Logic Check Please: VIsc From Oil Pressure & Temp?

[Jim Allen]

I now have an accurate engine oil pressure gauge and an accurate engine oil temp gauge. I wonder if I can roughly correlate "operational" viscosity from those two things? Here's my idea:

1) I will start with fresh 5W-20 of a known viscosity.

2) I will note the oil pressure at a specific set of circumstances, including load, rpm, and oil temp.

3) If I know, for example, that at 1788 rpm at 197 degree oil temp with a new 5W-20 oil will deliver 61 psi oil pressure. (Note these are actual numbers from 8K mile 5W20... I just got the last of the gauges hooked up last night < )

4) Say 7,500 miles later under the same conditions, the oil pressure has dropped to 58 psi, I will know that the oil has sheared. A used oil analysis will be able to tell me how much and, over time, I can roughly correlate the drop in pressure with the drop in viscosity.

5) I may even be able to produce and use a temp/ viscosity chart to plot viscosity loss over the course of the OCI.

I think? What do you think?

(Yes, yes! This is the ultimate in mental, tribology "self abuse" but it give me something to do on long trips. Watching the viscosity drop as you drive... only a BITOGer would ever consider it!)

 

[TAP]

logic makes sense. But sometimes science and logic don't correlate. Maybe someone else can chime in as it sounds interesting.

 

[mechtech2]

To make this work, you would have to have many readings with used up oil. and even more used up.
Is this really what you want to do?

 

[TallPaul]

I don't recall ever having a significant drop in oil pressure that would indicate sheering. I don't think you will see that either unless you are running it extremely hard.

I think rpm related oil pressure changes (due to pumping) will overwhelm any possible indication of actual viscosity.

You could try over several OCI 5w20, 10w30, 15w40 and see what the hot pressure at road speed is. I think 30 weight runs about 3-5 psi lower than 40 wt in my F150.

Then there is the upper limit where the oil pump relief valve dumps.

Now there are some diagnostics you can do with oil pressure. For example, get her hot (no double meaning meant) then record the oil pressure at 1000, 2000, 3000, 4000 (maybe every 500 rpm). Plot it and it should be convex (hump up). Sorry, I forgot what the problem is if it is not convex.

Another is to get it hot, run on the freeway and when you have a slight grade, put your foot into it some for maybe 30+ seconds, then let off suddenly and mark how much the pressure rises. If more than say 2-3 psi, it means loose bearings or something like that.

Of course if the pressure drops when you rev it, that means the rod big ends are leaky and the centrifigul force is throwing oil out.

Surely there is more, but I am scratching my head to remember. Maybe a Google search. But this is not where you were going and I have to get going, so ...

 

[Familyguy]

Wouldn't the oil filter gradually becoming less efficient at passing oil over time (until it bypasses) be the real fly in the ointment here?

 

[FowVay]

Which of your gauges will compensate for those pesky variables such as oxidation, nitration, fuel dilution, and just plain old volatility boil off?

 

[Craig in Canada]

I've seen pressure drop at idle as oil wears in (Porsche 968 running Mobil 1 15W50 red cap - probably in the SJ/SL timeframe in the early 2000s)

There are so many variables I would never consider any of the observations accurate or worth charting, however. I could just observe that full temperature (t-stat controlled oil/air cooler) idle RPM on fresh oil was 2 bar and more like 1.5-1.75 bar when coming up on an OC. I ran Mobil 1 0W40 once in that car too and the idle pressure was 1.25-1.5 IIRC shortly after OC, but my memory is a little foggy since it was only the once around 5 years ago.

Under running the pressure was always in the 4-5 bar region as regulated by the oil pressure regulator/bypass system.

Other interesting observations is that "cold" starts even in the 90s F resulted in pegged idle and operating pressures until 2-3 times as long as the coolant took to warm up. I used the idle oil pressure as a gauge of when the oil was up to temp and the fun could begin. Mobil 1 15W50 SJ was mighty thick although it came out just as clear as it went in it seems.

I exchanged the car for a baby in 2005 so I cannot make any further observations

 

[jstutz]

Awesome experiment. As and engineer, i find myself doing experiment around the house that may think im nuts for doing but it amuses me. Your experiment sounds pretty solid but i a question or two. Are you doing it for absolute accuracy or just generic curves. I think it is going to be difficult to get accurate numbers becuase the starting viscosity, tested viscosity after OCI are going to have a tolerance. I dont know what that is and it may be insignificant. I believe the stated viscosity on data sheets are avg numbers, i dont know how much they very from batch to batch. You would always have to use the same type of oil becuase each of them probably have there own viscosity breakdown over time. What is awesome about this is, even if the numbers are not perfect, you could get a really good idea what your viscosity breakdown will look like after doing several of them and loging the information. If you ever switch oils you will know if it holds up better or worse as far as viscosity. I think it will be very interesting to see if viscosity breakdown over time is linear or exponential. Please keep us posted on your results as i would love to see the graph after a couple years of OCI on it. Seems like it should work. If you can measure temp and oil pressure i dont see why it wouldnt work. Even if the numbers are thrown off a little by different filters or variation from batch to batch, the deltas should line up pretty close as long as driving habits dont change much. Good luck with your experiment. It is nice to see im not the only one that does nutty experiments for nothing more than the sake of knowing something that most dont give a [censored] about.

 

[oilyriser]

Sump temperature doesn't matter. Bearing temperatures are what you need to measure.

 

[Jim 5]

It stands to reason Jim. After all, when the oil is cold on every oil pressure gauge I've seen, you get dramatically higher oil pressure readings than when the oil temp has reached operating temperature equilibrium. On my bronco and my Duramax (without oil temp gauges), I use what I know about their oil pressure to determine when the oil has reached operating temp. This happens at a significantly later time than full operating temp has been reached for water temp.

In any case, as the fluid becomes thinner with heat, you can see it in the oil pressure. If the oil becomes thinner due to shearing, there is no theoretical reason (in my mind) why you couldn't see that too. You would see a change in oil pressure at a certain reference RPM once the oil has reached full/equilibrium operating temp. Having an oil temp gauge makes that measurement even more accurate. The only limiting factor would be if the amount of shear was less than your gauge's ability to detect it.

 

[CATERHAM]

I think it should work.

When I switched from GC 0w30 to Red Line 20wt in my Caterham the oil pressure dropped 10 psi @ 6,500 rpm and 95C. The drop in oil pressure was noticeable at all temps including start-up. There was also a noticeable drop in oil pressure when I switched from a 5W-40 to GC.

One comment I would make is that with a 20 wt oil having a nominal 100C vis of say 8.5 cSt if it sheared to 8.0 that's significant (6%) but it is still only 0.5 cSt. Is that enough to register on your pressure gauge? I'm not sure.

 

[Craig in Canada]

Originally Posted By: CATERHAM
I think it should work.

When I switched from GC 0w30 to Red Line 20wt in my Caterham the oil pressure dropped 10 psi @ 6,500 rpm and 95C. The drop in oil pressure was noticeable at all temps including start-up. There was also a noticeable drop in oil pressure when I switched from a 5W-40 to GC.


In my limited personal experience I'm a little surprised to hear about drops at operating RPM (particularly as high as 6500). Even with a thinner oil, the pressure available from running up in that range should still have the bypass system controlling the oil pressure. As I said in my Porsche I could only discern oil temperature or changes in weight in the idle-speed oil pressure. Operating pressure was always the same.

I wonder if the particular bypass/regulator mechanism on your car had its operation affected somehow by the change in grade as well.

 

[CATERHAM]

The oil pressure dropped from 83-84 psi to 72-74 psi.
The oil pressure with 5W-40 was about 90 psi under the same conditions. The oil pump goes into bypass at about 95 psi.

 

[Jim Allen]

Bumped this for easy review relative to new post.

 

[GMorg]

I think that you should be able to plot your pressure by temp as the engine heats (but only for the pressures that would not be in by-pass mode). Since you can plot the expected viscosity at each temperature along your graph, you should be able to correlate observed pressure and expected viscosity at each temperature. So, I am suggesting the you can validate your strategy on new oil without additional oil changes. The viscosity spread during engine heating is much larger than the viscosity spread across oil grades at operating temp.