Where are you getting this data on drive units failing? How old is that? I'll do my own investigation and get back on my sources. Please post yours.
Cross query Tesla Motors web forums and enthusiast sites. Drive unit failures are not a thing of the past. Tesla would never release data on their drive unit failures since they replace it themselves. They replace everything themselves. It isn't good for Tesla is they indicate they have issues with drive units failing or other serious issues that cause customers to turn their vehicles in for repair.
I'm more inclined to believe people who post about their cars needing work done or the issues they face with Tesla acknowledging there is a problem.
You do realize that the battery pack uses fluid that goes through it like lasagna to cool/heat the battery, right? It's not a "heat sink" like you suggest. I think your data is either out of date or just made up....please bring sources.
Is the fluid moving hot liquid to be cooled by fans? That's a heatsink. In the same language, an AiO cooler still uses a massive copper plate to remove heat from a heat source and on the other side fluid travels to radiator fans.
To quote this site:
https://avidtp.com/what-is-the-best-cooling-system-for-electric-vehicle-battery-packs/
The Tesla models S battery cooling system consists of a patented serpentine cooling pipe that winds through the battery pack and carries a flow of water-glycol coolant, thermal contact with the cells is through their sides by thermal transfer material. Again this will remove heat from the side of the cells rather than from the tabs, and overheating a Tesla battery pack under hard driving is easy to do.
This in basic English is cooling pipes which direct water glycol mixture (this is to prevent freezing, it is essentially anti-freeze) through a series of pipes that make contact with the battery packs using a thermal transfer material.
That is fancy talk for a thermal pad or sticky paste. It's likely to be a similar compound to CPU compounds. High end electronics equipment in the compute segments may have VRMs that reach 90C. Most high end mosfets in electronics including cars can hit 150C without issue and still function normally. Both those figures are far higher than what battery packs are able to handle comfortably. Sticky pastes are synthetic materials that allow heat transfer to occur. This is basic thermodynamics. Liquid metal also exists.
Tesla cooling isn't special. Your video discusses the ability to make more compact, energy dense batteries over what was possible decades ago.
Edit: Your lasagna comment got me intrigued. So I looked it up. Basically, they may use a thermal compound, but it's an aluminum wave sheet with tiny channels allowing liquid to flow through. This is a heatsink. The aluminum sheets are the heatsink, the channeling within that moves water makes the the car's system an AiO cooler. Ideally, they could have used copper since it's better for heat dissipation, but it adds more weight and much more cost. The cooling design Tesla uses is new and unique, but the theory in which the packs are cooled is old tech. Ideally, Tesla would be better off designing a cooling liquid that can allow batteries to be submerged and draw coolant through piping and cooling it off using intaked air and the cabin cooling system. Because of the way metal is machined, I'm almost certain some synthetic is used to allow for a better contact patch with the "lasagna' fins with the channels for better heat dissipation. The idea of a non-conductive fluid is good, but weight and replacement becomes a concern. It would cost more to repair batteries and increases waste in the long term.
