Pour most of the jug into the engine first (drain plug installed, lol) and then fill and install the filter last.
Freedom Worx ... video about pre-filling filters
- Thread starter dnewton3
- Start date
After watching the first 3/4 of the video about the pre-filled vs non pre-filled filters test and resulting UOA data, my concussion is the test is inconclusive one way or the other. If you think that one UOA is enough, then notice that the Aluminum and Iron PPM counts went down too between no prefill and prefill. Of course the change difference in PPM counts is so small you could say it's in the noise level ... so really inconclusive. With the amount of new make-up oil used, I think it made the test very insensitive using one UOA to any wear difference that could result from not pre-filling. I think LS Jr's test was more controlled than this one was.
dnewton3
Staff member
I agree the test is anecdotal in that the data is lacking. But the results do correlate with the study information he presented. And the logic he speaks to (regarding the drop in metal counts) aligns with both the results and the studies he discusses. I do believe that because the no-fill filters also necessitated "top off" volume for every change, he was grossly diluting the sump with fresh oil in such short FCIs; that's not hard to understand at all. He should have taken 60 UOAs (30 filters each), but he'd discover that the stdev is large enough that small singular UOA differences would be moot.
Also, the UOAs only come with whole-number return data; that truncates the ability to get better refinement in the analysis. The reality is that pre-fill or not, the data isn't going to distinguish itself in a statistically significant manner.
And that's kinda the point perhaps to take from all this ... If you can't tease out the variable from noise, then the variable doesn't matter. So ... whether or not you prefil makes no difference. And that is the conclusion he did come to, albeit perhaps the wrong path.
As for the very small counts, well, that's what HALTs get you if you don't intentionally magnify the intensity via manipulation of the inputs. (aka the infamous GM filter study ...). It's impossible to have it both ways. You can't get immediate data from long-term tests. And if you run quick HALTs intentionally manipulated, you'll never see the results manifest in reality, because you didn't test "real world" conditions. HALTs are great at proving concepts and at times, showing relative relationships. But they suck at proving anything applicable to real world conditions.
Also, the UOAs only come with whole-number return data; that truncates the ability to get better refinement in the analysis. The reality is that pre-fill or not, the data isn't going to distinguish itself in a statistically significant manner.
And that's kinda the point perhaps to take from all this ... If you can't tease out the variable from noise, then the variable doesn't matter. So ... whether or not you prefil makes no difference. And that is the conclusion he did come to, albeit perhaps the wrong path.
As for the very small counts, well, that's what HALTs get you if you don't intentionally magnify the intensity via manipulation of the inputs. (aka the infamous GM filter study ...). It's impossible to have it both ways. You can't get immediate data from long-term tests. And if you run quick HALTs intentionally manipulated, you'll never see the results manifest in reality, because you didn't test "real world" conditions. HALTs are great at proving concepts and at times, showing relative relationships. But they suck at proving anything applicable to real world conditions.
Last edited:
The test would have been more apt to show more difference in wear if he would have somehow not added any new oil over each test case scenario ... it was 3 gallons of new added oil over each of the 25 filters he ran in each test case. It's just another variable that shouldn't be there for a test like this. What if it wasn't exactly 3 gallons in each case. I think LS Jr did it a different way as to not introduce so much new oil over the test sequence. He might have not added any new oil, can't recall without watching that video again.
I would have put some kind of base adapter on the filter mount (or an appropriate remote mount) with a valve where I could open it and drain the filter and galleries above, or somehow suck out the oil in the filter and the galleries above with a clean pumping system to simulate a filter change. That oil would be pumped into a clean container and poured back into the engine each time. Something like that would not dilute and constantly decrease the particle count in the test oil volume. Of course that would probable cost more money and take more time than what he did.
I would have put some kind of base adapter on the filter mount (or an appropriate remote mount) with a valve where I could open it and drain the filter and galleries above, or somehow suck out the oil in the filter and the galleries above with a clean pumping system to simulate a filter change. That oil would be pumped into a clean container and poured back into the engine each time. Something like that would not dilute and constantly decrease the particle count in the test oil volume. Of course that would probable cost more money and take more time than what he did.
Last edited:
I’m pretty sure the LS Jr video had a different point to prove, and that is that some crazy people claimed the oil in the bottle had impurities and was “dirty” that an oil filter would filter out, hence pre-filling would help with that.
Why would they pick such an obviously idiotic claim to prove it wrong, I have no idea?
You don’t need a test to prove that it’s a baseless claim, and adding oil to the filter, to the return side at that, doesn’t actually filter it.
So clearly, most of these videos are made for entertainment rather than educational purposes IMO.
Why would they pick such an obviously idiotic claim to prove it wrong, I have no idea?
You don’t need a test to prove that it’s a baseless claim, and adding oil to the filter, to the return side at that, doesn’t actually filter it.
So clearly, most of these videos are made for entertainment rather than educational purposes IMO.
I just found and watched that video again. He talked about pre-filling filters, but only had some UOA data to try and support the claim that pre-filling may help reduce wear. They also showed the time it took to obtain oil pressure without pre-filling the filter ... obviously took more time when the filter was empty. The other point of the video was showing that new oil was clean enough to not worry about pre-filling the filter. Of course people should try not to be boneheads and introduce contamination themselves if pre-fillng a filter.
This Worx video was not controlled enough to show one way or the other if pre-filling helps reduce start-up wear or not. There was way too much new oil put through the engine over changing out 25 filters, thereby making a UOA totally insensitive to any real changes in wear metals. If it could be done with no new oil make-up then I'd believe the results more ... so the jury is still out. As said before, nothing wrong with pre-fillng the filter if that makes people feel better about an oil change. I like to see the oil pressure come up in 1~2 seconds vs 5~6 seconds regardless if it's helping or not. Also, if the filter's ADBV is constantly leaking and causing way more time to get oil pressure or engine noises on each start-up (ie, another form of a "dry start"), it would be a good idea to change out the oil filter ... it wouldn't hurt to be cautious.
If it isn't enough to show up after 25 dry filter starts, then it's probably not enough to worry about. Which I felt was the takeaway and why the last section of the video was about what actually contributes to wear in meaningful ways.
Well, if no oil was added then there would’ve been a cumulative effect of wear metals going on, and you would likely see a gradual increase in overall wear metals. So while it would confirm your side of thinking, it would still not be representative.
Ideally, fresh oil would have to be used, but we all know that would be quite an expense. And I have a feeling there would be no difference. The results the Worx guy got were statistically irrelevant anyways. A few PPM is within the measuring error.
dnewton3
Staff member
We have no idea if they were in the measuring error (gauge R&R) any more than we know the stdev from sample of 1.
Plus a “dry filter start” isn’t the same as the mythical “dry start”.
Agreed ... but that wasn't the point of my comment. My viewpoint is the way that testing was done it can't really accurately show the impact of doing 25 starts without pre-filling the filter. Just way too much clean make-up oil was used which diluted the particle counts too much to see small changes accurately. And beside, comparing just two UOAs could be argued to be not so accurate. He also should have sent out samples to 3 different UOA labs to compare results, and the noise in the system would probably not even show agreement between the different UOA test labs.
And in that case if it did show that there was a cumulative effect of wear metal from not pre-filling the oil filter then one could conclude that dry starts do increase wear to some degree ... regardless of how small it is. It either makes zero difference or it does make a difference regardless of how small. And if it does it's going to be a small additional sliver of wear added to the engine over it's life time. I already discussed earlier in this thread how any wear from a bad filter ADBV (or the filter not even having an ADBV) is probably a lot worse in the cumulative long run than not pre-fillng the filter on every filter change over the life of the engine. If an engine is ran to 200K miles and the average filter change is a relatively low 5000 miles interval, then that's 40 filter changes. I never said or even implied that's going to make or break the longevity of the engine. The debate is does pre-filling the filter reduce some wear, regardless of how small.
You assume there would not be a cumulative effect from the pre-filled filters, but it would also be there.
For the UOAs to have any meaning, this guy would have to do 50 of them and with fresh oil each time. Then we would have a large enough sample to determine if there are any trends in the wear metals or not.
The "regardless how small the difference" argument to me is odd. If, and that's a big "if" IMO, the difference is so small and overall inconsequential to engine longevity, why obsess over it?
I can actually use that same logic. The chances of you dropping something in that oil filter, while trying to pre-fill, it are certainly there, no matter how small that chance might be, it's definitely there and could happen. So why take it and risk engine damage?
I could assume there may be more of a cumulative effect difference from not pre-filling the filters. That's why if the test was done without refreshing 3 gallons of oil through the engine it could be easier to see the difference between pre-filling and not.
Starting by doing the test without introducing 3 gallons of new oil would be a good start.
I don't think anyone really "obsesses" over it ... no more than someone who argues that not pre-filling doesn't matter "obsesses" over it. People are either a filter pre-filler of they are not. And as said before, people can and will do what makes them feel better ... simple as that, regardless if it makes little or even any difference.
Like said ... don't be a bonehead when pre-filling an oil filter. It's not rocket science, lol.
Last edited:
- Joined
- Apr 1, 2024
- Messages
- 34
I have been watching this guy’s content for a while and really enjoy it. I thought the wear metals would have been measurably higher on the dry starts, esp after seeing that oil px guage take forever to climb.
In the past I would only fill oil filters if they were A) vertical and B) I was pouring directly from a new oil bottle. Now, I am fully convinced it doesn’t make a lick of difference and probably won’t bother.
Changing your mind with new information is fun, more people here should try it!
In the past I would only fill oil filters if they were A) vertical and B) I was pouring directly from a new oil bottle. Now, I am fully convinced it doesn’t make a lick of difference and probably won’t bother.
Changing your mind with new information is fun, more people here should try it!
Pretty hard to tell when there was constant new make-up oil between every filter change which added up to 3 gallons ... that's like over 2 full oil changes.
- Joined
- Apr 1, 2024
- Messages
- 34
Would you have the same issues with the test if it provided the results you were hoping for? I’m obviously assuming you wanted it to validate your prefilling — in all honesty, I did too.
I'm saying the test methodology was flawed to see any real effect. When the oil if being flushed with new oil throughout the test, it's going to make the oil cleaner and cleaner, so you can't really see the true effect of what the test was meant to try and find. Any test that has multiple variables involved is going to muddy and skew the results. I suggested earlier in this thread how the test could have been done to take out the constant new oil addition to the sump.
- Joined
- Apr 1, 2024
- Messages
- 34
You didn’t answer my question! Just put a filter on dry, I dare you.
Already said there is no conclusive answer because of that test methodology. Go read all the posts I've made in this thread and you'll see my viewpoints. I have put non-filled filters on lots of my vehicles, and I've also pre-filled a lot of them too.
Similar threads
- Sticky
