Axle Spline Fix! ... How to save $2.5K

Axle Spline Fix! ... How to save $2.5K

Summary update -

Rear axle click was totally fixed with $100 service and no new axles. Totally solid after extended use.

If you’re getting a 2.5k estimate to replace them, probably not needed. Give your service tech the info below.


I bought two Model S’s a few years ago, and super happy with both cars.

One recently developed a rear axle ‘clack’ sound on acceleration from a stop, or on shifting between regen and acceleration.

Problem was diagnosed to be slapping of the rear axle splines that connect the wheels to the drive unit.

This occurs when the grease packing between the male and female splines gets compromised. The bare metal spline teeth hit their counterpart, and you’ll hear a loud metal-to-metal clack.

Potentially it can be caused by wear or damage to the spline teeth, which would require a replacement of the rear axles. That runs about $1,250 for each rear wheel.

The protocol is usually to first try re-packing the grease, and replacing the axle nuts that keep the grease in place.

That’s low cost at about $100 total for both axles, parts and labor all-in.

When that didn’t work in my case, I researched it to understand the failure mode, and worked with the service center to see if it could be solved without a full replacement.

What I learned pleasantly surprised me, and I thought I’d share it here to help others.

Here’s a summary -

1. The male spline fits into the female spline at about a 3.5 inch deep engagement.

2. When re-greased quickly (as on first visit), the fresh grease only gets about halfway in.

3. Corrosion at the deeper end of the splines can displace the grease, and give rise to the clacking.

4. To fully resolve, techs need to get solvent and long narrow brushes deep into the female spline.

5. The solvent has to sit there long enough so that all the oxide particulate is loosened and cleaned out.

6. Then fresh molybdenum paste grease is packed deep inside, and also brushed liberally onto the male spline.

7. Once assembled, the excess grease squeezes out, and the new axle nuts are installed to prevent water ingress, or grease leaking.

8. The specific grease is about 65% pure molybdenum. I brought them some Loctite LB8012 from Amazon @ $32.

9. They stock a similar paste at Tesla, Dow Molykote M-77. The loctite stuff had a bit more molybdenum, but also a small amount of microscopic hard quartz. That helps enforce a minimum spacing to prevent the grease from being forced out between the teeth.

It really worked.

The difference was immediately obvious as I drove away. The clack is totally gone, and the axle is now silent, even when subjected to strong acceleration / deceleration cycling.

I think the key to it is really getting all the rust out at the full depth of the female spline coupling.

That made room for the new grease, and then everything worked very smoothly.

Kudos to awesome shop foreman Alfredo at Tesla West Los Angeles, for patiently studying it to figure out what’s needed.

Since I had no luck the first time around, the first data was discouraging. But now those very same parts are performing perfectly.

I’m guessing a lot of techs and customers give up on the first attempt, and think the axles need replacing.

But it looks like this comes down to getting deep into the female spline to fully clean it.

If you have this issue, please share this info with your tech.

It can save time, money, and customer service workload for Tesla.

NKYTA | September 22, 2019


Bighorn | September 22, 2019

Good stuff. Thanks. Maybe Tesla Tap can get this archived so it’s not lost to the recesses.

Mark K | September 22, 2019

Bighorn - one more thing. A big-picture postscript -

Years ago, my Mercedes SL500 got a warning light for its active body control electro hydraulic shocks.

Two weeks out of warranty, with low miles, they quoted $10K+ to fix them. I asked if they refilled the hydraulic fluid, and they said no. I bought a quart of fluid, put it in myself, and they worked fine.

That ended my 20 year relationship which Mercedes-Benz.

Contrast that with Tesla.

I went in, on a Sunday morning, and they fixed it the same day.

The foreman methodically and creatively worked through subtle details, to fix it for 5% of what it might have cost, had service profit been their objective.

No other car company on God’s green earth is doing this today. Only Tesla.

This ethic, this sense of collective interest, is why I expect to be buying their stuff 20 years from now.

And it’s why over time, their market cap will likely dwarf Apple.

Mathew98 | September 22, 2019

Thanks Mark! Did you troll the head technician to propagate this solution up the ladder so other customers can share your fruit of detective work?

So no upgrades anytime soon?

Mark K | September 22, 2019

Hopefully this forum, TMC, and TeslaTap will get the word out. Definitely suggested to staff to share it.

Next upgrade looks plaid to me, maybe with a dose of cold gas thrusters.

RedShift | September 22, 2019

Thanks Mark.

Mine had this at 36K miles, but under warranty Tesla replaced the DU. never got to know the reason for the clacking. Seems from your description, it might have been the splines.

Mark K | September 22, 2019

RedShift - under 48K, everything covered, regardless o whether they replace the entire DU or just one part.

After 48K miles, the 100K DU warranty takes over. But that actually stops at the axles, so per Tesla, they are not covered under the 8 year, 100K mile warranty.

That’s why this info can save a lot of time and money if you’re over 50K miles.

jordanrichard | September 23, 2019

I had the clink noise develop and they went with a cleaning and repacking method for a fix. No issues since.

Anthony J. Parisio | September 23, 2019

Wow thanks for sharing. I am sure this will come in handy.

rm760 | September 23, 2019

It would be great if there could be a sticky thread detailing these know issues and methods of repair.

sbeggs | September 23, 2019

Thanks, @Mark K. Useful to know.

Mark K | October 26, 2019

Posted longer term test results in summary edit of OP. Confirms the fix works great. Car drives awesome.

PBEndo | October 27, 2019

Is this problem more likely to occur in certain areas, such as where road salts are used?

Mark K | October 27, 2019

That seems likely, since the failure mode was corrosion, which then blocked and displaced grease.

The clacking was two oxidized, unlubricated steel surfaces slapping each other.

The other factor is whether the lock washers were loose, which can allow water ingress, and grease to bleed out.

Those washers are only a couple dollars, so they replaced them after cleaning the corrosion, and made sure they were sufficiently torqued.

Given how loud the clack was, I worried if corrosion had compromised enough metal, then the gaps may have grown large enough that grease couldn’t maintain the spacing.

That would force replacement of the metal with new axles.

But after thorough cleaning, and greasing with the very clingy 65% moly plus quartz grease, it was perfect.

The transformation was remarkable.

So a better visual model is not that oxide has removed much metal, but that it displaces the necessary grease.

Remember that the spline is not a simple cross, but actually like a full gear pinion inside a mating ring gear, so there is so much redundancy, it’s very tough and durable. It would be quite hard to cold work all the interface surfaces enough to make it irreparably sloppy with normal use. It was just noisy due to lack of grease in the right places.

Therefore, parts that sound pretty far gone, but are actually fine, and just need thorough cleaning before regreasing.

It‘s a bit counterintuitive from what’s audible, so often techs perceive wholesale replacement is needed, where it really isn’t. Just careful cleaning, and the right grease.

Since the relative cost is 25X cheaper to clean rather than replace, it saves a lot of hassle and needless expense to spread the word among customers and service centers. Ergo this thread to journal what happened.

The perception that Tesla axles wear faster because of the higher impulse torque of electric drive sounds plausible, but turns out not to be the cause.

So I was pretty thrilled with what was learned and how well it works.

Hopefully lots of owners are spared needless expense as the knowledge spreads through the service network.