High Torque, Low Viscosity Multi Vehicle CVT Fluid Tech Discussion

[jim302]

The vehicle in question is a 2023 Subaru Outback 2.4L Turbo with the TR690 CVT, but I'm interested in having a primarily technical discussion about fluid so this could apply to other vehicles which specify a "high torque" and/or low viscosity CVT fluid.

OE would be the "safe" choice, but Subaru likes to sell their fluid in 5 gallon buckets and they have 4 types. Yet there are aftermarket fluids that claim compatibility with multiple specs, with Amsoil, Valvoline and LubeGard claiming the widest range of matches.

So first, what makes a CVT fluid suitable for high torque applications? Initial theories:

• It is simply branding / marketing... but some fluids specifically state that they are not for use in high torque applications so that leads me to believe there is a real design difference
• Better quality base oils to resist oxidation and shear in high heat / high load applications
• Higher dose of friction modifiers and additives, or possibly a completely different additive chemistry

If the main difference is base oil, then clearly it would be quite easy for a fluid to match the HT and standard applications (for example, Subaru High Torque CVTF and Subaru CVTF II). Different additive chemistries seem like a higher risk, especially during a simple drain and fill where half of the original fluid is left behind.

The second piece of the puzzle is viscosity. While it is relatively easy to find aftermarket solutions for LV and ULV applications in step shift applications, CVT fluid seems to be a one size fits all thing in North America. Interestingly, there are several dedicated LV CVT fluids in the Australian market.

• How big of a deal is it to use a somewhat higher viscosity fluid (such as Amsoil, with a cSt of about 7.1@100°C where the OE fluid may be closer to 5.5 cSt)?
• The same applies in reverse, but seems like less of a concern considering that UOAs indicate that the OE fluids often shear to this level during normal use, and these fluids should be more shear stable.

I'm interested in hearing thoughts and theories, as well as experiences with aftermarket fluids in high torque and/or low viscosity applications.

• How can aftermarket companies claim such broad matches? Is it just that OE manufacturers only validate one fluid type and recommend that to be on the safe side?
• Are CVTs as finicky as people suggest on fluids? Fear Uncertainity and Doubt are quite common, but rarely if ever does anyone post any definitive evidence to back up their claims that Amsoil, Valvoline, etc harmed a transmission.
• Which would you rather have? Dirty, oxidized and sheared OE fluid or at least some fresh aftermarket fluid that is not an exact match in viscosity or additve chemistry?

 

[LazyDog]

I would first consider the internals of the CVT since it seems they where few variations of Jatco/ Nissan's CVT's.

I believe they where the belt and chain driven CVT's like on the Maxima cars.

This was just something else to consider.

 

[SC Maintenance]

Given a CVT can use a torque converter or not, or can use a use hydraulic or electric pulley actuators or not - the answer regarding viscosity might be CVT design specific?

Also given there is a chain involved, I would guess the add pack is also very important.

I generally don't like aftermarket AT fluids, at least not the ones that supposedly meet every spec on earth, for these reasons.

What do the subi boards say?

 

[Passport1]

Just my personal anecdote so it adds nothing empirically. I have a 2012 Suzuki bought new. Early on while driving on hwy 95 in south Florida one sultry summer night the CVT overheated and the car went into limp mode. Let it cool down and got to my destination. It happened again on the way back to NC but I caught it earlier. A dealer reflashed the computer and that helped a bit but the CVT (jatco) began to whine. I decided to change the fluid and subbed in 4 quarts of Valvoline CVT fluid (actually amber not green color) I did that about three more times and now I calculate about a 90% exchange of the OE fluid. I plan to do this again soon. All this cvt talk makes me want a newer car, this time with a third pedal.

No more CVT trouble.

 

[D60]

I just did a spill 'n fill on a '15 TR580 with the Castrol Universal "nearly everything" fluid. They do claim it's compatible for chains/CVTF-II

I only did a single drain. Got about 5.5qts out and put 6 back in per the proper procedure.

I dunno, it seems fine. Then again I rarely drive the vehicle (not mine) and that transmission always seems a little weird to me.

 

[jim302]

What do the subi boards say?

They aren't a great place for technical information about lubrication, like most brand/model specific forums. There have been success stories for Amsoil and Valvoline, but nothing about long term use vs. just leaving the factory fill in forever until the transmission dies or something else happens that is not worth fixing.

One poster claims to have seen multiple CVTs that were immediately destroyed by Amsoil (and also by dealers using the regular High Torque fluid in HT-LV applications). Of course he can't post any specific examples and is being really dramatic about everything, so I believe him about as much as I believe the random people on the Internet that claim to have seen dozens of engines (sometimes more) destroyed by FRAM filters... so not at all.

Still, it is a little hard to not be suspicious about claims of such broad matches without more information to back it up. I wouldn't hesitate to use these fluids in a Jatco push belt style CVT since Amsoil and others have actual test data on these. I don't want to give in to fear mongering on the Subaru forums, because then either I leave the factory fill in or have to pay $500 for a 5 gallon bucket (commonly out of stock), and hope I can sell the leftovers to someone else.

Dealer service is probably a no-go, since most refuse to touch the CVT and I honestly trust myself to follow the instructions in the factory service manual more than I would trust a flat rate tech I never met that is trying to beat book time.

 

[MolaKule]

The vehicle in question is a 2023 Subaru Outback 2.4L Turbo with the TR690 CVT, but I'm interested in having a primarily technical discussion about fluid so this could apply to other vehicles which specify a "high torque" and/or low viscosity CVT fluid.

What engines are considered high torque for CVT couplings?

Most CVT fluids in the US are around 7.1-7.3 cSt@100C. WHich Autralian CVT fluids are of lower viscosity?

 

[jim302]

What engines are considered high torque for CVT couplings?

Subaru uses the High Torque CVT fluid for when the engine is a H6 3.6R (EZ36D) or H4 2.4L (FA24) which produce between 247 and 277 lb⋅ft of torque. They claim up to 3,500 lbs of towing capacity on some models.

Most CVT fluids in the US are around 7.1-7.3 cSt@100C. WHich Autralian CVT fluids are of lower viscosity?

I may have mixed up some product data sheets. Atlantic Oil CVT Ultra-Syn Low Viscosity shows a cSt of 7.09 on the PDS but 5.6 on the SDS. Penrite claims a cSt of 5.6 but is not suitable for chain type CVTs.

Upon further review, the PDS for Valvoline's US market CVT fluid says it has a cSt of 5.6... so it is interesting to see the discrepancy in viscosity between brands claiming to meet the same specs. I can't find the viscosity of LubeGard.

Of course I'd really like to know the viscosity of the OE fluid, but that is impossible to find as the SDS from ENEOS just says ">20.5(mm2/sec)(40℃)" for both the High Torque and High Torque LV fluids. Not very useful...

 

[MolaKule]

Subaru uses the High Torque CVT fluid for when the engine is a H6 3.6R (EZ36D) or H4 2.4L (FA24) which produce between 247 and 277 lb⋅ft of torque. They claim up to 3,500 lbs of towing capacity on some models.

I may have mixed up some product data sheets. Atlantic Oil CVT Ultra-Syn Low Viscosity shows a cSt of 7.09 on the PDS but 5.6 on the SDS. Penrite claims a cSt of 5.6 but is not suitable for chain type CVTs.

Upon further review, the PDS for Valvoline's US market CVT fluid says it has a cSt of 5.6... so it is interesting to see the discrepancy in viscosity between brands claiming to meet the same specs. I can't find the viscosity of LubeGard.

Of course I'd really like to know the viscosity of the OE fluid, but that is impossible to find as the SDS from ENEOS just says ">20.5(mm2/sec)(40℃)" for both the High Torque and High Torque LV fluids. Not very useful...

Yes, there is a discrepancy between the Atlantic PDS and the SDS. It could be the formula was changed and the PDS has not yet been updated.

Our Nissan Pathfinder's 3.5L V6 produces about 265 lb.ft of torque and I have always used a CVT fluid of 7.1 to 7.3 cSt and the CVT has over 120,000 miles on it. I have it changed every 6 months.