So to start with you have a clear glass slide with two closely spaced magnetic strips that run down the middle (longitudinally) of the slide. The oil sample is mixed with a dilution (to reduce viscosity) and the slide is placed on a mount at a downward angle. The diluted sample is then precisely poured at the top of the strips and allowed to flow over the length of the slide into a drip container. Ferrous particles that are present in the oil (wear content) get collected by the magnetic strip along this path. At the top of the slide you will pick up the largest of the particles and as you get further down the slide the particles will be smaller and smaller. As long as you have ferrous material, you will also catch other non-ferrous metals and contaminants. Once the sample is poured, the slide is allowed to settle/dry. At this point the slide is ready for analysis.

Using a high powered microscope you can view the length of the slide to see what type of particles have been collected. At this point it will take a skilled analyst to interpret what is actually going on with the unit in question. Based on the appearance of the particles themselves, size, color and shape, you can determine types of wear conditions; rubbing wear, cutting wear, abrasive wear, corrosive wear, etc. Depending on the type of component you can also make recommendations on how to remedy the wear condition.

The images you see are particles collected from the oil viewed on the slide itself. The left two images are close-ups of the fine wear particles with a few larger particles. The top right shows a further away view of the fine, almost dust like appearance of the particles at the edge of the slide. Bottom right shows a close-up of some of the bigger particles present as a result of sliding and rubbing wear. These bigger particles, if left to swim around in the diff, further wear gear teeth they come in contact with.

Ferrography is often used in components that contain thousands of GALLONS of oil. Or components that, if they fail, can cost millions of dollars in downtime. Our little one quart differentials are hardly what this type of analysis is intended for, but it is still interesting to look at!

 

For those interested, I have done ferrography images on my BMW 325i track car engine oil in the past. This report shows some of the recent history from this car but please ignore the lube times and lube type as they are not accurate. I have been bad in the past about keeping accurate track of history. The bottom row shows the sample I am going to refer to. I think the sample is going to have around 2000 miles and is probably Mobil 1 20w-50 (the 50 is accurate in the history). This would have included a few autocrosses and one track day.

The most recent sample shows an increase in wear across the board, most notably the Cr which is from Piston rings. This report should have actually been a severity 3 with that kind of ring content but I backed it down to a 2 because, I don't know, I was in denial? Note the low Si content.

Because of the elevated wear, I made a slide and looked at it up close. What I saw was an eye-opener, dirt. Lots of large dirt particles that were causing abrasive wear that did not show up on the ICP metals results. (the particles were too large to be detected with inductively coupled plasma)


This image shows the large dark dirt particles and rubbing wear.


This image shows more dirt and some Aluminum alloy particles and rubbing wear.


This last image shows more dirt and some of the larger metallic wear particles including a giant Copper particle.

This car has a high beam removed with a metal screen in place of it, directly behind that is the air intake. The idea is that it acts as a cold air intake. This embarrassing mid-spin pic shows what I am talking about.


In addition to that, my K&N filter was found to be dry, it needed to be oiled. So I had direct air flow to a filter that was not capable of proper filtration. This is what I believe was at fault in this case and the result was increased wear.

The lesson learned here was to keep your K&N's oiled!

 

looks like a sandbox. But silicon is low. Isnt that the argument people are always using for how bad kn is? or is that low because of the time in service?

 

K&N's work great, as long as they are oiled, this one had gone too long and was "dry".

 

I was just expecting to see higher silicon values since an integral part of the filter was not there

 

The large particles shown on your ferrography slides surprise me. I would expect the engine oil filter to trap particles this large. Do you think they just settled to the oil pan instead of circulating?

So far I've used OEM PCX-003, K&N, and AFE Pro-Dry air filters on my S2000 with a stock air box. I've also done UOA's with each of them installed. I have never seen an increase in Si ppm. They all seem to fall into the fleet average of around 10ppm. There was some elevated wear using the OEM, which might have an assoiation to Si not seen on the UOA. But I leave those considerations up to Terry Dyson to interpret for me. I've gotten away from K&N air filters based on his advice. He's seen enough to recommend better options IMO.

 

I was a little hesitant to post these results because I can't fully explain why the dirt shows on the sample but not on the ICP. There was an increase, but not even a flaggable levels. The resulting wear did show up however.

From my experience and talking with some of my co-workers is that K&N's are fine, as long as they are oiled. If your running a dry K&N, you are not filtering properly and you are wearing your engine. As far as oil filters go, I am always weary of using non-oem stuff. Even if the filtering quality is better with the aftermarket brands, fit and function is what I worry about. Check out that thread over in the racing forum right now about filters backing off, scary.

My lab does offer filter debris analysis...lol, but no way. It's like a 3 day process that is pretty labor intensive and VERY expensive. As far as I know we have never done it on a passenger vehicle filter before, only big industry pays for this kind of thing. Maybe someday, if I want to devote an entire weekend to it!

 

What is the oil consumption rate on your Honda BirdShot? Did you add any oil during the 2000 mile OCI? If so, what was it?

 

I had no measurable loss of oil on the stick, checked I think 3 times over the 2000 miles although I forgot to check it at the time of the oil change.

 

150C / 302F sump oil temp would - for me - be a huge concern.
I'm not saying oil can not reach that temp in an engine but IMO not in the sump and not for a prolonged time.

My point about the HTHS was to show the "protection" differences betwen 20-30 and 40 weights.
The thicker oil will not always have the highest HTHS at 150C.
The thicker oil will however alwas have the least oil flow at sump temps of 100C / 212F.
Higher oil flow = more fresh / cooler oil to the journals.