Repair issues and fixes

It's really good to have new data like this at 142 km and the earlier one at 60,000 km (notably with no magnetic plug). The puzzle pieces are starting to fall into place.

The above photos even add the information that the oil was essentially clear from the start as in no prior photos have we ever been able to see the bottom of the container through the oil.

From a statistical perspective, the fact that essentially 100% of our very small sample of about 20 owners out of over 100,000 cars on the road have found black oil indicates to me a strong confidence level in this being caused by a design flaw rather than a random defect only affecting some examples. We would never have been lucky enough to randomly hear about only defective cars, as with the usual phenomenon that owners with car problems are the ones most likely to post in forums.

As such, when we see evidence from only one owner at 142 km and another at 60,000 km I feel that this new data has a good chance of applying to the entire fleet as well. If adding a magnetic plug from new turns out to prevent the wheel-of-misfortune noise, imagine the money Hyundai could have saved, to say nothing of the unnecessary stress caused to affected owners.

 

Changing the oil early and often initially is good. But that sounds more like treating a symptom. What is the problem that is causing the oil to go so black early on?

I doubt we will ever find out, as manufacturers don't like to admit a design flaw, and if there is a fix, they won't tell you what it is/was either.

 

As brief as I can describe it, steel particles normally shed from gear meshing need to be quickly removed while they are relatively large and undamaged. The longer they remain in circulation, the more they get beaten down in size and become much harder or impossible to catch because they become far less magnetically-attractive.

The particles also turn black from the momentary heat as an unsuspecting gear tooth or ball bearing crushes them to dust, the latter taking some damage in the process. The problem is not so much that the oil colour becomes black, but that it's indicative that particles are remaining in circulation. This process is well-understood in the field of machinery lubrication.

For whatever reason the stock internal magnet is not doing a good job and the evidence of that is the black oil. When we use a better magnet in perhaps a better location we will hopefully find out that the end result is less black oil, indicative of better particle sequestration. Soon, we will have evidence to point us in one direction or the other.

Yes, changing oil is "treating a symptom" but adding a much better magnet in a possibly-better location is an attempt to address the root cause.

Carrying on, as pure speculation on my part, at a certain higher level of particle content a runaway (snowball) effect can happen as damage to bearings and gears releases even more particles. That may lead to bearings jamming, outer race spin, on to added housing wear putting aluminium in the oil, a looser bearing fit and eventually a knocking noise. That's an entirely plausible chain of events but not yet proven to be true.

Hyundai don't need to explain the black oil because that doesn't stop our cars from being drivable. Complaints of noises however seem to have reached a threshold catching the attention of the (US) NHTSA, forcing Hyundai to publicly publish a TSB addressing 'motor rumble' which we have to assume is the same as the tapping noise. The contents of that TSB don't seem to address any of what I've mentioned, other than checking the gearbox for roughness. It remains to be seen if that is addressing the issue or not, and whether they actually understand the problem.

One day a new Kona owner will change their oil on my recommendation then complain right back that it's perfectly clean. That's when we'll know it's been addressed at the factory.

https://www.machinerylubrication.com/Read/794/magnetic-filtration

 

I've attached a photo of a bottle of the oil with a bright workshop LED behind it. As you can see the oil is black not clear after 142km. In my original photo you are seeing the reflection of the top of the container (cut up 2litre milk bottle) in the surface of the oil not the bottom of the container.

Maybe when OzKona does his oil change I give him my oil to get analysed also (share the cost).

 

Oops photo this time ...

 

+1 on 04/06/22, @38 months, 22K km trouble-free, no gearbox noises..
Completely opaque black..

 

Thanks for the clarification, no pun intended! That's an optical illusion that my own oil photo also had, but even knowing that I really did think that it wasn't the case here. I was hoping to confirm that the oil was clear from new.

One thought that occurred to me in the middle of the night is that this is the first time a Votex has been placed in the original oil while it's still in the gearbox. That metal found on the Votex was not generated in 5km, it was produced over the prior tens of kms and was just floating around. The key point here it that if the internal magnet was working as designed the Votex would have found almost nothing. If I'm not incorrect, this actually proves that the existing internal magnet is ineffective.

As you drive there is a steady stream of fresh particles shedding from the gears that are initially ferromagnetic. Particles that are not sequestered promptly will circulate through gears and bearings over and over becoming smaller, darker in colour and less magnetic and harder to catch in the process. In the stock configuration it seems a significant number of particles are not caught and that's why the oil turns black so quickly. The entire purpose of installing a magnet in a gearbox is to catch as many as possible and it's clear that the Votex is far superior at doing that.

I gather your first oil change and no magnetic plug installed? Thx for posting that ...

 

Yes and yes, but a free 2nd flush tomorrow after the technician put in Shell Spirax S6 AXME (GL-5) instead of S6 GMXE (GL-4) - Hyundai oil next...
I’m inclined to the view on an additional magnetic plug, that at 22K km the horse has largely bolted already, and regular flushes will suffice to satisfactorily minimise any impact from small amounts of further contamination, perhaps at 30K and 50K for my light use. DIY’ers can flush and check magnetic plugs as often as their interest allows, but typical owners would just think of the $’s so not get any feedback on plug effectiveness.
This scenario doesn’t seem any different from back in the day - an effective magnetic plug and an early flush sets a gearbox up well for long and reliable service, but failing that, regular oil changes are the best and simplest recommendation...

 

The GMXE has a viscosity of 9.5 @100°C, really getting a bit far off the 5.4 OEM spec anyway. The readily-available Penrite 70W-75, is 7.5, I think the thickest I'd be comfortable with in our climate and maximum road speeds. But the worse that can happen with thick oil is a temporary whining noise. GL-5 is not a good idea simply considering the brass grounding brush we seem to have discovered. Many GL-5 additives can corrode brass.

Certainly the whole thing about having other owners incurring moderate cost on my advice makes me cautious because in the big picture only a very few examples suffer the dreaded tapping noise and many would not recognise, consider important or even care about lessor levels of deterioration such as a higher general noise level.

But I'm also confident that the magnetic plug is correcting a now-proven design flaw and cannot help but reduce ongoing wear. Black oil is a bad sign whether there are tangible symptoms or not. I've actually spent over $500 on this project so far, but for me there is considerable value in the thrill of the game, in an unusual subject I happen (by good or bad luck) to be well-acquainted with.

Yuri, your comment sat in my brain for a few days after I checked that the oil you mention has a nearly identical viscosity as the factory fill. What could be different? Well, in the middle of the night it popped out. It most certainly could be the fine particle content of the oil. In thick oil (e.g. due to cold) tiny particles can trigger cavitation as the particles exit a high-pressure region such as gear tooth meshing. This was explained to me in the '80s by a UK bearing manufacturer I was visiting. This is also possibly why a few of us noticed quieter running after an oil change, a difficult thing to explain any other way.

 

Well, so I finally got around to my own reduction-gear oil change yesterday,
and changed the drain plug to the Votex magnetic one. After draining the case,
I took the opportunity to do a little spelunking with a cheap "endoscope"
camera I have, which was mostly unproductive, but I do think I found the
existing magnet that's been pointed out here. Going into the lower hole and
turning left as tightly as the camera gooseneck would allow, I could see this:



The camera's not good enough to let me really see if there's material stuck
to what I'm guessing is the magnet, the flattish round thing inside the alloy
boss, and I couldn't even determine if that construct is significantly
magnetic or not. I fished a bent paper clip in and over toward the area,
and did not feel it trying to stick to anything. So yeah, not so useful.

The old fluid was definitely dark, and full of *some* kind of metal particles.
As someone else has said, it looked very much like "metalflake" paint. Under
full sun and a little swishing around, the metallic content was clearly visible.
I played around in it with a small magnet at the end of a screwdriver:



Nothing noticeable stuck to the magnet, though. So is it non-ferrous? What
was really interesting was that if I hovered the magnet just above the liquid
surface, the metallic sheen seemed to get pushed *away* from it, making a
darker area right underneath. Not attracting the particles like I'd expect.
Or perhaps it was changing their alignment so they reflected less light. So
here's a puzzle: the particles *did* respond to magnetism in some way, but
did not race to stick to it. What kind of metal are they? Anyway, at a little
10,000 miles, it was clearly time to get that stuff out of there.

Refilling turned into a bit of a disaster. I have a funnel with a long
extension hose, which I can suspend from the hood latch to keep the whole
run going downhill, and after inserting the end in the fill hole, wadded
a rag in next to it to make sure the hose end stayed put. I only took off
the front under-cover, since that gave me enough access to both plugs. After
installing the Votex in the drain I dumped in about half of the replacement
Redline oil, and waited a bit to check for any leakage around the new drain
plug and "crush washer" [which I am loath to actually crush]. Then I dumped
the rest of the quart in, figuring I'd then borescope the upper hole to see
how far up that brought the level. But when I went back underneath, oil was
drooling out of the fill hole and all over the front edge of the second
under-cover. I yanked the hose and got the fill plug back in as fast as
I could, but basically it made a mess. The front of the car was up on ramps,
so clearly even at that slight upward angle, one quart [plus what didn't
originally drain out] is just a little too much. There's probably enough in
there nonetheless, and I figure I'll run this fill for another 10K or so and
then see what that and the Votex magnet looks like.

_H*

 

My second reduction gear oil R&R:

Since I am losing access to my hoist in about 45 days (we are moving), I figured I might as well swap the fluid out while I still had the tools at hand.

The first time I changed the reduction gear oil was on 30 Dec 2021, with 8,776 km on the vehicle, using the Redline GL-4 75W-80W.

At that time, I had the typical "black as molasses" OEM oil drain out of the car.

I created two magnetic plugs using rare earth magnets stuck (with 3M double side tape) to the OEM plugs.

Today I drained the oil and replaced it with the Redline GL-4 70W-75.

These pictures show what 3,576 km of distance has done to the oil that was installed on 30 Dec 2021:

The drain plug tip was covered in black micro-particles, which was rubbed off onto the white rag for clarity as shown:



The fluid was not black:



Another angle of the fluid:



The fluid did have the floating "sheen" (like metal flake, but not attractive to a magnet) that others have mentioned.

Details of my home made drain plug (typical with fill plug) showing the two rare earth magnets attached to same:



The fill plug also had black micro-particles that I rubbed off onto a rag.

It had less volume of the micro-particles than the drain plug did.



Barring any surprises, the next time I do this will be at 50,000 km, which may be in 10 years...

 

So it looks like your external magnet trick worked! Good to know.

_H*

 

A nice surprise to see two oil changes happened overnight. There's a lot of info to unpack!

That's adventurous. If I had placed just about anything inside the gearbox, with my luck I think I'd know what would have happened, lol! I even got my pinky finger stuck for a moment on the sharp edges. Really good info though to know that a paperclip wasn't attracted to Hyundai's idea of a magnet and that is just the evidence we needed. The image looks about what I would have expected and I'm not seeing much if any debris on the magnet itself. It does look the the magnet is a round-edged disk rather than a short cylinder as I had expected but they would have wanted to avoid sharp edges that could chip off.

Frankly, that readily explains the condition of the oil. Fresh steel wear particles lose their original ferromagnetic attractiveness as they are further crushed while circulating and become smaller and blacker. They have to be caught very quickly, and I mean in seconds or minutes of being ejected from gear teeth.

That is essentially the crux of the matter.

That's got to be annoying after all that great investigative work. Yes, the bulk of the gearbox volume is forward of the openings and ramps don't work for the refill operation unless you raise the rear of the car as well. I drive the car off the ramps after inserting the tube and then raise it on the left side only with a jack and jack stand. I pour in a measured amount from empty rather than rely on the overflow then crawl under and fit the plug finger-tight to I can put it back on the ramps. I hope you didn't lose too much oil.

 

That's about twice the distance I drove with the dual Votex setup and your homemade setup still kept the oil just as clean. I think the "sheen" is brass dust off the grounding brush, exactly what I saw. It's presumably a softer material and not as harmful to bearings, IMO, as the remnants of steel that your magnets picked up.
It really does demonstrate the value of magnetic particle sequestration, or "filtering" as the industry calls it. My ICE manual gearbox, transfer and diffs have relatively weak magnets yet the oil stays very clean. Manual gearboxes show bronze particles off the synchro rings and that produces a similar glitter.

 

I'm super-happy with the reports we've had recently and it's gratifying that the data is totally consistent with a defective OEM magnet. Big thanks to everybody who made the effort, especially @hobbit for bravely sticking a paperclip inside the gearbox!

It's clear that adding a magnetic plug of some form is essential but it also seems that the oil only needs an unscheduled change at that time. If the plug was fitted at under 100 km I think even that could be skipped.

Meanwhile my Chinese-manufactured DEEFILL plugs arrived today by DHL, packaged like three diamond rings in retail boxes with a magnetic catch, as they should for the price. Only problem is that one M18 was missing, the only one of the three boxes further packed inside a factory sealed plastic bag. Nothing had been tampered with and it looks like a factory 'mistake' rather than Amazon's. Very odd ...

But it looks like a very nice product and magnet is wicked-strong. The 14mm one is for my ICE engine and is made of the same 22mm diameter SST stock as the M18. So, yes the shoulder is very narrow on the M18 and I'll have to determine if I'm happy with that on my Kona at some point in the future, with an aluminium washer. I doubt it's a problem but it will need reviewing and an appropriate torque spec determined.

Amazon's frustrating automated resolution system said the item was "no longer in stock" (incorrect) but quickly processed a refund. I won't buy a replacement until I'm happy with the application. For initial testing the M18 will sit in a sample of gear oil for a few weeks and the M14 is going in my ICE today for "thermal cycling", lol.

 

Fear not, I retained a good hold on enough of the paper clip wire that was still outside the hole!
If I'd lost it, the borescope includes a little magnet-pickup attachment [that, indeed, I was swishing the
old oil around with].

Those "deefill" plugs look really long, almost enough to risk contact with the diff gear that's right inside
there... or at least be in the slung flow that the gear picks up? I didn't measure the depth from the plug
seating face, though, as it was visually clear that the Votex was fine.

Guess you'll find out when you try to drive...

_H*

 

Where can I buy those rare earth magnets?