Reduction Gear Oil Change - Benefits for Range

My Nissan Leaf Forum

Help Support My Nissan Leaf Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
denwood said:
Frankly I was surprised that the final drive oil was so thin on the LEAF as I'm quite accustomed to 75w oil being used in differentials on most cars I've worked on. These ICE diffs operate at far lower RPMs, so it's obviously the EV motor drive RPM that is dictating the use of highly engineered fluids. The Audi A3 TDI that I serviced recently had some pretty interesting 75W fluid used both for final drive and DSG trans operation. It also took a laptop and $700 of software to service correctly....249 ft/lbs of torque from that engine at 1200 RPM. So again, it's a bit weird to be putting such low viscosity fluid in what is essentially a differential.

Nissan Matic S - Kinematic Viscosity @ 100°C cSt ASTM D445 5.2

Typical 75w-90 Gear Oil - Kinematic Viscosity @ 100°C cSt ASTM D445 14.56

Not that thick in the Audi A3 TDI DSG transmission. The Audi-VW DSG transmission in the A3 uses spec G 052 182 A2, which has a kv100 6.8, which puts it close to the thin Dexron VI & Mercon LV spec, also close to Nissan Matic S too. I hope you didn't use 75w-90 GL-5 in the DSG by mistake.
https://pim.liqui-moly.de/pidoc/P000247/3640-DualClutchGearOil8100-42.0-us.pdf

The Audi A3 rear differential does use a typical 75w-90 high-viscosity GL-5 gear oil, high-visc & GL-5 due to a Hypoid gear with a lot of sliding friction as it turns rotation 90 degrees. The Leaf doesn't turn rotation vectors 90 degrees at all & hence doesn't need 75w-90 GL-5. ...
Our Leaf's gears & differential are all helical involutes, not Hypoid, and therefore don't use high-visc GL-5 stuff.

A good starting point for understanding why a Leaf gearbox wears more at low-speeds & high torque loads, read the summary below:
https://www.motioncontroltips.com/why-is-viscosity-important-for-bearing-lubrication/
 
denwood said:
DougWantsALeaf said:
So Denwood, are you saying Nissan is actually already pushing the envelope with the stock gearbox oil?

I'm assuming the Nissan engineers have done their homework :) What I will say is that I would stick to oils that match or exceed the Matic S spec.
Does the ULV exceed but in the wrong way? :lol:
 
Knight, maybe :)

Volt, I used this stuff. Note the picture is a bit weird but the name was so long, it wrapped half way around the bottle :)

dsg.jpg


... and yes you are 100% correct on the kv100 value at 6.8. That was my mistake on the cross reference. The process to change it was a bit painful as you had to fill it the case from underneath, (no dipstick) and then run the car on the hoist, check the trans fluid temp via Rosstech (laptop) and then let it drain on expansion via an overflow tube. The TDI (2010 in this case) A3 is FWD only, so no rear diff. The LEAF differential oil is a whole lot simpler to change :)

Good point on the hypoid vs helical (I'm no gear expert), however my interest in the oil viscosity was comparing an older transaxle (like the VW 020) which used 75w90, and Nissan final drive which is not so dissimilar (ok, no syncros), using a much thinner oil. Same thing on the DSG transaxle on the A3 TDI as you correctly pointed out...very thin oil. We do have some 90 degree gear action happening in the diff planetary but they are not moving much unless on corners, or spinning in the slippy stuff.

Of course this all illustrates why I truly appreciate the simplicity of an EV after swapping out transmissions, changing clutches, TDI fuel filters, O2 sensors, catalytics, leaky exhausts, timing belt changes, timing chain issues, leaking radiators, worn out starters etc. etc.
 
denwood said:
Of course this all illustrates why I truly appreciate the simplicity of an EV after swapping out transmissions, changing clutches, TDI fuel filters, O2 sensors, catalytics, leaky exhausts, timing belt changes, timing chain issues, leaking radiators, worn out starters etc. etc.
I love this... nothing is more worthwhile about an EV than mechanical simplicity.

If considering the use of slippery stuff additives, my favorite secret sauce has been Dyna-Maxx EP lube concentrate. I've used it in commercial buses and my own cars for over 25 years. I add it to everything from engine oil, trans, wet wheel bearings, diffs, etc.. Never lost a major component since using that stuff. Even had an incident where a turbocharger oil line on a 2-stroke Detroit Diesel failed while cruising at 65 mph where it ran so low the oil pressure shut-downs killed the engine. Fixed the broken line while on the shoulder of the road, refilled the oil and drove it away without any apparent damage to the turbo, or supercharger, or bearings. That engine should have grenaded, was totally amazed. That's what I plan to add to the gearbox oil change on my LEAF.
 
Roger, I will admit to not being a fan of additives, other than the odd bottle of injector cleaner. The truth of even injector cleaner is that you can get the same benefit from just running a few tanks of a T1 fuel through.

That said, I know better than to argue :)

This guy has done some interesting tests, great channel: https://www.youtube.com/c/ProjectFarm/videos

I did a fair bit of travel as a company owner and one thing that always stuck out on the car side were conversations with car service owner/drivers in New York. These guys would often rack up 500 000 miles on a fairly pedestrian vehicle, often with the original engine. The secret there was simple. Oil changes every 3000 miles and fastidious maintenance. As I've taken apart and built up a few engines (along with pretty much every other job, including full paint jobs) it becomes more obvious that careful maintenance is about 98% of the equation. The other 2% I'd say is manufacturing or engineering flaws that you can't really do anything about (depending on the car...some are much worse). I also keep our vehicles a long time.

I would agree 100% that an EV just makes a lot more sense with respect to simplicity :) I'd also agree 100% that changing out the final drive oil a bit more frequently can only help, and most certainly will not have negative consquences. If you do play with additives in there, I would 100% recommend a few oil analysis to make sure you're not causing damage.
 
This research study, using a Leaf, assumes the gears lose 3% of the power & energy driving around. ... with Nissan Matic S of course. ....
They lump both tire rolling resistance (rubber flex friction against the road & tire rubber hysteresis) & gear losses into the whole assumed 3% figure, which looks like it could be off a bit.

In the paper, they also show the Coast Down Test (Vehicle Load vs. Speed) graph the EPA did, which reveals the total energy (power) loss due to air drag, tire rolling resistance, the gears, & electric motor windage with support bearings dissipation, all lumped into one big loss.
The EPA threw it into Neutral at some speed, and then let it coast down to zero speed on level ground.

https://www.researchgate.net/profil...onne-National-Lab-data-on-the-Nissan-Leaf.pdf

denwood said:
If you do play with additives in there, I would 100% recommend a few oil analysis to make sure you're not causing damage.
Sounds like a good idea. The concern is that any added chemicals to a carefully formulated ATF gear oil might cause foaming (aeration) in all that high-RPM gear spinning, and/or it could interfere with the phosphor-boron EP tribofilm. About you can say about additive chemicals is "It might work, maybe, maybe not....".
 
denwood said:
As I've taken apart and built up a few engines (along with pretty much every other job, including full paint jobs) it becomes more obvious that careful maintenance is about 98% of the equation.
And 98% of maintenance is ... lubrication of mechanical parts.

My grandfather taught my 8 year old self that lesson. At the time I thought he was slightly BS'g me but I have come to learn just how right he was.
 
Surely cleaning is part of that 98%. As my neighbor used to say 'clean and lubricate'. That pretty much covers most user-serviceable mechanisms. I'd venture to say it even covers non-moving parts if lubricate stands in for 'oil for corrosion prevention'
 
Here is a comparison of a Tesla Model S (2015 Sport model), Bolt, & Leaf gears that turn at the same RPM as the wheels, that last big gear.
The width of those gear teeth varies depending on how much torque (teeth force) each has to handle.
The Tesla one is a monster. Even a lot more viscosity couldn't help it much since full torque gets applied at near-zero RPM, a real gear-grinding experience.
s1P70hw.jpg


It's from the youtube video https://www.youtube.com/watch?v=MQV3D8F6gvw if you want to hear John Kelly's explanations.

The video also shows a Model S (performance model) oil pump inside (no filter on the S) which keeps oil flowing to the gear bearing nearest the motor that turns at up to 17,919 RPM to cool it too. Our Leafs don't spin that fast, so no real need for a pump cooler/lubricater.

Nubo said:
And there was some kind of "milling noise" problem with the Model S too. So maybe they embraced the habit of over-designing their gearboxes and drivetrains.

I like the part where John Kelly mentions the Silicon Nitride ceramic motor shaft support bearings.
Nubo mentioned the "milling noises" some early Teslas were vexed with, & the ceramic bearings prevent micropitting from the induced current off the motor's fields (or maybe ground current), solving the problem. They are sealed grease bearings on the Tesla and don't share the Dexron VI fluid like the gearbox is bathed in.
 
I'm guessing that the other reason the Tesla drive is so much more robust is that the Model S is a substantially heavier car than either the Leaf or the Bolt. It has to push against all that additional mass whenever you step on the accelerator, and the harder you step the more force is being applied to it...
 
LeftieBiker said:
I'm guessing that the other reason the Tesla drive is so much more robust is that the Model S is a substantially heavier car than either the Leaf or the Bolt. It has to push against all that additional mass whenever you step on the accelerator, and the harder you step the more force is being applied to it...
You would still need the same Tesla heavy-duty transmission in a customized Leaf fitted with a Tesla motor. The way it works, a powerful electric motor will still reach high torque levels up to the back-EMF dominated higher RPM area, regardless of the body mass attached to it. The variable here is acceleration only. Accel = Thrust / mass

And that would be a Rocket Leaf ! F = MA means the lighter body Leaf would accelerate even faster than a heavier Tesla body. Wider stickier tires like a Tesla has, means it's possible to stuff a Tesla motor in Leaf. You might need some extra battery to supply full amps though. Teslas have 85 to 100 kWH ginormous go-boxes down under with wide parallel banks for high amps.

I wonder if anybody has tried to install a Tesla Model S battery, motor, etc. on the back wheels of a Chevy Silverado, that would be nice. No need for an ICE engine up front either. Big frunk there. Off topic.
I have heard of a Leaf powertrain in a Miata, which produced a fun hot-rod.
 
voltamps said:
...
I like the part where John Kelly mentions the Silicon Nitride ceramic motor shaft support bearings.
Nubo mentioned the "milling noises" some early Teslas were vexed with, & the ceramic bearings prevent micropitting from the induced current off the motor's fields (or maybe ground current), solving the problem. They are sealed grease bearings on the Tesla and don't share the Dexron VI fluid like the gearbox is bathed in.

Interesting. Iirc, it's common to use shaft brushes to try and make sure the bearings aren't the best "path" for current but maybe that wasn't sufficient.
 
I'm over +3,000 miles in my current gear oil change experiment, so at this rate, I should have a good amount of mileage by summer when I do the change. Will be able to inspect the magnets and send in another sample to Blackstone Lab to see if an abnormal bits of metal are floating around inside the changed oil. So far so good, my weekly efficiency keeps increasing as the weather gets warmer, starting to hit the 270's range on the GOM :mrgreen:
 
knightmb said:
efficiency keeps increasing as the weather gets warmer
It's possible to put in an adjustment due to temperature, accounting for hot or cold temps affecting mile/kWH usage. Approx, but close enough. .....
Average Energy Economy figure gets reduced by 0.05 miles/kWH "Energy Economy" per degree F. That way you can account for the different temperatures.

The perfect max economy driving temperature, assuming your thermostat is set at 70F for automatic cabin heating or cooling, is about 69F to 73F outside ambient or so. Outside of that ambient, adjust by 0.05 for every degree F deviation, either hotter or colder, to be approximate, based on an Idaho National Labs paper I read.

Instead of 0.05, could be 0.03 is more like what our 2020 models do. 'Splaining below:

Example: Say outside ambient temp is 60F, so that's about 9F from "ideal". 9 x 0.05 = 0.45 miles/kWH Energy Economy reduction.
If it's 30F outside, take 39 x 0.05 = 2 miles/kWH decrease adjustment. That almost sounds excessive, so let's look at the reference source:
From the paper at https://avt.inl.gov/sites/default/files/pdf/fsev/2015LeafColdWeatherTestJune2016.pdf for a 2015 Leaf S with a resistance (toaster oven) heater, so maybe our 2020 models with heat pumps are more like 0.03 for every degree F. So 39 x 0.03 = 1.2 miles/kWH Energy Economy reduction when driving in 30F ambient, delta 39F from the ideal 69F.

This would basically work for driving in hotter temperatures too, higher than the ideal 69f-73F range. If driving around with 85F outside, that would be 12 degrees higher than ideal, so 12 x 0.03 adjustment to account for air conditioner usage & a little less battery efficiency, people have found from experience.
 
Just to clarify the results from that Idaho National Labs paper, here is the procedure in a nut-shell:

Say you're out in TN driving and it's 83F out, and your climate control setting is at 70F, which means the AC is on most of the time.
Take 0.03, multiply that by the diff between 83F and the ideal 73F, which is 10F, to get 0.03 x 10 = 0.3 miles/kWH adjustment to ideal conditions. Adjust all readings back to ideal conditions to compare apples to apples going forward.

Of course you also want to have your tires set to whatever the placard on the door jamb cold psi (my '20 SV tires are a different size as yours, and mine is for 36 psi cold. I don't know what it says on the '20 SL Plus models with diff tires..)
 
The recommended tire pressures for the Leaf (any generation) are an automatic handicap. If I wanted to compare several EVs for range and/or efficiency, I'd inflate all of the tires to 90% of the tire manufacturers' maximum recommended sidewall pressure.
 
The benefits of going from a Nissan Matic S to a thinner Valvoline ULV fluid in ye ole gearbox can't be more than 1%, maybe 4% on very cold days.

Some carefully measured evidence found:
I found a paper at https://www.sae.org/publications/technical-papers/content/2019-01-1296/preview/
and Figure 7 in particular shows a comparison in transmission efficiency (planetary gearsets & bearings, with a torque convertor locked up so no losses there) between 2 different fluids, one a ULV thin one, and the other with a typical Nissan Matic S slightly thicker viscosity, at a typical Leaf gearbox temperature of 100F. At 100F there would be a signficant viscosity spread between those two fluids. In colder weather with city driving cycles, you could get a bigger spread, so in winter time the differences might balloon to 4% I'd say.

The planetary gearsets spinning on bearings resemble our setup close enough, and may even amplify the effects of viscosity a little, especially when you take into account the pumping losses in a typical locked-up automatic tranny compared to our Leaf's pure gears & bearings.

This may explain why, only now, is Ford going to Mercon ULV thin goo in the new Mustang Mach E. A small gain can be found, but small, ignored for durability concerns up to now.

There should be a small benefit in Range going from an old used fluid to a new fluid. Small metal particles could be raising friction in the ball & roller bearings a little. Can't be that much since the gearbox would overheat with too much friction.
 
^^
I'd like to convert the reported friction into work to have an idea of the oil heating kinetics and drivetrain efficiency loss. How do you convert friction into work where gears are involved ?

What is the distance ? I'm inclined to say it is the same as the distance the car traveled but I am not sure. If that is correct then the energy savings are trivial -- on the order of 500 joules/km or ~ 0.22 Wh/mile per gear
 
The paper is nice because it isolates a Range loss due to just the change in gearbox fluid viscosity alone.

Torque x Rotation Degrees is energy expended, and also Force x Distance at the tire interface to the road is energy expended, basically.
Energy Units can be kWH, joules, calories, all the same energy but in different units, like feet & meters r diff units for the same thing.

Power is the rate of change of energy (time derivative of energy) per time unit. If you integrate Power over time you get Energy expended.
All basic physics.

In our gearbox, the energy wasted is the integral of Power over time needed to overcome viscous (hydrodynamic) friction & boundary "metal-on-metal" friction inside, producing waste heat. A good analogy is a leaky garden hose: Some water leaks out, lost water, which reduces the water flow you get at the important nozzle.

When you integrate Power losses to compute Energy loss, the percent loss get preserved in the total Energy lost over time. So the lab results FEV ran for Ford, GM, & Chrysler (Stallantis) in the paper I cited above really is the Range percent lost due to gearbox friction from all sources.
 
voltamps said:
The paper is nice because it isolates a Range loss due to just the change in gearbox fluid viscosity alone.
Spell it out for the dummies like me, please.

Am I wrong in my ~ 0.22 Wh/mile traveled per gear estimate ? **


**
Rather than multiple force by angular velocity, I multiplied force by the car travel distance
 
Back
Top