98sub2500leg

Is the speedometer and RPM on the instrument cluster both controlled by the speed sensor? The RPM works, speedometer doesn't, no cruise control.

It is getting a 0500 code. I removed the cluster and tightened those screws. No change though. The instrument speedometer does seldomly come back on for short spurts. The fact that it does, although it is not solid as it should be. For example at 40mph, it flutters flutters at 40mph or close to it then completely drops out. It leads me to believe there is a connection issue.

Does anyone have a good picture of one. The one off the Internet shows a threaded emd. No flats for a socket or wrench that I can see. How do you remove it. Also, what is the best way to test it once removed. I have a variable power supply to test it. I would need the pinouts and voltage readings.

How do you remove the speed sensor from from the transmission housing, looks like there are threads on the sensor? Once removed how can you test it? Does anyone have any pictures?

How do you remove the speed sensor from from the transmission housing, looks like there are threads on the sensor? Once removed how can you test it? Does anyone have any pictures?

Also, the cruise control is somehow linked to this circuit and also stopped working.

My speedometer stopped working. It has intermittently came back but when I saw it come back on, it was not registering actual speed and dropped out again. I investigated the cluster. It seems to be ok. There was an older thread about a bad solder joint but it's on a 99 model. The panels on the flex circuit are totally different than the 98 model which I have. ODB2 shows speed sensor. Haven't checked that out yet. Any thoughts.

I'm in need of a lot of parts for my 98 Outback. I'm looking for a good recommendation for a parts supplier with competitive pricing.

Well, there isn't much clearance on the 98' to the sidewall that's for sure. I loosen the plug with the Gloyale combo. I'm sure a lot of us have many a plug sockets from the rubber coming out then having to purchase yet another one. On a few occasions I grabbed the wrong socket with the rubber out of the socket(lost) and found out after the plug was unscrewed. Lucky for me I have small enough hands to get the tips of 2 fingers into the holes and grab the end of the extension, angle the extension to grab the socket and maneuver it out-not fun. Dave's method with the rubber would have saved me some grief though.

I think I got it working ok now. I charged it a bit with a 12 oz. can of r134a got the numbers up to 25 and 180 and that is very relative as the pressures are always changing and trying to equalize with the delta T. The compressor stays on for about 50 seconds with the door open, about 20 sec with it closed. The outside temp. was 75 deg F. cools good but still cycles more than I expected but cooling is acceptable. One of the big problems I was having was that I was testing it at too low an ambient temperature. It seems the best way to properly charge it and test is to have a heat load on a warm day, not a cool day. I also tested our Caravan which has one of the best ac systems I have ever seen it really chills. It seems to cycle about the same.

I am also getting poor milage (about 20-21mpg). My machinist told me that even though the O2 sensor will work after you had a HG problem when you previously had a coolant leak into the engine. He said it may or may not throw a code, but will still work and be at the high end of the tolerance scale of the sensor. These sensors usually get plugged up with coolant in the exhaust after HG, or just high mileage. However, I believe if possible,the part should be tested properly before being replaced. There must be a tolerance for testing the resistance rating with a multimeter, but I would think to test it properly you it you need to take a reading while the engine is running. However, if you disconnect the sensor to test it wouldn't the mixture will be thrown off to the ECU thus giving an erroneous reading? I would think the expensive diagnostic equip. would plug into the ECU side and give it the voltage it is looking for so the sensor could be tested while the engine is running open loop. I can't think of any other reliable way it could be tested. OB, is this the case? He also recommended heating up the pipe before removing to make it easier if replacing it.

I have the same make, model, except I have a manual trans. I just got done doing mine, it's not difficult, there needs to be a lot of attention to fine details. If it has a leak, where is it exactly (has he properly diagnosed it). What was the compression of each of cylinder before he started the head gasket job and what are the numbers now? Did it make the knock before they started the job? One mistake I made when I did mine was before I started I didn't check the compression before I started the job. Not required but a good practice whenever tearing into any engine. If you can get access to an OBD2 tester you can check the engine codes. Some auto parts stores such as will test it if you bring the car to them or they have a loaner program where essentially you borrow it. The connector to plug it into is behind a cover to the lower left side of the steering wheel on the lower dash. If possible check the codes and report back. Just a recommendation, be careful if they recommend any parts, they seem to push parts you absolutely don't need. You shouldn't replace anything until you properly diagnose the problem by troubleshooting the root cause of the problem. I also can't help wonder if he was off on the timing belt. It's a fairly easy job, but again it's also easy to screw it up, especially on this car (DOHC). If I were you and you can't do this, either find another reputable shop to take it to get it diagnosed, or just pay a diagnose fee to the dealer to have them troubleshoot it. Then you will know instead of guessing. Before you do that though see if you get some good responses here to help diagnose it. Good luck, and report back on findings.

Very informative info, thanks to Scooby and OBW. I had a can of freon on the shelf from Interdynamics. It has a pressure gauge with limits. Example, 0-25-low, 26-45 filled, 46-65-alert, 66-225 warning. They have a system pressure chart as a general rule (depending on make/model). The method you described is more accurate OB and Scooby seems to have the right amount of freon to add for my car. Yesterday was too cool to do anything later today should be ok. I will add a bit of freon and repost. Functionally, it does cool and during the compressor cycle the evaporator cools the air inside rapidly. I can't help wonder if the sensing bulb is reaching its set point then cutting out. I doubt it though. http://www.efproducts.com/oneadmin/faqsupport/home.print.php?faq_id=53&category_id=12[/url]

Sorry, I just checked it again and the the answer to question #2 is it reaches equilibrium with in about 5 sec. The on cycle reads as following: 1st 50 sec cycle(on): Low=26 High=115 off for 8 sec then on cycle2 readings: Low=20 High=110 off 8 sec, then on cycle 3 readings: Low=20 High=105 off 8 sec, then on cycle 4 readings same as cycle 3 and the numbers are still slowly dropping when the 8 seconds is up and the compressor hits it off cycle.

Ok, here are the readings: Outside ambient is 62deg F Initial static pressures with engine off and system not run since last night: Low=68 high=65 AC compressor time cycle: Cycling first ran for 50 sec on then 5 sec off then 8 sec on and stayed at the 8 sec on and off for the rest of the cycling. AC compressor cycling on readings: Low=22 high= 105 It seems to stay at steady state with these numbers at 8 seconds on too short of a on cycle to say with any accuracy. It seems to stay at steady state after 6 seconds on and seems to stay there for the last 2 seconds. Compressor cycling off readings: Low=45 high=90

Thanks for the informative post OBW, Yes, I did replace the condenser, unfortunately not the drier on recommendation from another shop. That guy told me that since it was sealed from the factory charge it should be ok w/o replacing. The guy who did the job later quoted quite a bit lower and was recommended by the machinist who rebuilt my cylinder head so I took it to him. He originally told me that it would take 2-3 hours, but when I got there told me it should take 45min-1hr. I thought that didn't sound right. I now know the correct way is to vacuum for about an hour. Stop for 1/2 hr, check for leaks, then revac for another 1/2 hr then charge. My static pressures I measured at 70 deg ambient was 25PSI low and 75PSI high. Although my gauges scale at the bottom shows deg C. on both gauges. I have read a few posts showing the high side should be 4x higher than the low side as a general rule, not sure if this has merit. Unfortunately I don't have a vapor pressure chart. So 70deg F should read 77 or 21.2 deg C which is the scaling of my gauges. When you said the high side should be 7:1-8:1, is that relative again to ambient? If it is, then the numbers all show it to be good because again my gauges read in deg C. I will stick the gauges on again this morning and get back to you with some numbers. Thanks for the post OB.

Thanks Scooby, There seems to be a lot of conflicting info I am getting regarding this issue and that may be the problem. Also, whats strange I have found is that there isn't much info out there to confirm gauge readings or pressures at either static no load (engine off condition) or more importantly, at load condition and/or at during cycling. It seems like the pressures are relative to each other and output temperature differential is what seems to be used by a lot of mechanics with generally using between 1.5-2 lbs of R134a freon. In other words, some mechanics use 1.5lbs or 16oz. (max), others use 2lbs or 24oz. Our 2003 Dodge caravan is not the best to drive, but I appriciate the AC as it really chills the air down. I used that as a measure since it works so well and used a type K digital thermocouple and measured 33.8 deg F. I know that is pushing it, but with the constant cycling on the Suby, you can actually feel the temp and humidity immediately rise when it cycle every 10-12 sec. even if the doors are left open. All I know is I drove it for 11 years and this condition is not normal. Also, flustered after being quoted $79.95 which he said included freon and oil but was charged $147. He also claims he put the right amount in. I greatly appreciate the help. Bryan

Model: 98 OBW, 2.5L Manual Trans. I replaced my AC condenser and took it in to a shop to have the AC recharged. I am a bit weary that he actually did the job right because he did the job under an hour, he said 45m-1 hr. the AC doesn't seem to get as cool as it used to. I drove it back after driving down the road and testing it, he said he added he added 2oz of oil since it was a new condenser and 2lbs of freon, which he said was spec. The Haynes manual says 2lbs of freon but the guys at ackits.com told me that it should take 1.5lbs of freon max for that make and model. I put a set of gauges on it and measured 22PSI(low), and 200PSI high at 1500rpm. The service manager at the dealer said generally Subaru's should have a 35 deg. temp. differential from center vent to ambient. I am getting 22 deg differential which is a 13 deg. off of what it should be. Also, the Ac seems to cycle on and off too much (every 10-12 seconds). The fans cycle on when the compressor turn on and off when it is off. If ackits.com is right, then it is overcharged. They got their data from a Michell manual which is supposed to be an accurate compiled software that shows actual individual ac for any make and model. Does anyone know what kind of real pressures I should be getting on this from the actual factory service manual (not a Haynes or any other manual or general ac data)?

Just got back from test driving it all day. Stopped off this morning and bought several lengths of new hose (hadn't installed them yet), after the car clocked about somewhere over the 100 mile point the check engine light went off. I then remembered when I bought the car new, after about 5000 miles the check engine light came on. I took it to the dealer. They said it usually comes on from not building enough pressure in the fuel line. He also said if I had not brought it in that it would have cleared on it's own after so many miles (he didn't say how many miles). The high idle explains it because when I reinstalled the throttle control cables, I hadn't adjusted the nuts yet for idle. It started out about 1200 or 1300rpm, then got the check engine light then lowered the idle to 800RPM, it seems to have reset itself at some point during the test drive today. The car drove as good as it ever did before. The engine temp. was also very normal, not sure why it was giving off extra heat yesterday but it's good now. Thanks for the good info and update fairtax.

Yeah stupid thing to have done (more stupid to not have double checked it). I examined the old set and kept thinking back and came to the conclusion it had to have been the disc in backwards. There were past posts showing it wasn't possible, a local mechanic told me the bolts wouldn't have fit, but they did fit and really just as easily. I usually always inspect at least 3-10x before assy. depending how familiar I am with the particular job or part. Pretty sad ordeal it took as long as it did but at least it's finally done. 170miles, works great, and thanks for all the help Dave.

Coolant fill: Top rad hose removed from upper cross tube prefill until top rad. hose is full; Remove top plastic air lock screw from passenger side rad. side. Fill radiator, burp-refill, start until warm,cool down, open refill,repeat over & over until full. Did this over the course of about 6 hours, its full. Exhaust seems normal (no coolant smell anymore). The temp gauge is right at normal, the car hasn't run for several months and it was a pretty warm day, it may have just seemed like the inside of the engine was running warmer than normal with the hood open and the car at idle after driving it on the freeway and some crazy test driving. I don't think this is the problem. Both error codes have a common diagnostic of vacuum leak. The P0106 shows map/maf. The 0506 shows idle control rpm higher than expected. Fairtax, did you find your problem yet for your P0106? I saw your previous post from the other day? It may be a leak caused by the same vacuum hose in the same sensor as I had before last week when the small hose was off. After getting the P0106 code I noticed it was off then reconnected it. The error code went away as well as the check eng. light. I will go get some new hose for that as it is a press fit type and seems it has lost it's durometer and is too loose for the press on metallic fitting. I saw some older info (thread/posts) which shows it could be a hose reversed.

After finally finishing the HG job, test drove the car, first ran great with no check engine light, then I noticed it came on. Car starts normal at around 1200RPM, then drops to 800RPM after a min. or two as it always had before, idle does not fluctuate(is solid), runs and drives awesome otherwise. Seems like it runs hotter than it used to, all fluids are at normal levels. I have checked everything a dozen time for connectivity. Codes read P0507 and P0106. Before I pulled the engine 3rd time last week(when I had the clutch disc in backwards). I got a P0106, right as it was being read on the display, I noticed the small vacuum tube was off, reconnected & the engine light went off. Now if I push on that hose the engine idle goes up, but the engine light doesn't go out. Any ideas?

Finally got it together, car runs great now. There is still a coolant-exhaust smell but I believe it has to be the residual that spilled out from the 4x engine pulls. The clutch disc was in backwards, the lower plate was bent and the edges were caught near the clutch housing near the studs. Got a new slave cylinder (that was also bad). Still need to bleed the master also due to same crud in the fluid. A clutch disc can be installed backwards. The fork pins were ok as well as the fork. Still burping coolant but runs awesome, good strong clutch pedal. I realized it must have been the disc was in backwards, I bucked up, pulled the motor again within 5 hours had it back together. It went pretty quick, I made a new jig for pulling the motor. After the 2nd pull, I realized the 2 points of pull I was using (chain) wasn't giving me any side adjustment or front to back. I used steel stut-chain-quick link-turnbuckle-quick link-chain-engine. Using 4 turnbuckles is definitely the way to go. You can stand there and fine tune the pitch and side-side with each turnbuckle. Separation slides out quick, no pulling-no mucking with anything. Installation is a bit more difficult, but still pretty quick. It's all about getting the engine mount studs to just clear the cross member by not more than 1/4". Fine tune the angles, mating is a breeze. Thanks to everyone who helped out. Hopefully I won't have any further problems with it.

Is it easy to rebuild the master cylinder?

Definitely bled and cycled all the bad oil out of the reservoir 3x too many pumps to count. The rod pushes in when depressed and out. The rusty metal particles are all out of the system. Although I checked the fluid in the master cylinder and it seems to have the same crud in it (rusty fluid). Not sure how to cycle that out. The slave seems to be working ok now and the old was garbage. When you fully depress the clutch pedal, the rod extends and presses against the clutch fork, then the clutch pedal still stays at the floor (doesn't retract). If you manually pull the pedal back the rod retracts and moves back into the slave cylinder. I am not sure how much it should retract. I thought I read the total throw should be approx. 3/8", which I would guestimate it is. I believe there should be enough spring in the clutch fork to move the pedal back when the rod retracts and relieves pressure against the fork or is the rod just not retracting enough? I think it must be a bad clutch install. Could also be the fork clips as Nipper & Davebugs suggested although they appeared to be ok during inspection. It could also be the disc may be in backwards or a combination of the two, but won't pull the motor until I am conclusive.