Particle Count Comparison, 2/5/06

Status
Not open for further replies.
quote:

An understanding of what these codes mean requires a discussion of the structure and nomenclature of the codes. The structure can be broken down as follows:
16 / 15 / 12

The first number, "16", is the range code number that corresponds to a range of the number of particles present in an oil that are greater than 4 microns (µm or micrometers). That count is typically based on the number of particles per milliliter (ml). The second number, "15", is the range code number corresponding to the number of particles/ml that are greater than 6 microns and the third number corresponds to the number of particles/ml greater than 14 microns in size. It should be noted that the first range code includes the particles from both the second and third range codes. Likewise, the second range code includes the particles counted in the third range code.

The actual numbers of particles counted per milliliter in each size category (4, 6, and 14 microns) are converted to the appropriate ISO Code via the chart in Table 2.

For example, if a particle counter determined that there were 501 particles/ml greater than 4 microns, the corresponding ISO code would be 16. If 199 particles/ml were greater than 6 microns, the corresponding ISO code would be 15. Finally, if the counter determined that 27 particles/ml were greater than 14 microns, then the corresponding ISO code would be 12. The overall cleanliness of the lubricant would be reported as 16/15/12. Most particle counters do this conversion automatically. The following characteristics should be noted:

* The number of particles/ml will always correspond to the same ISO code independent of the particle size being measured.
* The overall ISO code structure - XX/YY/ZZ - will always correspond to the particle counts at the sizes:
o XX - greater than 4 microns
o YY - greater than 6 microns
o ZZ - greater than 14 microns
* Each ISO code increase represents roughly a doubling of the particle count
* However, an increase of only 1 particle/ml at a particular size can also give the appearance by the ISO code that the oil is twice as dirty

It should be noted that the ISO Cleanliness Code was revised in 1999. The earlier code used only two range values corresponding to particles greater than 5 microns and 15 microns. This would be reported in the following format: YY/ZZ. Additionally, some companies also used a range code for particles greater than 2 microns, and it would be reported similarly to the current code structure with the particle sizes corresponding to 2, 5, and 15 microns as opposed to the new 4, 6, and 14 microns. This change was due to the phasing out of AC Fine Test Dust (ACFTD) as the test method and replacing it with Medium Test Dust (MTD) and to improvements in the test procedure. There is no net effect on filter performance and minimal effect on how particle counts are reported with this change. The new reporting standard may be referred to as ISO 4406:1999 or ISO 4406 (MTD). The previous standard may be referred to as ISO 4406 (ACFTD). Results reported as a two "digit" code could be assumed to be the older standard at 5 and 15 microns. If results are reported as a three "digit" code it is assumed to be the new standard at 4, 6, and 14 microns. Many labs will report the appropriate sizes to avoid confusion.

Users should be aware of this as particle counters calibrated to the old standard are still widely used and will be for some time. Particle counters calibrated to the new standard are available and will become more common with time. Particle counters that are calibrated and report to the new standard will meet the ISO 11171 calibration standard.

The above came from ISO Code Info.

I found it enlightening as I had no clue what the three numbers were in the particle counts vs. the two numbers expained in the kleenoil link provided by Filterguy.
 
Winston, have you considered not using the "code" and applying the actual data? You can throw out anything below 15um ..or maybe 10um. How would that appear in comparison to the above table/graph
confused.gif
 
OK, here is another graph. I have plotted particle counts at 10u vs miles and 15u vs miles. Then I drew a line to represent the average slope of the data. From this graph you could interpret that a filter that had fewer paticles that the "average" would be a good filter and if it had more particles than the "average" it would be a bad filter.

I was not able to find an easy way to lable the data points, but the good filter is the Motorcraft and the bad one is the Fram X2.
grin.gif


 -
 
quote:

Winston, have you considered not using the "code" and applying the actual data? You can throw out anything below 15um ..or maybe 10um. How would that appear in comparison to the above table/graph

Whoa there. Why exclude the 5 microm rating? That is where the motorcraft really shines!!

Cool posts Winston. The first graph kinda skews the common thought that longer filter use means better filtration. That is, unless they go into bypass because they are plugged (which I doubt is what happened with X2 or the Pure One). Do another graph with the five micron data included...PLEASE.
cool.gif
 
So, I plotted miles vs Iso Code and there seems to be a pretty strong correlation. To me, this data does not show which filter is better. It only shows that the oil gets dirtier as more miles are added to the oil/filter.

 -
 
There is no ff filter that is rated for 5um nominal (50%). Few are rated for 10um nominal. The higher grade brands (wix, PureOne) seem to be about 15um nominal. M1 is probably a bit better. Any 5um data will probably have more to do with the engine and service duty then it will about the filter. You will be falsely assigning worth to something that the filter is incapable of having much, if any, effect on.

Here is a bypass filter offered by Wix. It has somewhere between a 10um and 12um nominal rating.

Part Number: 51660
UPC Number: 765809516601
Principal Application: Volvo Trucks, John Deere 744E
All Applications
Style: Spin-On Lube Filter
Service: Lube
Type: By-Pass
Media: Paper
Height: 10.365
Outer Diameter: 4.282
Thread Size: 1 3/8-16
By-Pass Valve Setting-PSI: None
Beta Ratio: 2/20=12/18
Nominal Micro Rating: 10


Gasket Diameters
Number O.D. I.D. Thk.
Attached 3.967 3.587 0.340

This filter will leave half of the 10um particles in suspension in the beta test. There is nothing under nominal for a beta ratio ..since you can't have a 1-1 % ..only the size of the nominal particle can vary.

If you rescan the posts ...I believe that there is no MC filter. Allegedly, according to Winston, Michael put Mobil 1 Extended Performance PC under the lable of MC.

[ March 03, 2006, 04:27 AM: Message edited by: Gary Allan ]
 
quote:

If you rescan the posts ...I believe that there is no MC filter. Allegedly, according to Winston, Michael put Mobil 1 Extended Performance PC under the lable of MC.

Gary, Michael corrected me. I reread the posts, the test data was for a Motorcraft filter. I was confused because the tester had just put on a Mobil 1 Extended Performance filter, but the test data was from his old Motorcraft filter.
 
If there is no Motorcraft filter in this study, then what the heck are we talking about?

I understand the beta ratio info well after reading articles posted by the National Tribology Service, Inc. website. Very good information there.

If you want to throw out the five micron data...be my guest. All aftermarket filters "meet or exceed" manufacturers recomendations. My quest is to find the "better" of that bunch and use it.

I contend the following, feel free to correct me if I'm wrong (like I need to give permission):
wink.gif


First, it is my contention that just because a filters "nominal" rating is at say 19 microns (which is the majority of the wix brand per their website), doesn't mean that it will not filter anything below the nominal rating.

Second, there are filters out there with a beta 2 ratio at or below TEN microns. Amsoil (seven microns), Baldwin B2-HPG (beta 2 = six microns), and M1 now claims beta 125 at 10 microns (yeah right), the Purolator numbers are out of sight as well.

Third, if the NTS website has an ISO 4409:1999 "goal" set for a diesel engine of 17/16/14 (that is at 4/6/14 microns per the new standard), then a gas engine should meet or exceed this cleanliness level "at the least" IMHO.

Forth, ALL of the engine/filter/oil combinations I believe met this ISO 4406 "cleanliness" criteria in this particle count comparison. However, again, which ones did "better." Per the NTS website, the cleaner the oil the longer the machine will last.

I find it interesting that the Trasko filter particle count that was run for 11,000 miles was left out. ISO code 14/10 (5/15 microns) which kicks ... compared to all the others and it was run LONGER than all of the others. The trasko is only supposed to be run a maximum of 10K miles per their website...so I wonder what the code would have been at say, at 9K miles or 8K as milage DOES seem to matter.

Fifth and last, the ONLY way to have any real data is to have multiple particle counts taken from different filters on the SAME engine (and hopefully the same level of ingested contaminants?). Also (per Winston's graphs), it appears, at or near the same mileage mark.

Bottom line....we need MORE data. Poney up for those particle counts. Yes, expensive, but it is the only way we will know with any degree of scientific basis.
 
quote:

Cool posts Winston.

Thanks.
smile.gif


Compliments will get you somewhere.

Graph with 5,10 and 15u.

 -


I like making graphs for data. It is often much easier to determine trends with a properly created graph. Especially, when you have logarithmic relationships, etc.

As far as the 5u data being relevant; I think that the pores in the filter media vary in size, so even with a nominal rating of 10u, a filter will filter some 5u or even 2u particles.

I hope we get some more particle data! We really need more to make any real conclusions.
 
If there is no Motorcraft filter in this study, then what the heck are we talking about?

Winston corrected his assertion. I reread beyond where he said that there was no MC filter. That's what I was talking about.

I understand the beta ratio info well after reading articles posted by the National Tribology Service, Inc. website. Very good information there.

If you want to throw out the five micron data...be my guest. All aftermarket filters "meet or exceed" manufacturers recomendations. My quest is to find the "better" of that bunch and use it.

As would we all. At least on a cost/benefit basis

I contend the following, feel free to correct me if I'm wrong (like I need to give permission):
wink.gif


First, it is my contention that just because a filters "nominal" rating is at say 19 microns (which is the majority of the wix brand per their website), doesn't mean that it will not filter anything below the nominal rating.

This is surely true. However, can you find a linear, expotential, or factorial relationship between over 10um particles and under 10um particles?

Second, there are filters out there with a beta 2 ratio at or below TEN microns. Amsoil (seven microns), Baldwin B2-HPG (beta 2 = six microns), and M1 now claims beta 125 at 10 microns (yeah right), the Purolator numbers are out of sight as well.

These numbers are surely manipulated for marketing purposes. It's not too much better then the battery industry performance ratings. There's no standard to the standard.

Third, if the NTS website has an ISO 4409:1999 "goal" set for a diesel engine of 17/16/14 (that is at 4/6/14 microns per the new standard), then a gas engine should meet or exceed this cleanliness level "at the least" IMHO.

Forth, ALL of the engine/filter/oil combinations I believe met this ISO 4406 "cleanliness" criteria in this particle count comparison. However, again, which ones did "better." Per the NTS website, the cleaner the oil the longer the machine will last.

Has anyone disputed that the cleaner the oil the longer the engine lasts? I actually think that it would be better stated that the dirtier the oil is ..the shorter an engines life may be. I think, just due to your own assesment of filter performance in this discussion, we're talking various states of what is considered "clean". There's not a dirty one in the bunch. Or is the standard rather "liberal" and allowing a vast distiniction beyond its scope of "authority and validity"??

I find it interesting that the Trasko filter particle count that was run for 11,000 miles was left out. ISO code 14/10 (5/15 microns) which kicks ... compared to all the others and it was run LONGER than all of the others. The trasko is only supposed to be run a maximum of 10K miles per their website...so I wonder what the code would have been at say, at 9K miles or 8K as milage DOES seem to matter.

Well, this is a comparison of popular ff filters. PC has shown to be substantially lower when bypass filtration is introduced. Oddly, elemental wear doesn't appear to have any direct relationship to PC where bypass filtration is concerned. This is what I base the transparency between soot/insolubles/
The point is that the under 10um is speculative and variable from engine to engine. An engine can have a wonderful filter and produce lots of soot/insolubles that will pass seamlessly through the filter. So just how much did any one filter incidentally trap? 10% ==12 ..22.68%. What % of what number?
confused.gif
Conversely a emissions clean engine may produce very little ..and still capture the same 10-12%. Sure you can cite the need for multiple PC on the same engine, but over the span of miles that we're looking at ..seasons change ..states of tune..service duties ..etc. These are things that will impact the 10um figure can be unchanged.



But, by all means, choose any criteria you want for determining what suits your needs.
smile.gif
 
What scares me is the tendency to put so much credibility to pc analysis. Years ago I was responsible for a power plant lab that ran pcs on turbine oil and hydraulic fluids. These numbers can jump all over the place, you can resample and run the numbers over and over and sometimes have difficulty believing the numbers, and that is with trained technicians taking the samples. Just touching the sample valve can mess up the sample. Now we are comparing numbers of samples taken by different people in different locations under various conditions and thinking they are comparable. Not to be a nay sayer but I guess I am. It is interesting to compare them, but some care should be taken in how 'absolute' these numbers are. I commend those trying to make sense of all the numbers and doing all the work though. I don't know how the samples are tested here, but I don't think we could have tested motor oil due to the opacity of the sample. Oils we tested were relatively clear, dirty motor oil wouldn't allow the light of the testing device to pass through the sample....
 
Third, if the NTS website has an ISO 4409:1999 "goal" set for a diesel engine of 17/16/14 (that is at 4/6/14 microns per the new standard), then a gas engine should meet or exceed this cleanliness level "at the least" IMHO.

Errrr...this was proposed before 1999 because Diesel engines had more soot in the oil due to the combustion process. Not only were oil formulations required to change but engine OEM's were required to find ways to eliminate soot from the oil. Of which..as one could expect.. no consensus by the engine manufacturers as to what is the best way to trap the soot.

Last I knew..gas engines don't have a soot problem..
wink.gif
 
I know regular readers of this forum know what a micron is and what a beta ratio is.
A micron is .000039 or 39 millionths of an inch.

When you see Bx=2, this is can be broken down to:
B --Beta Ratio
x --the micron as a unit ( example 10)
2 --a number assigned that relates to efficiency.

The corresponding efficiencies generally used are:

2--50% efficient
20--95% efficient
75--98.7% efficient

The above are the standards ones used for oil filtration.

then there are:
200--99.5% efficient
1000--99.9% efficient

The above two rating are used primarily for Hydraulic fluid filtration.

So one can see B10=2, B10=20, B10=75, B10=200, and B10=1000.

All of those are different ratings for a 10 micron Beta rated filter.

To put things in perspective, Suppose you had 1,000,000 particles 10 micron in the fluid.

B10=2....would have a particle sample count of 500,000 in the fluid at test sampling.

B10=20...would have a particle sample count of 50,000 in the fluid at test sampling.

B10=75...would have a particle sample count of 13,000 in the fluid at test sampling.

B10=200..would have a particle sample count of 5,000 in the fluid at test sampling.

B10=1000.would have a particle sample count of 1,000 in the fluid at test sampling.

Hope that helps...
patriot.gif


Now that people are discussing particle counts of used oil analysis, anyone want to determine the dirt holding amount and the # of particles actually trapped by the oil filter?
grin.gif
 
Gary Allan
quote:

Well, this is a comparison of popular ff filters.

Well then. Maybe the Trasko should be more "popular."
lol.gif
Anyway, I don't own one, but I'm thinking about it.

quote:

Oddly, elemental wear doesn't appear to have any direct relationship to PC where bypass filtration is concerned. This is what I base the transparency between soot/insolubles/
Please define "elemental wear." In this NTS article it talks about "particulate and water contamination, abrasive wear, erosive wear, adhesive wear, and fatigue wear." Is "elemental wear" defined as wear cause by particles?

quote:

The point is that the under 10um is speculative and variable from engine to engine. An engine can have a wonderful filter and produce lots of soot/insolubles that will pass seamlessly through the filter. So just how much did any one filter incidentally trap? 10% ==12 ..22.68%. What % of what number? Conversely a emissions clean engine may produce very little ..and still capture the same 10-12%.

I guess that is why NTS has a chart with varying ISO cleanliness code "GOALS" for different "machines/components." Clean (or good enough) for one application is not clean (or good enough) for a more "sensative" (meaning tighter tolerances) application.

Are you saying that particles under 10 microns do not matter? Or are you saying that UOA particle counts under 10 microns are meaningless? Or are you saying becuase popular FF filters don't filter below 10 microns that the values below 10 microns are worthless? Just what are you saying?
wink.gif


quote:

Sure you can cite the need for multiple PC on the same engine, but over the span of miles that we're looking at ..seasons change ..states of tune..service duties ..etc. These are things that will impact the 10um figure can be unchanged.

Do you think that the above uncontrolled variables will have no effect on particles greater than 10 microns? I guess that could be true if the filter is working properly. Yet, I find it disturbing that the particle counts GREW with milage on the filter instead of DROPPING due to increased effeciency of the filter over time. Granted, this is a very small, eclectic sample. Yet it makes me wonder about the assertion given the uncontrolled variables of engine tune etc. over time and its effect on FF filter performance at any micron level.


Again, I contend that if ALL of the UOA done and posted here on BITOG had particle counts we would defenitly have a much better idea of which filters were better. We are not filter manufacturers. We are not privy to their testing methods. So, what to do? The complexity of the situation is such that all we can do is find trends of cleanliness for each specific application (engine) with a given filter.

Yet, in oder to have "trends" you need to have data. LOTS and LOTS of DATA over YEARS of time. Or, the filter companies could standardize the testing proceedures and then we would know.
rolleyes.gif


I appreciate the studies done by those who have cut open filters so we could look at their innards. I have now cut a few open myself. Because of this, I now do my best to use filters that have a "front side" or "thread side" filter bypass instead of the dome end bypass. Even to the point of using a filter for a different application to obtain this feature.

Now I've moved on to the actual filtering "quality" of the filter in addition to its physical structure "quality."

Here we are talking about what is "clean" when as stated before, they technically all "passed" the cleanliness test. They were all "good enough," just some "gooder than others."
cool.gif


Thanks for the thought provoking exchange.

John K,

What would you have us do? If not particle counts...then what? I really don't have the time, money or expertise to tear down engines at will to check their wear status.

Winston, Thanks for the additional graph. Again, cool, because it does show "trends."
 
quote:

Now that people are discussing particle counts of used oil analysis, anyone want to determine the dirt holding amount and the # of particles actually trapped by the oil filter?
grin.gif


lol.gif
lol.gif
lol.gif


Just one baby step at a time Filter Guy.

Can anyone say...oversized filter? How about remote filter mount?
grin.gif
Will it every end?
shocked.gif
shocked.gif
freak2.gif
 
I guess that is why NTS has a chart with varying ISO cleanliness code "GOALS" for different "machines/components." Clean (or good enough) for one application is not clean (or good enough) for a more "sensative" (meaning tighter tolerances) application.

I don't quite integrate how this allows you to determine where b>

Are you saying that particles under 10 microns do not matter? Or are you saying that UOA particle counts under 10 microns are meaningless? Or are you saying becuase popular FF filters don't filter below 10 microns that the values below 10 microns are worthless? Just what are you saying? [Wink]

Did I ever say that 10um abrasives or insolubles. 10um particles that the 10um abrasive wear. You can attribute them to other origins. This is rather a simple concept.

Do you think that the above uncontrolled variables will have no effect on particles greater than 10 microns? I guess that could be true if the filter is working properly. Well, you got that point. We're getting somewhere here.

Yet, I find it disturbing that the particle counts GREW with milage on the filter instead of DROPPING due to increased effeciency of the filter over time. Then plan on a future of much "disturbance" if you are reading PC. Why would you not expect an increase of particles that are only trapped @ 50% rate as you accumulate mileage?? Looks like simple math to me
dunno.gif


What would you have us do? If not particle counts...then what? I really don't have the time, money or expertise to tear down engines at will to check their wear status.


Hey, pal ..that jeep PC on the end there is MINE. You somehow have integrated that because I said that you can throw out the statistic that has little to do with the filter ..and is an indicator of OTHER CONDITIONS ..that I said that PC was worthless ..when the only truth to that is in YOUR interpretation of my postings.


Thanks for the thought provoking exchange.

It's been a pleasure.
grin.gif
 
"Are you saying that particles under 10 microns do not matter? Or are you saying that UOA particle counts under 10 microns are meaningless? Or are you saying becuase popular FF filters don't filter below 10 microns that the values below 10 microns are worthless? Just what are you saying?"

I've always been told..brainwashed...by the filter industry that Engine manufacturers claim that particles in the 5-25 micron range are what causes the most engine wear.

Full flow filters do filter particles below 10 micron. Just not as efficiently as they "could" if they had better media.

The problem with having a --for example-- 5 micron beta 200 element in a spin on automotive can is that either the filter can would have to grow in size because the filter removes particles at a greater rate or the by-pass would open more often due to restriction as the filter plugs. Or you would have to change the filter much more often. As it's hard to get consumers to follow 3,000 mile change intervals, imagine asking them to be more alert to changing a filter sooner.

Now we sell, at work, a 3 and 6 micron spin on filter Beta 200 rated..it's 5 1/2" in diameter and 11" long...
shocked.gif
Would look good spun on your Corvette...
lol.gif
 
SWHeat - I don't have the answer, I simply wanted to point out that there are a lot of variables and just the one of taking the sample can be a daunting one. I just don't think people have any idea of how much the data can be messed up just in taking the sample. Surprisingly, the data shown 'seems' to be pretty good.
 
quote:

SWHeat: I guess that is why NTS has a chart with varying ISO cleanliness code "GOALS" for different "machines/components." Clean (or good enough) for one application is not clean (or good enough) for a more "sensative" (meaning tighter tolerances) application.

Gary Allan: I don't quite integrate how this allows you to determine where
Huh? The only assertion I've ever made is that the less particles the better, the cleaner the oil the better. I back this up with articles from the NTS website. I really don't know how the particles got there. Does it really matter? I don't think the filter cares. I think there is a misunderstanding here. Filter Guy just illustrated what what is stated in the NTS articles in an nice illustrative manner. I get it, really.

quote:

Then plan on a future of much "disturbance" if you are reading PC. Why would you not expect an increase of particles that are only trapped @ 50% rate as you accumulate mileage?? Looks like simple math to me.

Gary, before I say anything more...let me just say that I've been reading on here since May 2005 and have always been impressed with your responses and vast knowledge database. Now having prefaced my remark....I find no "simple math" involved in filter loading. If the filter is loading (plugging) it should be becomming more effecient (even for smaller particles below its nominal filter rating..am I wrong here filter guy?) and should show cleaner oil. The only other factor being the filter going into bypass in which NO filtering is happening. Complex I tell you. Nothing simple about it.
grin.gif


quote:

Hey, pal ..that jeep PC on the end there is MINE. You somehow have integrated that because I said that you can throw out the statistic that has little to do with the filter ..and is an indicator of OTHER CONDITIONS ..that I said that PC was worthless ..when the only truth to that is in YOUR interpretation of my postings.

I know it is yours. I actually think the purolater filter did well (as said before...it "passed" the ISO cleanliness test). I don't think you understand my postings. True, sometimes I don't know what your are trying to say as I don't have your knowledge database. Hence, the exchange. No personal attack made here.
cheers.gif
If the jeep had "OTHER CONDITIONS" maybe the filter was kicking into bypass more. Who knows. More data I tell you! Another UOA on the jeep with a PC with less miles with the same filter would answer many questions. What say you?

I still contend that the filter doesn't care if it is on a new engine or an engine with "other conditions," it is just supposed to filter out junk.

Filter guy said:
quote:

The problem with having a --for example-- 5 micron beta 200 element in a spin on automotive can is that either the filter can would have to grow in size because the filter removes particles at a greater rate or the by-pass would open more often due to restriction as the filter plugs. Or you would have to change the filter much more often. As it's hard to get consumers to follow 3,000 mile change intervals, imagine asking them to be more alert to changing a filter sooner.

Just what Ralph Wood has been saying all along. If you have an effecient filter it either needs to be huge, or changed more often. Now that is a simple concept I believe we can all agree upon. What "clean oil" is..... well, you know.
 
Status
Not open for further replies.
Back
Top Bottom