own car. Just for completeness, this is a: 1993 Subaru Loyale SW, 4WD, AT, A/C, mfgr'd 10/92 ECU: 22611-AA397 (from 9/92 thru 1/93) Processor spec.: 7 (Fed/49 state, A/T) Component ID: IC14 Component Part No.: NEC uPA1478H

Yes, this ECU has a one-quarter-failed NEC uPA1478H. Base = pin 8 (5v drive from somewhere else) Collector = pin 9 (to CEL lamp) Emitter = pin 10 (tied to ground) Base drive voltage toggles fine, but the C-E junction doesn't ever pass current. With open collector, I see base voltage less .3v at collector. Looks like a bad part to me. Unfortunately, I don't find a ready source for the part. I could call around on Monday, but I'm going to try to replace that one-fourth with a discrete transistor instead. Circuit design isn't my fortè -- I usually just fix 'em, I don't try to re-engineer them -- so I've posted the relevant info over on Usenet sci.electronics.repair, and I hope that someone over there will point me to a generic transistor that I can pull out of my junk bin and solder in. Unless someone's got a spare ECU from that era laying about from which I could rob that part? Would likely need to be a 26111 series ECU, but I doubt that it matters what the last five characters after that are, they probably all use the same part for that circuit. For reference only: Datasheet picture (PNG format, with my scribbled-in circuit notes) here: http://www.ncplus.net/~asavage/Subaru/uPA1478_01_.png (77k) http://www.ncplus.net/~asavage/Subaru/uPA1478_02.png (37k) Or, if you prefer JPEGs: http://www.ncplus.net/~asavage/Subaru/uPA1478_01_.jpg (147k) http://www.ncplus.net/~asavage/Subaru/uPA1478_02.jpg (74k) (Original NEC datasheet: http://www.necel.com/nesdis/image/IC-3566.pdf (59k) )

Not on mine I've swapped the CEL bulb with the Brake Fluid Level bulb, they're both good. The ECU is not providing a CEL lamp ground -- I checked. Sigh. Looks like I'll have to round up an ECU. ECU 22611-AA397 (from 9/92 thru 1/93) Plenty of used ones available: http://www.car-part.com Cheapest listed is $45, and it's on the wrong coast. And I'm seriously short on budget right now.

I'm chasing a driveabilty issue. My '93 Loyale (A/T) has a Check Engine Light on the lower right of the instrument cluster. I do not have an Owners Manual. My Alldata reference CD strongly suggests that, under normal conditions, this CEL should light with Key On, Engine Off (KOEO) (without D-check or Read-Mem connectors connected: that is, normal driving setup). Well, mine doesn't. I'm playing with a SnapOn Scanner (scantool) MT2500 and cursing it (because its onboard directions do not remotely match the paper instructions), but it also implies that the CEL light should come on in KOEO mode. I've backprobed the ECU -- actually, it's now sitting on the workbench in front of me, but I did a ground check on the appropriate circuit (Pin 2 of the center/18-pin connector, Red with Blue stripe) and it does not provide a ground at KOEO. So . . . do I have a faulty ECU? I need someone who owns a similar year/model to check that the CEL comes on with Key On, Engine Off (without playing with any connectors whatsoever). Thanks in advance. Email to asavage@iname.com if convenient, cc here of course.

Gervais from Scio? That's quite a commute. What, is the trash-burner hiring? I lived in Woodburn for five years, Salem for two.

Hello, Skip: Sometimes my writing seems more abrasive than I meant it Also, I really am not all that familiar with the Subaru culture etc. My previous Subaru was purchased because it was an eyesore in front of a house in Everett that I began renting in 1986. So, naturally, I bought it $50. It was a '73 (GL? DL?), pumpkin orange with a peeling black vinyl roof, and it had been punched pretty hard in back. The more I worked with it, the more impressed I was with it. Electric fuel pump, electric rad fan, in '73. I eventually ran it up and down from Everett to Salem every other week (long distance affair), got 31 mpg and it would scatch in 2nd, really fast off the line. I can recall one trip in December, with a storm blowing through, I was on I-5 in Centralia area, noticed that the speedo was moving up and down from 60-80 and back. Weird, I thought. Then, I realized that I was hydroplaning, but driving just as smooth as can be. Aaron's wrecking was the Subaru mecca for Seattlites at the time -- are they still in business? I recall very friendly staff, exceptionally cheap parts, and an impossible to get to location -- you could see it, but you couldn't actually drive to it! I found Aaron's from his sticker that someone had put inside the cover of the used Haynes manual I'd bought for that '73. Well, the car was pretty well worn, the (1300?) was a bit loose, and especially the trans input seal would dump about a quart of GL5 every week, so I sold it to a kid for $250 in early '87. And though about 40 cars/trucks/MCs have come and gone in the years between, this '93 Loyale wgn is only my 2nd Subaru, and I didn't buy it for myself, but for a friend whose wife always wanted one, and because the seller needed cash quick to move back to Pittsburg. Win-win. So, if I've stepped on some long-standing Subaru tradition WRT 20W50 oill, I'll act somewhat contrite -- for a while. I'm a fan of synthetic oils, specifically, and low viscosity oils in general. Back in the heyday of VWs (I also owned a bunch of Corvairs, loved 'em), I was in the thick of it, and Castrol GTX 20W50 was "the thing". I think, for the same reasons that it seems to be popular with other loose-tolerance engines: heavy base viscosity masks other issues, from a lack of oil volume (exhibited by a lack of pressure, naturally) to worn parts, to design flaws. While running a heavier oil will boost system pressure, it's not a cure for the root problem(s). I've owned enough beaters to know that it's a valid economic decision to run heavier-than-spec'd oil. I've also achieved a place where I don't have to settle for that fix -- I can afford to chase causes, invest time and effort to overcome them . . . or revert to 20W50, if I feel like it. Right now, I'm not tired enough to give up Qman: Yup, and Cameron's page is specific about replacing it. Mine was (as seems typical for the EA82) hard as a rock, and additionally was leaking external to the engine, which was one of the factors that led me to R&R the camcase. (The other side is leaking at the same point, and I might work up the gumption to R&R that side as well -- but probably not.] oregonloyale: Yup, I bought an oil pump install kit from Beck-Arnley, in-stock at my local parts place, even up here in the sticks! Something like $15 for the seal, o-ring, and "mickey mouse" profile gasket. The large o-ring was in decent shape, but the shaft seal was pretty stiff, though not crumbly-hard. ========================= Now, for the good news: I got three, quiet cold starts out of it today, after replacing the four right lifters last night. That's a first for this car, in the 10 days I've owned it. I'll be firing it up again in nine hours, and if it's quiet two days in a row, I'm gonna declare the noise banished. If so, then I must have had one of the rare faulty lifters, in addition to the camcase o-ring leak, the oil pump mtg gasket leak, and the varnished camcase drip tube regulator shuttle. MMO might have eventually freed up the shuttle, but as I disassembled and thoroughly cleaned those four lifters with no improvement, I've got to believe that MMO wouldn't have done a thing for them. Neither would running 20W50 have repaired the leaking camcase o-ring, nor the oil pump's multiple sealing issues, and it would have actively discouraged the MMO from doing its thing on the shuttle valve. When the goal is to remove deposits, you don't want thick oil. Now, if the bottom-end starts a-knocking, and I can't afford or don't want to delve into that level of repair, I'd switch to 20W50 or even 30 in a heartbeat. But not for this car, not for now, and not for this (noisy lifter) problem. Thanks for all the advice, folks. I found lots of good tips and helpful people, which is 3/4's the battle. [Now, where can I learn more about the 3AT's governor problem?]

Skip: No, I didn't check the oil pump rotor-to-hsg clearance, though I did pull it out and look at both, looking for obvious wear and scoring: none found that was disturbingly bad. On the oil: Gack! No, I'm not running 20W50. I'm a bit of an oil snob. I've just acquired this car last week, and have it running Chevron 10W30 for cleanout, and will switch to Mobil 1 10W30 after a couple more hundred miles -- I have 150 miles on this oil, and it's already looking lousy. edrach: While it's common for springs that are exercised frequently to change their rate and total loading, I am reluctant to believe that this little spring under the banjo fitting has seen a lot of exercise. Not much polishing of the (not very precision) plunger it operates, and even if it was sacked, I doubt that it would change the pressure regulation much: that whole passage is fairly small. I found a source for lifters at $28 ea (cost, Beck-Arnley, made in Spain, probably on the plain), so I ordered up a set of four and installed them tonight (three more hours: I'm getting faster at this). After initial bleed, I was surprised to find that the engine overall is quite a bit quieter, I must have thought that the general racket was "normal Subaru noise", but now I can actually hear the lifter(s) murmuring on the "other" (driver's) side now. Drove it home 15 miles, sounds nice. Now, if it'll be this quiet tomorrow at 6am, I may have it licked. I'll post tomorrow with results, good or bad. I was sure hoping that the bad gasket above was the cause!

Whoops! I read that as you were trying to use a '96 modulator, and I didn't think that that would work. Sure, I think a modulator from a '90 would likely work.

Don't use the condenser for either. It's not designed to be compatible with green (silicate-based anti-corrosion) coolant, and the organic-acid-based orange/pink coolants aren't compatible with your brass radiator (if yours is brass, that is). It won't flow enough coolant to do much good anyway. It'll just end up being your coolant filter: in a parallel-flow plumb job, there won't be enough pressure to keep fines in suspension, and they'll settle in the condenser. In a series-flow plumb job, the condenser is too much of a restriction You *might* use it for oil. It'll handle the pressure, there aren't any corrosion issues. It'll be hard to drain at oil change time, though. And, it *is* possible to over-cool oil, esp. organic oil. You really need it up over about 220° pretty quickly. Drives out water through the PCV system. The reason you want to cool oil over cooling coolant is the larger delta T of oil to air: The oil is hotter than air by a greater amount than the coolant is hotter than the air. Greater difference in hot to cool temp = greater heat flow. Therefore, you can dump more heat out of the oil than you can out of the coolant, assuming the same surface area and media flow rates -- this last is theoretical, of course. Also of course, this ignores the fact that once you dump the heat out of the oil, you've decreased the delta T of the coolant, because the condenser is dumping heat into the rad's airstream. On the topic of removing the thermostat: fine for experiments, lousy for engine and oil life on gasoline-fueled engines. It takes much longer to come up to operating temp (undoubtedly you've noticed this), condensate water in the crankcase (byproduct of combustion, gets past rings, and a lot of it is produced when cyl walls are cold) mixes with acids formed during combustion and present in non-synthetic oils, combine to form sludge, which doesn't lubricate very well. Keep those warmup times short and those temps up for long engine life. OEM = 195° for coolant, isn't it? [i just replaced my themostat, because somebody had put a 180° one in, and the dash gauge barely moved. It was regulating to an actual 167°. I used an infrared "gun" to measure. Changed to an OEM 195° t-shat, measured at the thermostat housing 193°. Nice.] (LPG fueled engines are a different story)

I don't know the '83 GL, so this is only generic advise, but I have seen this on three vehicles over the years. If it doesn't overheat at idle or low speed, but does at higher speeds, check the lower radiator hose. Many vehicles are supposed to have an anti-collapse spring in the lower hose, so that when the engine is revved and the pump is pulling hard, the lower hose doesn't collapse. You can check this easily: with a hot engine (coolant up to temp), open hood and watch the lower hose while opening the throttle with one hand. Rev it up to 3-4k or so and hold it there for five seconds or so. Hose should not change in diameter appreciably. If you see it start to flatten out, you've found your problem. I've had three cars over the years (non-Subarus) with the springs either missing or rusted out (from lack of regular coolant changes -- every two year!).

OK . . . pretend I don't know what's gone before: I assume there was previous hosting on ezboard, is there some way of accessing it still?

It sure sounds like a failed vacuum modulator to me. I can't answer whether the later unit will fit (I kind of doubt it) or how difficult the 3AT's is to change. I will say that oil is a lot lower octane than gasoline, and you can do permanent damage to one or more pistons if you run it hard in this condition.

While the modulator does affect shift points and clutch apply pressure, it doesn't (shouldn't) have any effect at a stop: even with the vaccum line pulled off completely, or power braking, the trans should shift to 1st if the wheels aren't moving. I think it might be a lazy governor, which is why I'm trying to find out where this "archive" or "manual article" on the malfunctioning governor is. Anybody care to enlighten me?

I cleaned that. Spring was in OK condition.

Just where is this info? I went here: http://www.usmb.net/repair/?FileType=Folder&CurrentDirectory=FOLDER_3f29b58f50d9b8.15492800/ but no mention of governor. So I used the "Search" for governor, got zero hits. My '93 Loyale 4WD 3AT has several times stayed in 2nd gear when I pull up to a stop, and then when you press on the accel, it downshifts into 1st and you take off. I've already flushed the AT and replaced the filter, so I'm hoping to find a cure. BTW, I had the Devil's own time finding the correct AT pan gasket for the 3sp. The one I've got has a very shallow dogleg on one side (about 1/4") and a larger divot on the other side. The pan gasket the parts stores kept trying to sell me didn't have the dogleg, and without it the gasket for sure wouldn't seal. It figures that both the correct gasket and the incorrect one both have 14 bolt holes. I couldn't get that gasket separately, but I found the correct filter kit that contains it: ATP B-154. The ATP catalog's picture is extremely misleading (wrong aspect ratio, and slanted ends!) but the gasket and filter are correct.

If that's the filter base I've seen before, that takes a std Pennzoil PZ1, Fram PH8A, Wix 41515 or NAPA 1515 filter. Yup, Ford used them for about 40 years. Popular on remote filter conversions for air-cooled VDubs, too. They hold almost a quart of oii, and you can buy them anywhere

Damn! It didn't. Got new timing belts & tensioners, replaced the idler bearing for the left belt, and the idler bearing for the A/C belt, fired it up, it did the same thing: tapping until warmed up, but the tapping comes & goes sudden-like, on the right side. So . . . since I have a pretty good oil leak going right at the point of the cam case where the famously-leaking o-ring resides, I bit the bullet and followed Cameron's advice and pulled off the right cam case, removed all the lifters, took them all apart and carefully cleaned them. Replaced that hard-as-a-rock o-ring that he points out, cleaned the oil relief valve under the banjo fitting, put it all back together . . . and it sounded pretty much the same when I fired it off. Drat! The real test will be in the morning, when it's cold again, and the oil system's been purged by a 15 minute run. However, I'm thinking that I really do have a bad lifter. Unfortunately, even with a stethesope, I can't tell which lifter it is -- I can't localize it. Conventional wisdom says to replace all four, but my cost on the damned things is $52 ea! Yikes!

It is. Something like an Impco CA100 or CA125 mixer (carburetor) for single-fuel use (ie not gasoline and LPG). The major parts are: DOT Highway-approved tank -- this is NOT just a barbecue tank! High-pressure line. About $3/lf Lockoff unit. Impco combines both the lockoff and filter features in their VFF30 vacuum-operated unit, which can be had on eBay for about $20 + rebuild kit for $10, or you can buy it new for about USD$60. Electrically-operated lockoffs are also available, as are separate filter unit, but the VFF30 is cheap, ubiquitous, sturdy & reliable. Converter/Regulator. Impco Model E is the standard of the industry in the US. eBay prices run around $30-70 depending upon conditiion. Plumbs into the heater hose to use engine heat to vaporize the liquid fuel into a gas, and has a sensitive two-stage pressure regulator to feed the . . . . Miixer (carb). Many manifold and air filter adapters are available. For a "modern" closed-loop (feedback controlled) system, you'll want to use a std O2 sensor and TPS, and an electronic controller that modulates the Model E's output to control the fuel mixture more accurately. But a few million of these setups are running around without electronic controls, so it's not really necessary. Just a bit easier to tune. The whole electronics package can be had for under USD$200. Trying to get good info in LPG conversions is like pulling teeth. I finally bought a bunch of old Impco catalogs.

What do you think: will replacing this mounting seal fix the cold lifter noise on the '93 Loyale? I replaced all three pump seals tonight (including the shaft seal, which was pretty stiff), and I have to round up new tensioners and idler for the new timing belts tomorrow. I'm hoping that this sucked gasket was the cause of the intermittant lifter noise when cold on the ol' EA82.