Transaxle oil, recent analysis

BACKGROUND

Jeffrey began changing his transaxle oil every 60,000 miles which now includes 120,000 and 180,000 miles. His is a 2004 Prius and now we have a clean record of Type WS history in the NHW20. I've just uploaded Jeffrey Denenberg's latest transaxle oil testing results to the Yahoogroup SAT2 files along with a spreadsheet with the data.

REPORT

(1) additives - after the first change, no effect. There isn't much we can do about them since they are mixed when the oil is formulated. There is evidence the first 60k change saw some usage but since then, nothing.


(2) wear materials - great profiles, we know Si decreases due to the sealant leaching finally being covered by less soluble material. The Fe decrease is likely, partially due to the sealant pigment, iron oxide, as well as gear surface polish. Speculation but a sealant with another pigment material might let the us track gear wear more accurately. Al, Cu, and Zn appear to be accurate wear materials for what is going on inside the transaxle. Everything else appears to be trace material likely left over from assembly and manufacture.


(3) viscosity - great profiles although the 100C rate of 'shear down' seems to trend downward. Still it met the 15% threshold.


(4) trace - the Na and Mg are worth watching but these are at the limits of detection. The concern is these may be the harbinger of some 'leakage' into the transaxle but at such low values as to be at the limits of detection. The concern is any coolant leaks into the transaxle would also increase or add water, a bad thing.


CONCLUSION

Based upon the Al, Cu, and Zn levels, increasing the oil change interval above 60,000 miles looks feasible. This is offset by the loss in 100C viscosity that has reached the 15% threshold. Extending the next oil change to 90,000 miles makes sense with testing to see what Al, Cu, Zn, and 100C viscosity levels go to. Adjusting change intervals to achieve level Al, Cu, and Zn makes a lot of sense.

Jeffrey reports,". . . the first interval should be at 30k followed by 60k (perhaps) then stretching it out . . . " which agrees somewhat with my plan for my wife's 2010 Prius, ZVW30.

I did the first transaxle oil change at 5k miles with an extended analysis. In addition to flushing the manufacturing and initial wear material this change showed a straight-line to a 15% viscosity loss at 15k miles. So now I'm waiting until 20k for the next change. I'll adjust the subsequent change intervals based upon the straight-line estimate to a 15% viscosity loss. Eventually we will find a stable distance, the interval that leads to reproducible, 15% viscosity loss.

Bob Wilson and Jeffrey Denenberg

 

We're seeing the 100C viscosity hitting on the 15% viscosity limit:

	Column 1	Column 2
0	40C %	100C %
1	13.4%	14.5%
2	9.3%	10.9%
3	9.8%	14.5%

Percent loss of viscosity for 40C and 100C measurements.

In practice, we'd like to see the viscosity loss average above and below the 15% target. Although the 40C viscosity has a lot of head room, the 100C is just at the 15% limit. The hotter temperature approaches the operational transaxle oil temperatures I've measured with my NHW11, 60C, on a hot day at high speed and more likely closer to the gear and bearing temperatures.

It makes sense to extend the interval and test again, say 90,000 miles. These are small steps reaching out towards what hopefully remains on the conservative side of the wear curve. Sad to say, we are just learning what works in our Prius:

• NHW11 - a few of us continue to study these older vehicles.
• NHW20 - the vast majority of today's USA fleet.
• ZVW30 - the growing fleet, this is my wife's car currently under test.

We'll let the data tell us what works and doesn't. For example, we already know Amsoil ATF is Type T-IV viscosity compatible but it eats yellow metal. We learned this with my NHW11.

This effort would move forward a little faster if we had more participants. Folks willing to send their transaxle oil samples off for the $20 testing expense and then sharing the results. Empirical data that can help others extend their vehicle performance and transaxle life.

Bob Wilson