The truth about Toyota WS transmission fluid

I was actually really interested in this, but gave up listening to this idiot at the 2:45 mark. I couldn't imagine what the next 13 minutes would be like

 

It's PEOPLE!

 

For the truth about ATF WS, it might be better to start with the technical papers:

• Yamamori, K., Saitou, K., Kobiki, Y., and Ogawa, A., “Development of New Automatic Transmission Fluid for Fuel Economy,” SAE Technical Paper 2003-01-3258, 2003.

• Umamori, N. and Kugimiya, T., “Study of Viscosity Index Improver for Fuel Economy ATF,” SAE Technical Paper 2003-01-3256, 2003.

• Kurosawa, O., Matsui, S., Komiya, K., Morita, E. et al., “Development of the Fuel Saving Low Viscosity ATF,” SAE Technical Paper 2003-01-3257, 2003.

As far as I know, Toyota’s ATF WS—unlike GM’s DEXRON or Ford’s MERCON, for example—remains a closed, proprietary specification, which Toyota has chosen not to license for sale under other brands. It’s hardly a secret, though, that it’s made for Toyota by other companies; indeed, Yamamori and his coauthors acknowledge collaboration with TonenGeneral Sekiyu (now part of ENEOS), Nippon Oil, and ExxonMobil.

As for the video, I simply couldn’t accept its conclusions that either of the products shown is defective or contaminated, that “Mobil did something wrong” in production, that the “new, improved formula is definitely not improved,” or that there has been a change in the ATF WS product.

It’s not clear to me how the third-party laboratory, or the creator of the video—neither of which seems to have had access to Toyota’s specifications—would have been able to know what should have been in the fluid samples, or in what amounts. The lab’s generic “contaminant metals” label may be useful to its customers in the context of a used oil analysis, but it’s hardly definitive.

Without knowing the lab’s measurement uncertainty, I couldn’t even agree that there are any significant differences between the samples. Is a difference in kinematic viscosity (at 100 °C) of less than two percent, 5.4 vs. 5.5 mm²/s (vs. 5.4 mm²/s in the Yamamori et al. paper), for example, something that the lab could even measure reliably? More importantly, for a fluid that’s described as having a viscosity index of 175, would such a difference matter at all in practice?

If you want to watch a video and learn something, I’d suggest Noise Vibration & Harshness instead.

 

By coincidence, just yesterday in another thread I reposted this video about a nifty app (from the same guy who does the Weber Auto videos often linked here) that applies the same principles. By entering some key parameters about the car, drive train, and tires, you can get it to analyze the NVH frequencies and direct you right to the likely suspects.