News
Tesla Model S Drive Unit Replacement by the Numbers
During my early days of researching the Tesla Model S, I had concerns about a number of things that were emerging in the press. Fires, Model S drive unit issues, and handling in the snow. Tesla quickly addressed the “firegate” issue by retrofitting the Model S with a titanium shield. My concerns around winter driving, in particular handling in the snow, also became a non-issue after watching several winter driving videos put together by Bjørn Nyland.
Within no time, I became a new Model S owner and quickly learned that driving in the winter could actually be quite fun with a good set of winter tires. Tesla later introduced the all-wheel drive “D” models putting to rest any remaining doubt that the Model S could be the best and safest handling car in the world.
But one question still remains for me, Has Tesla dealt with the Drive Unit issues?
Model S Drive Unit Replacement by the Numbers
There is a group of Model S owners across the forums, but also validated by major publications such as Edmunds, that have had one or more drive unit replacements.
Early reports of drive unit issues ranged from total failures that left the Model S inoperable, to more recent complaints of clunking and milling noises. Elon has said in the past that many of these pre-emptive drive unit replacements were unnecessary and a simple $0.50 shim was all that was really needed, yet Tesla Service centers have been proactively replacing drive units.
The following polls from TMC show that a majority of Model S owners have had their drive units replaced.
While the poll only represents a small sample of Model S owners, one might draw significance in the numbers as the density circles around owners needing a drive unit replacement at relatively low mileage, and without any harsh driving prior to failure.
The polls also give insight to the motive behind each drive unit replacement.
Though it appears the issues mainly center around earlier versions of the Model S, there’s been reports of newer Model S with Autopilot hardware hearing milling noises coming from the drive unit. One viewer described the sound as “sawing wood”.
Tesla’s Response to Drive Unit Issues
In typical Tesla fashion, the company acted fast to address the issues, and curb concerns by offering an unlimited mile/8 year warranty on the drive unit. I think this was a great response by Tesla which put to rest any concerns about the quality and long term durability of their drive units.
Elon is now talking about building a power train that will last a million miles and I believe Tesla continues to put significant energy into improving the drive units. Tesla continues to honor the warranty and is proactively replacing drive units that show early signs of problems.
My Experience
After 12 months of Model S ownership and 30k miles drive, I started hearing a humming noise from the rear of my Model S during highway cruising. The noise was unusual as I had gotten used to hearing only the noise of the tires and wind, but there was a new noise only noticeable when cruising or decelerating with regeneration at 65 MPH and over. The noise continued to become more audible over the next few months, but I learned to live with it and did not contact Tesla. My personal rule of thumb on noise related issues is to wait for passengers to comment on it first. It’s my self crazy check.
Two months after the original drive unit humming noises began, a higher pitched milling type noise started occuring when traveling at speeds of 20 MPH or less. The sound varied depending on how much power was being used. That’s when I reached out to Tesla.
Tesla service record
Tesla service had me come in for a test drive which they can do on demand without you needing to wait. We drove about a block before the Tesla service technician said the drive unit needed to be replaced.
The explanation was that tolerances inside the drive unit have led to metal particles getting into the fluid around the drive unit, and as the concentration of particles increased, it caused the noise from the drive unit to become louder when under load. The new drive units evidently have better tolerances and are less prone to having metal particles being shaved off.
At that time, Tesla service has been very backed up here in Massachusetts so my appointment for the drive unit replacement was scheduled six weeks out. I put on another 4,000 miles during that wait time, with an increasingly louder drive unit, before it was finally replaced.
When Tesla replaces a drive unit, they replace both the drive unit and inverter as you can see from a copy of the parts list. The last letter in the drive unit part number (“M” in my case) seems to indicate the generation of the drive unit. The later the letter the better, with the “Q”s seeming to have the best longevity according to the forums.
I’m happy to say that I’ve logged 8,000 miles of happy and noise-free driving after my drive unit replacement. The replacement was done recently so my guess is that I have a newer unit, and thus hope this will be my only replacement.
Thankfully for all of us, Tesla has a first class warranty and level of service. While it goes without saying that inconveniences such as this may happen, rest assured that Tesla has you covered and will always strive to make things right.
Tesla welcomed JPMorgan analysts to one of its factories earlier this month, with the Wall Street firm highlighting its findings in a new note to investors. One of the more pertinent pieces of information is that Tesla admitted to slowly integrating Model Y vehicles into its Robotaxi fleet, but it has a good reason.
JPMorgan analysts recently toured Tesla’s Fremont Factory and met with the company’s investor relations team, emerging with a clearer picture of the automaker’s Robotaxi strategy. According to the bank’s note, Tesla is intentionally limiting the addition of Model Y vehicles to its existing Robotaxi fleet.
The firm’s analysts said:
“Tesla indicated it is intentionally holding back on adding Model Y units to the robotaxi fleet, expressing confidence in its ability to scale Cybercab in the near-term. On FSD V15, Tesla views this release as a step-change in performance, comparable to the leap from V13 to V14. The V15 upgrade encompasses seven core technologies, with ~40% of those currently being tested in the robotaxi fleet, where initial feedback has been encouraging.”
JPMorgan after meeting with Tesla recently in Fremont:
“Tesla indicated it is intentionally holding back on adding Model Y units to the robotaxi fleet, expressing confidence in its ability to scale Cybercab in the near-term. On FSD V15, Tesla views this release as a step-change… pic.twitter.com/W9yGCWRT3C
— Sawyer Merritt (@SawyerMerritt) August 20, 2026
Far from signaling delays or doubts about autonomy, the move reflects strong management confidence in the near-term scalability of the purpose-built Cybercab.
Tesla has operated its Robotaxi service primarily with modified Model Ys since launching in Austin and expanding to other markets. Yet the company is now deliberately holding back further Model Y conversions. The rationale is straightforward: leadership believes the Cybercab, a two-seat, steering-wheel- and pedal-free vehicle optimized for high utilization, can ramp production and deployment more efficiently in the coming months.
This dedicated form factor promises better unit economics for the majority of rides, which typically involve one or two passengers, while freeing consumer Model Y inventory for retail sales.
Supporting this pivot is Full Self-Driving (FSD) software version 15, which Tesla describes as a genuine step-change in performance, comparable to the leap from V13 to V14. The update incorporates seven core technologies; roughly 40 percent are already undergoing real-world testing in the current Robotaxi fleet, with early feedback described as encouraging.
Tesla is carefully managing software development to minimize regressions in core driving functions as new capabilities are added. Management positions V15 as the primary gateway to scaling unsupervised FSD. Importantly, the existing AI and Hardware 4 stack is already capable of running V15 and supporting unsupervised operation.
Cybercab itself is only the first vehicle on the platform. Tesla reiterated that additional form factors will follow, pointing to concepts such as the earlier “Robovan” demonstration as examples of how the architecture can evolve.
Tesla’s mysterious Robovan makes a sneak peek with Optimus in Terafab video
Parallel progress continues on the Optimus humanoid robot, which remains on track for start of production in the coming months, with commercial sales possible as early as the second half of 2027. Generation 3 details will be revealed closer to production to preserve competitive advantages, while Generation 4 scope will draw on real-world Gen 3 experience.
JPMorgan left the meeting with a deeper appreciation for Tesla’s manufacturing automation and maintained its $475 price target. The decision to slow Model Y Robotaxi integration is therefore not a setback but a calculated prioritization of a more efficient, purpose-built solution that management believes is ready to scale.
SpaceX CEO Elon Musk announced today that the company will likely attempt to catch the Starship upper stage with its launch tower arms “in a few months.”
In a post on X, Musk wrote, “Looks like we will probably catch the ship with the tower in a few months. If there had been a tower out to sea where we practiced landing the ship, it would have been caught.” He added that the first reflight of a Starship vehicle is expected by the end of 2026 or early 2027, describing it as “a fork in the road of history for consciousness reaching the stars.”
Looks like we will probably catch the ship with the tower in a few months. If there had been a tower out to sea where we practiced landing the ship, it would have been caught.
First reflight of the ship will be either end of this year or early next. That will be a fork in the… https://t.co/O5g9pqrzyo
— Elon Musk (@elonmusk) August 20, 2026
Musk’s prediction comes amid ongoing progress toward full reusability of the Starship system, a two-stage rocket designed for rapid turnaround and dramatically lower launch costs. Catching the upper stage, known simply as “ship,” with the Mechazilla tower’s mechanical arms would mark a major milestone. It would allow both stages to return directly to the launch site for quick refurbishment and reuse, eliminating the need for ocean recovery.
Musk has previously signaled plans for a ship catch. In July, shortly after SpaceX’s wildly successful Starship 13 mission, he stated that the company would attempt to catch the ship with the tower on the next flight unless problems emerged in the mission data review. Earlier comments also outline conditions such as successful soft ocean landings before attempting a land recovery to minimize risk.
SpaceX has solved Starship’s biggest challenge, Elon Musk says
The latest update from Musk adjusts this timeline to a few months, reflecting the iterative nature of the test campaign.
SpaceX has already demonstrated the tower catch technique successfully with the Super Heavy booster on a couple of occasions. The first successful booster catch occurred during Flight 5 in October 2024, when the massive first stage returned to the Starbase pad in Texas and was plucked from the air by the tower arms.
Additional catches followed on later flights, including Flight 7, proving the concept for the booster and building confidence in the system as a whole.
Achieving a similar catch for the upper stage would represent a significant step forward. The ship returns from much higher speeds and greater heat loads after orbital or near-orbital flight. Success would advance SpaceX’s goal of full and rapid reusability, potentially reducing the cost of access to orbit by a factor of 100 or more and supporting ambitions for frequent satellite deployments, lunar missions, and eventual Mars flights.
Musk has long emphasized that true reusability, refueling rather than discarding hardware, is essential for making humanity a multi-planetary species.
As SpaceX continues refining Starship through successive test flights, the coming months will test whether the ambitious catch timeline can be met. The combination of prior booster successes and improving ship landing precision suggests the company is steadily closing in on this historic capability.
SpaceX has achieved an incredible milestone by launching its 11,000th Starlink satellite into orbit.
This accomplishment occurred during the Starlink Group 17-50 mission, which lifted off on August 19 at 04:01 UTC from Space Launch Complex 4 East at Vandenberg Space Force Base in California.
Falcon 9 launches 24 @Starlink satellites from California pic.twitter.com/UscpmAxDls
— SpaceX (@SpaceX) August 19, 2026
A Falcon 9 rocket carried 24 Starlink V2 Mini satellites on this flight, successfully deploying them into low Earth orbit approximately one hour after liftoff. The first stage booster, identified as B1097 on its twelfth flight, landed successfully on the droneship Of Course I Still Love You in the Pacific Ocean.
According to tracking data compiled around that date, this deployment brought the total number of Starlink satellites in orbit to just over 11,000.
The Starlink program began with test satellites known as Tintin A and B, launched on February 22, 2018. The first operational batch of 60 Starlink satellites followed on May 24, 2019, when a Falcon 9 rocket lifted off from Cape Canaveral. Those initial satellites marked the start of a rapid expansion that has continued for more than seven years.
SpaceX has conducted hundreds of dedicated Starlink missions since then, routinely launching batches of 20 to 30 satellites at a time using reusable Falcon 9 rockets. By mid-2026, the company had already surpassed 12,000 total satellites launched across all versions, with continuous replacements for units that deorbit as designed to manage space debris.
Looking ahead, SpaceX continues to expand the Starlink constellation to enhance global broadband coverage, capacity, and speed. The network already serves millions of users across more than 160 countries and supports applications ranging from residential internet to maritime, aviation, and emergency services.
Future plans center on next-generation hardware, including larger V3 satellites capable of delivering substantially higher throughput, which require the increased payload capacity of the Starship vehicle currently under development and testing.
In July, SpaceX submitted an application to the Federal Communications Commission seeking authority for a Gen3 constellation of up to 100,000 satellites. These spacecraft would operate in very low Earth orbit shells at altitudes near 325 kilometers and 475 kilometers. The filing requests use of existing Ku, Ka, V, and E band spectrum along with new greenfield W and D band frequencies between 92 and 275 GHz.
SpaceX states that the expanded system aims to deliver multi-gigabit symmetrical broadband to consumers, enterprises, governments, and billions of AI-powered devices worldwide while handling a majority of global internet traffic. Approval and subsequent deployment would depend on regulatory review and the operational readiness of Starship for high-volume launches.
This ambitious scale reflects SpaceX’s ongoing commitment to providing ubiquitous high-speed connectivity from space.
TESLARATI Premium Membership
$19.99 USD
Tesla admits to slow Model Y Robotaxi integration, but for a good reason
Elon Musk gives a timeline for SpaceX’s first Starship catch attempt
