why does oil 'disappear'?

[Thomas Pyrek]

This is a rather odd question, but IMO, has some merit...

My friend is runing an unscientific oil filter study. He using a constant oil (Pennzoil 20w-50) and a constant milage (2750) and a constant driving pattern (normal driving with a few rallycrosses thrown in for good measure). Car is a 1985 Nissan 300zx non turbo in good mechanical condition and has had regular changes in it's life.

The so far the results seemed odd. with the known poorer filters less oil has come out of the car and with the known good filters more oil has come out. He has saved all the oil and measured it and each time when he refilled the car he put in the same amount of oil and it was filled to the same level. No top off was added in the testing. The filters have not been cut open yet either.

K&N- 4.3 quarts removed. cleanest looking and not gritty
PureOne- 4.1 quarts removed. near K&N and not gritty.
Pennzoil (Fram)- 2.1 quarts removed. Oil very dirty looking


These were the only ones he told me specifically about. He also told me that Napa gold produced similar results as the FRAM. This one was tested immediately after the k&n was.


My question is, what would cause for there to be less oil with the "crappier" filters. I'm guessing it would be since the oil was more contaminated it burned off. Is this correct or is this logic as flawed as the experiment?

 

[Drew99GT]

The only thing I can think of is oil return ports becoming clogged with crud that the crappier filters are not catching? That is exactly what was happening to my Dad's truck before Auto-RX. After Auto-RX, the disappearing oil scenario was eliminated.

 

[Thomas Pyrek]

how come it's sporradic then? Do the passeges suddendly change their mind about being clogged, then change them later?

the K&N wasn't the first one tested. The pure one wasn't tested near the K&N either. It's inconsistant. that's why I ask.

 

[TR3-2001SE]

I would guess he didn't have the car in the same position each time he drained the oil. The car was not level each time. Only thing that really makes sense.

 

[Thomas Pyrek]

nope, car was drained in the same driveway using the same tools each time. What seemed odd was that the dirtier the oil was the less of it was left. that was always directly proportional.

 

[labman]

Strange. Is he checking the dip stick before draining the oil? Was he running it 2 quarts low with the Pennzoil filter? Could he be getting crud in the seals with the dirtier oil and losing oil? Then clean oil washes the crud out? Did he try the Pennzoil filter again and get the same results?

 

[slalom44]

Is it possible that you did a lot of idling and easy driving with the "good" filters, and harder, faster driving/accelerating with the Fram/Napa filters? Higher RPMs and faster accelerations tend to increase blow-by, therefore oil consumption. I know that excessive idling increases fuel dilution which will reduce the oil viscosity and (possibly but I'm not sure) increase oil volume. This would explain why the Fram oil looked dirtier.

It doesn't make sense that the Fram oil would be significantly dirtier unless the car was driven harder, the filter's bypass valve was stuck and/or the filter medium was torn. You may want to cut open those filters to see. There's a thread on another forum comparing particulate count with various filters, and Fram's filtering efficiency is in line with most others (now, their quality of construction is another issue).

Subconsciously, your friend may have wanted the Fram filter to be the worst and without realizing it, he was more aggressive behind the wheel.

Ambient temperatures may have had a role in this as well. Colder startup temps with some of the filters may have had an effect, however I'm not sure what that effect might be.

 

[White 03]

My first thought is he measureing the oil left in the filter?

My second thought, like Labman, is how could he get two quarts less one time than another time? Did the dipstick show two quarts low?

I would also look to see if some how the testing is not accurate.

Has he tried a back to back test with the same filters to get a kind of baseline?

[ April 05, 2004, 11:02 PM: Message edited by: White 03 ]

 

[T-Keith]

Is it just me or does this "test" not make any sense? So the Fram filter leaks oil, or what is this proving?

-T

 

[Schmoe]

Got to be a human error somewhere in his technique.

 

[TR3-2001SE]

OK I will make another guess. I know with my toyota that has 251,000 miles when the oil gets dirty it burns quicker. I was thinking this is the dino oil just thinning out which it does as it gets to 3000 miles etc. However maybe dirty oil just burns quicker. Maybe he has a filthy engine and the better filters are handling the dirt while the poor ones do not so the oil burns up.
Oh I guess this was your original idea of what was happening.

[ April 06, 2004, 11:23 AM: Message edited by: TR3-2001SE ]

 

[Kevin Hahn]

I might as well chime in since its my test thats being discussed.

Here is the testing procedure its not all that scientific, I preface my test as a subjective oil filter test.

*Every 2,700-2800 miles (first chance after 2,700 miles)
*The warming up of the car mainly with idling but some revving and about 30-60 seconds of 4,000 RPM levels for pulling up onto ramps. engine is always warmed to normal operating temperature.
*Same ramps and same piece of street used for every change
*Oil is drained into an open pan and the filter is drained into the pan for 30 minutes, then a piece of filter paper is dipped in, along with my hand in a tight latex glove and then ungloved hand. From these three samples my opinions on the oils color and other subjective impressions were taken and recorded.
*Penzoil 20w50 is the oil of choice for refill, I would have picked an oil with better temperature range for Colorado but I started the test in the spring and go to the 3rd filter on 20w50 before I thought about changing so I have maintained it.

The filters have been stored but as all my work is done at my parents house and I am at college, the filters have been moved from my storage location. I will search for the fitlers the first chance I get but I fear they have been thrown away in a poor attempt to clean the garage.

I still have my notes from all the tests.


A couple comments in regards to previous replies.

1) The oil dip stick on the 80s 300zxs is more or less useless, it is very difficult to read and hardly ever reads full even after an oil change.

2) While this may seem foolish to you I was not aware that the Penzoil filter was made by Fram therefore the results for the Penzoil fitler were not biased by personal feelings regarding Fram filters.

3) The testing order is: WIX (Napa GOLD), K&N, penzoil, PureOne, Purolator, I will probably retest ST to make all things even and fair as far as my criteria are concerned then test a motorcraft fitler, maybe a baldwin filter, and maybe a mobil 1 filter before I finish.

4) Engine compression was good when I tested it before the test began, I do experience a little bit of blow by on start ups with a cold engine and temperatures below 40 degrees.

5) Driving has remained a constant mix of highway and city miles, although the WIX filter recieved a bit more highway miles then the others.


Its not the most scientific test, I never have tried to make it appear as such, these are my own personal findings.

Say whatever you like, I doubt it will have any impact whatsoever on my results or the decisions I make because of them.

 

[jsharp]

Sounds like you 300ZX has a dipstick similar to my Pathfinder which uses the same basic engine. A semi-rigid "cable" with a piece on the end that allows you to see the level. Sometimes anyway.

Since different filters contain different amounts of media, you oil level will change slightly because of different amounts held in the filter can. It wouldn't make 2 quarts difference though. Unless you check the oil before a change, you'll have no idea how much oil is in the engine before you drain it, and so this "test" is meaningless...

[ April 06, 2004, 02:24 PM: Message edited by: jsharp ]

 

[Thomas Pyrek]

Calling it meaningless is going overboard, IMO. It's not the end all of research, but even you admit that size doesn't hold 2 quarts. Oil capacity is about 4.5 quarts, so some is missing.

Why it is important, IMO, is that it does call for further questioning. How come when less that oil comes out the dirtier (color and grit) it is? I would do this test more thuroughly, but I don't have the resources to test these filters and get oil analysis's of all the oils.

 

[jsharp]

quote:

---

Originally posted by Thomas Pyrek:
Calling it meaningless is going overboard, IMO. It's not the end all of research, but even you admit that size doesn't hold 2 quarts. Oil capacity is about 4.5 quarts, so some is missing.

Why it is important, IMO, is that it does call for further questioning. How come when less that oil comes out the dirtier (color and grit) it is? I would do this test more thuroughly, but I don't have the resources to test these filters and get oil analysis's of all the oils.

---

I agree it needs further investigation, but unless you know exactly how much oil is in the engine right before you drain it, any volume observations are meaningless...

edit: another thing I didn't mention. Even the dipstick reading is suspect. If you have a filter with a non-functional anti-drainback valve the dipstick oil level reading will be off and you still won't know how much oil is in the engine right before you drain it.

[ April 06, 2004, 02:51 PM: Message edited by: jsharp ]

 

[Kevin Hahn]

The problem is the unexplainable corrolation between how the oil appears out of the engien and how much comes out. It doesn't bother me that much, with almost all of the other filters I don't have nearly the loss and its the same type of driving.


Whats curious is that you trust the dipstick, I wouldn't trust the dipstick on any car to accurately indicate the amount of oil in an engine, maybe a ballpark, but never accurate.


Call it meaningless, call it flawed, call it whatever you want, Tom was puzzled and asked you to help him undertand something that I noticed, I don't want you guys to explain any results or even justify my results.


Also why do you feel that you have to know how much oil is in the engine before you drain it? What would that show you, the oil is either there or it isn't, a small variation wouldn't matter, it would be liekly explained by the filter. Unelss you are trying to say that I am not draining the engine I can't see where you could so heavily value the oil level in the engine before the drain when compared to afterwards.

[ April 06, 2004, 05:21 PM: Message edited by: Kevin Hahn ]

 

[TR3-2001SE]

Let us know how the regular Purolator did? The mystery continues.

 

[Kevin Hahn]

I actually want to retest the Purolator (going to be a while, the original retest got messed up due to an airfilter issue), there were no discernable observations to be made between the PureOne and the Purolator. When I looked at my notes I only got 3.9 qts with the Purolator but it was colder because of seasonal changes and I do when its cold out get a little blow by till the motor warms up.

But other then the .2 qt difference there is nothing that stands out that was different results wise then the PureOne which is why I want to retest it to make sure.

I will retest the ST probably also as its on there now but conditions won't be mimicked with the rest of the conditions in full.

Its not scientific but I am trying to keep atleast some things equal.

 

[jsharp]

quote:

---

Originally posted by Kevin Hahn:
The problem is the unexplainable corrolation between how the oil appears out of the engien and how much comes out. It doesn't bother me that much, with almost all of the other filters I don't have nearly the loss and its the same type of driving.


Whats curious is that you trust the dipstick, I wouldn't trust the dipstick on any car to accurately indicate the amount of oil in an engine, maybe a ballpark, but never accurate.


Call it meaningless, call it flawed, call it whatever you want, Tom was puzzled and asked you to help him undertand something that I noticed, I don't want you guys to explain any results or even justify my results.


Also why do you feel that you have to know how much oil is in the engine before you drain it? What would that show you, the oil is either there or it isn't, a small variation wouldn't matter, it would be liekly explained by the filter. Unelss you are trying to say that I am not draining the engine I can't see where you could so heavily value the oil level in the engine before the drain when compared to afterwards.

---

Unless I read this wrong, it seems simple to me but maybe I didn't explain my thinking well enough. You add about 4.5 quarts to the engine, then drive it a few thousand miles. You then drain it and look at how much oil drained out. But you don't know how much oil was used along the way.

Since you really don't know how much oil was left in the engine right before you drained it, you have no way to determine how much *should* drain out.

I can see a .2 quart difference possibly from filters, but a 2 quart difference seems very strange. The missing 2 quarts has to be somewhere. There are only 2 choices - it's still in the engine, at which point it would probably drain out the next time you drained it, or the engine used it in operation and it will never reappear.

I'm one of those guys that actually trust the dipstick, just like I trust the sightguage on a lot of motorcycle engines, and the overflow fitting on lawn and garden equipment. How else would you know if the engine is over/under filled...

[ April 06, 2004, 11:45 PM: Message edited by: jsharp ]

 

[javacontour]

Maybe I missed it, but do you have the dates each filter was in service.

You know, some in the summer, some in the winter.

I believe to eliminate weather, you would have to test each filter for a year and also indicate how many miles you drove in that year.

Perhaps summer driving causes more oil consumption. Or if you have cold winters, then perhaps extended idling causes you oil to break down faster and it is consumed.

There are so many variables.

However, I do believe I understand the question. You are trying to determine if there is a link between how well a filter filters and how much oil is consumed. Your hypothesis is that poor filtering leads to higher oil consumption.

I think you are on the right track, but need more than one data point for each filter.

TB