ATF fluid changes ARE Required.

So, how exactly WS gets contaminated when exposed to sunlight?

 

I don't know, never seen it happen. If it does, my guess would be chemical decomposition via UV.

 

I know it's old and you didn't start it, but where did the sealed story start?

Check out

PLUG, BREATHER(FOR TRANSAXLE). TOYOTA Prius (2010 - 2012) # 9093003201

 

I found the breather plug in the NHW11 is actually a one-way valve. As the transaxle heats and cools, the thermal action pumps a slight vacuum. When you open the fill plug first you can hear the inrush.

Bob Wilson

 

If the transmission gets to 180F during hot weather operation, and eventually sits in 40F weather without venting, that would be 4psi below atmospheric pressure in the transmission. That is showing a lot of confidence in their shaft seals, considering the consequences of sucking contaminants into the transmission.

 

Good question, just one of those things said and perpetuated. Made sense since Toyota is calling lifetime fluid.

Your temp analysis makes sense and it would seem likely to introduce contaminates into the system. But you would think that there would be more reports of leaking seals. What goes in, must come out eh?

 

It is a lifetime fluid, it lasts until the transmission fails, of course that transmission would probably last longer if the fluid was changed at reasonable intervals

Lip seals tend to work better in one direction than the other. A seal that doesn't leak any oil out could allow contaminants to be sucked in. That used to be a problem in off road vehicles that got their differentials and transmissions suddenly cooled by driving into a stream. Seal design and quality has come long ways in the past few years though.

 

I don't recall seeing moisture or dirt problems in UOAs on old Prii, so I think Toyota has a handle on the problem.

 

I suspect that everything conducts a little bit but I understand your point. Mineral oil does not conduct enough to be a problem even at over 500KV. I realize oil is commonly used as an insulator in HV transformers and power circuit breakers. But I am not sure about the conductivity of the additives in transmission fluids from various manufacturers.

The cost of type WS is about the same as other synthetic transmission fluids and it's readily available so why not use it? I also think Bob Wilson may have a good point about the viscosity.

 

True enough... I was just thinking about that over lunch. Plan on doing my oil this summer and will have about 46K on it. If my oil looks like these photos, I am sending it in for analysis. Probably will anyway, nice to have an Alaska benchmark.

 

I didn't have ready access to my files at lunch but here are some more background and details about the NHW11 transaxle, vent plug. In addition to testing the oil, I also laid out a 'roll down' test to get some idea of the effect on rolling drag.

As part of my first transaxle oil change, I captured a sample for oil analysis. It was nearly black, opaque, and smelled of paraffin:

The oil testing results were 'off the charts' bad. The silicon level was 227 ppm and only one other sample ever came in with 258 ppm, slightly higher. The analysis suggested dirt contamination and the high aluminum strongly suggested this was the case. But how did it get in the transmission.

My first thought was the negative air pressure, detected when the fill plug was removed, suggested dirt might have come in via the seals. So I decided to see if I could eliminate the pressure differential.

The transaxle vent plug is very difficult to reach, between the transaxle and firewall:

It is in a miserable location and a pain to reach.

I took it apart and found:

Nothing indicating it was a path for dust and dirt into the transaxle. It is just a simple, one-way valve:


Still, there are two gaskets on the half-shafts and one on the flywheel end but I was not interested in disassembling a perfectly fine transaxle to diagnose a dust leak problem. So I decided to replace the vent-plug with a clean-air source.

First I attached hose and a fitting to the breather body:


I then installed it on the transaxle:

and ran the hose to a clean air source:

It shares the air behind the engine air filter.

Sad to say, this later turned out not to be necessary. In another experiment, I coated the bottom of some shallow jelly jars with Toyota transaxle case sealant. After letting it fully set, I poured in Type WS, sealed it, and drove around for a month before sending the sample for analysis. Sure enough, the silicon levels were 'off the chart' and all that had been done was to be in contact with cured, Toyota case sealant.

I have not backed out my modification. I can find neither a problem nor advantage so I'm keeping it. The hypothesis of seal-leakage remains but given the absence of other seal testing, I don't think further seal investiation merits more effort.

Bob Wilson

 

It's made by vampires.

 

Thanks Bob, that answered my unasked question about whether there was a vacuum limit on the vent valve. I should have expected that you would have already torn one down.

 

Have you done UOA after installing the contraption? Wouldn't that increase humidity in the trans-axle?

 

Yes. No. But since you asked, let me explain.

Silicon and aluminum are two of the most abundant elements on earth and found in ordinary dirt. When it was found in great abundance in the first sample, the oil analysis service postulated 'dirt' had gotten in the transmission. I could find no mechanism except the lower, internal pressure might be sucking in dirt and grit through the seals. So I installed the open vent behind the engine air filter as part of an effort to eliminate this lower internal pressure and reduce the silicon and aluminum content in the subsequent oil tests.

Sure enough, the silicon and aluminum content went down with each subsequent change. However, the Amsoil ATF had a different ratio of additives and these were still evident. Using partial fractions, I found a ratio consistent with a ~15% carry forward of the previous oil load. In the NHW11, we drop and wipe the pan and all exposed surfaces but the NHW20 and subsequent models do not have an oil pan to drop and clean. I would expect them to have a greater carry forward.

Testing for water is standard in each oil test and none of them have shown evidence of water. In normal operation, the transmission warms up, initially at ~0.9 F/min (in my 10 mile urban commute) ~15F increase in the morning commute. On the highway, at 65-70 mph, there is ~50F increase above ambient to a constant temperature of around 135-140F on a warm, 88F day. This higher temperature in the transmission tends to drive out any water and the inlet is behind the engine air filter opposite the PVC outlet on the other side of the throttle inlet. Although there are air-drying, air filters used in manufacturing plants to trap water and dust, there is no evidence that my unique, open air vent is letting in dust and moisture.

Bob Wilson

 

And previous tests were made with what fluid? Original oil from the factory? Is that not type WS?

As I understand your previous tests you came to a complete stop on the markings on the map, and now you came at the end with 15 mph, I think this is very big difference. What is the actual new tyre size?

I sometimes start in EV and then shift to N (at 33 kph), as there is a small downhill just so much to throw me out of EV mode. This morning when I tried this I hit the N at 33 kmh the car was gaining speed very very slowly, and it reached only around 38 kph, on a summer day I can reach well over 40 kph and have to brake as there are speed bumps. But not today at 32° F. I am on original ATF from the factory and at 26k miles.

 

I first tested the original oil after 53,000 service miles, the factory Type T-IV.

It is a big difference:

• ~3,000 lbs -> 1,360 kg - vehicle weight to mass
• ~7 M - altitude change
• 1360 * 9.8 * 7 ~= 93,296 Joules, hill potential energy
• .5 * 1360 * (6.7*6.7) ~= 3,052 Joules, kinetic energy at 15 mph

Sumitomo T4, 195/70R14 - 849 rpm/mile - the largest that fits in the 2003 Prius wheel well

There are multiple effects:

• air density - colder air increases aerodynamic drag, 33 kmh is not terribly fast
• tire hysteresis - colder tires have greater loss, mitigated by maximum tire pressure
• oil viscosity - colder oil has higher stirring losses

However, it is likely the unchanged oil has a measurable, particle load increasing the frictional drag.

Bob Wilson

 

Whew, thanks for the executive summary at the end.

 

The air volume doesn't matter because the bulk modulus of the fluid and the transmission housing are so high compared to the air that they don't enter into a calculation or measurement.

PV=nRT V,n and R are constants so P absolute is directly proportional to T absolute