Mobil 1 Test Results
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.

Mobil 1 5W30 Super-Syn Formula


Down the hatch! Our LS1 gets its Mobil 1.
Mobil 1 currently enjoys the distinction of being the only readily available true PAO synthetic motor oil. You can get the stuff at Wal-Mart for Pete's sake! Perhaps because of its wide availability and comparatively low price, we entered this test with both hope and skepticism: hope that it would do well and vindicate millions of M1 users, and skepticism that it would pull it off. We were expecting it to look pretty beat up by 8,000 miles or so; to say we missed the mark would be polite understatement. Mobil 1 really delivered: 18,000 miles later, it was showing its age but was still hard at work protecting the engine.


Mobil 1 Honor Roll
0Brian Schreurs
1,000Gregory Wells
2,000Joshua Scott
3,000Charlie Nowlin
4,000Andrea Holladay
5,000Jay Rasberry
6,000"Patman"
7,000Sunny Garofalo
8,000"Orange 01SS"
9,000Tommy Lawson
10,000Timothy Bundrick
11,000Paul Seminara
12,000Neil Womack
13,000"Jerry"
14,000"Joee12"
15,000Gary Heidebrecht
16,000George Michaels
17,000Stephen Scheifler
18,000John Richardson


As our first phase of the Synthetic Oil Life Study, it was as much a learning experience for the process as it was for the results. We changed some things along the way -- moved the filter change from a mileage-based rule to a performance-based rule, altered the tolerable limits for viscosity, learned a lot about TBN, and a few other things -- which set precedents for the remainder of the study.

But the big question is, what have we learned so far? Here are a few points to ponder, based on our experience with the Mobil 1 phase.

  • Getting just one oil analysis only tells a tiny piece of the picture. It essentially would serve only as a pass/fail mechanism; without a trend to monitor, the most interesting parts of the analysis would be impossible to see.
  • Total base number is a moving target. There are multiple methods for testing it, which makes comparisons between laboratories worthless, and none of the methods have repeatability rates worth getting excited about. While TBN is worth considering as part of the larger picture, as a singular measure it is too flawed to rely on.
  • Engine wear actually decreases as oil ages. This has also been substantiated in testing conducted by Ford Motor Co. and ConocoPhillips, and reported in SAE Technical Paper 2003-01-3119. What this means is that compulsive oil changers are actually causing more engine wear than the people who let their engine's oil get some age on it.
  • Topping up the crankcase is a critical component of extended oil change intervals, and frequent filter changes are most likely the key to extreme-length intervals. The cumulative effect of even minor top-ups, let alone a filter change, substantially increases the longevity of the oil.

Based on the results we've got here, we'd recommend 8,000 miles between oil changes on an engine that uses no oil at all, perhaps 10,000 miles on an engine that uses some oil, and 15,000 miles or beyond with a filter change every 5,000 miles. This, of course, isn't any kind of guarantee, and you must evaluate for yourself what your engine requires. One thing we're pretty sure about though: 3,000-mile intervals is a huge waste of resources.

This is a long page with a lot of information. Here are a few bookmarks to help you navigate:

Test Sample Results
The Effect of Top-Up Oil
Interpreting Wear Metals


Test Sample Results


CLICK THE GRAPHIC FOR A FULL-SIZE VIEW

Oil Miles

0

0*

1000

2000

3000

4000

5000

6000

7000

8000

Aluminum

0

0

3

3

4

5

5

6

5

5

Chromium

0

0

1

1

1

2

2

2

2

2

Iron

2

2

10

12

14

19

23

26

27

35

Copper

0

0

58

75

81

99

102

105

101

111

Lead

0

0

5

4

4

7

8

11

12

13

Tin

0

0

4

5

4

2

5

4

3

5

Molybdenum

68

69

69

74

60

65

72

75

74

75

Nickel

0

0

0

0

0

1

1

1

1

1

Manganese

0

0

0

1

1

1

1

1

1

2

Silver

0

0

0

0

0

0

0

0

0

0

Titanium

0

0

0

0

0

0

0

0

0

0

Potassium

0

0

0

9

0

0

2

1

2

3

Boron

120

202

172

178

139

124

150

155

155

140

Silicon

4

4

9

9

9

9

10

11

11

12

Sodium

7

7

8

8

7

8

9

9

8

10

Calcium

2649

3014

2780

2812

2599

2429

2709

2821

2930

2988

Magnesium

18

25

73

69

66

71

66

68

61

64

Phosphorus

737

817

763

800

694

699

742

773

778

764

Zinc

819

921

895

1011

815

815

890

899

922

985

Barium

0

0

0

0

0

0

0

0

0

0

Viscosity

60.3

59.7

59.6

58.9

60.1

59.3

58.5

59.9

60.2

61.4

Flashpoint

430

435

390

380

390

395

415

415

420

430

Fuel

<0.5%

0%

3.0%

<0.5%

<0.5%

<0.5%

<0.5%

<0.5%

<0.5%

<0.5%

Antifreeze

0%

0%

0%

0%

0%

0%

0%

0%

0%

0%

Water

0%

0%

0%

0%

0%

0%

0%

0%

0%

0%

Insolubles

Trace

0%

0.3%

0.3%

0.3%

0.3%

0.3%

0.3%

0.4%

0.4%

TBN (old)

15.5

14.5

16.5

11.5

8.5

6.0

6.0

6.5

6.0

5.5

TBN (new)

 

11.8

 

 

 

 

 

 

 

 

 

Oil Miles

9000

10000

11000

12000

13000

14000

15000

16000

17000

18000

Aluminum

7

6

6

7

6

6

5

6

6

5

Chromium

2

2

3

2

2

2

2

2

2

2

Iron

38

36

42

41

33

37

36

36

48

47

Copper

124

110

127

124

106

110

105

108

114

102

Lead

16

17

22

26

20

25

25

26

31

31

Tin

4

4

3

3

4

5

7

6

3

1

Molybdenum

83

78

76

74

75

83

84

81

74

72

Nickel

1

1

1

1

1

1

1

1

1

1

Manganese

2

2

2

2

2

2

2

2

2

2

Silver

0

0

0

0

0

0

0

0

0

0

Titanium

0

0

0

0

0

0

0

0

0

0

Potassium

1

0

3

1

3

2

1

0

1

1

Boron

150

127

138

117

124

118

116

117

103

101

Silicon

13

13

15

15

13

13

13

13

14

12

Sodium

10

10

11

10

10

10

9

10

9

9

Calcium

2579

2906

3643

3103

3002

3097

3089

3084

3142

3239

Magnesium

68

57

61

58

49

51

44

46

45

39

Phosphorus

807

783

820

796

781

842

857

833

748

733

Zinc

993

955

1058

915

905

959

969

958

902

882

Barium

0

0

0

0

0

0

0

0

0

0

Viscosity

62.4

62.1

63.3

60.7

62.6

64.0

63.8

67.8

65.8

64.1

Flashpoint

415

400

380

420

410

390

SHORT

375

380

390

Fuel

<0.5%

<0.5%

<0.5%

<0.5%

<0.5%

<0.5%

<0.5%

<0.5%

<0.5%

Antifreeze

0%

0%

0%

0%

0%

0%

0%

0%

0%

0%

Water

0%

0%

0%

0%

0%

0%

0%

0%

0%

0%

Insolubles

0.3%

0.3%

0.5%

0.4%

0.4%

0.4%

0.4%

0.4%

0.5%

0.1%

TBN (old)

7.0

7.5

5.0

 

7.5

3.0

5.0

5.0

4.0

6.5

TBN (new)

 

 

 

3.2

2.5

2.5

2.4

2.4

1.9

3.3

Mile 0 -- Brian Schreurs, November 9, 2002.
Oil/Vehicle miles: 0 / 9,939
Oil added after sample: none

The baseline run (mile 0) indicates the composition of the additive package in Mobil 1 Super-Syn Formula, as well as the physical characteristics of the oil before spending time in an engine. This baseline will allow us to track the degradation in oil quality over the miles. Some people have suggested that the TBN is way too high and may be a testing error. However, the long-term TBN trend seems to indicate otherwise.

* Blackstone was kind enough to conduct a second virgin oil analysis because of the new TBN test method. We're glad they did a full test though, because in addition to seeing the relative TBN numbers it's also interesting to note that Mobil seems to have tweaked their formula a bit. Check out the significantly higher levels of boron and calcium, and marginally higher phosphorous and zinc. Now, some of this could just be production variances, but the fact that it's up across the board suggests to us, at least, a more robust package. It's too late to do anything about this for the current test, but the Amsoil test will not face the same problem: we bought all 18 quarts as one lot.

Mile 1000 -- Gregory Wells, December 3, 2002.
Oil/Vehicle miles: 1,070 / 11,009
Oil added after sample: none

So far the oil is showing virtually no appreciable degradation. Both viscosity and flashpoint are down somewhat, probably due to the fuel contamination. Blackstone Laboratories let us know that fuel contamination is typical for new engines, though we suspect we may not have allowed enough oil to drain before taking the sample; we'll improve the technique for next time. Copper is rather high but the lab says that is typical of the LS1 engine up to as far as 30,000 miles. The TBN is actually up slightly from the baseline, which is highly unusual. However, bear in mind that the crankcase oil is an average of roughly six quarts of oil, whereas the virgin sample was from just one quart; evidently, the one we sampled was on the low side of the average. All other numbers seem quite reasonable.

Mile 2000 -- Joshua Scott, December 12, 2002.
Oil/Vehicle miles: 2,044 / 11,983
Oil added after sample: none

We put a lot of extra miles on the car due to a trip to Washington, but it was still a healthy mix of city and highway. Copper is still high, which concerns Blackstone Laboratories somewhat. The fuel issue has gone away (we did a better job sampling this time); viscosity and flashpoint are down slightly while TBN has dropped significantly.

Mile 3000 -- Charlie Nowlin, December 31, 2002.
Oil/Vehicle miles: 3,082 / 13,021

Sidebar
LS1 Engine
Copper Levels

Oil added after sample: 1/2 quart

For the first time, we had to top off the oil. The oil is still holding up quite well, with the TBN far higher at 3,000 miles than standard oils even start out with. The lab is concerned with the copper levels and is recommending that we change the oil. However, our own research indicates that this level of copper is normal -- or at least acceptable -- for this engine. Therefore we are going to continue with the study. Anyone who has an LS1-powered vehicle with an oil analysis conducted on it, we would really like to know what your copper level is.


I think we can all agree that this here... this is nuts.
Mile 4000 -- Andrea Holladay, January 21, 2003.
Oil/Vehicle miles: 4,036 / 13,975
Oil added after sample: 1/2 quart

We retrieved this sample in the snow. It wasn't as fun as it sounds. Thanks to everyone who offered garage space for the next sample. For this sample, we again had to top off the oil, suggesting that somehow the engine is using more of it as the oil ages. Copper is still increasing, but not at a frightening pace -- it should look pretty normal after we flush the system. Viscosity and flashpoint are still commendably stable, but TBN is dropping like a rock. Of course, we could be surprised -- the TBN could level off. We'll see.

Mile 5000 -- Jay Rasberry, February 10, 2003.
Oil/Vehicle miles: 5,008 / 14,947
Oil added after sample: 1/2 quart

The Camaro seems to fairly consistently use half a quart of oil every 1,000 miles, typical but not terribly impressive. Major thanks to Ryan and his family, who generously provided garage space to pull this sample indoors. Interestingly, the oil indicator activated at 14,999 miles, just 52 miles after we drew this sample, which would seem to be extremely premature. TBN did in fact level off -- perfectly level, no degradation at all. Before you start typing those e-mails inquiring how this is possible, we'll give you our definitive answer: we have no idea. All other critical measurements remain in excellent shape, and even the copper is slowing down.

Mile 6000 -- "Patman", March 6, 2003.
Oil/Vehicle miles: 6,010 / 15,949
Oil added after sample: 1/2 quart

The TBN has again not dropped; indeed, it actually went up slightly. If it keeps this up, it'll last forever. We're doing a little research to learn how TBN works, as it certainly hasn't behaved the way we would expect. Meanwhile, copper seems to have leveled off, so whatever was slowly dying in the engine must be dead by now. Since all wear levels remain good, and the filter is still hard at work, we'll give it another thousand and see what happens!

Mile 7000 -- Sunny Garofalo, March 22, 2003.
Oil/Vehicle miles: 6,994 / 16,933
Oil added after sample: none

Much to our surprise, the oil level was only marginally below the full mark, so we didn't bother with a top-up. Nothing looked particularly unusual with this sample when we took it, and nothing particularly unusual showed up in the analysis results. Notice how copper is now completely flat; it'll be interesting to see what happens when we start over with Amsoil. Confused about TBN behavior? So were we, so we gathered a bunch of SAE reports to find out what we're missing. More to come on this fascinating subject. We'll probably have to change the filter soon, but otherwise this oil is holding up fine.

Mile 8000 -- "Orange 01SS", April 4, 2003.
Oil/Vehicle miles: 8,011 / 17,950
Oil added after sample: 1/2 quart

Here we go again -- the drop in oil consumption unfortunately proved to be nothing but a pause. Ah well, if it keeps using 1/4 quart per 1,000 miles, that's mostly from the samples anyway. The viscosity has crept upward close to the maximum allowable for this oil (62 SUS) and TBN is finally slipping a little, so that bears watching. Iron also spiked; Blackstone noted it but isn't worried yet. Is the end near? We've been fooled before, but this oil is starting to show some age.

Mile 9000 -- Tommy Lawson, April 24, 2003.
Oil/Vehicle miles: 9,011 / 18,950
Oil added after sample: 1/2 quart

Pretty freaky that we managed to hit exactly 1,000 miles eh? Man we're good. Visually the oil is pretty skanky at this point. If we were going on appearance alone we would have changed it a long time ago. Analysis shows this oil is reaching its limit; iron, lead, and copper are up again, and viscosity has thickened out of range for this specific oil (but not out of grade yet). That crazy TBN continues to behave unpredictably. We'll probably drain the oil soon.

Mile 10,000 -- Timothy Bundrick, May 8, 2003.
Oil/Vehicle miles: 10,089 / 20,028
Oil added after sample: none

Wow, here we are at 10,000 miles (or slightly more, due to the misfortunes of bad timing). At this point we hesitate to make predictions about the longevity of the oil, as we are constantly being surprised. We're as shocked as almost anyone else that at 10,000 miles the oil is still good, the filter is still good, everything really looks just fine in there. 3,000-mile oil changes? Puh-leeze. Never again for us. But just how long will it last?

Mile 11,000 -- Paul Seminara, May 24, 2003.
Oil/Vehicle miles: 11,002 / 20,941
Oil added after sample: 1/2 quart

Another sample down the drain. Ar ar ar. We know, we've said before that the oil is about done, but this time we're really getting antsy. Iron, copper and lead are all up noticeably -- particularly lead. Viscosity is creeping up. Insolubles are near their upper limit. TBN is down 2.5 points, a rather sharp drop, no doubt in part because this sample didn't benefit from any make-up oil but clearly showing that the equilibrium in this engine is in a very delicate state. Blackstone wants us to change. Overall, quite a shocker considering how good everything looked at the last sample.

Mile 12,000 -- Neil Womack, June 12, 2003.
Oil/Vehicle miles: 12,025 / 21,964
Oil added after sample: 1.5 quarts

Thanks to advice from Terry Dyson, we elected to carry on for a bit longer, but with a filter change. So, this sample officially ends the no-filter-change portion of the Mobil 1 test. Used to be that adding half a quart of top-up tended to prop the oil up for a bit longer, but no more -- or at least, if this oil is propped, then we shudder to think what it would have looked like without the top-up. The next sample will show us what a new filter and a large top-up is capable of doing for the oil, but on its own the oil here is definitely past its equilibrium point. In fairness it must be noted that the laboratory changed TBN test methods so this number is not directly comparable to previous numbers. We're going to double-test some samples to determine the conversion between the two methods.

Mile 13,000 -- "Jerry", July 3, 2003.
Oil/Vehicle miles: 13,030 / 22,969
Oil added after sample: none

After the larger-than-normal top-up, we expected the oil to have some new life in it. That certainly proved true, but it ended up being a curious mix. Many of the wear metals are down quite a bit, perhaps a result of changing the filter. If it is, it's the only significant result of doing so; insolubles were unaffected. The filter might have been able to keep going. Adding a quart and a half of make-up oil sure propped up the TBN. Even so, these new TBN numbers will take some getting used to. Overall, the filter change seems to have "bought" about 5,000 miles of wear back -- compare this sample to the 8,000-mile sample to see what we mean. Looks like we're going to at least 15,000 miles now.

Mile 14,000 -- "Joee12", July 16, 2003.
Oil/Vehicle miles: 13,976 / 23,915
Oil added after sample: 1/2 quart

We came in a little early on this one, but it was either that or come in a couple hundred miles late. We're putting an unusually high amount of highway miles on the car right now, which will accelerate the pace of this and the next sample. The oil still looks decent to our eyes, with wear metals continuing to increase but not at an alarming rate. Viscosity is finally thickening up but it's still a ways off from being out of grade. The difference in readings between the two TBN methods is troubling -- by the old method, it has really tanked; by the new method, it's steady. We keep expecting this oil to expire, but it keeps surprising us.

Mile 15,000 -- Gary Heidebrecht, August 4, 2003.
Oil/Vehicle miles: 14,986 / 24,925
Oil added after sample: 1/2 quart

We had a bit of a scare a few miles before pulling this sample. Sitting at a traffic light, oil smoke started billowing out from under the hood in all directions. "Holy crap we went too far!" was our immediate thought. Fortunately, it turned out to be the mosquito-killing Chrysler minivan in front of us. The sample itself had nothing unusual about it; thanks to Jon Satterfield's donation, we have a few TBN kits of our own now and gave one a try with this sample in addition to Blackstone's readings. We got a 3.5, which is nothing like Blackstone's 5.0, but it should be noted that they've run thousands of these things and we've run... two. One thing that's rather disappointing is the seeming total lack of correlation between the old method and the new. TBN drama aside, the oil soldiers on, barely harmed by this interval. The flashpoint and fuel readings means that they ran out of oil and couldn't test it.

On several internet forums discussing this study, there have been calls for us to stop now and move on to Amsoil. While we heartily sympathize with the impatience to give Mobil 1's arch rival a run, it is worth noting that at this juncture we have 15,000 miles and 10 months invested in this sample. We will never again have a chance to look at this data. It would be a shame to pull the plug now; please be patient.

Mile 16,000 -- George Michaels, August 23, 2003.
Oil/Vehicle miles: 16,026 / 25,965
Oil added after sample: none

Here we are at another sample. With the way Mobil 1 had been going, we instituted a maximum test run of one year. That's only a couple of months away, but it looks like Mobil 1 won't make it after all. The oil has thickened up in a serious way; 30-weight oil tops off with a viscosity of 12.49 cSt, and the 67.8 SUS of this sample converts to 12.36 cSt. The slightest increase in viscosity at the next interval will push it out of range and out of the study.

It's kind of odd, though, that otherwise the oil hasn't aged a bit. Perhaps there's something to all that balderdash about less wear in the summer.

Mile 17,000 -- Stephen Scheifler, September 17, 2003.
Oil/Vehicle miles: 17,031 / 26,970
Oil added after sample: 1/2 quart

Oil pressure is noticeably higher than it was before the viscosity went nuts on us, so we're looking forward to finishing this test and moving on. We've decided against using viscosity as a test-ending value, but we'll certainly flag the oil when it creeps out of grade. The viscosity has slipped back down a little, keeping it safely in grade, but the TBN sure has taken a dive. Also note the sudden increase in wear numbers, especially iron and lead. As of this writing we're waffling on whether to continue. Another wrinkle: the insolubles have again reached the trigger point for an oil filter change. Is it worth it, or is it time to put this to bed? Hm.

Mile 18,000 -- John Richardson, September 29, 2003.
Oil/Vehicle miles: 18,021 / 27,960
Oil added after sample: drained oil!

Well, it's done. At long last, we drained the oil. Considering that we've gone two and a half times longer than we thought we'd go, we're pretty impressed by Mobil 1's longevity. Our Dexsil TBN test resulted in a 6.0, comparing favorably to Blackstone's 6.5. But speaking of high TBNs, what happened to this oil? If it had posted these numbers at the last interval, we wouldn't have dumped it. This oil could have kept going based on the numbers we have now, but they sure weren't looking this good a thousand miles ago. Viscosity was dropping, TBN rising, and what's going on with the insolubles? Meanwhile, despite all these paper improvements, the engine still had higher oil pressure and wasn't running as smoothly as normal, so clearly not all was perfect with the oil.

Thanks to the help of reader Doug Hillary, we had the final sample analyzed by Mobil 1's laboratory. Their results matched Blackstone's analysis pretty well; we certainly didn't receive any surprises.

Honestly, this first phase of the study has probably raised as many questions as it has answered.

Presently, the Z28 is enjoying a relaxing flush fill of Amsoil "ASL" 5W30 synthetic oil, the same oil we will be using (with a refill of course) for the next phase of the study. We're going to begin the Amsoil test when the odometer turns 31,000 miles, which we currently anticipate to be early November.


The Effect of Top-Up Oil

The Mobil 1 oil in this test indisputably ran for 18,000 miles; 12,000 miles without a filter change.


Sidebar
Corrected Oil
Age Table

Some would point out, though, that the make-up oil along the way skewed the numbers. This is a valid point, and the effect is cumulative across the test, especially after the filter change. Fortunately, we know the quantity of oil we added, so conjuring up a hypothetical "corrected" oil age, based on the effect of refreshing the oil periodically, is simply a matter of tedious mathematics. This corrected oil age assumes zero oil consumption -- something that few if any engines can match, especially over 18,000 road miles -- so for most engines the truth will be somewhere in between. For us, the corrected oil age is 10,300 miles.

Here's a chart showing how the corrected oil age drops off compared to the actual oil age. The table we used to construct this chart is too large to display here; to see it in a separate window, follow the link in the sidebar.

More than anything, what this shows is the incredible effect top-up oil has on the longevity of the motor oil. Even with the minute amounts of oil we added over the course of the study, we extended the "age" of the oil by 8,000 miles compared to adding no oil at all. Indeed, the filter change at 12,000 miles refreshed the oil so well that its corrected age actually remained below the 12k sample's corrected age for almost 3,000 miles! It's no wonder extended-use oil change schemes demand frequent filter changes. The top-up that follows the fresh filter is practically a time machine for your oil.


Interpreting Wear Metals

Looking at a single oil sample's wear metals wouldn't impart much useful information. Frozen in time, with no trend to compare it to, it would serve solely as a pass or fail mechanism: either the wear metals are excessively high or they're not. While that's useful to know, it's only part of the story. The rate of accumulation is actually the more useful measure, as it allows us to see whether the oil's protective ability is improving or weakening. Iron, copper, and lead are the three most important wear metals, as they derive from the components most prone to wear, such as piston rings and various bearings. This chart shows the wear metal trend over the course of the Mobil 1 test. We've standardized the values so that they're directly comparable to each other.

The solid lines are the standardized cumulative totals of wear metals in parts per million for iron, copper, and lead. The shaded lines are the standardized totals of wear metals in ppm per mile -- in other words, the shaded lines represent how quickly the wear metals accumulate as compared to how quickly the miles accumulate.

While the wear metals all accumulated steadily over the course of the test, the highest concentrations of accumulation per mile occurred in the first 3,000 miles of the test! From the 3,000-mile mark all the way to 18,000 miles, only lead showed an increase in per-mile wear beyond 3,000 miles. Yet even with an increased wear rate, lead wore the least in terms of absolute wear. For iron and copper, the longer the oil remained in service, the lower the wear rate got.

In case it isn't obvious yet, this means that the most wear occurs in the first 3,000 miles.

The filter change at 12,000 miles also substantially affected the wear metals. Though changing the filter by itself wouldn't have a significant effect on the concentration of metals -- the metals picked up in spectrometric analysis are too small for a filter to capture -- the amount of top-up oil required to fill the crankcase after the filter swap substantially alters the chemistry of the oil. Iron and lead didn't reach their 12k levels for another four or five thousand miles, and copper never reached its 12k level in the remainder of the test. It's easy to see here why proponents of extreme oil changes demand filter changes at regular intervals. Indeed, one is forced to wonder whether an engine with a high-quality PAO synthetic combined with a bypass filtration system and regular filter changes would ever need its oil changed at all.