[bwilson4web]

Comments?

EXECUTIVE SUMMARY

Transaxle oil testing has shown the NHW11, 2001-03 Prius, transaxle oil, Type T-IV, will shear-down to nearly the Type WS viscosity range in just under 40,000 miles. Early testing with Amsoil ATF as a substitute revealed excessive copper wear and its use is strongly discouraged. Switching to Type WS, the same oil as the NHW20, reduced the copper wear but still about 4x the original Type T-IV and Fe and Al were 2x compared to Type T-IV. Based upon oil analysis from half a dozen vehicles and our NHW03, we recommend:

NHW11: 30,000 mile oil change interval - which is half the Toyota recommended 60,000 miles.

Early reports of contamination based upon Si levels were traced to the sealant leaching.

BACKGROUND

Transaxle friction losses along with tire rolling resistance are the primary sources of rolling losses. The transaxle loss is primarily a function of the lubricant that needs to balance viscosity 'stirring' losses versus friction and wear loss.

Viscosity Shear-Down

Toyota has two transaxle oils for the Prius and these are the measured, viscosity wear rates:

• 40C / 100C - Type T-IV
• 35.3 cTs / 7.4 cTs - 0 mile Type T-IV
• 30.0 cTs / 6.3 cTs - 15% viscosity loss - oil needs replacement
• 26.5 cTs / 5.1 cTs - 40k mile Type T-IV
• 26.1 cTs / 5.2 cTs - 53k mile Type T-IV

Oil testing services recommend that a 15% viscosity loss means the oil should be changed, absent any other metrics.

The NHW11 transaxle was redesigned in the 2004 with the NHW20, 2004-09, to replace some bearings with roller bearings. Using the Type WS was cited as a friction reduction strategy. However, the similarity of used NHW11 oil to new NHW20 oil suggested testing was in order:

• 40C / 100C - Type WS
• 24.5 cTs / 5.5 cTs - 0 mile Type WS
• 21.3 cTs / 4.7 cTs - 61k mile Type WS
• 20.1 cTs / 4.6 cTs - 15% viscosity loss - oil needs replacement

.

Contamination

Foreign material, dirt and grit, are usually identified by high Si and Al rates. The first samples had high Si with higher rates found in more rural, agricultural and dusty areas. But testing with cured sealant in a sample container with virgin transaxle oil revealed rapid Si leaching from the sealant. Since leaching is temperature related, hotter climates may have accelerated the observed correlation to agricultural and dusty areas.

Oil testing of a new, ZVW30 at 5,000 miles revealed early wear material including small bits of sealant. Furthermore, an NHW20 that has gone over 180,000 miles revealed the wear rates were substantially lower suggesting a 90,000 mile change rate in the future.

I will be backing out my transaxle oil breather modification that let the transaxle stay at atmospheric pressure. Temperature cycling causes a reduction in pressure in the transaxle and was suspect in the early, high Si rate. This is the one-way valve I will replace:

Because of the obscure location, the back out will be coordinated with possibly two other experiments: (1) solar block heater and non-radiator cooling, and/or; (2) Prius inverter to AC house power.

Material Wear Rates

Testing Type WS in the NHW11 transaxle designed to use the Type T-IV revealed:

• 2x - Fe and Al wear rates
• 4x - Cu wear rate

Like the thinning of brake shoes, loss of iron and aluminum are associated with mechanical removal. This is an indirect measure of friction removing material. Type WS in an NHW11 transaxle saves viscosity losses but at the expense of higher mechanical friction and wear losses.

Viscosity Temperature

It takes about 30 minutes to bring the transaxle oil temperature to operating temperature. Measured warm-up:

• 88F/31C - starting temperature
• 125F/52C - warmed temperature 30 minutes later

Notice these temperature nearly bracket the 40C viscosity reading.



What this means is oil viscosity testing at 40C is all that is really needed to understand the operating range of transaxle oil. There will be point sources with higher temperatures between the bearing and contact surfaces but these are a small fraction of the total oil.

Oil viscosity is a strong function of oil temperature:


Given the effect of cold temperatures, we can observe viscosity effects in freezing weather with a hill roll-down test:

The longer rolling distances were as a result of an oil change that also removed a substantial amount of wear material in spite of the lower viscosity of the used oil.

CONCLUSIONS

This initial study took from 2005 to 2011 but by no means is the last answer.

Recommendations

1. Early transaxle oil change - a new car should have the transaxle oil changed early with the first engine oil change. This will flush the early wear material and loss sealant bits.
2. NHW11 - should use Type T-IV over Type WS as the wear patterns suggest the lower viscosity losses may be more than offset by higher wear, friction losses. NEVER USE AMSOIL ATF DUE TO EXCESSIVE COPPER WEAR. Do an early change and then every 30k miles.
3. NHW20/ZVW30 transaxle oil should be changed early: at 5-30k miles, at 60k miles, and then 90k miles thereafter.
4. Use of a transaxle heater to bring the oil closer to operating temperature should receive equal weight to an engine block heater in colder weather.

Future Studies

• Blending Type T-IV and WS may find a transaxle oil with optimum viscosity and wear losses. However, it is difficult to measure viscosity losses apart from friction. We can see friction losses in the wear material. We need a better model to figure out viscosity losses.
• Boron CLS - oil analysis suggests the commercial product does not introduce material not already found in Type T-IV and WS. This may provide a mechanism for post-treatment to reduce friction wear losses. However, there is an ongoing engine friction analysis and this is a risk only experimenters should assume.

Special Thanks

R&G Labs of Tampa Florida are my oil testing service with timely advice and affordable, rapid turn around. Also thanks to those who provided oil samples for testing including Jerry Jorgenson, Woodson Moore, Dr. Denenberg, Doug Schaefer, good Prius friend Hobbit and many others who hang out in "Prius_Technical_Stuff" with suggestions and comments. Thanks also to Toyota Motor Corporation who brought these wonderful machines, these Prius, to our shores. Finally, special thanks to our Japanese friends who wrote:

• Development of Electric Motors for the Toyota Hybrid Vehicle "PRIUS" - Kazuaki Shingo, Kaoru Kubo, Toshiaki Katsu, Yuji Hata (Toyota Motor Corp)
• Development of New-Generation Hybrid System THS II - Drastic Improvement of Power Performance and Fuel Economy - Koichiro Muta, Makoto Yamazaki and Junji Tokieda (Toyota Motor Corp)

This six year project would not have been possible without the help, assistance and sharing of these and others whose names my poor memory may have forgotten. Thank you everyone.

Bob Wilson

[jdenenberg]

bwilson4web said:

CONCLUSIONS

This initial study took from 2005 to 2011 but by no means is the last answer.

Recommendations

1. Early transaxle oil change - a new car should have the transaxle oil changed early with the first engine oil change. This will flush the early wear material and loss sealant bits.
2. NHW11 - should use Type T-IV over Type WS as the wear patterns suggest the lower viscosity losses may be more than offset by higher wear, friction losses. NEVER USE AMSOIL ATF DUE TO EXCESSIVE COPPER WEAR. Do an early change and then every 30k miles.
3. NHW20 transaxle oil should be changed early, at 60k miles, and then 90k miles thereafter.
4. Use of a transaxle heater to bring the oil closer to operating temperature should receive equal weight to an engine block heater in colder weather.

I guess I'm a bit more conservative. In a NHW20 (or Gen3 until real data exists) I would suggest that an early refresh of Type WS ATF at 30k miles followed by a second 60k miles afterwards (at 90k total). You could then consider stretching out to 90k mile intervals thereafter. I will do my next refresh at 240k ( the fourth at 60k intervals) miles G-d and my 2004 Prius permitting.

There is an individual doing testing of Redline ATF in his Prius and reporting the results on PriusChat. I do see a little CU in his Blackstone Lab test reports so I suggest sticking with the Toyota Type WS in the later models for now.

JeffD

[bwilson4web]

jdenenberg said:

I guess I'm a bit more conservative. In a NHW20 (or Gen3 until real data exists) I would suggest that an early refresh of Type WS ATF at 30k miles followed by a second 60k miles afterwards (at 90k total). You could then consider stretching out to 90k mile intervals thereafter. I will do my next refresh at 240k ( the fourth at 60k intervals) miles G-d and my 2004 Prius permitting.

I like it and also combined it with the ZVW30 recommendation. I've already done one ZVW30 change at 5k miles and given the similar transaxle designs, I think this is an improved recommendation. Do an early change to flush out the left over manufacturing debris and sealant beads. Do a second change at 60k to get the residuals and finally 90k miles thereafter.

I will be doing a 15k change on my wife's ZVW30 in about a month. This is part of my new vehicle studies but the 5k sample was pretty definitive about the need for an early flush. Get the stuff out before the differential has too many chances to powder the left-over debris.

jdenenberg said:

. . .
There is an individual doing testing of Redline ATF in his Prius and reporting the results on PriusChat. I do see a little CU in his Blackstone Lab test reports so I suggest sticking with the Toyota Type WS in the later models for now.

Good for him! Experimentation should be encouraged along with collecting empirical data.

BTW, I periodically check the Gen I forum and am pleased to see NHW11 problems are being competently addressed. The net has many outlets for Prius advocacy and it is kinda nice to go back to my 'Prius lab' work. <GRINS>

Thanks!
Bob Wilson

[rhcev6]

I just did my Trans fluid for the first time (I don't think it was ever done before) at 120K miles. When I went to Galaxy Toyota in Eatontown, NJ (closest deal to 07701 zip code) they INSISTED that the right fluid for my 2003 Prius is the WS fluid and that the T-IV fluid was for 2002 and earlier. The old fluid looked (form my amature point of view) seemed to look like the same fluid but was several shades darker. It did not have too bad of an odor either. My drain and fill plugs both looks the same (24MM and no flat or magnetic parts). I do not see any heavy signs of debree either. It looks like regular car trans fluid that just went a bit longer than it should have.

I have the pan gasket that I need to pick up from the dealer this week so I can do the change again and this time drop the pan and inspect what has accumulated on the bottow (if much at all). After the fluid change I did not notice much of a difference in the running of the car.

I was suprided at just how easy this work was t do! It took me longer to get the car up on the ramps and jacked up the rear to level it out in the garage than it did to do the change!

[bwilson4web]

The only difference I found before and after the first change was in freezing temperature, hill roll-down testing. It was small but reproducible.

Bob Wilson

[rhcev6]

So the WS fluid will not harm the 2003 prius? Does anyone have any idea why all my manuals seem to say T-IV but the dealer said their specs say WS? Did Toyota discover something that made them say their specs call for WS? Why only the 2003?

bwilson4web said:

The only difference I found before and after the first change was in freezing temperature, hill roll-down testing. It was small but reproducible.

Bob Wilson