Oil Analysis - good read

[mvweebles]

Originally Posted by a64pilot

You think oil analysis will find the cause of a rough idle?
That is exactly my point, you made it well.

Actually I wasn’t comparing aircraft engines, actually the transmissions, since GE said oil analysis of the T-701 engine was of no value, we didn’t sample them.

BTW - a rough idle could be caused by low compression on one cylinder. Could be a blown head gasket. And yes, that would show up on an engine oil analysis as glycol in the oil.

[Wotname]

Originally Posted by mvweebles

It's the right question - does an oil analysis give meaningful data that can be actioned. Unfortunately, it's proving a negative - the theory is that by identifying a small problem now you prevent a large problem later. I am conformable saying that you have a much greater chance of identifying salt water intrusion or advanced bearing wear via an engine oil analysis than through any other method as there are none.

I agree with A64 in some ways - if the fault initiates immediately after the oil analysis, the issue may cause failure before the next change. But given the modest cost of the analysis and the hi cost and inconvenience of a failure, this one is a no-brainer for me.

It's not quite like proving a negative (IMO), it is more like has a positive result (i.e. something untoward in the oil) prevented an incipient serious engine failure???

I do agree that the oil analysis tells one a lot about the condition of the oil at time of taking the sample and that could (would?) be a good thing.

To my mind, monitoring (and trending) the temps, pressures and sounds is one's primary diagnostic tool with the oil analysis trends being a nice to have bedtime reading.

[mvweebles]

Originally Posted by Wotname

It's not quite like proving a negative (IMO), it is more like has a positive result (i.e. something untoward in the oil) prevented an incipient serious engine failure???

I do agree that the oil analysis tells one a lot about the condition of the oil at time of taking the sample and that could (would?) be a good thing.

To my mind, monitoring (and trending) the temps, pressures and sounds is one's primary diagnostic tool with the oil analysis trends being a nice to have bedtime reading.

Below is a decent article on oil analysis. Describes how oil analysis gives info on wear limits, contaminants, oil condition (not just lubricity), particulates, water, etc. These are beyond the ear and are "canaries in the coal mine." These are actionable indicators, meaning presence of out-of-band readings will prompt specific remediation actions before the condition makes the situation worse and more expensive to repair. If course, there's a chance you won't get the information in time. But if you don't do the test, you have no chance of getting the information in time. As an example, oil analysis will show lower bearing wear long before there's an audible knock.

https://www.machinerylubrication.com...alysis-reports

[skipperpete]

The OP makes some good points but from my experience in the marine industry, the posts/ observations by A64pilot echo some of my experiences with lube oil analysis. I did Cat scheduled oil sampling for years and in all that time the only advantage was being informed about when to NOT change the lube oil, never when to open up an engine because of a bad SOS report.
Yacht buyers who request an oil test at survey are generally wasting money because a single isolated sample can give very confusing results...... especially on an engine that may have had serious health issues in the past and still shows disturbingly high levels of critical metals even though a rebuild has been done and those trace levels are no longer relevant.
Also , at what point in the oil analysis readout result do you pull the engine out and rebuild it. a rebuild done on the basis of a high lead indium or babbit spike might just be showing thrust bearing wear that requires no intervention.
Pete.

[Steve DAntonio]

Originally Posted by slug

Oil analysis only gives reliable results if you have a regular test log from new at scheduled intervals

You can be forgiven for saying this because it's a common misconception, one often perpetuated by brokers.

Having taken hundreds of fluid samples (crankcase oil, coolant, hydraulic fluid and gear oil), and having reviewed many hundreds if not thousands of analysis reports, I have, on many occasions, in doing a single analysis, identified potentially fatal flaws. You don't need a trend to know that coolant, excessive fuel (over 10%) or sodium in the oil sample is a serious problem. For something like this it would be common to retest, and maybe even use another lab. If the result is the same, it's actionable.

Here's a better link to that article https://stevedmarineconsulting.com/oil-analysis/ If you ever get this plane jane version of one of my articles, clicking on the link at the top will get the fully formatted version. My IT specialist is currently working on why this happens and correcting it.

And a more detailed version written for the marine industry https://stevedmarineconsulting.com/w...s143_Final.pdf

Needless to say, I'm an advocate for and strong believer in fluid analysis, the primary problem, and lack of faith in fluid analysis arises from the inability to interpret the results, a subject I delve into in the articles. The average user, and even most mechanics, have no idea how to interpret results, and simply scanning the report for red or yellow alerts can be misleading at best.

There are week-long classes that teach fluid analysis (I've taken them), however, there's ample material available for self-study. At the very least you can call the lab for assistance in interpreting results. However, labs are unlikely to catch their own data errors, you need to be prepared to do that.

[mvweebles]

Originally Posted by Steve DAntonio

You can be forgiven for saying this because it's a common misconception, one often perpetuated by brokers.

Having taken hundreds of fluid samples (crankcase oil, coolant, hydraulic fluid and gear oil), and having reviewed many hundreds if not thousands of analysis reports, I have, on many occasions, in doing a single analysis, identified potentially fatal flaws. You don't need a trend to know that coolant, excessive fuel (over 10%) or sodium in the oil sample is a serious problem. For something like this it would be common to retest, and maybe even use another lab. If the result is the same, it's actionable.

Here's a better link to that article https://stevedmarineconsulting.com/oil-analysis/ If you ever get this plane jane version of one of my articles, clicking on the link at the top will get the fully formatted version. My IT specialist is currently working on why this happens and correcting it.

And a more detailed version written for the marine industry https://stevedmarineconsulting.com/w...s143_Final.pdf

Needless to say, I'm an advocate for and strong believer in fluid analysis, the primary problem, and lack of faith in fluid analysis arises from the inability to interpret the results, a subject I delve into in the articles. The average user, and even most mechanics, have no idea how to interpret results, and simply scanning the report for red or yellow alerts can be misleading at best.

There are week-long classes that teach fluid analysis (I've taken them), however, there's ample material available for self-study. At the very least you can call the lab for assistance in interpreting results. However, labs are unlikely to catch their own data errors, you need to be prepared to do that.

Thanks so much Steve.

Question : any guidance on how to mitigate risk of an owner changing oil immediately prior to sea trial? Best I can think of is long sea trial 4+ hours. But even that isnt too long. Will the lab be able to calibrate based on that few hours? Other thoughts?

Peter

[a64pilot]

So you have an oil sample interval of every 250 hours when most change oil and you are relying on that to determine if you have water in your fuel or fuel in your oil?
You can see even low amounts of water in oil long before it becomes catastrophic, and there are cheap tests that detect combustible gasses in coolant if you think it’s coolant.
Fuel you can smell, and most commonly it’s called “making oil” as of course as the fuel is added to the oil level the oil level increases or you engine making oil.
It’s a lot more common than you think especially in bigger outboard engines that do a lot of trolling, but mechanical fuel pumps can leak and take your engine to run away pretty quick, but bad injectors can also be the source, for that it’s almost always a change in idle and an oily smoky exhaust, more than the common Yanmar.
If your oil level is increasing and the oil isn’t milky, its fuel.

You need to be watching these things all of the time, not sitting back thinking my oil analysis that’s done every oil change will catch any problems. It’s a snap shot of one moment of time, and that moment is often spaced hundreds of hours apart, you may get lucky and that head start leaking just before the sample or that fuel pump, but statistically probably not, and these issues rarely go on un corrected for hundreds of hours without serious mechanical breakdowns.

That is why on Ultra expensive Military aircraft our sample interval was 25 hours plus or minus 10%. Cause the likely hood of catching a serious problem when looking every few hundred hours is slim.

With an oil pressure gauge you will see lower oil pressures at idle with fuel diluted oil, cause of course fuel is thinner than oil, on your before every run oil check you will see water in the oil and if you don’t check you will see lower oil pressures too, cause emulsified water and oil has less viscosity than straight oil.

So if an oil analysis is what tells you of the presence of fuel or water, first you got very lucky and the failure timed right before the analysis, and secondly your not very observant of what’s going on in your engine.

[dannc]

I have been doing UOA on my truck and tractor for years with Blackstone.

Has the UOA identified issues that allowed me to make changes and make said issues go away. Why, yes, yes it has. Did UOA prevent an blown engine? No way to prove that one.

First of all, at least with Blackstone, one does NOT need a bunch of tests to be of value. Blackstone has a data base of engines and UOA results from every engine they test. With YOUR oil test, they include these universal averages, along with your test results. One can easily compare how your engine tests compare with other engine test results. Even with only one test.

Having a bunch of UOA results is certainly goodness and does a better job of providing long term specific values for YOUR engine but it is not the end of the world if one has not done a bunch of UOA. And you have to start somewhere.

So how has UOA help me?

I will start with the iffy one first. My truck is almost 19 years old and I have owned it since new. At one time, I was buying JD oil in 5 gallon buckets because they have a very good 0Wx40 oil at a very good price. By the book, the oil is supposed to be changed every 5,000 miles. I change about 15,000 miles. I can hear the gasps! The screams! The horror. Well UOA, shows that when I am changing at 15K miles I am throwing away perfectly good oil even though I ran it three times more than the manuals specifies. Furthermore, my UOA test results are BETTER than the universal averages for my engine.

Why?

My daily drive is 70ish miles and mostly highway driving and very little stop and go traffic. At least that is the only reason to explain why my UOA is so good and I am throwing away perfectly good oil.

But. There has to be a but, I had an UOA that show much higher engine wear that appeared out of no where. If I remember right, the numbers, for the most part, where within spec but some where out of spec, and most were around the universal averages. But, the numbers where really higher than my engines normal numbers. Over the next few oil changes and UOA, those wear numbers declined and returned to normal. What the heck happened?

My best guess is the engine oil. I had bought the JD oil before the switch to ULSD and that "old" oil had very high TBN numbers which caused some wear on the engine when we went to ULSD fuel. At least that is my best guess.

Where the UOA really helped me was when fuel started showing up in the oil. WTF. The numbers where in an acceptable range but I don't want/need fuel in my engine oil. Even though most of my driving is highway, at the time, I was having to drive into an area where there could be some very bad stop and go traffic. I think the fuel in the oil problem was due to excessive idling. Thankfully, that was short lived and the fuel in oil problem went away.

Ironically, at the same time, my tractor engine, a Yanmar 48 HP engine, showed fuel in the oil. That one was easy, I just moved up the RPMs when running the tractor doing front end loader work and the problem went away. I would run the engine around 1600 RPM so I moved up to 1800-2000. Problem solved.

The way I see some boat's run at low RPMs and lots of idling, I would suspect they have fuel in the oil. Is it bad? Only a UOA will show it.

My guess is that if one changed their oil per the book one could get away with fuel in the oil due to excessive idling but that is just a guess. Only a UOA will show the numbers.

I will keep on using UOA for my engines. The itty bitty amount of money it costs is worth it, not only for knowing how the engine is aging but also showing up issues.

Later,
Dan

[a64pilot]

Dan, your oil tested good at 15,000 miles first because it was a good oil, secondly the average Diesel engine just isn’t hard on oil, they don’t run at real high outputs and real high temps very often if at all, the oil lives an easy life.
But oil wearing out or breaking down is not why we change Diesel engine oil frequently, as you said and your correct, the oil is fine, it can go much longer.
But it’s loaded with soot and the longer it’s run the more soot it becomes loaded with, and soot is abrasive, and much much smaller than any full flow oil filter can filter.
We change oil to remove the soot, the oil itself is still fine. It’s why I try to tell people that expensive synthetic oil may not really be necessary in a Diesel, sure it doesn’t hurt, just maybe it’s not really necessary.

If you want to run extended oil intervals in a Diesel, first do as your doing, use a good oil, but secondly look really hard into a high bypass oil filtration system. It’s a big ole filter that only filters a small amount of the oil flow, it does that so that it can filter very fine particles out, a good one will filter out pretty much anything large enough to cause wear, there is a size at which smaller than that and very little if any wear is caused. Of course even though it’s only filtering a fraction of the total oil over time it’s all filtered.

With a high bypass oil filter, then you can use your oil analysis to determine when to change the oil, often this is done by watching the TBN / TAN numbers. Ever a super fine filter can’t filter out acid.
TBN is of course total base number which is a measurement of how much acid the oil can neutralize, TAN is the Total Acid Number, when TAN equals or exceeds TBN, most consider the oil done, a higher TBN oil can of course go longer before the TBN is consumed.

Since the introduction of ULSD pretty much all oils are formulated with a lower TBN, one to save money as secondly it’s just not required, plus often the metals and or chemicals used to increase TBN are toxic to modern pollution controls, so the old high TBN oil with a lot of zinc will over time poison a catalyst.
So with ultra low sulphur Diesel fuel TBN isn’t as necessary or as important as it was back in the day.

But unless it’s insanely high TBN will not cause any wear, if stupid high it may cause ash, and that ash cause wear, but unless you get some ship oil or used be locomotive oil which burned real high sulphur fuels and required real high TBN, you can’t buy an oil with too high a TBN, it’s just not sold.

Oil analysis never hurts, if it makes yourself feel good go for it, just do NOT change your other good habits like watching oil pressure and temps and checking the oil before every day etc., because you now are doing analysis and don’t need to do those things anymore.


So far as high metals occasionally etc, a whole lot of that has to do with how the oil was sampled, how long was the engine run prior to sampling, how long was it off, did you sample mid flow on the drain etc. and was everything real clean, on the aircraft it’s wasn’t uncommon to get a high silicon reading, because there was some dirt on the cap or whatever that fell in when the sample was taken, it doesn’t take much.

Also if your getting fuel in your oil, you really should have your injectors serviced as opposed to running the engine harder, almost certainly the source of that fuel is an injector not properly atomizing it and there is unburned fuel. Instead of a fine spray it’s pattern has what we call streakers.

If anyone cares, turbine aircraft engines fuel nozzles are subject to the exact same varnishing up etc as a Diesel engine, and on a turbine it can cause hot spots in the hot section, greatly reducing the engines life, so regular “hot end” inspections where the hot end is inspected and nozzles cleaned is very important.

[mvweebles]

Originally Posted by a64pilot

So you have an oil sample interval of every 250 hours when most change oil and you are relying on that to determine if you have water in your fuel or fuel in your oil?
You can see even low amounts of water in oil long before it becomes catastrophic, and there are cheap tests that detect combustible gasses in coolant if you think it’s coolant.
Fuel you can smell, and most commonly it’s called “making oil” as of course as the fuel is added to the oil level the oil level increases or you engine making oil.
It’s a lot more common than you think especially in bigger outboard engines that do a lot of trolling, but mechanical fuel pumps can leak and take your engine to run away pretty quick, but bad injectors can also be the source, for that it’s almost always a change in idle and an oily smoky exhaust, more than the common Yanmar.
If your oil level is increasing and the oil isn’t milky, its fuel.

You need to be watching these things all of the time, not sitting back thinking my oil analysis that’s done every oil change will catch any problems. It’s a snap shot of one moment of time, and that moment is often spaced hundreds of hours apart, you may get lucky and that head start leaking just before the sample or that fuel pump, but statistically probably not, and these issues rarely go on un corrected for hundreds of hours without serious mechanical breakdowns.

That is why on Ultra expensive Military aircraft our sample interval was 25 hours plus or minus 10%. Cause the likely hood of catching a serious problem when looking every few hundred hours is slim.

With an oil pressure gauge you will see lower oil pressures at idle with fuel diluted oil, cause of course fuel is thinner than oil, on your before every run oil check you will see water in the oil and if you don’t check you will see lower oil pressures too, cause emulsified water and oil has less viscosity than straight oil.

So if an oil analysis is what tells you of the presence of fuel or water, first you got very lucky and the failure timed right before the analysis, and secondly your not very observant of what’s going on in your engine.

I assume that folks who use oil analysis use it augment a diligent engine operating monitoring regime, not replace good maintenance and operating practices with occasional oil analysis. It's not an either/or. I would agree that oil analysis is not as useful to detect the potential of an instantaneous catastrophic failure. Fortunately, majority of failures of a well cared for diesel develop over time, not instantaneously. A64pilot is correct though - for the rare instantaneous failure, diligence while monitoring gauges and alerts are your best defense.

Dannc's post before mine describes a good scenario on how he's used oil analysis over time. Despite the age of his truck, I would think his practice of regular oil analysis would more than pay for itself at resale - I know I would be much more inclined to buy an old truck with that type of service record than the typical Craigslist truck.

Engine oil analysis is done for the same reason you have blood work done or check your blood pressure. There is a direct correlation to the results and the underlying health. Sure, a chunk of something could break free a day later and cause a massive stroke (god forbid), but it's good data and helpful over time. Just depends on your perspective.

For those unsure or are new to trawlers and diesels, please read Steve D's articles and make up your own mind. He's a recognized industry expert with very detailed explanations that make sense, not emotional appeals. His only business in life is as a buyer's advocate - he has no other dog in this fight except to give his best thinking on how we can maximize our pleasure, safety, and investments.

Peter

[a64pilot]

Curious, how do you think Dan’s UOA helped make his truck last longer?
Personally I’d want the truck that had its oil changed every 5,000 miles as opposed to one that based on a UOA went to 15,000 miles.

If you pay extra Blackstone will include in the analysis the amount of soot in your oil, 2% is considered the redline if you will, but it’s effect on wear is linear, the more their is the more wear it causes, so is 2% the max?
The fuel Dan was getting in his oil was almost certainly from incomplete combustion, and guess what is a major contributor in soot?
It’s sort of like cycling a battery, the deeper the shorter the lifespan, many think that there is a limit at 50% but in truth it’s not, it was an arbitrary number picked long ago by someone, the 2% soot is the same.

[a64pilot]

Google Diesel engine wear due to soot or similar, there is a lot of real scientific data.
For you guys that are UOA proponents, here is a short article from Blackstone.

Point is, if your considering extended oil change intervals, for goodness sake monitor the soot, cause the oil will be fine, long after it’s overly contaminated with wear causing soot.
https://www.blackstone-labs.com/soot...50kgjulrb))%2f

[mvweebles]

Originally Posted by a64pilot

Curious, how do you think Dan’s UOA helped make his truck last longer?

I don't know Dan, but from his post

"But. There has to be a but, I had an UOA that show much higher engine wear that appeared out of no where.

He goes on to say it was a bit of a false alarm as it worked its way out over a few more oil changes. So he was alerted to a condition that may have been a problem well in advance, but ultimately proved otherwise. To suggest that Oil Analysis was a waste of time/effort/money because it didn't turn out to be a critical condition (but could have been) is sort like saying if a biopsy reveals a benign growth instead of cancer, was a waste of time/money to have the test.

Do I sense a bit of contrarian here?

Peter

[a64pilot]

If you read Dans post you’ll find that all UOA did was to give supporting data for him to run extended oil changes.
Which is pretty much exactly my point, unless he knew to and paid extra for the soot level to be determined then it’s likely that data he was given to support extended oil change interval while not false, was misleading in that it didn’t contain all available data. But it could have. Analyzing oil is exactly what a UOA is good for. Maybe he did know to and paid for the soot test, if he did, good.
By now if UOA was so good at saving engines I would have expected several stories about how people had their #3 bearing wiped out and due to UOA they were alerted in time to disassemble their engines and saved the crankshaft.
UOA is not new,
However I’ve never actually heard a single credible story of that happening, some my cousin knew a friend who heard kind of thing, but thats it, and in my 15 years of maintaining a fleet of 24 helicopters all over the world bears that out, we lost many components of course, but not one single time were we alerted to impending failure from an oil analysis, and that was with as low as a 25 hour sample interval. Nose gearboxes as they were problematic were on a 25 hour interval, main transmissions I believe were 50, but don’t quote me, that is coming up on two decades ago.

As I keep saying and you don’t want to hear it, cause you truly want to believe that an oil analysis is excellent for determining engine health, but it’s not so good from an engine monitoring perspective, and you and many, many others swear that it is, and at least one other person who it seems cared for construction equipment pretty much said the same thing I’m saying, that it’s excellent for monitoring oil condition, engines, not so much.

It’s a tool, if you want to do an UOA, go ahead it’s not hurting anything, but do NOT base any of your maintenance decisions on it, if your engine starts to pick up a hollow knock, but the UOA is good, don’t head off on a passage believing that your engine is healthy, that hollow knock could be a big end bearing.
Also if you buying a boat and the engine clearly hasn’t been well maintained as evidenced by old belts and dirty etc, don’t be persuaded by a good UOA that the engine is good.

Me, personally I don’t bother, I change my oil every 100 hours and almost always am out cruising and can’t send it in for analysis if I wanted to.

Now for BIG motors that measure their oil capacity in dozens of gallons especially, an oil analysis is likely a very good tool to determine oil change interval. That’s what the Army does with all of its ground equipment, and there is no telling over the years how many maybe millions of gallons of oil they have saved and how many tax dollars.
Oil analysis is not new by any means, US ARMY began their oil analysis program in 1961.
If it were so good at determining engine health, then surely all the major engine manufacturers publish in their maintenance manuals the max allowable parts per million of each kind of metal and tell you when the engine needs to be torn down and which parts to inspect?

[mvweebles]

Originally Posted by a64pilot

If you read Dans post you’ll find that all UOA did was to give supporting data for him to run extended oil changes.
Which is pretty much exactly my point, unless he knew to and paid extra for the soot level to be determined then it’s likely that data he was given to support extended oil change interval while not false, was misleading in that it didn’t contain all available data. But it could have. Analyzing oil is exactly what a UOA is good for. Maybe he did know to and paid for the soot test, if he did, good.
By now if UOA was so good at saving engines I would have expected several stories about how people had their #3 bearing wiped out and due to UOA they were alerted in time to disassemble their engines and saved the crankshaft.
UOA is not new,
However I’ve never actually heard a single credible story of that happening, some my cousin knew a friend who heard kind of thing, but thats it, and in my 15 years of maintaining a fleet of 24 helicopters all over the world bears that out, we lost many components of course, but not one single time were we alerted to impending failure from an oil analysis, and that was with as low as a 25 hour sample interval. Nose gearboxes as they were problematic were on a 25 hour interval, main transmissions I believe were 50, but don’t quote me, that is coming up on two decades ago.

As I keep saying and you don’t want to hear it, cause you truly want to believe that an oil analysis is excellent for determining engine health, but it’s not so good from an engine monitoring perspective, and you and many, many others swear that it is, and at least one other person who it seems cared for construction equipment pretty much said the same thing I’m saying, that it’s excellent for monitoring oil condition, engines, not so much.

It’s a tool, if you want to do an UOA, go ahead it’s not hurting anything, but do NOT base any of your maintenance decisions on it, if your engine starts to pick up a hollow knock, but the UOA is good, don’t head off on a passage believing that your engine is healthy, that hollow knock could be a big end bearing.
Also if you buying a boat and the engine clearly hasn’t been well maintained as evidenced by old belts and dirty etc, don’t be persuaded by a good UOA that the engine is good.

Me, personally I don’t bother, I change my oil every 100 hours and almost always am out cruising and can’t send it in for analysis if I wanted to.

Now for BIG motors that measure their oil capacity in dozens of gallons especially, an oil analysis is likely a very good tool to determine oil change interval. That’s what the Army does with all of its ground equipment, and there is no telling over the years how many maybe millions of gallons of oil they have saved and how many tax dollars.
Oil analysis is not new by any means, US ARMY began their oil analysis program in 1961.
If it were so good at determining engine health, then surely all the major engine manufacturers publish in their maintenance manuals the max allowable parts per million of each kind of metal and tell you when the engine needs to be torn down and which parts to inspect?

You're right. Everyone else is wrong. Including the experts. Oil testing is a total waste of time and money (well suited for the Govt I suppose). Everything in articles from industry experts such as Steve D, Nigel Calder, and BoatUS is misleading. Knowing if there is diesel or coolant or sea water or soot or bearing filings in your oil doesn't mean beans - it ain't broke until it starts banging, blowing smoke, seizes, or triggers an alarm, so why bother?

Gotcha. I have now seen the light. No need for no stinking test!

BTW - you're the only person who's suggested anyone might use an oil test as a proxy for maintenance. No one else has uttered a word anywhere close to "no need to change your oil or other maintenance until a test says the oil is shot." I'm not sure why the constant drumbeat. But hey, you've sold me. Anyone want to buy a used oil test kit?

Peter