The 10-minute diagnostic check before you buy a used VW or Audi

A ten-minute scan will not tell you whether a car is a good buy. What it will tell you is what the seller has and has not fixed, and whether the fault memory was wiped an hour before you arrived — which is often the more useful information. A large part of that routine is covered by an inexpensive generic reader, because the OBD2/EOBD standard includes emissions fault codes, readiness monitors and, on many petrol cars, per-cylinder misfire counts. The gearbox, airbag, ABS and comfort modules, the service history in the instrument cluster, DSG clutch adaptation and DPF soot and ash loads are all manufacturer-specific, and need a VAG-level tool. Below is the routine, in order, and what each reading honestly proves.

Before you plug in anything

Two things decide whether the visit is worth making. The first is that the engine should be cold when you arrive. A warmed-up car hides a cold start — the rattle, the smoke, the rough thirty seconds — and a seller who has 'just popped to the shops' before every viewing may be doing exactly that. Say when you book that you would like to start it from cold. If you turn up and the bonnet is warm, that is itself a data point.

The second is permission. You are plugging a device into the car's diagnostic socket, usually under the dash on the driver's side. Ask, plainly, before you get the laptop out. A seller with nothing to hide almost never minds; hesitation is worth noticing, though it is fair to say some people are simply nervous about a stranger's electronics in their car.

Do the free checks the night before. In the UK the government's online MOT history service is free and public: it lists past results, advisories and the mileage recorded at each test. Mileage that jumps backwards, or a long gap with no test, tells you where to point your questions. Most countries with periodic inspection keep some equivalent record — worth finding out what yours publishes before you drive anywhere. A separate provenance check for outstanding finance and write-off records is a different service again, and cheap relative to the car.

• Note the odometer reading before you connect — you will want to compare it with the mileage stamps inside stored faults
• Full scan of every module the tool can reach; save or screenshot the report
• Readiness monitor status
• Service history stored in the instrument cluster
• Gearbox adaptation and clutch values if it is a DSG
• DPF soot and ash load, and regeneration history, if it is a diesel
• Misfire counters and fuel trims on a petrol car — at cold idle, then again after the test drive
• Read only. Do not clear anything on a car you do not own

The full module scan, and the words that matter

A generic reader talks to the engine management and reports emissions-related codes. A manufacturer-level tool walks the whole car: gearbox, ABS, airbag, steering, climate, central electrics, cluster, and whatever else is fitted. On a used purchase the interesting faults are rarely in the engine — they are the airbag fault the dashboard lamp has been taped over for, the ABS sensor, the module that no longer answers at all.

Two labels matter. A static or present fault is happening now. A sporadic or intermittent fault is not present at this moment but has been logged — and those are the ones people clear before a viewing, because they are the ones that come back. Do not dismiss sporadic faults. An intermittent ABS or gearbox fault is still a fault; it is just waiting for the right conditions.

The detail most people skip is the environmental data attached to each fault. On many VW group modules a stored fault carries a mileage reading, sometimes a date, and a counter for how many times it has occurred. This is quietly powerful. A fault logged at a mileage higher than the odometer now reads is worth a hard question, and a fault with a high occurrence count is a recurring problem rather than a one-off. Two honest caveats: a second-hand cluster or control module fitted to the car brings its donor's mileage stamps with it, so a high stamp is not automatically a wound-back clock; and a scan cannot prove tampering on its own. Combined with the inspection mileage record, though, it is a good deal harder to argue with.

Readiness monitors: what a freshly cleared memory looks like

Readiness monitors are the car's own self-tests — catalyst, oxygen sensors, evaporative system, EGR and so on, depending on the engine. Diesels run a different set from petrols, and on any given car some monitors simply do not apply and will report as not supported rather than not ready; do not read that as a fault. The supported ones complete gradually as the car is driven across a mix of conditions. Clearing the fault memory resets them all, and so does disconnecting the battery.

So: a car that has just been driven to meet you, showing no stored faults and no completed monitors, has had its memory cleared, its battery disconnected, or a control module replaced or reprogrammed recently. That is the single most useful red flag in the whole ten minutes, because it is objective and it is hard to explain away casually.

Be honest about what it means, though. It is not proof of dishonesty. A new battery, a recent repair, a car that has sat unused — all reset the same counters. What it does mean is that your evidence has been erased, so you are now buying on the seller's word rather than the car's own record. The reasonable response is to ask what was done and to ask for the invoice, then drive the car and rescan. A genuinely live fault will usually reappear quickly; some monitors take longer than a test drive to complete, so an incomplete set after twenty minutes proves nothing either way. In some markets an incomplete readiness set can also fail an emissions test — the rules vary by country, so check the ones that apply to you.

Service history in the instrument cluster

Many VW group cars record services in the instrument cluster: date, mileage, and what type of service was carried out. How many entries are kept, and whether they can be read at all, varies quite a lot by model and year, and newer cars increasingly hold the record in the manufacturer's online system rather than in the car itself. So this is a check that either works or does not, depending on what you are standing in front of.

Where it does work, it is a cross-check on the folder of stamps rather than a replacement for it. You are looking for entries whose dates and mileages line up with the paperwork and the inspection record. Gaps, or a set of entries that appear suspiciously tidy and recent, are worth a question.

The honest limit: those entries can be written by anyone with the right equipment, and resetting the service interval indicator proves only that a button was pressed, not that oil was changed. Treat the cluster record as corroboration. Receipts with a garage's name on them remain better evidence than anything the car can tell you.

DSG: adaptation values, clutch wear and the test drive

DSG boxes come in wet-clutch and dry-clutch forms across the range, and they behave differently. Both use a mechatronic unit — the valve body and control electronics — which is the expensive part when it goes wrong. A manufacturer-level tool can read the gearbox module's fault memory and its live measuring values, including clutch adaptation and, on some units, a clutch wear figure.

Resist the urge to treat any single number as a verdict. These values move with gearbox oil temperature and with how recently adaptation was performed, and the sensible way to read them is against a known-good reference for that exact box or with a specialist's experience behind you. Anyone quoting you a universal pass/fail threshold for a DSG clutch is guessing, and so would I be.

What genuinely tells you something is the combination: stored faults in the gearbox module — particularly clutch, pressure or mechatronic faults, sporadic ones included — together with how the car actually drives. Hesitation or judder pulling away from a standstill, a shunt on the low-gear shifts, a refusal to select a gear. If there are gearbox faults stored and the seller's answer is 'it just needs an adaptation doing', price the car as though it needs a mechatronic repair, because that is the downside case and it is a specialist job. Also ask when the gearbox oil and filter were last done on a wet-clutch box; dry-clutch boxes are generally not serviced the same way, though the mechatronic unit has its own fluid.

Diesel: soot, ash, and what a scan cannot see

Soot is what the DPF collects and burns off during a regeneration. Ash is what is left behind afterwards, and it does not burn off — it accumulates for the life of the filter and eventually determines when the filter needs replacing. A manufacturer-level tool reads both as measuring values, along with the distance and time since the last successful regeneration and, on many cars, a count of regenerations performed.

The useful reading is the relationship between them. A high ash load on a car with modest mileage means the filter has had a hard life — a lot of short journeys, or an engine consuming oil. Frequent regenerations, or a regeneration that keeps starting and never completes, points at something upstream rather than at the filter itself. A car used mostly for short cold trips is the classic case, and it is worth asking directly what journeys the car has done.

Three honest limits. First, these are manufacturer-specific values; a generic reader will typically not show them, though some display a couple of standard DPF temperature or pressure readings. Second, on most cars soot and ash are values the ECU calculates from a model rather than measures directly, and they can be reset with a diagnostic tool — so a low figure is a claim by the software, not a look inside the filter. Third, and following from that, a car whose DPF has been removed or hollowed out, with the software altered so nothing complains, can present a very calm-looking set of numbers. If the readings look implausibly clean for the mileage and there is no regeneration history at all, look underneath at the actual filter. In the UK the MOT includes a visual check for it, and using a road vehicle whose emissions equipment has been removed is an offence; emissions and road-use rules vary by country and are the owner's responsibility. Do not buy a car with a missing filter on the assumption you can keep it that way.

Misfires and fuel trims: the part a cheap reader does well

On a petrol car this is where the standard is genuinely on your side. OBD2 includes on-board monitoring test results — usually called Mode 06 — and on many petrol cars that includes misfire counts per cylinder. An inexpensive reader paired with a decent app will show it. VAG-level tools show the same counters as measuring blocks, often with a bit more context, but the underlying information is available to anyone. Diesels are a different picture: they run a different monitor set and generally do not expose per-cylinder misfire data this way, so on a diesel this part of the check leans on the DPF and fault-memory sections above instead.

Read it twice: at cold idle before you drive, and again after the test drive. A count that climbs on one cylinder points at that cylinder's coil, plug, injector or, less happily, its compression. Counts spread evenly across all cylinders more often point at fuelling or a sensor.

Live fuel trims are the other free win, again on petrol engines — trim is a product of lambda-based fuelling control, and a diesel will not report it in the same form. Look at long-term trim at idle and at a steady cruise. A large positive number means the engine is having to add fuel, which points towards unmetered air getting in or a fuelling shortfall; a large negative number is the reverse. On a car whose memory has just been cleared, misfire counters and fuel trims are your best short-term evidence, because a real problem starts writing new data within minutes of the engine running.

Cheap reader, VAG-level tool, or a specialist?

Be honest with yourself about which situation you are in. If you want to know whether there is a stored engine fault and whether the memory has just been wiped, an inexpensive generic reader covers it and you should buy nothing else. If you are looking at a DSG or a diesel, or you want the whole-car picture including airbag, ABS and the cluster's service log, that is manufacturer-specific territory and a generic reader will not get you there — not because it is a bad tool, but because those modules are outside what the emissions standard defines.

And if the car is expensive, unfamiliar, or has a cambelt, rust, suspension or accident-damage question hanging over it, pay a marque specialist for a pre-purchase inspection. A scan tool does not put the car on a lift, measure paint depth or check a chain tensioner. For a single purchase, an independent inspection is usually better value than any tool you could buy — that is true even though we sell one.

Where a tool does pay off is if you are keeping the car afterwards: service interval resets, retracting an electric parking brake to change pads, coding a replaced part, reading live data when something goes wrong on a Sunday. VAGPULSE is that kind of tool — Windows software with a cable, bought outright rather than by subscription, for module scans, live measuring blocks, coding, adaptation and service functions. What it covers on a specific car, and current pricing and dispatch times, belong on the product page rather than in a guide. Worth saying plainly: if you are viewing a car this weekend and do not already have a cable in a drawer, no tool purchase solves that. Borrow one or book a specialist.

Two rules whatever you end up using. First, during a viewing, stay read-only: scan, read values, save the report, and leave. Do not clear fault codes on a car you do not own — you destroy the evidence you came for, reset the readiness monitors, and leave the next buyer blind. Second, coding, adaptation, immobiliser work and flashing are not driveway activities at all. They belong on a car you own, with a stable power supply — a healthy battery plus a proper charger or support unit, because a voltage dip part-way through a write can leave a module needing bench recovery or replacement — and with the original file and the original coding values recorded before you change anything. Coding and adaptation are usually recoverable if you wrote down what was there before. A failed flash may not be. Immobiliser and component-protection functions are tied to the car's security system and are frequently a genuine dealer or specialist job, whatever a forum post tells you.

Common questions