The scenario is becoming common in independent bays: a customer pulls in with a five-year-old EV. They're frustrated, maybe a little angry. When they bought the car, the dashboard promised 250 miles of range on a full charge. Lately they're barely scraping 180, and they're terrified that the most expensive component in the vehicle is dying.

They look across the service counter and ask: "Is my battery going bad? Can you fix it?"

Automotive diagnostics has always revolved around acute failures: a misfiring cylinder, a leaking strut, a shorted CAN network. EV battery degradation is a different animal—a slow, creeping loss of capability—and diagnosing it means changing how we think about vehicle health.

It starts with a question. When a customer complains about lost range, how does an independent shop prove whether the battery is aging normally or actively failing without taking the BMS's word for it?

Part 1: The Chemistry of Getting Old

Before anyone plugs in a scan tool, we have to understand what we're measuring. The metric that matters is state of health (SOH).

Don't confuse it with state of charge (SOC). If SOC is the amount of gas currently in the tank, SOH is the size of the tank itself. Over time, an EV's tank shrinks. A battery that started with 75 kilowatt-hours (kWh) of usable capacity might have 60 kWh left after five years of hard use.

The tank shrinks for two reasons: capacity fade and power fade.

Capacity Fade: Losing Lithium Inventory and Active Material

Lithium-ion batteries work by shuttling lithium ions between the cathode and the anode during charging and discharging, a process called intercalation. The inside of a cell, though, is a chemically reactive place, and that reactivity is where aging starts.

When a lithium-ion battery is first manufactured, the first few charge cycles build a protective film on the anode called the solid electrolyte interphase (SEI) layer. The layer is essential for stability, but it consumes a small amount of lithium to form. As the battery ages—especially with high heat or aggressive fast charging—the SEI layer keeps slowly growing and thickening, and every bit of growth traps more lithium ions permanently. This is loss of lithium inventory (LLI). Ions locked in the SEI layer can't shuttle back and forth to store energy anymore. Your customer's gas tank just got smaller.

At the same time, the physical structure of the electrodes degrades. Lithium ions forcing their way into the crystalline structure of the cathode and anode make the materials expand and contract, and over thousands of cycles, that mechanical stress micro-cracks the electrode materials. Bits of the cathode can dissolve into the electrolyte, isolating active material from the circuit entirely. This is loss of active material (LAM).

Power Fade: The Rise of Internal Resistance

Capacity fade limits how much energy the battery holds. Power fade limits how well it delivers that energy. As the SEI layer thickens and the electrolyte degrades, the internal resistance (IR) of the cells climbs.

Think of IR as a clogging fuel filter. The tank might still hold a decent amount of fuel, but when you stomp on the accelerator, fluid can't flow fast enough to meet demand. High internal resistance causes deep voltage sag when the battery goes under heavy load. If voltage drops too far, the battery management system (BMS) throttles power output to protect the pack, and the customer feels it as sluggish acceleration or a sudden, artificial drop in the dashboard SOC reading.

The Fast-Charging Killer: Lithium Plating

Then there's lithium plating. Push DC fast-charge current (what most of us still call Level 3) into a cold battery, or into one already above roughly 80% state of charge, and lithium ions arrive at the anode faster than the graphite can absorb them. Instead of intercalating, they pile up on the surface as metallic lithium. That permanently removes lithium from the usable inventory, and the deposits can grow needle-like structures called dendrites, which can eventually pierce the separator and cause a catastrophic internal short.

Part 2: How to Test SOH

That's the microscopic level. Here's how to prove it in the service bay.

Step 1: Read the Onboard BMS Data (and Why It Lies)

Connect a capable aftermarket scan tool, navigate to the high-voltage battery data stream, and find the SOH PID. You might see SOH: 84%.

Treat that number as a claim, not a verdict. The BMS calculates SOH with algorithms built mostly on Coulomb counting, tracking exact amps in versus amps out over time, cross-checked against voltage curves. The algorithm drifts when the customer has sloppy charging habits. A driver who only ever runs the car from 80% down to 40% and plugs it straight back in never lets the BMS see the true top or bottom of the pack, and after months of shallow cycling the calculated SOH can wander a long way from reality. (The drift is worst on LFP packs, whose flat voltage curve gives the BMS almost nothing to recalibrate against. See the sidebar.)

One more reality check: what you can see varies by make. A Nissan Leaf hands an aftermarket tool its cell data willingly; the used-Leaf market practically runs on LeafSpy screenshots. GM and Hyundai/Kia expose usable BMS data through the DLC. Tesla is its own world. The in-vehicle Service Mode will show you pack data, but the deeper tools sit behind locks, and the SOH field doesn't always populate. And with no standard OBD-II connector on Model 3 and Model Y, scan-tool access means a CAN adapter harness and third-party software before you see a single cell voltage. And some OEMs don't publish an SOH PID at all, which means the number on your screen is the tool's estimate, not the BMS's. Know which one you're reading before you quote it to a customer.

Either way, the calculated percentage is the starting point. The truth lives in the raw data.

Step 2: The Static Voltage Check

With the vehicle in Ready mode, in park, HVAC off, pull up the PIDs for individual cell group voltages. An EV pack is hundreds or thousands of cells wired in series and parallel into groups and modules.

Look at max cell voltage, min cell voltage, and the cell voltage delta (ΔV). A healthy pack at rest should be closely balanced; during and near the top of each charge, the BMS bleeds down high cells to match the low ones. Expect a resting delta under 20 to 30 millivolts (0.020 to 0.030V). Note the SOC when you record it, because spreads widen naturally near the bottom of the pack; a delta taken at 10% isn't comparable to one taken at 60%.

A resting delta of 100 mV or more is a glaring finding. The BMS is struggling to hold a degraded or damaged cell group in line with the rest of the pack. (On LFP chemistry, these thresholds don't transfer. See the sidebar.)

Step 3: The Dynamic Load Test

A static check is a baseline. High internal resistance hides when current isn't flowing, so to see the truth you have to stress the pack. Set the test up for repeatable numbers: pack warm, SOC somewhere between roughly 50% and 80%. A cold pack sags everywhere and proves nothing, and a pack down at 15% shows wide deltas that are normal at that depth.

Set the scan tool to graph max cell voltage, min cell voltage, and overall delta. On a safe, clear stretch of road, roll into wide-open throttle from low speed and demand maximum amperage from the pack.

Watch the graph. Under heavy load, every cell's voltage drops. That's normal. They should drop together.

A cell group with serious power fade falls off a cliff compared to its neighbors. If the delta spikes from 20 mV at rest to 250 or 300 mV under hard acceleration, you've isolated a failing module.

Then lift off the accelerator and let regenerative braking shove high amperage back into the pack. Watch again. The weak, high-resistance group often spikes higher than the rest of the pack on regen because it can't absorb the energy efficiently. Sag under discharge, overshoot under regen: that's the signature of internal resistance.

Step 4: Thermal Mapping

Internal resistance makes heat. Pull up the temperature PIDs for the battery modules. Under normal driving, pack temperatures should be close to uniform, usually within 3 to 5 degrees Celsius (5 to 9 degrees Fahrenheit) across modules.

If the dynamic test showed a cell group sagging, check its temperature sensor. A module running 15 degrees Fahrenheit hotter than its neighbors after a hard test drive is strong confirmation. One caution before you condemn it: temperature alone isn't proof, because a restricted coolant channel can overheat a perfectly healthy module. The voltage sag and the heat together are what close the case.

Step 5: Third-Party Capacity Testing

If the cells stay balanced under load but the customer still reports terrible range, the whole pack may have degraded uniformly. To prove that, shops are turning to independent test platforms. Aviloo sells two distinct products: a short plug-in check that reads BMS data in minutes, and a full test that logs the pack across a real drive from full charge down to low SOC, then benchmarks the measured energy against a database of the same model. Either way, you end up with an SOH certificate that doesn't depend on the vehicle's own bookkeeping—the gold standard for proving capacity fade.

Part 3: Interpreting the Data

With the data in hand, you can answer the customer's actual question: dying, or just old?

Scenario A: Normal Calendar and Cycle Aging

The scan tool shows pack SOH of 78%. During the dynamic load test, the delta never exceeds 50 mV, and every temperature sensor reads within 4 degrees Fahrenheit of the others. There's no broken part to fix.

The battery is aging. The chemistry has experienced uniform loss of lithium inventory; the whole tank shrank together. For context, large fleet studies put average degradation around 1.8% of capacity per year, which lands a typical five-year-old pack in the high 80s to low 90s. A 78% pack aged faster than average, but if it aged evenly, it aged honestly. This is normal wear—the EV equivalent of an engine giving up a little compression at 150,000 miles.

Scenario B: Localized Module Failure

The scan tool shows pack SOH of 88%, but under the WOT load test, cell group 43 drops 350 mV below the rest of the pack and runs 12 degrees Fahrenheit hotter. That's a localized failure.

Overall capacity might still be decent, but the one high-resistance group is a bottleneck. The BMS limits the whole pack to protect its weakest link, so one failing module cripples the car's range and performance. Depending on the OEM, the pack can be dropped, opened, and the failing module replaced. That's a real repair path on some vehicles and a dead end on others; a few manufacturers only sell complete packs, and structural or foam-potted designs—like Tesla's 4680 pack—aren't built to come apart. Check parts availability and the OEM's service position before you quote the job.

Scenario C: The Guess-O-Meter Effect

SOH reads 95%, the cells stay balanced under load, temperatures are even, and the customer still swears range has dropped 30%.

Now look at the environment and the driver. The dashboard range estimate leans heavily on recent efficiency in miles per kWh. Sticky aftermarket tires, an 85 mph commute, and a 20 degrees Fahrenheit morning with the cabin heat cranked will crater that number; AAA measured roughly a 40% range loss at 20 degrees Fahrenheit with the heater running, and that's on cars with resistive PTC heat (heat pumps soften the hit without erasing it). The battery is healthy. Consumption is just high, and the Guess-O-Meter is telling the truth about it.

Part 4: Explaining It to the Customer

Diagnosis is half the job. The other half is translating the data into something the owner can act on, and if you start talking about solid electrolyte interphase thickening and millivolt deltas, you'll lose them. Use analogies.

Explaining Capacity Fade

Think of your battery like the gas tank in your old car. When it rolled off the line, it held 15 gallons. Inside an EV battery, chemical reactions slowly build residue on the tank walls. After five years, that residue has thickened, and the tank physically holds 12 gallons. It doesn't matter how long you leave it plugged in; it will never hold 15 again. Our test shows your tank is uniformly at 12 gallons. That's normal aging, not a defect.

Explaining Power Fade

Your battery pack is hundreds of small batteries wired together, like a giant flashlight. We test them under hard acceleration to see how they behave under stress. Most of yours flow electricity smoothly, but one group has developed high internal resistance. Think of a badly clogged fuel filter: when you step on the gas, the car wants power, and that one filter chokes the flow. The computer protects the pack by going only as fast as its weakest link, so that one bad module is cutting your range and power. The good news is we found it.

The Warranty Talk

One of the most valuable services you can offer is warranty validation. Nearly every EV sold in the U.S. carries a manufacturer battery warranty of at least eight years or 100,000 miles. That's the OEM's contract, not federal law, and the distinction matters because terms vary, and the variation is where your customer's money lives. Hyundai and Kia run 10 years. Tesla runs eight years with mileage caps from 100,000 to 150,000 depending on model. Rivian goes to 175,000 miles on some trucks.

Then there's the degradation threshold—the number to check before you promise anyone a free pack. Most OEMs will repair or replace a battery that falls below 70% of original capacity inside the warranty window. GM's threshold is 60%. A Mustang Mach-E you've documented at 65% is a warranty claim; a Chevy Bolt at the same 65% is not.

If your testing proves the pack sits below the OEM's threshold within the warranty period, or you've captured a massive voltage delta pointing to a dead cell group, you've just handed your customer a $15,000 to $20,000 gift. Print the scan tool graphs, print the SOH certificate, and send the documentation to the dealership with them. Expect the dealer to run the OEM's own capacity test before honoring the claim; your job is making sure the customer walks in armed. You become their advocate, and they'll come back to you for brakes, suspension, and cooling system work for years.

Preserving What's Left

Finally, answer the question they haven't asked yet: "How do I keep this from getting worse?" Give them advice that slows SEI growth and avoids plating.

The 80% Rule (NMC/NCA Packs). Set the daily charge limit to 80% for commuting and save 100% charges for road-trip mornings. LFP packs play by different rules; see the sidebar.

Avoid Heat Soak. Don't leave the car parked in the sun for days at 100% charge. Heat plus high voltage is the recipe that accelerates calendar aging.

Fast-Charge Smart. The real plating risk is high current into a cold pack or a nearly full one. Modern EVs manage this well; charge taper and battery preconditioning exist for exactly this reason, and fleet data shows frequently fast-charged cars aging about the same as rarely fast-charged ones when thermal management does its job. The advice that matters: navigate to the charger in the car's own system so preconditioning warms the pack on the way, and don't hammer DC current into a frozen battery that hasn't had that chance.

Back to the Counter

So, is the battery going bad, and can you fix it? Now you can answer. If the pack aged evenly, you can show the customer the number, set it against the fleet average, and explain why nothing is broken. If one module is dragging the pack down, you've isolated it on a graph, confirmed it with a temperature reading, and either quoted the repair or armed them for a warranty claim. And if the battery is healthy, you've saved them from paying to fix winter.

None of it takes a Ph.D. in electrochemistry. It takes a workflow: read the BMS but don't take its word, check the deltas at rest, load the pack and watch which cells sag, map the heat, and when the whole pack is suspect, measure capacity for real. The propulsion technology changed. The rules of diagnostics didn't: understand the system, stress the components, find the weak link, and educate the customer.