Several methods of providing a magnetic plug function to the Kona Electric gearbox have been suggested, apart from my own choice of using two Votex DP007.
As background to this, we know that the non-serviceable round OEM magnet spins and rattles in the aluminium pocket provided in the gearbox housing. Particles caught on the magnet's surface are worn down against the aluminium surfaces, creating even smaller particles which are released back into the oil. Steel particles lose their ferrous attraction due to that working while aluminium particles make the oil black. There are are (15) used-oil analysis results provided by Kona and Niro EV owners that support this conclusion.
Because the Votex and similar aftermarket magnetic plugs raise compliance questions regarding the vehicle's warranty, I thought I would carry out some bench tests on the various alternatives to supplement similar tests others have done.
These tests will be done in agitated water with small amount of added iron filings. It's a cleaner and easier experiment to conduct than it would be using gear oil, especially in my carpeted spare room.
I installed and sealed each of three magnetic plug options into their own 400 ml clear plastic container such that only their top surfaces were exposed to the interior, as they would be when installed in the aluminium-cased Kona gearbox. Those plugs tested are (1) the DeFill M18 with neodymium magnetic, (2) the Toyota OEM 70-series Landcruiser diff plug with ceramic magnetic, and (3) the popular home-brew solution - neodymium disc magnets attached to the outer surface of the Hyundai OEM steel plug, in this case a stack of three, each Ø18x3mm thick.
I’d consider the DeFill to be similar enough to the Votex DP007 to perform the same, but the DeFill stainless-steel body is 3mm longer. It hasn’t been tested in the Kona so I can’t specifically recommend it as an alternative to the Votex. I'm using this plug because I have one on hand and it's close enough in design and performance to emulate the Votex installed in my EV.
The agitator is an 18mm spade drill and the target agitation speed is 1500 RPM. One container lid was drilled to fit the spade drill shaft so water doesn’t splash out when agitating.
The iron filings were made with mild steel on a slow-turning drill-mounted grinding wheel to avoid overheating the particles and damaging their ferromagnetic properties. I filtered the particles magnetically to remove non-ferrous debris and divided them up about equally for the three plugs. I’m using a quantity for each test that I’d describe as about what I’d expect to see as normal wear in the Kona’s gearbox over perhaps 50,000 km.
One downside of using water as the fluid medium instead of oil is that the particles rust in about 30 minutes so the entire experiment must be performed and disassembled expeditiously to keep the plugs in good condition. The upside is that the video images are clearer, there’s less risk of a nasty mess if something leaks and the final cleanup is far easier.
The experimental method was to fill the container with water, add the particles and allow them time to find their happy place. I quickly found out however that if dropped in gradually nothing much happens until agitation is started, but in two of the three cases the particles inadvertently clumped together when dropped in and subsequently that 'clump' found the plug immediately.
In all three plug tests I replaced the water after the initial agitation to clear out any remaining non-ferrous debris. The contents were agitated for about a minute then allowed to settle before being agitated again to see if any break loose. The video is edited to minimise boredom, so intervals appear to be much shorter than I actually used. The top agitation speed generates tiny air bubbles that should be not confused as being loose ferrous particles. There is some inevitable foreign debris floating around as well.
The important outcome for all three options is that no amount of agitation could visibly knock any particles loose and all options work acceptably well. I'll note that in reality, actual wear particles coming off gears may have been subject to intense heat and as a result may have lost some ferromagnetic attraction.
Additional discussion:
It's worth noting that agitation markedly improves the percent of particles captured over that in still water, perhaps close to 100% in a matter of seconds. Taking into account the larger fluid volume in the actual gearbox and the far-greater viscosity of 70W oil over water, I would anticipate the same process would take place, but over a longer time frame. The actual gearbox fluid volume v.s. test volume ratio is 1/0.35 = 2.9 and the viscosity ratio (at 45°C) is 27.5/1 = 27.5, for what that's worth. We could be looking at a time period ranging from 10 minutes to perhaps up to an hour for the same effect to take place.