Many people don't understand the true usefullness, and limitations, of Used Oil Analysis. I want to help put some perspective on this topic, and how it relates to oil group types.

First, let me assure you that no matter what other people tell you, you are in control of your own destiny. My opinion, or others, might help guide you in your decisions, but ultimately you should do what you feel is best and most comfortable with. As is often touted, beauty is in the eye of the beholder.

Next, let me assure you that I'm not completely shooting from the hip. I am a statistical quality engineer at work. I do NOT work for a lubricant company; I have no bias towards or discontent with any brands I might ever mention. Because of my training and education, I feel I have good insight that others may not, but I am NOT a tribologist.

For the purpose of these topics, I'm going to focus on engines and engine oils, but the concepts can be applied to trannys, diffs, etc. There are, as always, small variances due to specific applications, but the general concepts apply.

Let's get to the meat of the matter; lubrication and UOAs.
There are two ways to know how well your vehicle is fairing in daily use.
1) Tangible environmental events; it feels either smooth or rough, it makes funny smells, it sounds normal or abnormal, it leaks, etc. These are always present, but depend upon your sensitivity and desires. The upside is that it costs nothing to observe, and it provides immediate results. The down side is that it's hard to quantify, and sometimes misleading. If you're a gearhead, you can use these "tools" to help you; if you're not a gearhead, some of these things might go unnoticed. There is not much I can do for you in this regard; you either have the gift, or you don't.
2) UOA analysis; testing your lubricant for trace elements and possible contamination. The upside is that it's quantifiable, traceable, and directed at specific tasks. The downside is that they tend to lag in response time (from a few days to weeks) and they do have costs involved. Here's where I want to educate those of you willing to listen in.

UOAs provide two main thrusts of information. One is how well your equipment is experiencing life, the other is how well your lubricant is experiencing life. That is a subtle but distinct difference. Trace metals in UOAs can give insight as to how your engine is wearing; but only a true engine tear down with component measurement would actually tell you if a bearing is wearing well, or out of tolerance. Since teardowns are costly in time and money, we all hope to avoid this! We use UOAs to let us know now much metal is being shed, to guide us to certain conclusions of relative engine health. UOAs can also tell us how well the lubricant itself is doing. We can check contamination, additive packages, etc. So the concept to understand is that UOAs are a indirect reference as to engine health, but a direct reference to lubricant health.

I'm going to upload a couple of graphs shortly. You'll have to study them in regard to my text to understand where the concepts come from.

In proper UOA usage, you would control as many variables as possible to develop an understanding of how well both the engine and oil are doing. Swapping from brand to brand, grade to grade, does not lend itself to the establishment of ranges and trends in data collection and analysis; I think we can all understand that. So, the goal would be to use one oil for several oil changes, and then track all the indicies of a UOA such as wear metals, additive packages, contamination, etc. You will always see some movement up and down; this is normal variation, but you well also see ranges. You might also see trends (shift of items that continue in the same direction for more than 3 events). With all this in mind, we look at wear metals to let us know how much "protection" a lubricant is providing, and we look at additives & contamination to tell us how much life cycle is left in a lubricant.

The real concept to move towards is that some amount of wear (shed metals) is normal. It's just a fact, and we accept it. That's why we look at iron (Fe), aluminum (Al), copper (Cu) and lead (Pb), among other metals. We establish, through professional sources and many years of tracking, we most would consider "normal" amounts of metals shed. This is what you see in a UOA report, and it's reported in ppm (parts per million) of a sample. We also need to know the exposure duration (how long the lubricant was in service). So, with all these things, we can say that 10ppm of Fe is normal in 8k miles, but it would not be normal after 1k miles of use. The concepts of wear total, and wear rate, are relevant. But we don't change oil based upon a UOA wear rate, we look at wear as a total. While it would be more appropriate to look at wear rate, it's more difficult to develop the raw data into useful information for most laymen; it's much more of a geek engineer thing (like what I do). That in mind, we look at the total ppm for each item, and compare it to either a specific reference (our own past UOAs) or a general reference (the collective of UOAs for other vehicles, as referred to as "universal averages" such as what Blackstone provides).

Ok, now I'm going deep here, so follow along. If you know that XX ppm of a wear metal is ok and considered "normal wear", then anything over that would be "abnormal" and indicate a cause for concern, or perhaps just a change of lubricant. As long as your trends as slow and predictable, then the oil is fine, and you're just accumulating wear metals, and at some point, you're due for an oil change. The point to understand is that you don't want to change an oil before its time is up; why waste a good fluid before you've got your money's worth, right? Why even track ppm of wear metals if you're not going to use it as one of the several indicators of engine health and oil health? So, "normal" is XX ppm of Fe, Pb, etc. They are often different levels for each trace element, but they all have "normal" limits. It might be 10 ppm of Fe, 5 ppm Pb, etc. Further, different engine families have certain traits. Dmax engines wear well, but a known for high Cu (copper) when newer. VW diesels are know for shedding iron more than others. Ford/International Powerstrokes are known for shearing oil viscosity. You get the point, right?

OK, brace yourself, here it comes. You only want to change oil when it's time has come; when it's nearly at the end of its useful life cycle. To do that, you're going to run up the oil UOAs until the wear metal ppm's and contamination results say "OK Boss, time to consider a change. I've given you all I can, taken all the abuse I can, and I'm nearly expired. Change me BEFORE X,xxxx miles and you'll be fine; go past that, and you're in danger!" Makes sense, right. Use the oil to it's fullest, then change it. But here is where everyone gets lost. They loose sight of the concept of wear results versus exposure. They begin to think of mileage as the limiting factor, when it fact it should be ppm as the limiting factor.

Think of two people running, yet judged differently. One person run the 100meter sprint, and is times in seconds. The other person is given 20 seconds, and then measure by how far he runs. They are both running and being judged, BUT NOT BY THE SAME VARIABLES!

Time for the graphs. There are a few presumptions here; don't argue about them, just accept them. For one, I assumed that there would always be a bit of residual ppm of wear metals because you can never truly completely flush an engine clean with a typcial oil change. Further, the slopes of the graphs differ, but they are to indicate AVERAGES, not true results of any one engine. The concept is that wear metals increase with duration, but you'll not change oil until the ppm indicates a change is due. We're only looking at PPM here; other items like contamination and additives can have similar graphs, but to keep things simple, wer're just looking at one metal on one graph. Please don't be tempted to throw in the "yeay, but ..." monkey wrench.

Here is a conventional oil (dino oil) run up to the limit of 10ppm, and then changed in 5k miles. Notice a fairly consistent pattern of wear and results; that's exactly what we want. Predictable ranges and trends. Now, look at the area shaded in yellow; it represents the total amount of metal shed for the duration we're looking at. I multiply ppm by exposure to get a total of wear limits. The total wear metals shed is 180 units of metal.
76205

Next lets look at a synthtetic oil run up to the limit of 10ppm. It does what most would perceive of a "better" job of protecting. The rate at which the wear metals accumulate is slower, so the slopes are lesser. But, again, remember that the concept to UOA use and information is to run ANY oil up to the WEAR METER LIMIT of 10ppm. So to get the fair use of the synthetic oil and our money's worth, we do the same. Take a good long look at this next graph. Although there are fewer oil changes, the TOTAL wear, represented by "units" is the same at 180 units!
[See attachment below.]


Do you start to see the concept I'm trying to impart? Both fluids have the same amount of total wear. Neither engine is experiencing more or less wear; they are experiencing a different wear rate. The TOTAL amount of wear is the same, assuming you are commited to getting your value out of the fluid.

SYNTHETIC LUBRICANTS DON'T PROTECT "BETTER", THEY PROTECT "LONGER". If there is on thing I want you to learn from this topic, it's this concept.

With this all in mind, it is incorrect to say that synthetic fluids protect an engine better, they protect an engine longer. So, if you are capable of extending drain intervals, you can benfit from running synthetics. If you cannot extend drain intervals, then you're wasting money by throwing away good synthetic oil before it's time is up.

Now some might say "Yeah, but Dave, I get get lower wear by changing synthetic oil more often". True, but what you're actually doing by changing oil before it's reached the set limit, is change the desired ppm value. If you do that, you can apply the same concept to the "dino" oil as well. By shortening any OCI, you can effect the wear metal ppm value to a lower or higher limit. Again, remember my analogy to the two runners; it's unfair to judge one oil based upon exposure, and then the other upon a limit. Keep the variables and constants the same, and you can apply common sense to this, and come out with a fiscal approach.

Now, as I said in the beginning, you should do what makes you feel most comfortable. For some people, I'll never get them to understand that synthetics aren't the "miracle" fluid. Synthetics are awesome products, but like nearly all things in life, they provide the "best" service when used with the understanding of their benefits, and limitations.

You as individuals must decide what is "best" for your application based upon critieria that you deem important, based upon priorities. Cost, availability, performance results, ease of use, maintenance habits, etc all play into choices of products, be they oil, filters, etc.

There is no "best" fluid; there are "better" ones for given applications. Only you can decided which meets your needs.

nice wrie up/explanation, thanks

they must protect the engine better, synthetics that is, during cold start up when most of the engine wear occurs. wait, i think i just killed your theory...:eek:

Dnewton
excellent write up as usual.

Now how do you explain the fact to people that oil actually gets better as a lubricant as it is run. To a point at least...

A UOA will indicate when a change is needed because the additive package has broken down or there is contamination from wear metals, fuel, coolants or silicone/sludge.

Changing oil on a 5,000 mile interval instead of using extended oil drain intervals and UOA is not necessarily going to get you longevity out of your engine. I learned this the hard way running over the road trucks. Our longevity was increased by NOT changing oil on the engine manufacturers recommendation for mileage but buy using UOA's to determine when to change. I have 30 plus years in the truck maintenance industry.

Engine overhaul history for a fleet of transcontinental trucks (my experience)

250k to 300k using straight weight oil using eng mfg's mileage recommendation for oil drain.

Changed to multi vis oil= 500k to 600k using eng mfg's mileage recommendation for oil drain.

Using UOA's to establish oil drain frequency and multi vis oil= 750k PLUS... in most instances 1 million miles is not uncommon.

Newton good post as always.

So I am thinking going 10K or 12 months on Mobil 1 TDT and the Mobil 1 extended performance oil filter.

Would your recommend a filter change at 5K or is that a waste too?

:)

Kansas buggy - you are correct that short intervals can occasionally actually increase wear metal results. This is documented by Ford and ConnocoPhillips in an SAE sponsored technical white paper back several years ago. Details are minute and overwhelming in this respect, and I find it facinating. But most people give you that deer-in-the-headlights-look when you say that you can actually change oil TOO often. For most people and most engines, getting past the first few thousand miles is good enough that your over the break even point. But including that info in my post would have just added in some stuff that some people can't wrap their noodle around, so I just left it out. But it is a valid point.

Jimmy - there is just as much information available for filters as there is for oil. Completely revelant subjects that are interlinked, but not always intuitive. For example, filters will generally get BETTER at filtration efficiency the more you use them. As the particulate is caught in the filter, it ever decreases the opening size for the next particle, as a generalization. In theory, your filter works best just before it can't pass enough volume, and then the bypass would open up, negating all the filters abilities.
As a generalization, changing oil and filter together is a good practice. I don't see much value, for example, in a 10k mile oil change, but doing filters every 5k. In fact, some company products suggest just the oposite. An example would be Honda motorcycles for awhile had the oil changes set at 4k miles, but the filters set at 8k miles (every other oil change). Overall, I'd just stick with the OEM recommendation. If you greatly extend the OCI, then UOAs are a must, and you can use them somewhat as a guide. However, you don't want to get fooled into complacency. The filter and the oil should be matched in capabilties. For example, Amsoil has oil products that supprot a 25k mile OCI if you also use their premium EaO filters. Your choice of Mobil 1 TDT and the Mobil 1 filter can easily surpass the 10k mile mark, I would expect. However, I always suggest the use of UOAs past that point. The goal, after all, is to use the UOA as a tool for guidance in making informed decisions.

I believe that there is plenty of proof to show that a conventional HDEO oil with a decent quality filter can easily run 10k miles in a Dmax. Longer runs mandate UOAs, premium filtration, and synthetic fluids.

Dnewton
Again you are correct on the filters.

In my experience we found that changing the filters at a mid point was the best solution for increasing change intervals.

But with that said there is a potential impact to the oil change interval in changing filters at a mid point.

Look at it this way when you change filters you must add make up oil. Any (lost with filter change or used by the engine) make up oil has an impact on the UOA. It replenishes the additive package to some extent and adds new oil for the next UOA. This Ad's to the potential oil life.

So you see just looking at an UOA to determine change interval is only part of the story. The complete amount of information needed to manage an extended oil drain interval is-UOA's, oil filter change strategy, limit idling and clean intake air.