UOA for Hyundai Kona Electric, high aluminium and silicon levels

Joined
Oct 24, 2008
Messages
1,356
Location
New Zealand
This is not my car although I have the same model but my concerns prompted the owner to carry out an oil change and analysis at only 3,700 km for his 2021 model. The scheduled interval is 120,000 km. I changed my own at 19,400 km on my 2018 model but could not easily obtain an analysis.

The coloration of the used oil is black, almost as dark as black coffee. I'll also mention that this is a splash-lubricated single-speed 2-stage parallel shaft reduction gearbox with steel helical gears and a conventional open diff. All bearings are ball except the output shaft which uses tapered rollers. The housing is cast aluminium. There is no magnetic drain plug installed, nor any internal magnet as best as I can determine from a teardown video. The oil is a 70W GL-4.

The particulate and iron content is no surprise based on the absence of a magnet and I'm relatively certain that's also why the oil is black. This gearbox does not run hot, barely even warm. The aluminium is of great concern because IMO it indicates outer race spin. The silicon level is a mystery. Although the gearbox is conventionally vented, this particular car has not been used on dirt roads.

Technical comments welcome, please!

OzKona's analysis marked up.jpg
 
Silicon is probably an anti-foaming agent and the aluminum is probably from a housing that was poorly cleaned before assembly.

No reason to worry until you have a worrisome trend, and a single data point is not a trend.
 
This is not my car although I have the same model but my concerns prompted the owner to carry out an oil change and analysis at only 3,700 km for his 2021 model. The scheduled interval is 120,000 km. I changed my own at 19,400 km on my 2018 model but could not easily obtain an analysis.

The coloration of the used oil is black, almost as dark as black coffee. I'll also mention that this is a splash-lubricated single-speed 2-stage parallel shaft reduction gearbox with steel helical gears and a conventional open diff. All bearings are ball except the output shaft which uses tapered rollers. The housing is cast aluminium. There is no magnetic drain plug installed, nor any internal magnet as best as I can determine from a teardown video. The oil is a 70W GL-4.

The particulate and iron content is no surprise based on the absence of a magnet and I'm relatively certain that's also why the oil is black. This gearbox does not run hot, barely even warm. The aluminium is of great concern because IMO it indicates outer race spin. The silicon level is a mystery. Although the gearbox is conventionally vented, this particular car has not been used on dirt roads.

Technical comments welcome, please!

View attachment 80359
Document this with Hyundai as these diffs are failing early. They get dirty REAL fast and I have heard of 3 having had to be replaced already. I think these need GL-5 and not GL-4.
Hyundai AWD rear diffs get surprising dirty fast as they only have a hair under a 1/2 Qt of oil. I posted my finding in my post ohn his Youtube channel. look for Ted you cant miss my post.

This video is about the EV diff, by a Hyundai mechanic.



.
 
Yeah, that video is well-known among us owners but since the oil was out of a failed gearbox it wasn't immediately obvious if the dirty oil was a result, or reason for the failure.

He's a good source though and quickly answered my question about the plug torque specs, info not easily found elsewhere. Every Kona EV owner I know who has changed their oil it's similarly black. I personally believe that the black oil is solely a result of the missing ferrous particle filtering, AKA an internal magnet. Other EVs have way more sophisticated gearbox lubrication systems but that's partly because most others use the gear oil as a cooling medium between the electric motor and coolant. The Kona/Niro/Soul/Ioniq use coolant directly.

I don't see any reason why it would need a GL-5 because there are no hypoid gears present, the main reason that gear oil category exists. The Nissan Leaf has an almost identical gear reducer but has two plug magnets. There are a few videos of oil changes and the oil looks normal to me.

As for silicon v.s. silicone surely they are not related? At least Google tells me that. I will note though that one person mentioned that the analysis methods can confuse the two, but I would have thought the report would have mentioned that sort of anomaly.

But I also agree that one datapoint is not a trend and I'm not going to panic about the silicon just yet. I'll just add the vendor's comments even though they are generic and assume that this is a conventional automatic.

1638928096204.jpg

Thanks everyone so far for reading and responding.
 
We've just had a second report come in, this time from a Canadian owner. I haven't looked closely at the numbers yet but it has the same high aluminium and silicon. I'm assuming the odometer is 15,600 km. Unsure what ISO COE is.

Are those numbers going to be PPM, so directly comparable to the previous report?

2nd oil analysis -small.jpg
 
So, in summary I'm going to take control of the particle issue by adding two magnetic drain plugs, see photo. Two oil changes got the oil color back to normal. The high silicon and aluminium are going to be ignored for now because there is no obvious source for the silicon and if the aluminium is coming from where I think it is (second photo), nothing practical can be done about it. Thanks for the comments.

IMG_1625.jpeg


cross section spun bearing.jpg
 
It's only been 2,000 km since I refreshed the oil but I drained it out and to my surprise it's good as new. The magnetic plugs both had particle loading appropriate to the short distance traveled. I filtered the oil through an N95 face mask and put it right back in. Yes, synthetic gear oil is expensive here, this is about $38 worth for 1 litre so I'm not going to waste it if it's clean.

My conclusion is that the factory-fitted non-serviceable ceramic magnet (see photo) is too weak or in the wrong location to properly sequester the apparently-prolific generation of ferrous break-in particles by the gears. I'm going to suggest to owners of new Kona (and classic Ioniq and Kia Niro) EVs that they carry out preemptive oil changes at 2,000 and 10,000 km to clear out wear particles before they are ground down to dust by the gears, nevermind the innocent and unwary ball and tapered-roller bearings.

These useful pics were provided by an Ioniq owner in Poland who rebuilt his own gearbox to replace noisy bearings. I learned from these photos that there is an OEM captive ceramic magnet installed (yellow arrow).

One day we will be tearing apart and repairing EV gearboxes without a second thought. Note also the bronze "brush" on the input shaft tail end. That's to bleed off any voltages that conduct through from the motor shaft, a common issue with 3-phase motors driven from pulse-modulated ("VSD") electronics. The Nissan Leaf uses brushes just like an alternator, but located on the intermediate shaft.

1 with magnet identified.png
 
... the factory-fitted non-serviceable ceramic magnet (see photo) is too weak or in the wrong location to properly sequester the apparently-prolific generation of ferrous break-in particles by the gears.
My latest theory is that fresh wear particles are being blown off the weak internal magnet by oil windage. As particles continue to circulate throughout bearings and gears, over time they lose their affinity to stick to a magnet and make the oil black. The drain plug location for an aftermarket magnet (such as the Votex) is somewhat more protected within the threaded boss.

kona OEM magnet location.png

... The high silicon and aluminium are going to be ignored for now because there is no obvious source for the silicon and if the aluminium is coming from where I think it is (second photo), nothing practical can be done about it.
Furthermore, I now believe that particles jammed in the input shaft bearings are causing intermittent outer race spin which could account for the high aluminium in the UOA. The motor-side bearing is lightly loaded because the splined coupling is helping to support that end via the motor. The outer race is installed as a slip-fit and that's not the industry-recommended fit for light radial loads. It should be a press fit or clamped in place.

The eventual outcome is a knocking due to excessive outer race to housing clearance.

Meanwhile, it seems Hyundai have been forced to post a TSB by NHTSA due to the volume of knocking noise complaints. Reading between the lines, my impression is that Hyundai still have no idea what the root cause is.

EDIT: as owners report getting their gearboxes replaced under warranty I'm suggesting the addition of a magnetic drain plug from the start. I'm expecting that this will avoid a repeat performance.
 
Last edited:
I'll add the latest news today, a Kona EV owner in Boston found that the OEM internal magnet in the gear reducer is non-magnetic by bravely poking a paperclip into the drain opening to reach it. He also photographed the magnet to prove it's there, the black thing in the photo. We all know that these sort of experiments don't alway end well, lol, so I wouldn't try it!

Perhaps, it's a cost-cutting move by Hyundai, but I'm accepting that this is the root cause of the dirty gearbox oil problem I've been chasing down for six months.

We've had a half dozen owner oil changes reported in the last few weeks, about 20 in total since last November. Every owner who had a magnetic drain plug installed returned clean oil, those that didn't all returned dirty oil.

I'm happy with those stats and I'm calling this done.
mboss.jpg
 
We just a diligent new Kona EV owner change their gear oil at only 142 km, barely out of the showroom. As reported, the oil looked tainted black initially then the original clear brown colour party came back a few days later as particles settled out. Placing a magnet on the side of the bottle attracted the majority of the particles but not all. (To reiterate, we've established that the Kona EV gearbox in OEM form has no working magnet.)

In the first photo the bottom of the bottle is not visible, that's a reflection of the side you're seeing. In the second photo it appears that some particles are not drawn to the large magnet on the right.

We know that particles are initially ferromagnetic then lose that property as they get beaten to dust by continual circulation through bearings and gears. In this case most particles seem to be in the early stages of the progression. (In my own Kona, at 19,000 km there was no appreciable volume of ferromagnetic particles left and the oil colour remained black over six months.)

Since the total running time was about 2 hours and well over half the particles were still ferromagnetic, that tells me that an internal magnet has about 1 hour to catch fresh wear particles before they are no longer attracted. Of course those particles will still cause wear to the rotating parts during that time.

Given the high level of turbulence present in a simple 2-stage gear reducer (or any diff), I would (statistically) expect every drop of oil to visit every corner of the housing interior easily within a period of 1 hour, so it may not matter much where the magnet or magnetic plug is located and there's no reason to risk placing it in an area of high oil flow where particles may struggle to hang on.

Meanwhile, I bought a new stainless steel magnetic plug to evaluate, DEEFILL, a brand that just appeared on Amazon. It's looks good quality (yes, it's Chinese made) but might be too long (5mm over the Votex) and stick out into the oil flow, but at least it will clear the diff gear. In any case I'm going to test it submerged in a bottle of oil for a few weeks hanging from the magnet to ensure that the glue isn't softened by the oil.

I had ordered two but amusingly one of the factory-sealed retail boxes was empty! Must be a new marketing idea, here's a new box, use your old plug.

Oil_after_142km.jpg
Oil_vs_magnet.jpg

IMG_1982.jpeg
 
Most gearbox need a good break in drain/refill. Nothing in the UOA is worth worrying about. Normal break-in wear and manufacturer cleaning.

Toyota has some flush magnet drainplugs that can be used. Besides Amazon Deefil/Votex, don't forget about GoldPlug or DimplePlug.

Typical 18mm x 1.5 or M18x1.5? Common to Honda, Hyundai, Kia, Toyota...... But, OE magnets tend to be on the 'weak' side, but still pick up a bunch of magnetic particles. How much clearance is available? Some magnets can be the longer type.

Hyundai/Kia PN's:
0K95B17121
4532439000
0081017121

Toyota Pn's
10mm allen flush-magnet 90341-18021
10mm allen long magnet 90341-18035
24mm bolt flush-magnet 90341-18057 or Ford E3TZ-7A010-A
24mm bolt long magnet 90341-18040

Mitsubishi MB569390

Honda 90081-PX4-003




 
Yeah, at this stage the UOA with no working magnet present doesn't worry me. Later this month one owner will report back with a 10,000 km drain interval with a magnetic plug installed. A good report will confirm that there are no other underlying issues.

Thanks for tracking down those useful part numbers. I've looked through most of them and the shorter ones (Toyota app) would technically work although the Votex DP007 seems to have a slightly longer thread length (15mm) that gives me more confidence given the cast aluminium housing and my comfort level hoping nobody strips one out based on advice I provided! The longer ones would technically clear (30mm thread flange to main gear) but I'm slightly nervous about ceramic magnets hanging out in the breeze even though the Nissan Leaf does this without problem.

Some time ago I thought I had found the perfect Hyundai OEM plug and several owners ordered one from their Australian and Canadian dealers. All reported back that it was labeled as a magnetic plug but the part was not magnetic.
Hyundai 00810-17121.jpg


Another Kona owner did this, stick two disk magnets to the stock plug and it actually kept the oil clean over nearly 3,600 km.
1655254877072.jpg


On open task is to determine a suitable torque spec for the Votex because the OEM plug spec is too high, 33-44. The Votex flange diameter is smaller than OEM plus the supplied washer is thicker and softer than OEM. The number will be between 24 and 30 lb-ft, so I will have to get under the car again to refine that or hope that someone else does it for me.
 
I'll add the latest news today, a Kona EV owner in Boston found that the OEM internal magnet in the gear reducer is non-magnetic by bravely poking a paperclip into the drain opening to reach it. He also photographed the magnet to prove it's there, the black thing in the photo. We all know that these sort of experiments don't alway end well, lol, so I wouldn't try it!...
Huh?
 
The Votex I use is torqued to 35ftlbs in the 18mm thread... not at all worried about gasket thickness or flange diameter. Just torque it to center of spec.

I haven't lost a paper clip in a gearbox yet.... just other things.
 
The elevated silicon combined with aluminium could be an indication of dirt contamination - but they could also be not related to each other ie the Aluminium is a wear material and the silicon could be sealant leaching into the oil. The ISO codes are reasonably high in my opinion confirming that the oil is indeed dirty. To have significant dirt contamination so soon I would think is unlikely, however there's a chance Hyundai are not using the cleanest of oil to start with. What's the breather setup on these like?
Capture.JPG
 
The breather labyrinth is visible in the photos higher up, doesn't look to be unusual. The oil Hyundai use I understand is Ravenol, a German brand. I'm fairly confident it's high quality and it's quite expensive at the dealer, $70+ a litre by several reports. As I've mentioned before the iron levels are likely due to the puzzling lack of magnetic filtering and the aluminium levels I can only deduce come from a ball bearing that is spinning in the housing. There is a plausible failure path that could lead to that as the bearing closest to the motor may not be staying radially loaded as required by the design choice of an outer-race slip-fit. The nearby motor bearing will provide radial support via the spline when under torque. Most newer EVs gear trains avoid this mistake by integrating the motor more closely into the gearbox with just 2 or 3 bearings across, even Hyundai/Kia.

We've had nearly 30 owners report first oil changes from 90 to 100,000 mile and every one has been black. Changes from 1,000 -3,000 miles typically have visible particles while those later tend to have fewer very fine black particles that seem to stay in suspension. I did a test on my Kona with 2x Votex plugs in place over 2,000 km and the oil stayed clean. I didn't have it analysed. We had another owner with the same Votex setup return black oil over 6,000 km and we are awaiting the results of a UOA. He's also starting to hear a slight knock, a common complaint with this model, which usually requires both a new motor and gearbox. I suspect that's the result of the bearing spin but I'm not certain yet. I'm not clear as to whether both aluminium and iron can equally blacken oil at similar contamination levels.

Several observers have mentioned the leaching of silicone sealer and I'll accept that. You can see it bulging out in the monochrome photo above of the "magnet" Hyundai secure inside the housing.

I've bought a Toyota magnetic drain plug and am going to evaluate it alongside the Votex. It cost me $50 at the dealer but I am on an island and anything off Amazon is much the same cost. The reason I like this plug is that I can recommend it to owners globally with less worry than the Votex because the magnet is mechanically secured. It looks similar if not identical to OEM, uses the same 24mm socket and torque value. The magnetic is weaker but in my experience that's not a problem. Several owners are having problems convincing dealers to install the Votex. The dealers don't understand the technical reasons, don't think it will fit, and/or have said it will void any related warranty claims. The stories I hear are almost hilarious, see 2 below. These happen to both be female owners, not that it's relevant to the shocking level of service they are being subjected to.

After changing oil on cars for 100 years it seems all of a sudden dealer's service depts forget how to do this simple task.

IMG_2006 2.jpeg


FB conversation.jpg


This is an update to my June 13th posting regarding my plans to have the reduction gear oil and drain plug replaced in my 2020 Kona Limited. I wrote up a page for the Service Department to use. I left them with the Votex drain plug and a plastic container to give me some of the oil back to look at. When I returned to the dealership I saw my car was done but I was disappointed to find out they had done nothing.

First they told me there was no such gear box or oil. OK ,we got past that but - then they told me the drain plug didn't fit. It was way to small. I said OK what size do I need? They referred me to the parts department. I had more luck there. The Parts Mgr. called Hyundai - they were almost as clueless as the Service Dept. had been. But the Parts Mgr. was determined to get to the bottom of this. He said "how do they know that drain plug you have doesn't fit if they didn't even try screwing it in?" Good thought! So back and forth he went on the phone, looking at diagrams and manuals online.

Eventually who ever he was talking to agreed that the car does in fact have Gear Oil. It is even listed in the 75K service as needing replacement. But the drain plug does not get replaced - only the gasket (which I call a washer). So we got somewhere. They didn't have the gaskets in stock - so he said they would order them. The drain plugs might have to be ordered from South Korea (not that I suggested he needed them) but I don't think they are planning to do that anyway. But the extra gaskets are in case when they do remove the plug they can put it back in, in case the Votex plug does not fit. So there we are. I plan to bring the car back and try again when they have the gaskets (not that I plan to need one since I am supplying my own). They did have the 70 weight fluid in stock.
But I have a greater concern about the car. Every time I have been to the dealership I feel like I just pulled up in an alien spaceship. Do they really have any experience with these cars? One has to wonder. To be continued.....
 
Hate to say it, but has anyone tried thicker gear oil? Maybe a 75w, 75w75, or 75w80?
Or maybe Toyota's LV GL4 75w manual transmission fluid, or LF 75w transfer case oil.
 
Hate to say it, but has anyone tried thicker gear oil? Maybe a 75w, 75w75, or 75w80?
I use the Penrite 70W-75 (7.5) because it's inexpensive and readily available off the shelf "down-under" while many others who DIY in N. America use the Redline MT-LV (5.4) which has an identical viscosity to the OEM 70W oil (5.4). European owners are using Liqui Moly 70W-75 (5.9).

The input pinion runs up to 10,000 rpm so the viscosity range around 5.4 - 7.5 (@100°C) seems appropriate. There's no sign that these lubricants are insufficient regarding gear contact pressure but minor signs of cavitation when cold. EVs are quiet enough at low speeds that you can hear such faint noises, whether that's a good thing or not ...

I'm comfortable with the slightly thicker Penrite because the local ambient doesn't drop below 5°C but some Kona EV owners in sub-zero regions have reported noticeable gear whine when cold even with the thinner OEM oil. I notice a slight whine at at around 7°C but it's gone in minutes.

Hyundai/Kia's E-GMP lineup (Ioniq 5 and EV6 so far) have moved to an ATF pretty much like every other EV uses. The E-GMP is using the oil to also cool the motor so it's pumped and filtered as Tesla and GM do.

BTW, thanks for reminding me about the Toyota magnetic plug as an option.
 
Back
Top