Synthetic Oil Life Study
Paradise Garage
http://oilstudy.spacebears.com




We recommend Internet Explorer set to 1024x768.

© 2002-2005 Brian F. Schreurs
Even we have a disclaimer.

Truth costs money.

 

 

 

Check out our book:
The Hot Rodder's Quick Reference

 

 

  Come visit us in West Virginia:

Introductory Remarks

Since 2002, this website has been conducting a long-term study on the longevity of synthetic motor oil.


Down the hatch! Our LS1 gets its Mobil 1.
We accept that for most people in most cars, regular old "dinosaur juice" mineral-based motor oil will work just fine. We have seen plenty of cars run beyond 150,000 miles and 200,000 miles using the stuff as long as the oil is changed regularly. However, not everyone is content with "good enough". Some want extra protection for their engines because they drive hard, or live in the city, or go racing, or because they have a crazy expensive engine. Others consider 150,000 miles to be broken-in, and seek to double or triple that vehicle service life. We hear from owners of fleet vehicles where saving a few oil changes per year on an entire fleet of vehicles equates to thousands of dollars saved. And we hear from those concerned about reducing the environmental impact of used motor oil by reducing the number of oil changes for all our cars.

When you have a product that appeals equally to tightwads, racers, industrialists, and environmentalists, it's worth a closer look, no?

But then there's a lot of trouble getting any useful information about synthetic motor oil. Which is best? Which is good enough? How much protection do they really offer? How long can you really use them? The answers are generally shrouded in marketing-speak, buried under superlatives like "proven technology" or "industry leader", accompanied by bar graphs of results from esoteric laboratory tests that seem to have little to do with real-world driving. Indeed, they mostly seem to promote the quality of the oil when new, a condition that lasts for about the length of time it takes for it to go from the mouth of the bottle to the end of the funnel.

Those curious about the finer details of motor oil have found the available material wanting. After some bickering in a Jaguar community forum about motor oil, where most of the available arguments started with "My pappy always..." (no doubt other enthusiasts are familiar with this line of logic), this website launched the Synthetic Oil Life Study, on the premise that we would continue to test as long as readers continued to help fund. The readers certainly have held up their end of the deal. We hope that our end proves satisfactory.

It is our hope that this study eventually answers these questions:

  1. What happens to synthetic oil after it leaves the laboratory tests and gets poured into an actual engine out on real roads?
  2. In general terms, is synthetic oil capable of living up to the hype?
  3. Which one is best? (Or, Which one is best for a given application?)


Sidebar
Laboratory
Comparison

We started this study with Mobil 1 because it was what we were already using. Then we moved on to Amsoil, the independent brand most critical of Mobil 1. At present we are re-testing Mobil 1 to see what effect break-in may have had on our prior results. Once this phase is done, we will test Red Line. After that, Royal Purple, or perhaps Castrol GTX (to see how a quality mineral oil stacks up).

At the beginning of the study, we went to some effort to select an independent oil analysis firm. After examining the credentials, background, and reputation of several companies, we have selected Blackstone Laboratories as our analysts. The selection has proven fortuitous, as their service has been excellent. They are not providing us any consideration or service that you couldn't get just the same as us. We're even paying for the tests just like anybody else would.


Sample Test Honor Roll
F1000James Ashbaucher
F3000James Ashbaucher
0James Ashbaucher
1,000Peter Childers
2,000Ted Hoefer
3,000Ted Hoefer
4,000"Tochatihu"
5,000Kenneth Thompson
6,000Mark Koch
7,000Larry Shute
8,000Henry Gonshor
9,000Peter Childers
10,000John Hofstetter
11,000Alexander Komarov
12,000George Mertz
13,000"Richsaab"
14,000"Richsaab"
15,000Mr. K.K.
16,000Mr. K.K.
17,000Tom Johnson
18,000Lee "Stryder" Bussy
19,000Steve Williams
20,000David Kira
21,000Daniel Kirsch
22,000Jeff Blundell
23,000Keith Kochman
24,000Von Odenwald
25,000Robert Miller
26,000Robert Miller
27,000Robert Miller
28,000"RTH"
29,000"RTH"
30,000Glenn Schroeder
31,000Louis Kwan
32,000Ralph Coleman
33,000Roy O'Kelly
34,000James Ashbaucher
35,000Karl Knab
36,000Karl Knab
37,000Karl Knab
38,000Karl Knab
39,000Karl Knab
40,000A.E. Mouse
41,000You? Please?
Mobil 1 Purchase Honor Roll
Bryan Savage
Bryan Savage
Bryan Savage


Blatant Plea for Help

Speaking of paying for things, each analysis costs us 20 bucks. While we like to provide technical info for the public good,this particular project costs us a lot of money, essentially negating any economic benefit we will ever see from our own results. You readers, however, get all the benefit of this study at no cost.

If you find this article informative, if you think it might save you a lot of money down the road (and we hope it does!), then we ask that you consider paying for one of the oil samples. It's $20.00, a low cost if each person just buys one. We'll put your name next to one of the analysis results so everyone will know what a nice person you are. Write us if you're willing to help defray the costs of this study. We're PayPal-friendly.

Links to this page are also much appreciated.

At this point it's safe to say the Mobil 1 re-test phase of the study is completely paid for. We're still grateful for donations, don't get us wrong, but we're getting into the Red Line test by now. We still appreciate the support, but please understand that "your" test results may not be posted for a long time.

Now that the Mobil 1 and Amsoil phases of the study are over, the primary author is available for online Q&A sessions in your discussion forums. If you're interested, write us for details. We see you guys discussing this study, so c'mon, why not ask questions directly instead of just guessing!

Also, if you notice that we are sometimes running late in posting new information, it's because we're busily trying to operate our start-up business, West Virginia Riders. Please cut us a little slack. New businesses are time-consuming, but we'll post the latest oil study data just as fast as we can.


Test Procedures


The test car. We can live with it.
The test vehicle is a 2002 Chevrolet Camaro Z28. It has an aluminum 5.7L V8 (dyno tested at 307 hp) and a six-speed manual transmission. Its miles are split nearly 50/50 between city and highway, about 60 to 80 miles a day in spurts of 15 to 30 miles per trip. It tends to get around 20 mpg over the course of a month.

Its regimen prior to this study was Mobil 1 5W30 oil and a NAPA Gold filter every 5,000 miles. This study began at the 10,000-mile mark; at that point, we filled it again with Mobil 1 5W30 and installed a new NAPA Gold filter. We sent a sample of new oil as a baseline. Then, we've been sampling the oil every 1,000 miles to clearly establish trends. We change the filter whenever the insolubles level reaches 0.5% (hey, we want to keep this test as objective as possible, but not at the

Sidebar
Questions &
Answers

cost of our engine!). The oil stays in the engine until analysis shows it is no longer safe to use.

Keeping the crankcase filled to the right level can definitely be an issue with change intervals far beyond the standard 3,000 miles. All cars naturally consume a small amount of oil; for this test, we also lose oil to the analysis samples and to filter changes. The samples and filters are a static quantity so they will be equal across all brands of synthetic; however, if one type of synthetic is more prone to consumption than another, it might appear to have an advantage from more frequent refreshing. To mitigate this, we will return the oil level to the full mark immediately following each sample extraction. We will also record the amount of oil required to top it off and provide this information along with the test results. If a synthetic oil has a "long life" because it loses an inordinate amount and must be topped off frequently, that's important to know as it significantly alters the cost and longevity of the oil.

If any brand of oil should make it to a second filter change -- which so far has never happened -- then it will naturally get an extra boost from the post-filter top-off. However, this is not a concern because it reflects the real-world economics of using

Sidebar
Oil Change
Intervals

that particular brand of oil. We will, of course, report all filter changes.

To prevent a previous test oil from contaminating a new test oil, we run a 3,000 mile buffer of the new oil between tests. During the buffer interval we put in fresh spark plugs. After 3,000 miles, we drain the crankcase and refill with fresh oil, change the oil and air filters, then resume the study.

Also, just for grins, we reset the car's oil change indicator to see how GM's engineers did when they calibrated the oil life computer.


Sidebar
Margins of Error

Oil Analysis

The oil is analyzed for many different things. The most critical characteristics for the purposes of oil longevity are:

  • Viscosity: This is the SUS viscosity at 210 degrees. It is the standard SAE viscosity measure; SUS is the unit of viscosity. This value should remain within its grade, or plus or minus 20%, whichever is greater.
  • TBN: The Total Base Number. This is a measure of acid-combatting additives. Oil is essentially expired once it reaches a TBN around 1.
  • Insolubles: This is the percentage of the test sample consisting of solids. Solids are always bad; the value should be less than 0.6%.
  • Wear Metals: The presence of wear metals is a contaminant and the rate of accumulation is an indicator of the oil's protective condition. The exact values representing oil change time is rather subjective; we use Blackstone's conservative guidance in conjunction with other industry experts. Iron, copper, and lead are normally the ones to watch.


Think your corner station mechanic is gonna do this?
The oil analysis also checks for fuel, antifreeze, and water in the oil. Of course, ideally all of these values should be zero, but realistically the tolerable limits are: fuel, 2%; antifreeze, 0%; and water, 0.05%.

Also, there's the analysis of elements expressed in parts per million. Some of these elements are indicators of engine wear, while others are additives in the oil. For any of these, the presence or absence of any element is less important than how the values change over time.

  • Engine Wear: aluminum, chromium, iron, copper, lead, tin, molybdenum, nickel, manganese, silver, titanium, posassium, and silicon.
  • Anti-Wear Additive: molybdenum, phosphorus, and zinc.
  • Antifreeze Inhibitor: potassium, boron, silicon, and sodium.
  • Detergent Additive: boron, calcium, magnesium, and barium.

Note that molybdenum and silicon can be both an additive and an indicator of engine wear.


Past Results

To date we have completed our tests of Mobil 1 and Amsoil. Mobil 1 held on for 18,000 miles, and didn't need its first oil filter until 12,000 miles. For all the details, visit our Mobil 1 Test Results page. Amsoil was time-limited to 14,000 miles (its year ran out), but it made it the whole way without a replacement filter. It's not a total success story, though, as it had some trouble with its viscosity. To read all about it, visit our Amsoil Test Results page.


Mobil 1 5W30 SuperSyn

At present we are re-testing Mobil 1 to determine what effect the engine's break-in period may have had on our previous test results. The first two samples, F1000 and F3000, are from the flush period.

Oil Miles

F1000

F3000

0

1000

2000

3000

4000

5000

6000

Aluminum

1

2

0

1

 

 

 

 

 

Chromium

0

1

0

0

 

 

 

 

 

Iron

5

9

1

4

 

 

 

 

 

Copper

14

30

0

10

 

 

 

 

 

Lead

2

4

0

1

 

 

 

 

 

Tin

0

1

0

0

 

 

 

 

 

Molybdenum

53

58

78

63

 

 

 

 

 

Nickel

0

0

0

0

 

 

 

 

 

Manganese

0

0

0

0

 

 

 

 

 

Silver

0

0

0

0

 

 

 

 

 

Titanium

0

0

0

0

 

 

 

 

 

Potassium

0

0

0

0

 

 

 

 

 

Boron

122

130

59

53

 

 

 

 

 

Silicon

8

10

4

4

 

 

 

 

 

Sodium

5

5

4

3

 

 

 

 

 

Calcium

2318

2374

2507

2248

 

 

 

 

 

Magnesium

63

88

11

23

 

 

 

 

 

Phosphorus

640

733

599

651

 

 

 

 

 

Zinc

732

883

710

808

 

 

 

 

 

Barium

0

0

0

0

 

 

 

 

 

Viscosity

60.2

61.3

65.1

61.0

 

 

 

 

 

Flashpoint

380

415

425

405

 

 

 

 

 

Fuel

<0.5%

<0.5%

0%

<0.5%

 

 

 

 

 

Antifreeze

0%

0%

0%

0%

 

 

 

 

 

Water

0%

0%

0%

0%

 

 

 

 

 

Insolubles

0.3%

0.3%

0%

0.2%

 

 

 

 

 

TBN

8.3

4.9

9.6

7.8

 

 

 

 

 

Mile F1000 -- James Ashbaucher, December 27, 2004.
Oil/Vehicle miles: 1,030 / 46,000
Oil added after sample: none

There remains a lot of debate about the cause of the initial surge in wear metals in our samples. To help learn something about that, we've taken a sample at the 1,000-mile mark in our flush oil, and we plan on taking a sample again when we drain the flush oil at 3,000 miles. Early on, though, the level of wear metals looks promising -- just a fraction of what we saw last time we tested Mobil 1, even with nearly identical viscosity.

Mile F3000 -- James Ashbaucher, January 28, 2005.
Oil/Vehicle miles: 2,995 / 47,995
Oil added after sample: none

How about that -- in the ensuing mileage, the wear metals doubled! We saw the same thing with copper in the Amsoil test, but to have it double across the board is a first. Of course, it's worth noting that the absolute values remain miniscule -- among the lowest ever recorded in the course of the study.

Notice how each time we change the oil, there is still an initial spike of wear metals, but the spike is smaller at each phase. We are working on a hypothesis: that the true break-in of an engine is at least 20,000 miles, and that, perhaps, during the first year of an engine's use it would be wise to change more frequently, and only start moving toward extended oil changes once the engine is older. It'll be interesting to see what comes up in the Mobil 1 re-test phase.

Mile 0 -- James Ashbaucher, January 28, 2005.
Oil/Vehicle miles: 0 / 47,995
Oil added after sample: none

This sample is our virgin for the newest version of Mobil 1 SuperSyn 5W30. Look at that viscosity! Crazy. Hopefully this is just a fluke from the bottle we happened to select for testing.

Looks like they've tweaked the formula a bit since last time: boron is about a fourth of its previous value, and phosphorus and zinc are down about 25%. Nothing seems to have risen appreciably, though, which means they're putting less "stuff" in their oil -- or they're using something we're not testing.

This should make for an interesting re-test!

Mile 1000 -- Peter Childers, February 21, 2005.
Oil/Vehicle miles: 993 / 48,988
Oil added after sample: none

Here we are at our first sample in the Mobil 1 re-test. Nothing unusual so far. It sure is nice to have a smooth-running engine again! The extra vibration from the over-thick Amsoil was getting annoying.

Note how the wear particles are lower right out of the gate from even just a couple thousand miles ago. We are beginning to like the sound of our hypothesis. But the next sample will really tell the tale. Do the particulates double again? It's also nice to see the viscosity back down to a more reasonable level.


Well folks, the dread day has finally come. We all knew this couldn't go on forever. Here is the cold reality: it is now May 24, and since February 21 we have only put 954 miles on the test car. Even worse, we are about to embark on a cross-country journey by motorcycle, so we won't get that last 46 miles until... hmmm... early July probably. This lack of mileage is due to many factors, not really relevant here, but it is obvious that six-month sample intervals are in no way comparable to the three or four week intervals we used to see. It is unfair to pretend that the data we collect now is still valid under the terms of the study.

We thank everyone who has helped us over the last two years. We have another study in the works that is not so mileage-dependent, and we hope to release the details in July. Though we stopped accepting donations months ago, those that are still outstanding will be rolled into this new study.

Over the next month, while we are away, we would greatly appreciate hearing your take on the information we've gathered so far. In particular we would love to see some essay-length pieces (over 1,000 words) that we might include with a future update on this study, to wrap it up and provide some analysis. Some of you are very knowledgeable (you know who you are!) so here's your chance to say what you think needs to be said. In July, we'll update the study pages to fill them with the collected wisdom and insights. You can use the e-mail address on the Cave Drawings page.

Thanks to everyone, and especially Miro.