RV Problem Solver

This post is as much confession as a warning over several mistakes I’ve made while using repurposed Tesla batteries in a home solar RV power configuration, and why I would not use them in this application again. There is a great deal of varying information out there about using these batteries for “secondary” purposes, and in this post I highlight my own experience in doing so and what I’ve learned—sometimes the hard way— about using them in this way.

If you’d prefer the TLDR version, do not use a Tesla battery (or any comparable repurposed EV battery) in an RV power storage application, unless you’re obsessive and want to spend a bunch of time tending to them.

Regardless of one’s opinion of Tesla as a company, there is little doubt of their ability to design a battery module. The ones I’m talking about here are the model S and X type rated at 24volts and around 233 amp-hours of capacity each. The engineers have certainly outdone themselves with these battery modules. Built like brick houses, Tesla batteries are constructed to withstand quite harsh environments. And while far from indestructible, a great deal of well-crafted engineering went into their design and build, not just for performance but also safety. Therefore, after servicing their original intended purpose, it’s no wonder these modules have plenty of life left for secondary use purposes as power storage devices.

And there’s no shortage of used Tesla model S modules on eBay. Entire secondary markets have even cropped up around “legitimate” vendors selling recovered modules from decommissioned Tesla vehicles. Of course, why not? A single one of these modules stores over 5 kilowatts of power. That’s 5,000 watts from a device slightly larger than a briefcase! Compare that to any lead-acid battery and there is frankly no comparison. Compare it to other lithium-ion options –drop-in replacement, roll-your-own, or another repurposed battery technology – and these modules still hold their own on price, performance, and size.

So, with such a glowing opinion of them, why would I not use these batteries in an RV storage solution (again)? To answer that question, we need to take a slight dive into battery chemistry. If you’d like a more in-depth understanding of what a lithium-ion battery is and how it works, see this post. So, I won’t get too deep in the weeds here, but most batteries, lithium-ion or otherwise, are a chemical soup of electrolyte and metals. This soup converts electrical energy –from solar, generator, or shore power—into charged ionic potential and then releases it for later use. It does this through a series of clever tricks of chemistry that trap positively charged ions (in this case Lithium) at the anode of the battery. When an electrical connection is made between them, the positively charged ions are pushed from the anode (negative) to the cathode (positive) through the electrolyte. This triggers a release of now freed electrons flowing in the same direction, but through your electrical devices, all in an effort to rebalance the charge by producing electrical potential. This clever bit of chemistry obviously involves the use of chemicals, in various forms and configurations, to accomplish its magic trick. The properties and behavior of these chemicals are at the core of concerns over these batteries.

I’m sure by now you’ve seen a fair share of articles referencing EV fires and how all signs point to the batteries used in them. While these news reports make for sensational headlines, it’s important to keep the hype in proper context. In over a decade of large-scale use of lithium-ion batteries in vehicles, there have been several dozen reports of such incidents. This seems like a lot. But consider the fact that roughly every five minutes a gasoline powered vehicle bursts into flames for one reason or another, with nearly 300 such incidents a day, and lithium battery fires are not even a statistical margin of error by comparison, even when accounting for the difference in the number of units in use. We’ve simply gotten used to the fact that gasoline and other fuels burn, often for the simplest and most benign of causes, and we don’t give a second thought that every day hundreds of millions of vehicles traverse billions of miles powered by trillions of explosions. But statistical error or not, a fire is still a fire and be damned the stats, how does a lithium one start?

To answer that question, it’s important to know that there is a lot of potential energy stored in a lithium-ion battery. That energy is contained between the anode, cathode, electrolyte, and a separator all housed within a protective, convenient, and portable case package. When these components are operating at their intended design, lithium-ion batteries are one of the safest, cleanest, and most efficient forms of energy storage we have developed to date. But when things go wrong, they do so in a hurry, and with often tragic consequences. The good news is that in the majority of cases, “things go wrong” under the most extreme of scenarios—such as severe physical damage to the cell casing or misuse and abuse of the chemical magic trick inside, that causes the sudden and catastrophic release of that energy as heat (and sparks). Much like with gasoline, the core of many lithium-ion batteries is a contained chemical fire, and that heat is sufficient to ignite the plastic separator and organic solvents used in the electrolyte. Once one cell goes, the reaction is self-perpetuating enough to ignite its neighboring cells in what is called “thermal runaway.”

So, all lithium batteries are bad and I should stay away from them? Not at all. Quite the opposite, actually. All lithium batteries, when handled and managed properly, are far safer than say a gasoline powered generator. With gallons of highly volatile fuel, fed through a series of tubes and hoses, a combustion engine produces thousands of very hot, contained explosions per minute. But all forms of making and storing power involve a certain level of risk and the proper management of that risk. So far, I’ve referenced “lithium-ion” as a singular technology with ubiquitous behavior. But here is where we must separate the lithium from the electrolyte.

Tesla batteries (along with most other EVs and portable devices) use chemical compounds designed to derive the absolute most from the above referenced chemical magic trick, in the smallest package possible. While this design makes for extremely efficient and compact batteries, these compounds (formed with Nickel, Manganese, and Cobalt) are highly reactive and toxic, and the electrolyte used in them is itself a fire risk. And there lies the heart of the problem. When things are working as they should, these compounds do splendid work powering anything from headphones, to laptops, to semitrucks. But when they fail, the resulting fire is often quite intense and very difficult to extinguish –as a matter of fact, most guidance for first responders to these types of fires is to let them burn themselves out, only containing and minimizing incidental contact or damage to their surroundings.

And here is where I feel these types of modules may not be appropriate for an RV environment. If you’ve ever seen an RV burn, it is truly terrifying just how quickly one of these things goes up. And there is usually very little that can be done, with a quick exit from the rig being the highest priority and often only option. So, in my opinion, minimizing as many sources of ignition as possible is important. And to be clear, having gallons of propane stretched out in rubber hoses to various appliances or an “RV” refrigerator powered by a super-heated ammonia compound qualify as equal if not greater risks. So, reducing or eliminating these risks remains paramount. With the latter, a regular inspection of all propane systems and hoses and ammonia-charged refrigerators is essential. But the risk from lithium batteries doesn’t necessarily come from the batteries themselves alone. While these modules do, and have, spontaneously caught fire, in many more cases the people who use them, and how they use them, and the minimal understanding of the above risks and issues are a factor.

This is where I think many people might make some crucial mistakes, and I count myself among them. When I first got these batteries, I was quite excited by their capacity, size, and amazing performance that I overlooked their potential downsides. Many who utilize these batteries in an RV can often get away with just one or two of them to fulfill their power needs. And thus, they believe, as I did, that you can just manage them “manually” using a simple voltage meter. Along with that first mistake, however, many relegate these batteries to some hardly seen location in a bay down below somewhere, constantly exposed to temperature extremes that further degrade their components. Add to it the increased possibility of vibration and impact of things bumping around in a bay and couple that with modules five, six, or more years into their lifecycle, and the potential for disaster mounts ever greater. I’ll discuss these mistakes and why they’re easy to do yet can be disastrous.

Management: Many have the impression that just one or two of these modules somehow involves “less management” than say twelve of them. If, like me (and the countless wandering nomads I’ve met along the way), you’ve thought to yourself: “Self, I just have one or two of these batteries. I can certainly keep an eye on them myself, and not worry about a big, expensive, and overly… obtuse management solution for these batteries,” in reality, nothing can be further from the truth. The proper management of these modules involves the crucial need of managing the individual cells that constitute a whole battery. And a system with only one or two modules/batteries has cells that are actually working much harder and more often than one with a dozen modules. All of that aside, no matter how well you watch them, you cannot overcome the laws of physics.

A Tesla module consists of six cell-packs, each composed of dozens of individual cells in parallel, all connected in series. As such, power flows into and out of these cell-packs unevenly. This is true of any battery, lithium or not, when connected especially in series, but also due just to chemistry. In a lithium-ion battery these differences are small, but no matter how good/fancy/expensive your charger is, electrons will have an easier time entering the anode (negative) side as voltage is raised on the cathode side and the battery is being charged. In my experience, this causes a disparity on the last couple of cells closest to the cathode side, causing those cell packs to reach max voltage before the remaining packs further down the line. When the module is newish, and hopefully already well balanced, that may not be much of an issue as the differences will be quite small. But in as little as a couple of months of regular usage, the cells will drift further apart, if they’re not being balanced. This can pose a very dangerous situation, in a small window of time, if all you’re doing is passively “watching” the batteries on occasion, looking for problems. Because if those cell-packs reach and exceed rated voltage (4.2v), while the remaining cell-packs are still below voltage, your charger will have no idea anything is wrong. The charger is just watching for overall maximum system voltage (25.2v), which could easily still be in range, regardless of what the individual cells are doing. Imagine cell voltages at 3.9, 3.9, 4.0, 4.2, 4.3, and 4.3. The sum voltage of 24.6v is well below total voltage limits (with room to spare), yet the higher-voltage cells at the end of the series are undergoing significant stress and being pushed beyond their tolerances. This might only be evident when it’s being charged, as visibility to it goes away when the battery is later being discharged. In the short-term best case, this can reduce the capacity and available cycles of those cells over time, further exacerbating the problem. But in the worst case, this can cause one or more of the hundreds of cells within the cell-packs to swell under the stress and burst/explode, starting an unstoppable cascade. So, if like me, you thought that a simple, passive, voltage monitor that you reference “often enough to keep an eye on things” is enough, please rethink that plan immediately and install a fully functional Battery Management System(BMS) that not only actively watches each individual cell-pack closely, but balances each cell-pack individually and has direct and complete ability to remove any and all modules from the circuit at the first sign of trouble. If you take nothing else from this post, please make it this one.

Mounting and temperature: These modules are built so well that it’s hard to envision how fragile they can be. Many people, by now, know that you shouldn’t charge any lithium battery in cold temperatures (below freezing). And that is true. Doing so causes irreparable damage to the internal cell structure that will eventually render the affected cells, and subsequent battery, unusable. And I’ve seen most people account for that with some sort of heating system for colder temps. But what a lot of people might miss is that hot weather can be as bad, if not worse, for these batteries in particular. Heat is a killer of all electronics, but with these batteries, heat adds to the stress load discussed earlier and reduces their ability to perform. Under excessive heat – think Arizona in the summer, then stuffing these modules in a bay with little to no air movement, while charging them as you soak up that searing, mid-summer sun, and you have a sure-fire recipe for disaster. While the modules are engineered for extremes, and in the use-case of a typical RV power system they’re hardly being pushed at all, compound the above conditions with 444 possibilities (total number of individual cells in a single module) that one will go poof and start what cannot be stopped. Here again, a BMS is your friend. Any decent BMS has the ability to monitor at least two temperature locations on each module and act accordingly if the temps exceed specific tolerances –high or low. That action can be as simple as disconnecting the battery to reduce its stress-load or as complicated as also activating a secondary cooling and ventilation system.

Mounting location: This might not seem as obvious, but I believe it to be a significant risk. In the original electric vehicle the modules are installed, lying flat, side by side in a single layer, bolted to and encased in a protective metal box, and cooled (or heated) using a glycol-based coolant much like an average car. But in a “secondary” aftermarket use-case, glycol cooling is often unnecessary, messy, and cumbersome. Add rack rails and mounting points, along with an overall solid feeling frame of each module, and it’s easy to see how one would be tempted to mount them in all sorts of ways. I’ve seen many mounting configurations of these batteries, and I applaud the creativity it took to squeeze them into some very interesting places. But, mounting them in an orientation other than original – right-side up and flat— is taking your own risks as to the engineering and build tolerances of their design. Using the batteries in locations where it’s mounted to a wall, or hanging from a ceiling, etc. adds new stress variables to the frame of the module and all the interconnected components inside, stresses the designers likely had no intention to account for the full extent of these tolerances. And while Tesla’s Powerwall uses basically the same modules in a “wall mounted” configuration, even then, that is only one layer of two modules, bolted into the metal frame using robust mounting brackets, and is not expected to be moved again. An RV is a constantly rolling earthquake that might add a whole new set of loads to the frame and rails of a module, in particular to the later modules utilizing plastic rails. Doing so might be just fine, or it might be completely devastating, and there is no way to know until you have a problem as you become a beta tester of this clever bit of engineering.

In addition to the orientation of installation is the concern of surrounding things in the installed location (like boxes, or tools, or random bit of whatever banging around a bay). Right-side-up and pretty might look great, and you’ve even done a wonderful job running all the wires nice and clean and it truly is a showpiece. But now try to imagine what that exact spot would look like if it were upside-down and spinning like a washing machine, as would be the case in an accident, or a tornado, or… invading, power-crazed zombie hoards. If that module is not firmly attached to its mount points, it will go flying who knows where and do who knows what on its way there. And if the bay or location where it’s mounted is also your tool storage, lawn chair storage, jack storage, 100lb dumbbell storage, or… zombie survival kit storage, what happens when you make a tight turn and bump over a high curb trying to settle into that perfect boondocking spot? It’s important to keep these factors in mind as it would take minimal force for something like a jack-stand or a toolbox to shatter through the top cover, smash into one of the cells, and trigger a cascade.

If you’ve read this far, I applaud your perseverance. I realize there’s quite a bit up there to go through, but I wanted to be as thorough as I can, without writing a dissertation. As I’ve stated, these battery modules are well designed and built. But using them beyond their original design carries some risks. While I believe these risks are manageable, they are likely beyond what the average RV user wants to mess with. Below I highlight a few of the steps I’ve taken to mitigate these risks.

1. I have the modules mounted inside the rig. While it certainly would be more convenient (and prettier) to shove them into one of the bays below, I want to be able to keep a constant, physical eye on these modules, especially as they age. I also want them kept in a temperature-controlled environment. Our batteries experience the same temperatures we do, hot or cold. So, as long as we’re comfortable, I’m comfortable they’re comfortable.
2. I have them mounted in a ground isolated, aluminum rack that allows for ample air movement around all sides of each module, and each module is firmly secured to the rack using the original mount points, with the rack itself bolted and strapped to the floor.
3. I have fuses on top of fuses, fusing fuses. OK, maybe not. But I do have a dedicated fuse per module installed on the positive terminal –as close as possible to isolate all wiring. I also have an additional master fuse for the entire system, isolating the complete battery system from all loads (inverter, charger, converter). I have each connection running into and out of the battery system on its own fuse (solar charger, secondary DC charger). And I have a large, master switch that shuts off the entire system within easy access.
4. I have each individual module on its own BMS. Each BMS runs independently of the charger or any other component in the system. And each has the ability to completely isolate any single module from the system for any trigger condition such as high or low temps or voltage, among several other variables. More advanced BMSes can link with the charger and “tell” it about itself and how to regulate power flow, and I’m generally a fan of more data, more logs, and more control, but there is also something to functional simplicity. In addition to the BMSes, I have added a dedicated active balancer to each module. If you’re not familiar, these units balance each cell within the battery by passing it around from high to low cells, rather than wasting it on a big heater/resistor and a matching fan to carry all that heat away to do it, as is the case with most “fancy” BMSes.
5. I have added an arsenal of fire-fighting gear within arm’s reach including several fire extinguishers, a fire blanket, and gloves. I also have a dedicated combo photocell and ion-based smoke detector mounted next to the battery location.
6. I do NOT charge the modules to their full-spec voltage. A couple of years ago Tesla got some flack for overriding the default settings of their cars, without bothering to tell, let alone ask, the actual owners of these cars they did so. The settings reduced the peak-charge voltage of each module to 95% of original. Along with that, they made a recommendation to their users to avoid keeping the batteries at full charge fulltime, if full vehicle range is not required. Like Tesla or not, I’ve taken these settings changes to heart, and I have my entire system derated to roughly 90% of the batteries’ actual capacity. Along with that, I have my system configured to keep the batteries charged to no more than 70% of capacity when we’re stationary somewhere and have ready access to shore power.           

As you can see, I’ve taken this matter quite seriously with several steps to minimize what risks I can. Many of these actions came after realizing just how wrong I had been doing things. I’m a hacker at heart and I love figuring out how things work and devising new ways to use existing things. Fortunately for me, I am also obsessive, and had been closely watching the performance and behavior of these modules and witnessed firsthand as things slid further and further into sketchy risk territory. There is no way to eliminate all risk, but I feel confident I’ve done proper due diligence to minimize and manage such risks. And hopefully I’ve helped you to do the same. But if what I’ve outlined seems excessive, or if it’s something you’re unwilling to put similar effort into, I completely understand. But along with that, I would strongly advise that you NOT use these modules, whether from Tesla or Nissan Leaf or Chevy Volt in an RV power system installation.

By now, I’m sure some have been screaming at their screen, “What about Lithium Iron Phosphate?” And I agree. LiFePo batteries are a fantastic alternative to these reused modules, what I had before installing these modules, and frankly what I would do if I rebuild this system again. LiFePo chemistry suffers little of the fire concerns of repurposed batteries, as its chemical bits and pieces are less flammable/reactive, and has more headroom for error. But they are also not near as power dense –often 60% that of the above modules relative to size, making them somewhat heavier and larger, are often quite pricey (if you’re not willing to “roll your own”, which gets into a whole sourcing of the cells issue, and reliability, and configuration, and…) and they carry their own concerns and issues. But they are a topic all their own and beyond this particular post, and if one is not willing to go through most of the above, then LiFePo batteries (from a legitimate and trusted source) are absolutely the right decision when considering lithium power storage in your rig.

In conclusion, repurposed battery modules from Tesla (or Nissan Leaf, or Chevy Volt) are a great platform for adding high quality and efficient power storage. But they are also serious pieces of tech, and they require serious attention. That attention is essential to not only keeping them running at their peak performance for a long time, but to keep them doing so safely. If, like me, you’ve committed to a path of using these modules, please take a few minutes to review your installation and see if any of the above concerns apply to you and whether you should consider a change.