Tesla battery refurbishment: when Tesla offers a €20,000 invoice, we dive deep into analysis to understand the root cause.
Our focus is not on being a conventional workshop. We are primarily a laboratory, digging into failures to develop sustainable and reliable solutions that can be deployed across all EV Clinic locations in Europe, and soon worldwide.
Almost six years have passed since our first Model S teardown, when the first S85 batteries started going out of warranty. Yet the data and experience we have collected, including external guidance from Tesla engineering, are still not complete.
One of the most notorious failures, responsible for repeated warranty claims, is BMS_u029. We see it mostly in Tesla-remanufactured packs that were previously repaired in extremely questionable ways. Tesla owners received these packs under warranty, only for them to fail again with BMS_u029. We were among the first to identify this problem and find its root cause in 2021. To reliably pinpoint the issue, you may need around 20,000 measurements and up to 160 labor hours to repair what was previously missed.
We are transparent about this. Even we may fail on the first attempt. The repair is covered by our two-year warranty, but there can still be a 50/50 chance that we will see each other again. We don’t simply replace modules, BMBs or the BMS. We repair them while trying to preserve the integrity, resistance balance and capacity of the battery pack.
Many conventional workshops are cannibalizing what remains of the small early Tesla fleet. We are trying to save it by going deeper into the root cause. This requires time and money. The data-learning curve is painfully expensive.
Another common issue is BMS_F123, which indicates an isolation problem. It can be caused by an internal coolant leak following inexperienced repair work, water ingress through a rusty fuse cover, or dirty and clogged umbrella valves. Once cells have been exposed to water, the outcome becomes uncertain. Water can damage the insulator around the shoulder of the cell, with delayed consequences. A cell failure or another problem can appear even a year later, creating a potential warranty claim long after the original repair.
BMS_f107 and f098/f097 are often signs of prolonged water ingress. Galvanic corrosion of the copper/aluminum bond gradually degrades the contact and increases sense-line resistance. The BMS can then receive incorrect voltage measurements, increasing the safety risk of cell overvoltage. So, to be clear: f107 is not simply “bad welding.” It is often the result of slow water ingress over six or more years of battery life.
This brings us back to something we have repeated for years: the best way to repair a battery pack is to prevent the failure in the first place. That is why proper resealing adds enormous value to every Model S and Model X up to 2022.
But what brought us this far is research, and we cannot publicly explain everything we have discovered. Much of it is a business secret, and sometimes what we ask from customers is simply a minimum level of trust. For example, the root cause of prolonged water ingress is not always an umbrella valve or fuse cover. It can be a cracked weld in the battery enclosure.
The crack can create a capillary path that slowly draws external moisture into the pack, eventually damaging components such as the C27 capacitor. So when you repair C27, you may have repaired the consequence while completely missing the root cause. These cracks can be practically invisible and may not be detected by Tesla’s standard pressure-test procedure, which for this failure mechanism is simply the wrong diagnostic method. Proper resealing does not mean simply replacing umbrella valves and the fuse cover. There are deeper root-cause problems that must also be addressed.
Interestingly, repairing a Tesla battery pack is often relatively cheap in spare parts. Only around 1% of the packs we inspect have what we classify as parasitic cell failure. Around 30% show overheating related to irresponsible operation combined with an unserviced HVAC system. Many remaining failures are related to sudden or prolonged water ingress, or damage caused by previous repair attempts at inexperienced, non-engineering workshops.
The expensive part is labor, sometimes repeated several times before the true root cause is identified. We have received many previously attempted and ultimately unrepairable packs from Serbia, Bosnia, Montenegro and Macedonia. We have seen erased data logs, mixed modules, repeated water ingress and multiple previous interventions. In some cases, even we were unable to recover them.
Why do we work this way? Because our goal is to create tools, procedures and repair methods that allow us to deploy the same standard in every EV Clinic laboratory and eliminate every known weakness. We attempted our first individual cell swaps years ago, using carefully matched cells from donor packs. Today, we can open a Model S module without destroying it and replace individual cells. Step by step, we are getting closer to our ultimate goal: producing a completely renewed battery pack with new cells, even when Tesla eventually declares the original battery obsolete and stops selling it.
One disappointing moment came during our Zagreb-Frankfurt-Zagreb road trip. Over approximately 2,000 km, we saw only one early Model S. Too many failed repair attempts and too little accumulated know-how have already removed many of these “kings of the road” from European roads.
Battery repair looks easy and can even look cheap. But collecting the data required to repair these packs properly, and learning from unsuccessful results, is far more expensive. Nobody wants to buy a new battery pack. At the same time, once a pack has been opened outside Tesla, replacing it through Tesla can become problematic because the exchange process depends on approval of the old pack for the correction/exchange program. In some cases, vehicles can also face service restrictions after third-party battery interventions.
Can we keep these masterpieces on the road alone? Can we repair every battery pack? No. This requires collective awareness, not only from a safety perspective, but because these cars are part of automotive history. They changed the industry and deserve to remain a statement on the road for many more years.
After everything we have seen in the data, another point is important: Tesla did not simply “limit charging speed with software” for no reason. In many cases, those changes improved safety. Behind what appears to the owner as a software limitation can be an invisible physical root cause, including increased internal cell resistance.
The 18650 battery architecture remains, in our experience, one of the best cell technologies we have worked with in terms of cell-failure PPM, repairability and overall system design. Even compared with Model 3/Y cell quality, the old 18650 technology continues to demonstrate remarkable quality and an exceptionally low failure rate.
All EV Clinic laboratories work more than eight hours a day, often including weekends. We are doing everything we can to develop technical solutions while remaining transparent about what we find. We need more locations and more engineers. Some of us genuinely lose sleep over this mission.
What we need in return is understanding, especially with warranty cases. Every component or battery under EV Clinic warranty needs to return to us untouched so we can preserve the first forensic opportunity, identify the root cause and prevent the same failure from being repeated for other Tesla owners.
That is why, in selected cases, we offer two years and unlimited mileage on LDUs, despite an observed 3-5% return rate. For battery packs in suitable condition, we may also extend the standard one-year coverage to two years with unlimited mileage. But we cannot accept a warranty claim after another party has opened or repaired a battery pack or LDU still under our warranty. At that moment, we lose the root cause, the data and the first forensic opportunity to understand what actually failed.
To all Tesla community worldwide – thanks for your support – You built a EV Clinic and we will keep improving!
**we never mentioned other workshop who have their owne strugles, we only support them with the tooling and solutions. We never metnion other workshops in a bad way or with bad intentions. This is forbiden in EV CLINIC network – while other workshops sometimes are using that bad habbit to collect points with customer.
