1. Attachments are working again! Check out this thread for more details and to report any other bugs.

Fuel Dilution of Engine Oil

Discussion in 'Gen 2 Prius Technical Discussion' started by brick, Dec 24, 2008.

  1. brick

    brick Active Member

    Joined:
    Dec 10, 2006
    1,083
    78
    0
    Location:
    Upstate NY
    Vehicle:
    Other Non-Hybrid
    Model:
    N/A
    My most recent oil analysis indicated 1.0% fuel dilution of the M1 5W30 that I had run for the last 8300mi (since August). Wear indicators and TBN were just fine, but I would like to have a discussion about what may be causing this. Like others who have reported similar results (even running shorter oil change intervals), my driving conditions are not conducive to false-positives on fuel dilution. I commute 20 miles each way on the highway, so the engine always gets up to temp. I almost never let it idle. And I always take my oil samples hot, typically within 20 minutes or so of having driven home from work.

    So if it isn't me, it must be the car. Could this be an inherent condition of the Prius drive system? Since the engine typically runs at 80% load, the intake charge is usually dense with fuel. That creates ample opportunity for gasoline to be forced past the rings and into the crank case on the compression stroke.

    The wide throttle opening also means that vacuum in the intake should be relatively feeble. I suggest that crank case ventilation may be inadequate relative to your average non-hybrid engine. There is also the fact that some of the intake charge is forced back through the valves and into the manifold as part of the Atkinson cycle.

    I guess the first question is, should we be concerned about it? It stands to reason that this is no problem if there is no indication of excessive wear as a result. So far so good in my case. But it still bugs me. The trend is in the upward direction and I don't really want to get to the point where it **does** start to impact wear. A good car should be able to last decades if well-maintained, if not for the original owner then for the next person looking for economical transportation.

    So the second question is how to diagnose it. Has anyone looked at things like oil temperature in the Prius vs. other cars? Perhaps someone has already studied any impact of high engine load on crankcase ventillation? Perhaps there are other possibilities we haven't looked at? The Prius is such a well-instrumented car that I think we should be able to isolate data that might tell us something valuable.
     
  2. dogfriend

    dogfriend Human - Animal Hybrid

    Joined:
    Feb 26, 2007
    7,512
    1,185
    0
    Location:
    Carmichael, CA
    Vehicle:
    2007 Prius
    I wonder if environmental conditions may also have an effect - e.g. in the summer, the engine gets up to operating temp rapidly, but with the colder weather I notice (via ScanGauge) that it is taking much longer for the car to reach operating temp and on some trips it doesn't get there at all.

    I don't think it is going to cause any damage as long as you maintain OCI to minimize the fuel dilution. Perhaps I will submit a sample on the next oil change at 30k - I haven't had the oil analyzed but I have tested the trans fluid a couple of times just to see what was going on there.
     
  3. Celtic Blue

    Celtic Blue New Member

    Joined:
    Aug 4, 2008
    2,224
    139
    0
    Location:
    Midwest
    Vehicle:
    2008 Prius
    For many of us the question might be does fuel dilution quickly reach a steady state or does it continue to increase with mileage between changes? 1% fuel dilution in 3.5 L would be 35 mL. If sampled at 1,000, 5,000, and 10,000 miles would there be an increase in fuel concentration or would it hit ~1% by 1,000 or 2,000 miles then plateau?

    The effect on viscosity and lubricity are the real concerns. From what I saw in a quick web check dilution is not a concern until it exceeds 2.5%. At that point one is supposed to investigate mechanical reasons for the fuel concentration.

    The way the Prius runs I would expect higher fuel concentration in the oil in winter than summer even if the engine is nominally warm during every drive.
     
  4. Patrick Wong

    Patrick Wong DIY Enthusiast

    Joined:
    Mar 8, 2008
    18,198
    6,461
    0
    Location:
    Green Valley, AZ
    Vehicle:
    2015 Prius
    Model:
    Two
    Hi brick,

    What is your odometer reading now? Have you replaced the PCV valve yet (if odometer exceeds 60K miles)?

    I'd be very surprised to see Prius lasting "decades" which implies 20 years or more of life. I don't expect ICE failure to be a big issue; rather the complexity of the electronics, intermittent connections developing, etc. will result in very costly repairs (that exceed vehicle market value) as time passes.
     
  5. brick

    brick Active Member

    Joined:
    Dec 10, 2006
    1,083
    78
    0
    Location:
    Upstate NY
    Vehicle:
    Other Non-Hybrid
    Model:
    N/A
    That oil change was at 39,000mi. I have been reading a lot about PCV valves lately and will certainly look into it around the 60k mark, or if the fuel dilution starts to get excessive.

    As for the general topic of longevity, I guess I have high standards. My first car was an old Volvo that I didn't get until it was 13 years into its life. My family still has that car (currently 23 years old...only a few years younger than me!) and it would start and run just fine as a beater if one spent some money on a clutch and a set of road-worthy tires, maybe a 12V battery. Based on that experience, I think I am justified in expecting a lifespan on the scale of decades if a vehicle is well-designed, well-built, and well-maintained. (And if it isn't the first two things, I don't want it!)

    Except for the battery and the other wear items, there isn't anything on the Prius that shouldn't last in excess of 20 years. Electronics and electrical pathways don't really wear out the same way that mechanical parts do. Heat can kill them, but you take care of that through design and taking care of things like the inverter coolant pump. If you have a 20 year old computer that doesn't work, the most likely falure point is the hard drive.
     
  6. tochatihu

    tochatihu Senior Member

    Joined:
    Apr 10, 2004
    8,972
    3,501
    0
    Location:
    Kunming Yunnan China
    Vehicle:
    2001 Prius
    I had several UOA on 2001 Prius and the highest % fuel was 0.5 as I recall. Mild efforts to preheat the engine oil before draining seemed to have no effect. This was up to (about) 90 k miles on the original PCV valve.

    In my case there was no obvious upward tread through time. My guess is that letting it cool off after a very long hot cycle would yield the most meaningful % fuel number. Has anyone done exactly that?

    Years ago there was one German Prius reporting an oil drain that resembled a water/oil emulsion. If I ever saw something like that, then I'd worry.
     
  7. Patrick Wong

    Patrick Wong DIY Enthusiast

    Joined:
    Mar 8, 2008
    18,198
    6,461
    0
    Location:
    Green Valley, AZ
    Vehicle:
    2015 Prius
    Model:
    Two
    Hi brick,

    Well, the Volvo that you mentioned above likely relies upon a very low level of technology. As long as parts are available, it doesn't take a rocket scientist to maintain that car.

    Since you mention a PC, think of a network of ten 80386-vintage PCs that you are trying to keep running. The good news is that the PCs don't have hard drives or rely on the MS-DOS operating system. The bad news is that the network is mobile, subject to daily shaking and G-forces, and also daily heating and cooling 30-40 F degrees or more. The network is connected via a very complex wiring harness with literally hundreds of connections, we are not talking about an Ethernet hub. Further, each PC has its own custom design so parts are not interchangeable - imagine an IBM PC hooked to a Commodore 64, an old Apple, proprietary microcomputer systems like Burroughs, Control Data Corp, etc.

    How hard would it be to maintain that network, obtain spare parts, etc. Then add the electrical interfaces to the mechanical components and think about how hard it will be to maintain the total package.

    How many 1988 Toyota Supras do you see driving around your area? My guess is none or if you are really lucky maybe one, owned by a fanatic car collector who is willing to invest more money in keeping the car running than its market value.

    Now think about Prius and the fact that Toyota dealer service is barely adequate now, to deal with HID headlight issues, Smart Key fob programming, and the occasional braking problem. Think ahead 20 years when the techs no longer have expertise or diagnostic equipment for the model, and an owner can't get the immobilizer ECU to recognize the fob, the car refuses to stop when you press on the brake pedal, the MFD fails so you can't turn on the heating/AC or the stereo, etc. We haven't even talked about a traction battery or transaxle failure.

    If you personally are an electronics expert and are good friends with your local salvage yard, then you have a chance to keep your Prius running after 20 years without expending big $$s. Otherwise, forget it.

    If you do a few searches here and on other Prius forums, you can easily find many stories from Classic owners who already have the problem of being out of warranty and having to pay $3-4K to their friendly dealer's service dept to replace the traction battery, then facing the likelihood that the transaxle, electric steering, MFD, etc may also fail one year later. Usually someone driving a 7 year old car is not happy to pay four-digit repair bills with no end in sight.
     
  8. jayman

    jayman Senior Member

    Joined:
    Oct 21, 2004
    13,439
    639
    0
    Location:
    Winnipeg Manitoba
    Vehicle:
    2004 Prius
    I think that's true of any modern car made since OBD II came along. Once those cars are 20 years old, imagine how flaky the wiring will be.

    I have a 1984 Ford work truck at my hobby farm. Despite the fact it was kept inside most of the time when not in use, the wiring is literally falling apart.

    I've had to rewire almost everything on that truck, that I wanted to keep working. Since the truck came with a factory carb and no emissions controls - 1984 was the last year for Ford trucks to run on leaded gas in Canada - it was easy to retrofit a different ignition system

    All of the factory guages have quit, so I have an assortment of mechanical gauges for oil pressure, coolant temp, and an aftermarket fuel gauge

    A modern fuel injected motor with o2 sensors, what happens once the wiring goes flaky and the o2 readings get erratic? How about the MAF or camshaft position sensor

    Near my hobby farm, a neighbor has a small fleet of Kenworth highway trucks, including a few newer T2000. One of the older trucks with a Detroit 60 motor kept quitting on him. There was a glitch in the crank sensor, and whenever that happened, the motor quit

    I think the reason wny it's possible to find cars from the 1950's to early 1970's kept running, assuming they haven't crashed or rusted out, is because mechanically, they are very simple. Once you learn how to replace points in a distributor, they are very easy to keep tuned
     
  9. jayman

    jayman Senior Member

    Joined:
    Oct 21, 2004
    13,439
    639
    0
    Location:
    Winnipeg Manitoba
    Vehicle:
    2004 Prius
    It's interesting how some Prius have fuel dilution issues, but others don't. There must be a combination of ambient temp, duty cycle, perhaps even fuel used, that is contributing to this

    I'm as baffled as you are. I have no idea what could be causing this. I had zero fuel dilution at 10,000 miles
     
  10. dogfriend

    dogfriend Human - Animal Hybrid

    Joined:
    Feb 26, 2007
    7,512
    1,185
    0
    Location:
    Carmichael, CA
    Vehicle:
    2007 Prius

    That's why you need to move to a nice dry warm climate where they never have any reason to salt the roads. Even the OBD1 and OBD2 Fords survive in Sacramento. :D
     
  11. brick

    brick Active Member

    Joined:
    Dec 10, 2006
    1,083
    78
    0
    Location:
    Upstate NY
    Vehicle:
    Other Non-Hybrid
    Model:
    N/A
    We might be able to figure this one out with existing data. I know there is a spreadsheet full of oil analyses here in the file library, and there should be large quantities of more general data available at places like BITOG (though I don't know what form it would be in.) A little data mining might reveal a correlation between OCI and dilution. This is a good question and worth investigation.
     
  12. Celtic Blue

    Celtic Blue New Member

    Joined:
    Aug 4, 2008
    2,224
    139
    0
    Location:
    Midwest
    Vehicle:
    2008 Prius
    It might be beneficial to determine if the labs themselves correlate. If I sent the same oil out to two different labs, would one trend significantly high or low compared to actual?

    As I understand it, fuel dilution is determined by GC (gas chromatography). I've done some process GC method development in the past and am hesitant to rely on any lab GC results I can't crosscheck with blind known samples.
     
  13. Celtic Blue

    Celtic Blue New Member

    Joined:
    Aug 4, 2008
    2,224
    139
    0
    Location:
    Midwest
    Vehicle:
    2008 Prius
    I took a look at the thread in the PC file library. I could find no obvious correlations upon quick inspection.

    To get anything definitive would require pulling samples from the same oil charge at intervals with approximately the same temps over an extended drive (same profile.) The weather would likely be the most difficult factor to control for over 10,000 miles. Obviously having several engines and oil types in such a test would be necessary for confirmation of observed trends (or lack thereof.)
     
  14. Patrick Wong

    Patrick Wong DIY Enthusiast

    Joined:
    Mar 8, 2008
    18,198
    6,461
    0
    Location:
    Green Valley, AZ
    Vehicle:
    2015 Prius
    Model:
    Two
    Since other posters report lower gasoline content in the used engine oil tests than you, the problem doesn't seem to be the basic design of the Prius ICE. Although your odometer is only at 39K miles, it would be interesting if you could have the engine compression checked to see if you have a problem with one cylinder being slightly leaky. However this requires use of the Toyota diagnostic laptop or hand held device, so you would have to pay for an hour of dealer labor if you wanted to have this done. Does the question of 1% gasoline content in your engine oil bug you enough to pay ~$100 for that test to be performed?

    Regarding excessive wear, you won't have any indication of that until you do a full-throttle acceleration and see blue smoke in your rear view mirror.

    You also indicated that the "trend is in the upward direction". Does that mean that you previously had other used engine oil samples taken that reported lower gasoline content? If so, what were the prior results, and did your commute pattern remain unchanged over that period?
     
  15. jayman

    jayman Senior Member

    Joined:
    Oct 21, 2004
    13,439
    639
    0
    Location:
    Winnipeg Manitoba
    Vehicle:
    2004 Prius
    That's it, rub salt in my wounds

    Actually, I always thought baking desert heat ruined wiring a lot faster than bitter cold did. As far as salt, that ruins everything. If there is a break in the wiring insulation, or some Bubba used taploks to home wire a trailer harness, the wiring literally falls apart in under two years
     
  16. jayman

    jayman Senior Member

    Joined:
    Oct 21, 2004
    13,439
    639
    0
    Location:
    Winnipeg Manitoba
    Vehicle:
    2004 Prius
    That's the wrench in the gears. I would suggest that two different labs would return different results. Even the same lab would have statistical variation, which could only be determined by blind sampling

    Correct
     
  17. hill

    hill High Fiber Member

    Joined:
    Jun 23, 2005
    19,604
    8,036
    54
    Location:
    Montana & Nashville, TN
    Vehicle:
    2004 Prius
    Model:
    IV
    I've always expected to find some trace amounts of fuel in the oil after an ICE gets over 36K miles. On the other hand, one of my first 'green' vehicles was a Ford Van that I converted to run 100% Propane back in the late 1970's. The conversion took place at 56K miles and I kept the van for over 160K miles. Once it hit 50K miles, it began burning about a quart of oil, every 3K miles, btw. I never changed the oil after the conversion, although I had to add a quart to conpensate for the oil getting past the oil control rings / burning. I kid you not ... the oil retained its original consistency and honey color on the dipstick. The regular filter change was all it needed. Gasoline (especially compared to propane fuel) truly acts like a solvent.
     
  18. brick

    brick Active Member

    Joined:
    Dec 10, 2006
    1,083
    78
    0
    Location:
    Upstate NY
    Vehicle:
    Other Non-Hybrid
    Model:
    N/A
    Looking at the UOA results (updated and attached below) my results are not isolated by any means. Having browsed through the BITOG forums it would seem that even higher levels (up to 3%) have been reported by other Prius drivers.


    A compression test is a good idea. Is it worth $100 to me right now? No. If the condition (since it really isn't a "problem" at the current levels) gets worse I would consider it. Probably not until after I check simpler things like the PCV, though.

    ...Or the UOA results come back with abnormally high wear metals. I sample at every change so it's probably a better indicator. Blue smoke (burning oil) could be other things such as a head gasket or worn valve guides. Black smoke would be indicative, too, in the event that the A/F ratios get screwed up. OBD should catch that, though.

    It means that fuel dilution is becoming greater over time. At the moment I don't know if it is just because of the longer drain intervals or because something is really changing. It does fluctuate so it's tough to get a read from so few data points.

    Speakig of fluctuation, feel free to play with the spreadsheet that I have attached. I plotted flash point and fuel concentration against a couple of different independent variables. I think the most interesting is fuel % vs. month. It's pretty clear that the highest fuel dilution levels tend to be observed in the winter months. Not surprising by any means, but worth noting.
     

    Attached Files:

  19. Patrick Wong

    Patrick Wong DIY Enthusiast

    Joined:
    Mar 8, 2008
    18,198
    6,461
    0
    Location:
    Green Valley, AZ
    Vehicle:
    2015 Prius
    Model:
    Two
    Hi brick,

    Thanks for compiling the spreadsheet. I note that your car's used oil test results occupy columns AB through AF.

    See cell AF35 which shows 0.1%. Is that correct or should that cell be 1.0%?

    Looking at columns AB through AE, row 35: Are three samples really at 0.1% and one sample at 0.8% (or are there other errors?) If the values are correct then it is weird that the results jump up and down so much.
     
  20. brick

    brick Active Member

    Joined:
    Dec 10, 2006
    1,083
    78
    0
    Location:
    Upstate NY
    Vehicle:
    Other Non-Hybrid
    Model:
    N/A
    Patrick,
    You're right, that's a typo. Should be 1%.

    Also, I'll explain what I did. Blackstone likes to use imprecise measures at times, such as "TR" or "<0.5%" which cause havoc when generating a plot. As a remedy (albeit imperfect) I assigned 0.1% in place of "TR" and 0.25% in place of "<0.5%".

    In the interest of accuracy, the spreadsheet that I placed in the UOA thread of the Files section was NOT modified in this fashion. That one has the raw data as printed on the UOAs that users posted to that thread.

    I agree that the fluctuations for my car are a bit odd. I would be more comfortable with it if we could perform a measurement system analysis but I'm not sure how to do that without going to considerable expense.