How Realistic Is This Theory?
- Thread starter Red91
- Start date
- Status
Interesting argument (about a flat rate, plus contaminants from residual oil), producing "decreasing" wear metal rates. Before anyone objects that 10 ppm/000 miles is unrealistically high, the math also works if you assume less iron in the residual oil (than 11 ppm), and much less iron, say 2ppm, for each 1000 miles of driving. You still see a decline in wear rates, even if you assume an even amount of wear for every 1000 miles.
I do have one point I'd like to hear your perspective on, Ed. David also mentioned (somewhere) an SAE article/study about Las Vegas taxis: http://papers.sae.org/2007-01-4133/ Not only did they record lower wear rates, which you have evidently accounted for, but they in that study, they state that they have seen less wear in engine parts lubricated with used oil (see page 4 of the preview). The study linked to in Shannow's post also argues that point. In fact, the study in Shannow's link actually measured engine parts, and found less weight loss with used oil.
So what would your response be to that?
I do have one point I'd like to hear your perspective on, Ed. David also mentioned (somewhere) an SAE article/study about Las Vegas taxis: http://papers.sae.org/2007-01-4133/ Not only did they record lower wear rates, which you have evidently accounted for, but they in that study, they state that they have seen less wear in engine parts lubricated with used oil (see page 4 of the preview). The study linked to in Shannow's post also argues that point. In fact, the study in Shannow's link actually measured engine parts, and found less weight loss with used oil.
So what would your response be to that?
The theory at first glance is somewhat plausible.
However, I think the reality is quite different.
Consider the engine that is mechanically sound but requires frequent top ups of oil where the additive pack is constantly refreshed/replenished. Used Oil analysis typically reveals the oil to be in very good condition and close to the specs revealed in a VOA at the manufacturers scheduled OCI with absolutely no evidence or indication of accelerated wear or documented engine failures.
Also the evidence regarding accelerated wear and tear in engines is weighted much more toward the wear and tear associated with initial start ups and the initial warm up phase.
In summary, there are many more factors which affect engine wear which are more plausible and proven than the frequent oil change theory.
Personally, I believe the overly frequent oil changes are an absolute waste of time and resources with no measurable benefit what so ever, and more problematic than is necessary even if there's no measurable or apparent harm to the engine.
However, I think the reality is quite different.
Consider the engine that is mechanically sound but requires frequent top ups of oil where the additive pack is constantly refreshed/replenished. Used Oil analysis typically reveals the oil to be in very good condition and close to the specs revealed in a VOA at the manufacturers scheduled OCI with absolutely no evidence or indication of accelerated wear or documented engine failures.
Also the evidence regarding accelerated wear and tear in engines is weighted much more toward the wear and tear associated with initial start ups and the initial warm up phase.
In summary, there are many more factors which affect engine wear which are more plausible and proven than the frequent oil change theory.
Personally, I believe the overly frequent oil changes are an absolute waste of time and resources with no measurable benefit what so ever, and more problematic than is necessary even if there's no measurable or apparent harm to the engine.
Originally Posted By: Shannow
My current routine is to change the filter every second oil change, and do the filter a couple of weeks after the oil, so that it gets whatever's floating around in the fresh oil (some of them are pretty dirty) out.
I'm a little thick, how does this work ?
My current routine is to change the filter every second oil change, and do the filter a couple of weeks after the oil, so that it gets whatever's floating around in the fresh oil (some of them are pretty dirty) out.
I'm a little thick, how does this work ?
Originally Posted By: paulri
Interesting argument (about a flat rate, plus contaminants from residual oil), producing "decreasing" wear metal rates. Before anyone objects that 10 ppm/000 miles is unrealistically high, the math also works if you assume less iron in the residual oil (than 11 ppm), and much less iron, say 2ppm, for each 1000 miles of driving. You still see a decline in wear rates, even if you assume an even amount of wear for every 1000 miles.
I do have one point I'd like to hear your perspective on, Ed. David also mentioned (somewhere) an SAE article/study about Las Vegas taxis: http://papers.sae.org/2007-01-4133/ Not only did they record lower wear rates, which you have evidently accounted for, but they in that study, they state that they have seen less wear in engine parts lubricated with used oil (see page 4 of the preview). The study linked to in Shannow's post also argues that point. In fact, the study in Shannow's link actually measured engine parts, and found less weight loss with used oil.
So what would your response be to that?
I did use larger numbers than we see in passenger car oils for illustrative purposes. Numbers of that magnitude are not out of line for certain large OTR diesel families.
Here is a thread where this was discussed.
http://www.bobistheoilguy.com/forums/ubbthreads.php/topics/3262930/3
My replies begin on page 3 with post #3256418.
In summary:
1. Experiment not designed to measure what Dave claims. There were uncontrolled variables that preclude Dave's conclusions.
The new oil was always run on a new shim. The aged oil was run on broken-in shims. Break-in wear of the shims was an uncontrolled variable.
2. The wear was measured outside of an engine. The paper showed that combustion byproducts were part of the lower friction surface layers. The new oil was never exposed to combustion byproducts. In an engine, fresh oil is immediately mixed with residual oil containing combustion by-products and immediately exposed to fresh combustion byproducts.
3. Removal of anti-wear layers was from another study using non fully formulated oil. It has no bearing what we do with oil changes.
The above issues do not compromise the scope of the paper, which was to see how long an oil could form a viable protective surface layer. They tested to 15,000 miles. While the oils would form a protective anti-wear layer, all of the oils had long passed condemnation points for oxidative thickening, TBN, and TAN.
To be candid, Dave has read into this paper what he wants it to say, not what it actually says.
Ed
Interesting argument (about a flat rate, plus contaminants from residual oil), producing "decreasing" wear metal rates. Before anyone objects that 10 ppm/000 miles is unrealistically high, the math also works if you assume less iron in the residual oil (than 11 ppm), and much less iron, say 2ppm, for each 1000 miles of driving. You still see a decline in wear rates, even if you assume an even amount of wear for every 1000 miles.
I do have one point I'd like to hear your perspective on, Ed. David also mentioned (somewhere) an SAE article/study about Las Vegas taxis: http://papers.sae.org/2007-01-4133/ Not only did they record lower wear rates, which you have evidently accounted for, but they in that study, they state that they have seen less wear in engine parts lubricated with used oil (see page 4 of the preview). The study linked to in Shannow's post also argues that point. In fact, the study in Shannow's link actually measured engine parts, and found less weight loss with used oil.
So what would your response be to that?
I did use larger numbers than we see in passenger car oils for illustrative purposes. Numbers of that magnitude are not out of line for certain large OTR diesel families.
Here is a thread where this was discussed.
http://www.bobistheoilguy.com/forums/ubbthreads.php/topics/3262930/3
My replies begin on page 3 with post #3256418.
In summary:
1. Experiment not designed to measure what Dave claims. There were uncontrolled variables that preclude Dave's conclusions.
The new oil was always run on a new shim. The aged oil was run on broken-in shims. Break-in wear of the shims was an uncontrolled variable.
2. The wear was measured outside of an engine. The paper showed that combustion byproducts were part of the lower friction surface layers. The new oil was never exposed to combustion byproducts. In an engine, fresh oil is immediately mixed with residual oil containing combustion by-products and immediately exposed to fresh combustion byproducts.
3. Removal of anti-wear layers was from another study using non fully formulated oil. It has no bearing what we do with oil changes.
The above issues do not compromise the scope of the paper, which was to see how long an oil could form a viable protective surface layer. They tested to 15,000 miles. While the oils would form a protective anti-wear layer, all of the oils had long passed condemnation points for oxidative thickening, TBN, and TAN.
To be candid, Dave has read into this paper what he wants it to say, not what it actually says.
Ed
Originally Posted By: paulri
I do have one point I'd like to hear your perspective on, Ed. David also mentioned (somewhere) an SAE article/study about Las Vegas taxis: http://papers.sae.org/2007-01-4133/ Not only did they record lower wear rates, which you have evidently accounted for, but they in that study, they state that they have seen less wear in engine parts lubricated with used oil (see page 4 of the preview). The study linked to in Shannow's post also argues that point. In fact, the study in Shannow's link actually measured engine parts, and found less weight loss with used oil.
So what would your response be to that?
Not going to talk for Ed, but here's one case where the topic was discussed.
http://www.bobistheoilguy.com/forums/ubbthreads.php/topics/3755150/Re:_used_oil_better_than_new?#Post3755150
The paper used various used oils, which were well beyond their use by date, had thickened out of grade, and were oxidised heavily.
When applied to a cam/disk arrangement, they produced lower friction and wear than new oil did...that's all the paper offers, the establishment of tribofilms with used oils...and some of those used oils would probably have done more damage than good, as they were shagged.
Thus my comment in my last post about "virgin" materials, and that the tribofilms are able to last some tens of hours in unadditised oil once formed.
I do have one point I'd like to hear your perspective on, Ed. David also mentioned (somewhere) an SAE article/study about Las Vegas taxis: http://papers.sae.org/2007-01-4133/ Not only did they record lower wear rates, which you have evidently accounted for, but they in that study, they state that they have seen less wear in engine parts lubricated with used oil (see page 4 of the preview). The study linked to in Shannow's post also argues that point. In fact, the study in Shannow's link actually measured engine parts, and found less weight loss with used oil.
So what would your response be to that?
Not going to talk for Ed, but here's one case where the topic was discussed.
http://www.bobistheoilguy.com/forums/ubbthreads.php/topics/3755150/Re:_used_oil_better_than_new?#Post3755150
The paper used various used oils, which were well beyond their use by date, had thickened out of grade, and were oxidised heavily.
When applied to a cam/disk arrangement, they produced lower friction and wear than new oil did...that's all the paper offers, the establishment of tribofilms with used oils...and some of those used oils would probably have done more damage than good, as they were shagged.
Thus my comment in my last post about "virgin" materials, and that the tribofilms are able to last some tens of hours in unadditised oil once formed.
I'm thinking that if oils with different additive packs are used successively, there may be a period of shift in the dominance of the favoured AW regime, and in this transition period, there may be a measurable compromise during the 'handoff'. I can't see how replacement of the same oil with the same additive package would disrupt the existing tribolayers.
Isn't this just 7th grade algebra?
y = mx + b.
y = wear metals.
m = rate of wear per mile.
x = miles
b = contribution from the residual oil.
Because of "b," short OCIs look bad, but they actually aren't.
y = mx + b.
y = wear metals.
m = rate of wear per mile.
x = miles
b = contribution from the residual oil.
Because of "b," short OCIs look bad, but they actually aren't.
You make a number of points that provide alternative interpretations of the data we've seen.
I especially like the "simple math" example.
This provides all of us with some food for thought, although DN's meta analysis is still pretty compelling.
Of course, it may also lead us to false conclusions, as you've shown in your post.
Thanks for a well reasoned and well argued post.
I especially like the "simple math" example.
This provides all of us with some food for thought, although DN's meta analysis is still pretty compelling.
Of course, it may also lead us to false conclusions, as you've shown in your post.
Thanks for a well reasoned and well argued post.
Originally Posted By: Skid
...
y = mx + b.
y = wear metals.
m = rate of wear per mile.
x = miles
b = contribution from the residual oil.
Because of "b," short OCIs look bad, but they actually aren't.
But if "x" is limited to a small number, "b" will be small the next time. We need to concentrate on "m."
...
y = mx + b.
y = wear metals.
m = rate of wear per mile.
x = miles
b = contribution from the residual oil.
Because of "b," short OCIs look bad, but they actually aren't.
But if "x" is limited to a small number, "b" will be small the next time. We need to concentrate on "m."
Originally Posted By: CR94
We need to concentrate on "m."
to know what exactly ?
The radiotracer method in the various papers uses radioactivated metals and detectors to know exactly how much wear material is circulating...a UOA, as per the edhackett's post only measures the tiny circulating parts, and can/does miss the macro wear debris.
We need to concentrate on "m."
to know what exactly ?
The radiotracer method in the various papers uses radioactivated metals and detectors to know exactly how much wear material is circulating...a UOA, as per the edhackett's post only measures the tiny circulating parts, and can/does miss the macro wear debris.
And macro is relative, as the mass of the particles is important. Lead =/= iron =/= aluminium as far as detected particle sizes in a UOA goes.
Macro Fe debris is what magnetic drain plugs are for ...
It's also what the filter is for.
If it's bigger than either of those two mitigators, it's carnage and the motor is on life support ...
It's also what the filter is for.
If it's bigger than either of those two mitigators, it's carnage and the motor is on life support ...
I'll see if I can find the post describing what particle sizes can be detected, but from memory a particle with the density of water should be under 4 µm. and a lead particle would be much much smaller being 12 times as dense.
none will be trapped by the filter
none will be trapped by the filter
Originally Posted By: Shannow
Originally Posted By: CR94
We need to concentrate on "m."
to know what exactly ?
The radiotracer method in the various papers uses radioactivated metals and detectors to know exactly how much wear material is circulating...a UOA, as per the edhackett's post only measures the tiny circulating parts, and can/does miss the macro wear debris.
That's a good point about the UOAs, I've also wondered if the particles can also fall out of view of standard UOA if they get too small, can that happen?
Originally Posted By: CR94
We need to concentrate on "m."
to know what exactly ?
The radiotracer method in the various papers uses radioactivated metals and detectors to know exactly how much wear material is circulating...a UOA, as per the edhackett's post only measures the tiny circulating parts, and can/does miss the macro wear debris.
That's a good point about the UOAs, I've also wondered if the particles can also fall out of view of standard UOA if they get too small, can that happen?
Originally Posted By: BrocLuno
Macro Fe debris is what magnetic drain plugs are for ...
It's also what the filter is for.
If it's bigger than either of those two mitigators, it's carnage and the motor is on life support ...
Agreed.
I would add that the smaller particles of wear metals will always be produced in addition to the larger particles in the lead up to an imminent failure so there will always be a variable proportion depending upon how far down the path toward a catastrophic failure point.
In summary.
Normal wear metal particle sizes will be in the very smallest size range.
Toward the end of life of the assembly, the particles will become progressively larger but small particles will always be present.
An accurate UOA from a reputable lab will always have some benefit to varying degrees.
And a Filter Patch test will always reveal more information, with the cost effectiveness to be determined by the end user.
Macro Fe debris is what magnetic drain plugs are for ...
It's also what the filter is for.
If it's bigger than either of those two mitigators, it's carnage and the motor is on life support ...
Agreed.
I would add that the smaller particles of wear metals will always be produced in addition to the larger particles in the lead up to an imminent failure so there will always be a variable proportion depending upon how far down the path toward a catastrophic failure point.
In summary.
Normal wear metal particle sizes will be in the very smallest size range.
Toward the end of life of the assembly, the particles will become progressively larger but small particles will always be present.
An accurate UOA from a reputable lab will always have some benefit to varying degrees.
And a Filter Patch test will always reveal more information, with the cost effectiveness to be determined by the end user.
Originally Posted By: PeterPolyol
Originally Posted By: Shannow
Originally Posted By: CR94
We need to concentrate on "m."
to know what exactly ?
The radiotracer method in the various papers uses radioactivated metals and detectors to know exactly how much wear material is circulating...a UOA, as per the edhackett's post only measures the tiny circulating parts, and can/does miss the macro wear debris.
That's a good point about the UOAs, I've also wondered if the particles can also fall out of view of standard UOA if they get too small, can that happen?
I think it's reasonable to assume the negative effects of any of the smallest particles is not an issue in the grand scheme of things and almost impossible to measure in real terms, with the end of life of the machinery only measured by it's obsolescence or other factors.
Originally Posted By: Shannow
Originally Posted By: CR94
We need to concentrate on "m."
to know what exactly ?
The radiotracer method in the various papers uses radioactivated metals and detectors to know exactly how much wear material is circulating...a UOA, as per the edhackett's post only measures the tiny circulating parts, and can/does miss the macro wear debris.
That's a good point about the UOAs, I've also wondered if the particles can also fall out of view of standard UOA if they get too small, can that happen?
I think it's reasonable to assume the negative effects of any of the smallest particles is not an issue in the grand scheme of things and almost impossible to measure in real terms, with the end of life of the machinery only measured by it's obsolescence or other factors.
Originally Posted By: Jetronic
I'll see if I can find the post describing what particle sizes can be detected, but from memory a particle with the density of water should be under 4 µm. and a lead particle would be much much smaller being 12 times as dense.
none will be trapped by the filter
I know this thread is in relation to engine oil changes.
It depends upon the wear metal and it's origin.
I would assume that a soft metal like Lead, Copper or Aluminium would be less problematic overall than Iron depending on the assembly.
For example normal amounts of Iron in any engine type/design, will be far less problematic overall than normal amounts of Iron in any Differential type/design.
Of course it's all dependant upon the duty cycle of the relevant assembly.
I'll see if I can find the post describing what particle sizes can be detected, but from memory a particle with the density of water should be under 4 µm. and a lead particle would be much much smaller being 12 times as dense.
none will be trapped by the filter
I know this thread is in relation to engine oil changes.
It depends upon the wear metal and it's origin.
I would assume that a soft metal like Lead, Copper or Aluminium would be less problematic overall than Iron depending on the assembly.
For example normal amounts of Iron in any engine type/design, will be far less problematic overall than normal amounts of Iron in any Differential type/design.
Of course it's all dependant upon the duty cycle of the relevant assembly.
Last edited:
Originally Posted By: PeterPolyol
That's a good point about the UOAs, I've also wondered if the particles can also fall out of view of standard UOA if they get too small, can that happen?
No. ICP-OES works at the atomic level. The technique is designed to analyze elements fully dissolved in a solvent. The only reason it quantifies the small particles that make it to the plasma relatively accurately is that the plasma is hot enough to dissociate the particle into its individual atoms.
The heat from the plasma, 6,000K-10,000K, causes an electron to be kicked into a higher energy orbit. When that electron falls back to its ground state it emits a photon of light at a specific wavelength. The amount of light emitted is proportional to the concentration. This light is amplified by the instrument and converted to an electrical signal that is used to calculate concentration.
Ed
That's a good point about the UOAs, I've also wondered if the particles can also fall out of view of standard UOA if they get too small, can that happen?
No. ICP-OES works at the atomic level. The technique is designed to analyze elements fully dissolved in a solvent. The only reason it quantifies the small particles that make it to the plasma relatively accurately is that the plasma is hot enough to dissociate the particle into its individual atoms.
The heat from the plasma, 6,000K-10,000K, causes an electron to be kicked into a higher energy orbit. When that electron falls back to its ground state it emits a photon of light at a specific wavelength. The amount of light emitted is proportional to the concentration. This light is amplified by the instrument and converted to an electrical signal that is used to calculate concentration.
Ed
- Status
