Repair issues and fixes

A much-shorter and slightly less-tedious version of my previous video. Again, details in the description.

 

Some days it just pays to stay off the interwebs.

However, I didn't and ran across someone with a Niro EV on Reddit who was casually asking the Reddit masses for "gearbox bearing numbers" to DIY a repair, as though someone might have this information in their back pocket. Perhaps surprisingly to some I didn't, so did a quick YouTube search for "niro bearing" and found a teardown video of a Niro gearbox in Ukraine titled "Factory defect or coincidence. Kia Niro EV", noting that the video has English subtitles but the translation is sometimes lacking. The car had developed an intermittent knocking noise at 77,258 km.

There were two interesting discoveries made by the diligent mechanic. The first at 3:20 is that a steel plug on the input shaft tail-end had dropped off and was rattling around in the very narrow space between the nearby bearing and the housing. The soft steel plug is exactly like a freeze plug and is intended to keep gear oil out of the hollow input shaft, which leads to the splined coupling and motor shaft O-ring past that. The loose plug would undoubtedly produce significant iron and aluminium particles, as if I needed another factory defect to worry about!

The other discovery he made at 8:13 is that the input shaft bearing on the motor side was showing signs of outer raceway spin, the exact issue I expected long ago would happen due to a design deficiency and which might lead to the knocking noise. I'm not clear on whether he addressed that issue but I think he didn't.

When the gearbox was re-assembled he had replaced the input and intermediate shaft ball bearings but left in place the slightly-damaged output shaft tapered rollers due to the lack of parts. The car's owner apparently wanted it back on the road ASAP.

There are good views of the hollow internal magnet but it seems more secure and confined than the example in my Kona. In one scene he lifts up the casing and the magnet drops out with a clunk. He picks it up and after a time casually places it back in its slot.

 

I think I've misinterpreted the translated subtitles and rather than outer race spin he was indicating that the input shaft had excessive end float. Once the plug fell off, the shaft was striking it with every torque reversal (noting that happens with helical gears) causing the knocking noise.

The same mechanic has a teardown video of the Kia Soul EV gearbox. You can see that it's an earlier generation, not the same as ours. This gearbox suffered a catastrophic failure of the parking pawl but more relevant to us is the internal magnet. It's a rectangular shape which won't spin. It seems they kept the retention pocket but perhaps as a cost savings installed a round magnet.

 

With what I've learned the last week I think that one cause of the tapping noise / wheel of fortune / motor rumble could be the result of this cap falling off the input shaft.

It's also likely that there were noise issues with the motor bearings that the part number revisions were intended to address.

We don't hear of many noise reports now, while around 2019-2020 it was quite common, so I think both issues have been dealt with by Hyundai.

The cap keeps oil collecting in the bearing pocket from traveling down the hollow shaft and reaching the grease-lubricated splined coupling. Its absence may not result in any other problems, except that perhaps washing the spline's grease away might accentuate any noise at the spline. Excess end play in that shaft may be a contributing factor.

The carbon brush on the motor shaft is very sensitive to oil contamination which could reduce its effectiveness so some effort is made to keep it dry.





However the black oil and generally high levels of iron and aluminium contamination seem to be ongoing and I'm confident that it's due to the internal magnet being left loose. There may be variations in magnet diameter as installed and that might explain why we see some cases worse than others, including a few where the oil remains clear. Another possibility is that a high level of early wear particles stuck to it jams the magnet in place.

Fortunately this issue can be managed with more frequent oil changes and the addition of an owner-added magnet or two. The first at no later than 4,000 km (2,500 miles) and every second service subsequently should be practical. The primary value obtained from this added diligence is quiet running.

After several months I'll know if my attempt to reduce the movement of the internal magnet has been successful.

 

Найслабше місце електромобілів Kia Niro EV і Hyundai Kona Electric

 

For me getting a video like this is like an early birthday present and this one from our friendly Ukrainian mechanic is hot off the press and very detailed! We would never see this sort of analysis from global regions were the GRU and motor are replaced as units, as our dealers would always do.

Some salient points, assuming I fully understood the translation:

• The mechanic mentions that the oil on this Kona with about 30,000 km was discoloured. I'll comment that the input shaft turns at high RPM so oil cleanliness is important to avoid bearing damage.
  
  
• He points out the noticeable end play at the GRU input shaft but that the tail-end cap was intact (the one that detached in the previous video). It's not clear if that play is how it is from the factory, or a result of damage or wear on the axial load surfaces, e.g. from dirty oil.

• He test-drives the car and finding that a noise is still present removes the motor. The mechanic points out that the output-end motor bearing outer race is secured from rotation while the tail-end bearing is not, because it must slide in the housing to accommodate axial thermal variations. I'll comment that this is a feature normally found in every electric motor design. The tail bearing has a wavy-type thrust spring which is intended to stop the outer race from turning but clearly that's not working because there's visible wear on all relevant items: the spring faces, the housing bore and the bearing outer surface. It's worth noting that a greased-lubricated bearing running at high speed does have a slight drag which will encourage outer-race spin in the housing if that is not tightly held in place.
  
  
  
  
  
• He suggests that the resulting dust is rusting and (rather oddly, IMO) getting past the bearing seal. He points out what he describes as somewhat-minor internal bearing damage but also the moderately-contaminated grease. I'll add that the bearing damage looks somewhat like it could be from stray shaft currents, which the carbon brush on the other end is supposed to drain off. There's almost no radial loading on this bearing that could cause such wear.
  
  
  
  
• When installing a new tail-end motor bearing the mechanic applies some grease to the outer surface so that any rotation is at least lubricated as long as the grease stays intact, which IMO won't be long. Clearly his options are limited as really this is a fundamental design issue that doesn't have an easy work-around without experimenting on the customer's car at their expense. In my experience thrust springs are normally stronger than this one appears to be. I might have tried to locate a thicker variation or add a second spring to assist this one. Unfortunately, once you have some housing wear there will be more clearance for the bearing to spin. (I'll just add that one YouTube commenter says that this is the change Hyundai made in the later motor part number.)
  
  
• Near the end of the video he replaces the motor bearings on a Niro with similar kms, 35,000, and comments that there was no rust around the tail-end bearing despite the noise. He shows us the bearing components under magnification which show nothing much of interest and indicates that the cause of the noise is still not certain.

So, there we have it, more blame placed on the motor this time which would explain why both items usually need replacement when there is a noise. I would consider more closely current discharge across bearings because this is often the explanation for early bearing failure on industrial motors when driven by inverters (which cause these stray currents due to slight current imbalances across phases). As I've mentioned before, the circular carbon ring mitigation method is considered generally inferior to spring-loaded carbon brush solutions, at least by those companies that manufacture the latter type.

 

Kiwi, I always appreciate your thorough analysis. Reading through your takeaways left me feeling like the motor and reduction gear are doomed for premature failure despite oil changes, etc. Are there any ways to mitigate these design shortfalls?

 

It's rather depressing for myself as well but we can only recognise that statistically-speaking the vast majority of the fleet seem to carry on free of these symptoms. I don't see any easily-applied precautions to help with the motor deficiencies but for the GRU the periodic oil changes and added magnetic drain plug are still going to be highly beneficial in reducing the overall risk.

Most of us here have either long warranties (N. America) or government consumer protection laws (NZ, Australia) that might help with costs if we were affected by any of the multiple issues found. Clearly that's not the case in Ukraine and we're benefiting from the resourcefulness demonstrated here, not to mention the freedom to provide information we'd never be able to obtain from Hyundai.

I'll note that the E-GMP cars (Ioniq 5, EV6, etc) have a different motor design, being internally 'wet' with oil for cooling and bearing lubrication. The GRU also improves on at least two of the deficiencies we are living with, so Hyundai must be aware of these mistakes. It is notable that the fix in all these cases is nothing more than bearing replacements, but it's just unfortunate that they're so hard to access.

If the tail end of the motor was easy to remove I would check it out but unfortunately there's a massive motor mount in the way.

 

Has anyone brought their Kona EV in yet, for the 36 month service, to remove and replace the blue (non conductive, very expensive) coolant?

If so, have there been any “issues” since that service procedure took place? Thanks. Cheers.

 

In Europe they moved this service to double mileage. As per normal coolant.

 

Does anyone know if this is also true in the US?

 

A quick check on the Hyundai USA website indicates that the maintenance interval is still 40,000 miles.

 

Also posted this in AEVA forum.

Had the 15000km service in my 2022 Kona purchased in May. I asked for the oil to be changed and the new magnetic plug installed. Dealer didn't even blink and I received a sample, though small and not in the container I intended. The oil was black but not heavy and the meniscus was clearish. There was some detritus in the bottom but no iron was picked up by a magnet I swirled through the oil. There were noticeable small glints of aluminium particles.

Interestingly the replacement oil the dealer used was Penrite

I will have the sample tested after Xmas madness and the labs reopen.

I must also comment that I took the car to the dealer with my Froot setup, Vess disconnected, driving lights and spare wheel and not a single comment was made

 

Sounds much like my first change at 19,000 km, break-in particles are fully crushed by then to a fine dust and don't have any significant ferrous attraction left.

I seems your dealer is customer-focused compared to some of the stories that we've heard over the past 2 years, especially out of the US.

 

They're a multi brand country dealer and I think a lot of their customers customise. They are reluctant, however, to help me communicate with Hyundai over a problem I have with a fragrance in the fresh air ventilation system that is a health problem for me. We have 35 degrees coming up over Xmas and I won't be able to drive the car to see my family as I can't run the Aircon.

Working on that but certainly agree about the difficulties our Northern cousins have with their dealers.
I'll post the results when I get them. I think mine is one of the most recent to be tested so will be interesting.

A merry Xmas and happy festive season to all!

 

Random internet guys observation:

If you are getting a mouldy smell because of stagnant water collecting inside the air conditioning evaporator core because the condensate is not totally draining away from said core, an ad hoc patch until you can get it addressed is always run the HVAC in a semi manual mode and always force fresh air only at all times.

Never allow the system to run in recirculating mode, always have it draw in outside air. YMMV.

 

Not that I'm fully conversant with Ferenc's concern, but my hack is to always switch off the AC and crank the fan up 5 minutes before arriving at my destination to help it dry out before stopping. I haven't had a moldy AC smell in any of my cars since.

 

Thanks guys. I know that smell and have used KiwiMe's approach at times. This fragrance has been there since new. I have developed multiple chemical sensitivities as an old age present. I know most people can't detect what can knock me over. When I first smelt it, it was like that dreadful stuff they use to demould the car aircon, but now it's more like an old ladies' musk perfume.