Is oil analysis a good indicator of bearing wear?

Status
Not open for further replies.
Joined
Sep 22, 2007
Messages
12
Location
Austin, TX
Hello,

I have a 1992 Porsche 968 (thus my user ID) with about 107K very well-cared for miles. The precedessor to the 968 (the 944) was known to have issues with #2 and #3 rod bearing wear, though this seems to be less of an issue in the 968 due to improvements in oil pan baffling.

Anyway, I did an oil analysis with Blackstone a year ago at my last oil change, and it showed no sign of elevated copper or tin, though lead was a little high (but Blackstone said this could be due to a particle that could be temporary). I'm about to do another OA, and am wondering how reliable oil analysis is in determining the health of the bearings. In other words, if there's still no sign of elevated bearing metal, should I rest easy and not worry about doing a pre-emptive bearing replacement?

My concern is that on the one hand, bearing failure on these engines can lead to catastrophic results, but on the other hand, replacing the bearings on this engine is a HUGE and very expensive job, so I don't want to do it if I don't have to. I use Redline 10W40 synthetic in the engine, and since the car isn't a daily driver, I change the oil once a year, which works out to every 5000 miles or so. From the records, the previous owners kept up a very regular schedule of oil changes.

Thanks in advance.
 
If you are really concerned about failure then I suggest you also read up on trending. Random/occasional oil analysis may not catch a failure in time to be useful. Oil analysis as performed by most members on this site is only checking the condition of the oil and not done frequently enough to maximize its usefulness.

Analyzing the filter would also indicate impending failure.

You also need to determine the benefit of knowing when the wear rate elevates. If the rod bearing is wearing excessively and you catch it what is your plan of action? a complete rebuild? how is that different than if you had a failure and caught it late? Just tossing it out there for something to think about.
 
Thanks for the responses. Yes, I understand that I need to establish a trend to get a reliable read on the state of my bearings. I'll have my second data point soon; if it still shows no sign of elevated copper or tin, then this, combined with my solid oil pressure, will make me feel better about going awhile longer before undertaking the huge bearing replacement job (you have to take everything but the paint off this car to get to the bearings
mad.gif
). But I also understand that the OA gives no indication of how close the bearings may be to giving out. Oh well, walking across the street also invloves risk, right?
 
First off, nice car. A friend of mine had a 944S2 and a 951. The S2 ran 14.0@100mph bone stock. Very impressive for a stock non-turbo 4 banger.

Here's my experience with UOAs. Keep in mind it's just one experience so take it with a grain of salt.....

In my GN, UOAs were looking ok but oil pressure was falling steadily and I could see "sparkles" in the oil when I changed it. After a while, oil pressure was 0 at idle and it had so much blowby that I had to install a second crankcase breather. UOAs were still acceptable. Upon teardown the rod bearings were nearly past the copper, mains were ugly, no hone marks on the cylinders, and a nice big wear ridge on the top. Pistons were frozen on their pins, etc.

This engine had major problems, the UOAs didn't catch it. This happened twice. I think that when there's a lot of wear and the wear particles are larger the UOAs won't pick it up. When it's small wear and fine particles the UOAs will pick it up. I agree that trending is the only way to use it properly but I would not lose a bit of sleep over slightly eleveated metals. IMO a UOA is good at looking at how much life the oil has left or catching coolant leaks into the oil but not for wear.

I'll catch a lot of grief for saying this, but an accurate oil pressure guage and trending hot idle oil pressures may be a better indicator for rod bearing wear.
27.gif
 
Quote:
(you have to take everything but the paint off this car to get to the bearings ).


I know what you mean, I did a clutch on a 924. The front engine/rear transaxle certainly makes for a lot of work. Then again there is nothing common about the design of any Porsche, part of what makes them great.
 
Originally Posted By: tom slick
Quote:
(you have to take everything but the paint off this car to get to the bearings ).


I know what you mean, I did a clutch on a 924. The front engine/rear transaxle certainly makes for a lot of work. Then again there is nothing common about the design of any Porsche, part of what makes them great.

Actually, Porsche did its owners a favor with the 968 by going to a removable bell housing, which (I'm told) makes a clutch job relatively painless. But getting the oil pan off is a whole 'nother beast. The engine mounts sit on the cross-member, to which is also attached the castor blocks. So, to get the pan off, your have to pretty much remove the front suspension, unbolt the engine mounts, support the engine from above, and move the power steering pump and AC compressor out of the way. But at least the paint gets to stay on
grin2.gif
). And when you get everything back together, you need a wheel alignment. So, as I said, I'll do it if I can convince myself it's necessary, but I'd like to put it off as long as it's safe.
 
Originally Posted By: tom slick
If you are really concerned about failure then
You also need to determine the benefit of knowing when the wear rate elevates. If the rod bearing is wearing excessively and you catch it what is your plan of action? a complete rebuild? how is that different than if you had a failure and caught it late? Just tossing it out there for something to think about.

Thanks; If I can put together evidence that the bearings are getting nesr the end of their life, I'd replace them (and install a windage tray while I'm in there). From my understanding, bearing failure on these engines tends to be spectacularly catastrophic - the bearing bonds itself to the crank, causing the rod to break, and bandsawing the crankcase, destroying a priceless (to my bank account, anyway), no-longer-made block. But bearing replacement is a huge job on these engines, thus my quest for solid evidence that I need to do it.
 
Since you appear to be interested in using oil analysis for the intended purpose, I recommend you start using Dyson Analysis for this car. You will get a professional interpretation of the data instead of lots of questions. Since you value this car and seem to be intent on keeping it for awhile, Terry Dyson will get you on the right track.
 
UOA, for the average driver, is almost impossible to catch gradual normal wear over time that can eventually lead to catastrophic failure. Even trending will be tough to really see a failure coming soon and one does not know what normal is for your car or drving habits, how much wear is needed before failure etc After many years of doing UOA I came to the conclusion that unless looking for a major problem, coolant leak etc, (which you can problably diagnose by other symptoms) is a waste of money.
 
Originally Posted By: Spector
UOA, for the average driver, is almost impossible to catch gradual normal wear over time that can eventually lead to catastrophic failure. Even trending will be tough to really see a failure coming soon and one does not know what normal is for your car or drving habits, how much wear is needed before failure etc After many years of doing UOA I came to the conclusion that unless looking for a major problem, coolant leak etc, (which you can problably diagnose by other symptoms) is a waste of money.



Well said and why I've never had one. I've read 100's of them, enjoy the discussion, but at the end of the day feel they are a waste of money for "me".

Then I read about BuickGN experiences with wiped out bearings and his UOA's showing no sign what so ever. Upon his visual inspection he discovered they were shot. It was at that point I said I'd save my money and use it to change the oil a little sooner. All of my cars enjoyed a long life without a UOA, I'll continue to pass for now.
 
Cool, my son is an Airman attached to a Prowler Squadron with those planes. They jam radar as mentioned. I wonder if his buddies that actually work on them know anything about that company and filter analysis. Next chance I get to ask him I will. The EA6B is being retired, and replaced with an F-18 IIRC.
 
Originally Posted By: INDYMAC
Since you appear to be interested in using oil analysis for the intended purpose, I recommend you start using Dyson Analysis for this car. You will get a professional interpretation of the data instead of lots of questions. Since you value this car and seem to be intent on keeping it for awhile, Terry Dyson will get you on the right track.


The other difference you'll have with Dyson is Rotrode Spectroscope vs. the ICP Spectroscopy that most other oil analysis services use. Rotrode has a significantly larger particle detection size window. This helps in catching those 5-10 micron particles that can indicate abnormal wear issues.
 
Originally Posted By: Spector
UOA, for the average driver, is almost impossible to catch gradual normal wear over time that can eventually lead to catastrophic failure. Even trending will be tough to really see a failure coming soon and one does not know what normal is for your car or drving habits, how much wear is needed before failure etc After many years of doing UOA I came to the conclusion that unless looking for a major problem, coolant leak etc, (which you can problably diagnose by other symptoms) is a waste of money.



Spector, your statement is way too general. Whether you can detect an event depends on your sample interval and the maximum particle size that the oil analysis can detect. ICP spectroscope, which most people use here, is deficient in that respect, which is why I like the Rotrode method.
 
Last edited:
Originally Posted By: RI_RS4
Originally Posted By: INDYMAC
Since you appear to be interested in using oil analysis for the intended purpose, I recommend you start using Dyson Analysis for this car. You will get a professional interpretation of the data instead of lots of questions. Since you value this car and seem to be intent on keeping it for awhile, Terry Dyson will get you on the right track.


The other difference you'll have with Dyson is Rotrode Spectroscope vs. the ICP Spectroscopy that most other oil analysis services use. Rotrode has a significantly larger particle detection size window. This helps in catching those 5-10 micron particles that can indicate abnormal wear issues.


Good info. I'll keep that in mind if I ever decide to do another. 10 microns is about twice as large as Blackstone is able to capture, right?
 
Porsche uses their own in-house method to determine bearing wear.

Quote:
Porsche 996FL Engine test. This test will last 203 hours. The engine, and the oil, will go through: - 4 times the simulation of 35 hours of summer driving, - 4 times the simulation of 13.5 hours of winter driving, - 40 cold starts, - 5 times the simulation of 1-hour sessions on the “Nürburgring” racetrack, - 3.5 hours of “running-in” program Measurements on the engine and on the oil will be done at regular intervals, and the following parameter will be taken into account to grant the approval or not: - torque curve (internal friction), - oxidation of the oil, - Piston cleanliness and ring sticking, - Valve train wear protection. Cam & tappet wear must be less than 10 µm. - Engine cleanliness and sludge: after 203 hours, no deposits must be visible. - Bearing wear protection: visual rating according to Porsche in-house method.
 
Thanks for the interesting discussion, and varied opinions. I'm going to contact Dyson about my situation. Although, I do see the point several are making about UOA likely not really being able to detect an impending catastrophic failure, which, from what I gather, is the only kind my type of engine tends to experience...
 
Status
Not open for further replies.
Back
Top