34 replies to this topic

[DirtRoadRunner]

I've been having oil consumption issues with my 2000 Forster 5MT for a couple of months. A few days ago it burned though 2 quarts over a 600 mile highway trip. Here is the background on the car:

-136k miles
-Replaced the headgaskets and had the cylinder heads rebuilt in early September, due to external oil leakage
-Car runs great, smooth, has plenty of power, typically gets 24-27 mpg in mixed city/highway driving
-All of the bores appeared smooth when I tore down the engine for the headgasket change
-After the headgasket change, the engine consumed no oil for about 2,000 miles, then rapidly burned through an entire pan til the oil light came on (while turning into my driveway) within a few hundred miles of highway driving
-I replaced the PCV at that point, refilled the oil (4 quarts), and the problem seemed solved. I could not blow through the old PCV valve (an aftermarket) so I think it was clogged.
-It happened again, about a monthl later, though not as quickly (refilled it before it got really low), so I put in a genuine Subaru PCV valve which seemed to fix it again, and changed the oil, also adding a can of Restore.
-That seemed to fix it until the last trip, when it burned about 2 quarts
-It seems that when it isn't burning oil, I get the P0420 code every couple of days and I have spark knock under light-throttle acceleration. When it starts burning oil, both of those problems go away.
-Oil is Valvoline 5W30, which I've used on the car for the past 3+ years with no consumption issues, only leakage before replacing the headgaskets
-The underside of the car and engine is bone dry. I do not believe I have any oil leaks.
-All four spark plugs look normal and none are oily or smell oily.
-Oil does not appear to be entering the coolant, the coolant appears clean and not oily.
-I did a compression test today, and here are the results (cylinder: wet (dry))
1: 182 (210)
3: 184 (230)
2: 182 (225)
4: 195 (230)

I'm stumped. I've also asked this question on Subarforester.org. A leakdown test was recommended next. Since my compression appeared good, I'm not sure if it would tell me anything but I may try one anyway. Could some other problem, possibly in my PCV system, cause this? The valve is clear right now, I pulled off the hose with the engine idling and it was pulling in vacuum. All of the hoses appear OK but are 13+ years old.

Edited by DirtRoadRunner, 29 December 2012 - 05:04 PM.

[GeneralDisorder]

Change the PCV valve and insure all the hoses and ports are clean.

If the problem is oil control rings (poor oil change history and not using synthetic) then there's really nothing you can do short of pulling the pistons, bead blasting the carbon from the ring grooves and putting them back in with new rings - do not hone it as that will just create another oil consumption issue.

Oil control ring issues will not show up on compression tests or leak down tests.

GD

[1 Lucky Texan]

triple check routing of valve cover breather hoses and pcv hose. someone once actually heard gurgling in the dipstick tube due to misrouted hoses.

also, I've read that MMO and seafoam and maybe other treatments in the oil can free-up gummy oil control valves.

[nipper]

There is a problem. The compression numbers, assuming psi, do not make sense.

Factory spec is 185 psi dry. Wet test helps seals the cylinders and gives you an idea of ring wear. Wet should never exceed the dry spec. Compression is way too high. The oil consumption is not a ring issue right now, though if we don't find out why the compression is so high it can be in the future.

Your usage makes sense given the compression. With the higher compression the pistons are hotter then normal. Since oil is used to cool the psiton skirts, you are using more oil then usual on long runs. The ping is from the higher then normal compression, and out of range of what the knock sensor can deal with. You are clogging PCV valves because of the higher oil evaporation due to the higher then normal compression. The O2 sensor may or may not be fluke as the car is old enough to need one.

This is a puzzlement.

What process did you use to take a compression test/ Usally when someone makes a mistake on this it usually results in a low reading not a off the scale high one.


How much material was machined off those heads. Thats the only place I can think of where this can happen as it is too even for a jumped timing belt.

Edited by nipper, 29 December 2012 - 07:17 PM.

[1 Lucky Texan]

I think I've read that oil from the PCV hose can lower effective octane so, perhaps that is a possible reason for pinging?

[nipper]

HIs engine compression is dangerously high.

[1 Lucky Texan]

any blue smoke? at start-up, deceleration or on acceleration?

[efseiler]

My guess is that the oil ring on the piston is not scraping the oil effectively from the cylinder inner wall on each stroke. That tiny residual amount likely goes up in smoke on each revolution.

I'm told that if even a single pin-drop of oil is wasted on each engine revolution the pan would be bone dry within like...a few miles of driving.

I get a similar problem from time to time to time. 150 miles of highway driving and the pan would be down a quart. Other times absolutely very little. Sometimes I can go a thousand miles and the pan would only be down hardly anything. What seemed to exacerbate the problem was driving in trying conditions (snow, slush, rain).


What seemed to alleviate the problem is using additives. I usually put in a few ounces of the Lucas stuff in the pan per oil change and I also use their gasoline additive, too. Once I spilled a bunch of the gas additive and, seizing upon that opportunity I lighted the stuff I cleaned up with paper towels. It was definitely combustible but not volatile. It burned smoothly and uniformly. I took a whiff of it and it smelled a lot like plastic. So it's a petroleum byproduct but not an oil or simple hydrocarbon.

With those additives in use I really seem to be able to drive a full 3+ thousand miles and typically I only have to add a quart at most.

I also use synthetic oils a lot, too. Mobil 1 was purported to be on the of 'good ones'. One interesting property of Mobil 1 is that is expands about 6-7% (by volume) when heated to a typical operating temperature.
I also use 'Startron' as well in the tank.

You could be driving a lot on an odd end of a curve, too. If your engine is running hotter than normal (due to driving conditions or a heavier load) and the oil is 'iffy' it could be just the right set of conditions to cause a serious symptom. A lot of ethanol in the tank may also exacerbate the problem somehow.

Such high temperature and pressure as is found within the combustion chamber introduces many additional potential factors that may have unforseen and unexpected chemical consequences, anyway. It's really difficult to 'nail it' without lots of experimentation and/or consulting experts.

Bear in mind that the 'boxer' design of the Subaru engines may help cause just that kind of side-effect (among others).


Hope that input helps,

--Damien

[efseiler]

HIs engine compression is dangerously high.

ok...but what could be causing that? Are you sure those instruments are properly calibrated?

[efseiler]

so yeah...chances are that's probably it. The timing belt is probably off a tooth or two...

[nipper]

Or the heads were machined too much when the HG was replaced.

Thats my vote.

[GeneralDisorder]

It's probably just gauge inaccuracy. Remember the number you get doesn't matter - it's the difference between them that indicates engine condition. 190 to 195 is normal for most gauges at sea level on a Subaru engine tested cold and dry. His gauge is probably just cheap and a wet test will throw the numbers up higher than 195 pretty easily. Especially if he really pumped the oil into the cylinders - that takes up space and increases the compression. His dry tests are normal if not a tad low.

Compression numbers don't control oil consumption. They indicate a problem with compression rings (rings 1 and 2). Oil control is the sole job of the #3 ring and you can still have excellent compression with totally clogged up oil control rings due to poor oil change history and cheap oil.

GD

[nipper]

Remember, oil is not burned in the traditional sense in a modern engine people. Oil is sprayed on the pistons to cool them from underneath. Oil evaporates due to this process, which is why modern engines "consume" or "use" oil, they do not always burn it.

Lets see if we can get another set of compression numbers with another gauge if at all possible.

And yes the actual numbers DO matter when they are this high, if they are the actual numbers, otherwise yes GD you are correct the variation between numbers are important, but they do have to be within a reasonable spec.

[DirtRoadRunner]

It's probably just gauge inaccuracy. Remember the number you get doesn't matter - it's the difference between them that indicates engine condition. 190 to 195 is normal for most gauges at sea level on a Subaru engine tested cold and dry. His gauge is probably just cheap and a wet test will throw the numbers up higher than 195 pretty easily. Especially if he really pumped the oil into the cylinders - that takes up space and increases the compression. His dry tests are normal if not a tad low.

Compression numbers don't control oil consumption. They indicate a problem with compression rings (rings 1 and 2). Oil control is the sole job of the #3 ring and you can still have excellent compression with totally clogged up oil control rings due to poor oil change history and cheap oil.

GD

I did the test this morning. It was about 30 degrees out, and I let the car warm up for about 10 minutes. I live around El. 550 ft above sea level. Coolant temp gauge was up a few ticks, but not all the way up to where it normally sits. I would say it was warm, but not "normal operating temperature". I put a fair amount of oil in the cylinders for the wet tests, about 2.5 squirts from my tin oil can. It smoked like mad when I started it up after the compression test. The gauge was a $30 O'Reilly tool, definitely not the highest quality. The engine was cranked for 5-7 revs or so for each test.

Oil change history is good. I bought the car from my uncle, who is very particular about regular maintenance and he bought it new. The oil consumption issue came about quickly and suddenly. I am not sure how much material the machine shop took off when the heads were milled, however it is a highly regarded shop so I would assume they would have told me if they took off a lot. Remember it burned NO oil for the first few thousand miles after the HG change.

I replaced the TB during the HG change and got it spot-on. Unless it somehow jumped time since then, it should be spot on. Besides the oil consumption, the car runs great, has plenty of power, and gets 24-27 mpg in mixed highway/city driving. I have noticed NO smoke, other than one time when I noticed a puff when shifting from 2nd to 3rd at WOT. The tailpipe, however, is sooty so the oil is getting burned somehow.

So, how would I properly diagnose a stuck oil control ring? I'll triple-check the PCV hose routing tomorrow vs. my FSM. The last thing I want to do is tear the engine apart for no reason.

[Fairtax4me]

Restore does no good for this issue and may have made it worse.
The oil control rings may not be the issue. There are small holes in the pistons that allow oil to drain out from the oil rings and the holes can tend to clog with infrequent oil changes.
I would recommend switching to Valvoline High Mileage synthetic blend oil. 5w30 is all I run in mine. With 205k miles and a scored #4 cylinder wall I burn almost NO oil between changes. I run 3000 - 4000 mile intervals and MIGHT lose 1/4 quart in that span. It's probably not even that much.

Change the oil to something with stronger detergents to clean the rings, and change it more frequently. Your oil should not come out black when it is drained, it should be dark, more like a molasses color, but not black. If its coming out black you have oil building varnish in the engine before it gets drained. This leads to clogged ports and stuck rings.

[efseiler]

Or the heads were machined too much when the HG was replaced.

Thats my vote.

That's most like it then....but who knows...


Again that slight change of a physical characterteristic can put the chemistry into a whole other part of the curve so maybe if you use synthetic and additives you can eliminate that symptom.


Don't knock them until you try them.

Cheers!

--Damien

[efseiler]

Change the oil to something with stronger detergents to clean the rings, and change it more frequently. Your oil should not come out black when it is drained, it should be dark, more like a molasses color, but not black. If its coming out black you have oil building varnish in the engine before it gets drained. This leads to clogged ports and stuck rings.

So yeah you may have get the Seafoam stuff anyway.

Believe me...it's a very effective organic solvent.

[DirtRoadRunner]

So yeah you may have get the Seafoam stuff anyway.

Believe me...it's a very effective organic solvent.

I've been changing it every 3-4k miles as long as I've had it (about 54,000 miles), with normal Valvoline 5w30. I have all maintenance records from before I bought the car and it has NEVER had infrequent oil changes. However, the oil has always been black when I drained the pan. That could be due to driving conditions though, I drive a lot of dusty gravel roads and like having off-road fun in my Subaru.

I Seafoamed the car back in May, but will go ahead and try it again along with an oil change to high-mileage synthetic oil.

When I replaced the HG's, the engine was fairly clean but did have carbon buildup on the pistons. I soaked all of them in Seafoam and brake cleaner for a few days and scraped off what I could. When I got the car running, it smoked under heavy throttle for the first half hour or so. I think it was burning off all of the crap loosened by the Seafoam. Maybe a piece of loosened gunk is clogging something.

Edited by DirtRoadRunner, 30 December 2012 - 09:48 AM.

[1 Lucky Texan]

were the heads off the car for a long time?

I once had a Honda engine mostly disassembled but still in the car, in my driveway. and, after I put it back together, the oil pressure relief valve was stuck. I suppose I'm wondering if just being exposed for a while could lead to some gummy/varnish forming on the pistons/oil control rings.

[ivans imports]

the small holes behind the oil rings did you clean them ? and new rings or old ? i see them burn oil when the top side of the piston oil ring holes are pluged or the piston is scratched or rings are glazed. The boxers natualy burn oil is normal for it to burn some. if exsesive then rings or bore problem a hone and new rings may fix it up.

[GeneralDisorder]

Yep - the holes behind the oil control rings are the problem. That's what almost always causes oil consumption. The holes are TINY and non-synthetic tends to clog them.

I wouldn't hone it. Just clean the piston ring grooves and install new rings.

GD

[DirtRoadRunner]

were the heads off the car for a long time?

I once had a Honda engine mostly disassembled but still in the car, in my driveway. and, after I put it back together, the oil pressure relief valve was stuck. I suppose I'm wondering if just being exposed for a while could lead to some gummy/varnish forming on the pistons/oil control rings.

They off for one week (I did the HG change over two consecutive weekends). The bock was sitting on my engine stand during that time in a shop, so it is conceivable something got gummed up.

Just finished dumping half a can of seafoam in through the PCV valve. Also got a jug of Valvoline Max Life 5W30 to try out. I'll use the rest of the seafoam before my next oil change (I'll do it at 3,000 miles this time).

Also, I inspected the PCV system. Everything is connected properly and nothing is plugged. However, I noticed oily residue in every hose except the one from the DS valve cover to airbox. Also, I found oily residue inside the air box where both the hose from the PS valve cover and from the crankcase enters. I would expect it where the hose from the crankcase enters, but not from the PS valve cover (the FSM shows that flow should only go into the engine from these hoses....). I cleaned off the residue inside the airbox and will inspect it again at the next oil change.

With the engine idling, I pulled off all of the PCV hoses one at a time. There was positive flow from every hose at idle, but it went away if I revved the engine. Not sure if that is normal or not.

Now, back out to finish my Seafoam treatment by driving the car, and I'll change the oil afterward.

[DirtRoadRunner]

BTW - tearing the engine down is a last resort. Not until I get a place with a garage and a second car to drive to work (hopefully both of those things will be coming soon!). So I will just live with it for at least the next few months.

The bottom end of the car is stock. I just replaced the HG, I haven't touched the pistons or anything else.

[DirtRoadRunner]

Just changed the oil. ~3.75 quarts came out. I put ~4 quarts (from a 5 quart jug) in it ~1500 miles ago when I last changed it, the added the remaining ~1 quart on Friday. Assuming that ~0.25 quarts can be counted a spillage and oil remaining in the filter, I only burned 1 quart in 1500 miles, not 2 in 600.

After the 600 mile trip a few days ago, the dipstick was nearly bone dry (despite being parked level, and checked cold). After adding the last ~1 quart from the 5 quart jug, it was just barely to the "L" hole on the dipstick (where I got the 2 quarts in 600 miles). Maybe most of my problem is fiction and can be blamed on the crappy Subaru dipstick.

I also checked it this morning to find it nearly off the dipstick again (it wasn't quite parked level, the car was angled toward the PS), despite ~4 quarts of oil being in the car.

How in the heck can you get a reliable, repeatable oil measurment on these cars? I always wipe it off, then slowly reinsert and remove it, and still get wildly varying readings.

Also - it blew a LOT of smoke from the Seafoam treatment, almost as much as it did in May when it got its first-ever Seafoam with 130k on the clock. The smoke went away after ~10 min of driving.

[ivans imports]

I tell all subaru customers you can only check oil cold before startup after has been sitting min of 3 hours or it will lie to you by 1 litre