Tachyon_flux

Hi RDX pals,

Bone stock RDX family car with 36300 km (22600 miles) with up to date maintenance as per the MM. Recently did an oil change. Had 11300 km on the filter (7000 miles) and decided to cut it open just for fun. This is the 4th oil change for the car but my first time cutting open any of the filters. I found a few metal flakes, not more than 10. The vast majority were so small that unless you had just the right angle in bright sunlight, you probably wouldn't see it. 1-2 flakes were unequivocal shiny metal flakes, silver color, not copper. The largest metal flake is about the size of a "." if you are reading this sentence on a typical computer screen...I would guess 1/3 or 1/4 of a mm. The oil that came out didn't have any obvious metallic sheen and the car has been driving just fine with nothing abnormal that I can notice. The bits of metal were a mix of magnetic and non magnetic. It's possible that some of the really tiny pieces didn't jump to the magnet because it was embedded in the fibers and sufficiently small that the fibers were able to hold them in place against the pull of the magnet.

Since this was my first time cutting open a used oil filter, I had nothing to compare with and there's all kinds of opinions of what's normal online, everything from there should be no metal at all to a few flakes is normal. Someone on BITOG who seems pretty knowledgeable said that after the break in period, there should be no visible non magnetic metal at all in the filter, ever. His reasoning is that aluminium surfaces are not designed to be worn like that in engines. If there is wear, it will be tiny bits that you can't see with the naked eye. They might form some goop on a magnetic drain plug or something, but there shouldn't be discrete chunks that you can see with the naked eye. Apparently the naked eye can't see stuff below about 30-50 microns, so if you see metal flakes, they are bigger than that which is abnormal.

So, my question for you guys is, how normal is this? For those who have cut open used oil filters, what have you seen and at what mileage (and how carefully do you look). If this was the first oil change at 3000 miles and there were a dozen flakes in the filter, I could write that off as break in wear probably. But at 22600 miles, should I still be seeing stuff like this? Most car manuals say the engine is broken in after 500-1000 miles, but then I see other people saying that for an engine to "truly" break in, it could take 20-40000 miles. That's an insane range of numbers. What does it mean to be truly broken in anyway? When should one reasonably expect to see zero metal flakes in an oil filter and should I be concerned in my particular case?

Thank you for any insights/discussion.

thoiboi

Tachyon_flux

Will try to get a picture tonight. Can’t remember if the bigger flakes came off with the magnet already though.

Tomtwtwtw

I'd recommend sending a sample off to Blackstone for proper analysis vs what the eyeball test might show you. Do it for at least a few consecutive oil changes, as that'll give you a baseline and hopefully identify any possible trends.

Tachyon_flux

Thanks Tom. Next oil change will be in a month or so and I do plan to send for a UOA. I've read on BITOG that UOA doesn't necessary detect everything or predict failure. Lots of stories about cars with perfect UOAs but have destroyed bearings or suddenly seized up. I guess you can have large chunks of metal without super tiny particles and UOA only detects stuff within a certain size range.

John from PA

You did an oil change “recently” and “the next oil change will be in a month or so”. What is your change interval?

For your next change, you might consider using a magnetic drain plug like that shown at Magnetic drain plug

Tachyon_flux

Thanks John. The "recent" change was actually a couple months ago, I just didn't have a chance to dispose of the filter. We have a road trip coming up next week and I'll probably change the oil shortly thereafter. I anticipate the car will be just over 5000km/3000miles and the oil life will be at 50% on the MM. I used to change the oil when it got close to 0% on the MM (last filter had about 7000 miles on it) but read recently about possible oil dilution issues for our engines. So this time I'm going to do a change earlier and send that off to check the fuel dilution. The oil level on the dipstick looks fine but the oil does have a really strong gas smell, though I know that's not a good indication of actual % dilution and all DI cars will have some gas smell in the oil.

RDX-Rick

My first thought was, how do you cut open an oil filter without creating metal shavings? I'm picturing some type of saw being used, perhaps a hacksaw or a cutoff wheel on a Dremel tool. Perhaps tin snips where used!

Tachyon_flux

No, there are these special filter cutting tools. Do a quick search on Amazon or YouTube. Some bits do make it into the oil but usually they are curved slivers instead of flakes. But to be safe, I only look at metal in the filter media itself.

Tachyon_flux

I hope this shows up ok on screen. These are the biggest particles. Others are so small my phone either doesn’t pick it up or it just looks like filter media fibres reflecting light.

RDX-Rick

😯😲I never would have guessed such a thing existed. I will pick one up next time I plan on opening a filter👍.

Thanks.

Tachyon_flux

They are like $30. It’s a good tool to have. Some people just use big ass pipe cutters.

russianDude

I hope you are using 5w-30? 😀

Tachyon_flux

Not yet. Will wait for UOA results to see if there is bad fuel dilution. I will post the results when I can.

russianDude

just so you know, owner manual in Mexico for RDX 2019 has 5w-30 as one of the choices. Its on honda.mx website, there is a link somewhere for manual download

Tachyon_flux

Yup I know thanks. We talked about it before in another thread.

russianDude

Btw, Blackstone is a horrible place to measure fuel dilution, but you can still use their viscosity numbers at 100C to see how much viscosity is lost

Tachyon_flux

Yes I know they infer it using flash point depression. I am sending it to get analysed by gas chromatography.

attofarad

I sent samples to both Blackstone and OSI for analysis. Blackstone reported very little fuel dilution, and OSI reported more than 8%. That 8% is ludicrously inaccurate, as the viscosity had barely dropped from virgin oil value. The Blackstone level is probably lower than the true value. Both services got virtually the same viscosity result. I just change the oil when the MM hits 50% and don't worry about it.

I wouldn't worry about that tiny metal fleck in your filter. On my other car, with a cartridge filter so I don't have to cut it open to look at it every time, I probably had that much on my third oil change and nothing that I detected on the 4th. The first oil change had a couple of shavings is large as half millimeter by 2 or 3 mm and maybe 20 flecks like your fleck.

Tachyon_flux

Thanks Attofarad. May I ask, do you know how OSI measures fuel dilution? Is it with gas chromatography or some other method? My understanding is that GC should be the most accurate.

russianDude

What is OSI? I use Oil Analyzers, which uses Polaris lab and does fuel dilution using GC method that is accurate

attofarad

Yes, this was Oil Analyzers Inc also, GC method. I had misremembered it as OSI. The result that they gave me was 8.7% fuel dilution, and 7.8 cSt viscosity (center of range for 0w-20, but lower than starting value).

russianDude

I dont think they show above 5, they simply say its greater than 5%

attofarad

​​​​​Mine said 8.7%.

RDX-Rick

The recommended oil is 0w-20. What is the advantage of changing to 5w-30?

I have heard it said many times that higher viscosity oil provides better protection, and car manufacturers recommend lower viscosity oils, only to increase fuel economy ratings. But if that was the case, wouldn't better fuel economy also indicate less friction between moving parts?

Or is your recommendation based on the idea that starting with a thicker oil, will result in the recommended viscosity after fuel dilution takes place?

russianDude

Thicker oil protects better under load and higher temperatures. But if that is not enough for you, the 5w-30 will louse viscosity thanks to Honda Turbo GDI fuel dilution and will actually come out as Xw-20 viscosity, thats of course if you do UOA or do some reading about Honda 2.0t fuel dilution.
secondly, 5w-30 is allowed in mexico for the same car, just Mexican version of the manual.