A scan tool tells you a leak exists. It doesn't tell you where. Smoke testing pumps low-pressure visible vapor through a sealed system and lets us watch where it escapes — pinpointing EVAP, vacuum, intake, boost, and exhaust leaks down to a 0.010" hole. The single most powerful diagnostic tool for a half-dozen check-engine-light scenarios.
The roots of smoke testing trace back to the mid-1990s, driven by two converging pressures. First, the EPA and California Air Resources Board (CARB) were tightening evaporative emissions standards on a year-by-year basis throughout the 1990s — the rules around how much hydrocarbon vapor a vehicle was allowed to release into the atmosphere kept getting stricter. Second, OBD-II diagnostics became mandatory on all U.S.-market vehicles in 1996, which meant every car on the road would self-monitor its evaporative emissions system and set fault codes when leaks occurred.
The combination created a problem: vehicles were now routinely flagging EVAP leak codes (P0440, P0442, P0455, P0456) that technicians had no efficient way to diagnose. Visual inspection might miss a hairline crack. Soapy water sprays were messy and unreliable. Pressure-decay tests confirmed a leak existed but didn't show its location. The industry needed a tool that could pinpoint leaks small enough to be invisible but large enough to set codes — typically in the 0.020" to 0.040" range for "small leak" codes.
The first commercial automotive smoke testers emerged in the late 1990s. U.S. Patent 5,763,764 (filed 1995) covers an evaporative emission tester closely related to modern smoke machines. U.S. Patent 6,526,808 (issued 2003) describes a portable smoke and clean-air machine for detecting leaks in fluid systems including EVAP and brake systems. By the early 2000s, smoke machines from manufacturers like Redline, OTC LeakTamer, and Star EnviroTech were standard equipment in well-equipped independent shops and dealer service departments. Today, the technique has expanded far beyond its EVAP-only origins to cover virtually any sealed system on a modern vehicle — and roughly 80% of the most common check-engine-light scenarios involving leaks can be diagnosed with smoke.
Smoke testing is not a single-purpose EVAP tool. Modern automotive smoke machines diagnose six distinct systems, each with its own failure patterns and its own repair implications.
The original use case. Diagnoses leaks in the sealed fuel vapor system — gas cap seal, filler neck, charcoal canister, vent valve, purge valve, and all the hoses connecting them. Triggers codes P0440, P0441, P0442, P0446, P0455, P0456, P0457. Smoke is introduced through the EVAP service port (or the gas cap location) and we watch for vapor at any escape point. Most modern vehicles have integrated service ports specifically for smoke testing.
Diagnoses vacuum leaks at intake manifold gaskets, throttle body gaskets, cracked vacuum hoses, brake booster lines, PCV system components, and various sensor seals. Triggers codes P0171, P0174 (lean fuel trim), rough idle, and engine performance issues. Vacuum leaks are notoriously hard to find without smoke — they can be smaller than a pencil tip and located in cluttered, hard-to-see areas of the engine bay.
Diagnoses leaks in the pressurized intake side of turbocharged engines — intercooler piping, charge pipe O-rings, intercooler core integrity, throttle body connections, and turbo discharge couplers. Triggers boost-related codes (P0299 underboost), reduced power, and audible whistling. Critical on modern turbo engines (Ford EcoBoost, GM 1.4T, BMW N20/N54/N55, VW/Audi EA888, the various 2.0T applications across brands) where boost leaks are a leading cause of performance complaints.
Diagnoses pre-catalytic-converter exhaust leaks at manifold gaskets, header connections, exhaust pipe joints, and small cracks in welded sections. Triggers codes P0420 catalyst efficiency, oxygen sensor codes, and audible exhaust noise. Pre-cat exhaust leaks distort oxygen sensor readings and can mimic catalytic converter failures — replacing the cat without finding the upstream exhaust leak first is a common expensive mistake.
Less common but valuable: introducing smoke through the oil dipstick tube or PCV port to find external oil leak sources. The smoke escapes through valve cover gasket failures, oil pan gasket leaks, rear main seal seepage, oil cooler housing cracks (common on 3.6L Pentastar V6 — see our Jeep page), and timing cover gasket leaks. Identifies leak source when multiple gaskets are suspect.
Specialty application: smoke testing the cooling system to find leaks at water pump weep holes, heater core seals, radiator end-tank cracks, hose connections, and head gasket compression leaks. Identifies leak source when coolant loss is documented but visual inspection finds nothing. Particularly valuable on European vehicles with complex cooling systems (BMW, Mercedes, Audi) where cooling system overhauls are labor-intensive.
If your check engine light is on with one of these codes, smoke testing is likely the fastest path to a definitive answer.
Why your scanner alone isn't enough: An OBD-II scanner reads the fault code, which tells you the system has detected an out-of-spec condition — like a pressure decay rate that exceeds the threshold for a sealed system. The scanner does not tell you which of the dozens of components in that system is actually leaking. A P0442 code on a 2015 Toyota Camry could mean the gas cap O-ring, a cracked EVAP hose at the canister, a degraded purge valve seal, a leak at the fuel filler neck, a hairline crack in the charcoal canister body, or any of a half-dozen other components. Smoke testing tells you which one in about 15 minutes.
OBD-II self-tests use pressure decay as their detection method. The vehicle's powertrain control module (PCM) commands the EVAP system valves to seal it off — closing the vent valve, opening the purge valve in a controlled sequence — and then watches the fuel tank pressure sensor to see how quickly pressure changes. A perfectly sealed system holds pressure (or vacuum, depending on the manufacturer's strategy) within a narrow window over a defined time period. A leaking system loses pressure faster than the threshold allows. The PCM compares the rate of pressure change to its programmed thresholds and, if the rate is too fast for the threshold, it sets the corresponding code (P0442 for moderate leaks, P0456 for very small ones, P0455 for large ones).
The thresholds the PCM uses are extraordinarily small. The federal standard requires modern vehicles to detect EVAP leaks down to 0.020 inches in diameter — that's a hole roughly the size of the head of a pin. Some manufacturers' systems detect leaks down to 0.010 inches. The PCM is genuinely sensitive to micro-leaks that can't be seen by visual inspection or felt with a hand near the suspect area.
This is where the smoke test inverts the diagnostic logic. Instead of measuring pressure decay over time, we pressurize the sealed system with visible vapor and look for the location where vapor escapes. The vapor exits at exactly the same rate the air would have leaked during the PCM's pressure-decay test, but now we can see it. The smoke machine pressurizes the system to roughly 0.5 to 1.0 PSI — well below pressures that would damage anything — and dense white-grey vapor pours out of the leak point in seconds. Most leaks become visible in under five minutes once the system reaches full pressure; the most stubborn micro-leaks can take longer and may need UV dye for confirmation.
Roughly 30-45 minutes shop time, sometimes longer for very small leaks or hard-to-access systems.
Connect our scan tool to the OBD-II port and pull all stored and pending codes. Freeze-frame data shows the conditions when the code was set — engine temperature, fuel level, RPM — which helps narrow down where to focus. We document everything in writing before disturbing the system.
EVAP codes route us to the EVAP system. Lean codes (P0171, P0174) and rough idle complaints route us to the intake/vacuum system. Boost codes (P0299) route us to the turbo system. Catalyst efficiency codes (P0420) get a pre-cat exhaust check. Some vehicles need multiple system tests to isolate the cause when symptoms span multiple systems.
For EVAP: every modern vehicle has an EVAP service port (often near the engine, sometimes under the vehicle near the charcoal canister). For intake leaks: we typically pull the intake hose at the throttle body and seal it. For boost leaks: charge pipe disconnect at the throttle body or intercooler. For exhaust pre-cat leaks: O2 sensor port. The right access point matters — incorrect access introduces variability.
Some test points require sealing other openings to make the system pressurize properly. EVAP testing typically requires commanding the vent valve closed via scan tool. Intake testing requires plugging the IAC port and any vacuum lines that don't lead to where you want to find leaks. The setup is more time-consuming than the actual test on most vehicles.
Connect the smoke machine to the access point and turn it on. The internal heater vaporizes the smoke fluid, an internal pump pressurizes the vapor, and the system fills in 30 seconds to 2 minutes depending on volume. Pressure regulation keeps the system at 0.5–1.0 PSI — enough to find leaks, low enough to avoid damaging seals or sensors.
With the system pressurized and smoke flowing, we systematically inspect every part of the system with a flashlight. Smoke is most visible when backlit by a strong light source from the opposite side. We work from one end of the system to the other, looking for the telltale plume of vapor at any escape point. Mirror tools and borescope cameras help inspect areas that aren't directly visible.
For micro-leaks that produce very small smoke plumes, UV dye in the smoke fluid leaves a fluorescent residue at the leak point that's visible with a UV flashlight. This is particularly useful for documenting the exact leak location for the customer (we photograph it) and for confirming leaks in areas where the smoke plume is too small to be obviously visible.
Once the leak is located, we photograph it, document the finding in the inspection report, and quote the repair. Sometimes the repair is a $20 hose; sometimes it's a $400 charcoal canister or $700 intake manifold gasket job. Either way, you get a written quote based on a confirmed leak location, not a guess.
Before smoke testing was widely available, technicians diagnosed leaks using a handful of older techniques — each with significant limitations.
Visual inspection alone works only for the most obvious failures. A disconnected hose, a torn intake boot, a missing gas cap — visual inspection finds these. But for the 80% of leaks that fall in the 0.010" to 0.040" range — the smaller end of leaks that still throw codes — visual inspection has roughly a coin-flip success rate. The leaks are too small to see, often hidden behind harnesses and accessories, and easy to miss even with a thorough hands-on check.
Soapy water spray can find leaks in pressurized systems by producing visible bubbles wherever air escapes. It works on some applications (particularly tire and large boost-side leaks) but has serious limitations for most automotive diagnostics. Soap water is messy, can't reach into tight spaces, can corrode electrical components, and only works when the system is naturally pressurized — meaning it doesn't help for EVAP testing where the system is at near-atmospheric pressure during normal operation.
Pressure decay testing (which the OBD-II self-test itself uses) confirms a leak exists but provides no location information. Adding a manual pressure decay test with a more sensitive gauge can refine the leak size estimate but still doesn't tell you where the leak is.
Parts swapping ("the parts cannon") is the worst alternative — replacing the most likely failed component, then the next most likely, then the next, until the code clears. It works eventually but costs the customer money for parts that didn't need replacement and time spent driving back and forth. We've had Tedious Auto Repair customers come to us after spending $400-$600 at other shops on parts swaps for EVAP codes that we then resolved with a $25 gas cap O-ring after a 30-minute smoke test.
Smoke testing won the diagnostic war for these failures because it's the only method that gives a definitive answer. The leak location appears in front of you. There's no guesswork, no probability ranking of likely failed components, no parts cannon. The leak is where the smoke is.
After years of running smoke tests on vehicles across Butte County, certain leak locations come up over and over. If you're seeing one of the EVAP codes above, the leak is statistically likely to be in one of these spots — though we never assume; we always confirm with smoke.
$199.99 for a complete smoke test including diagnostic scan, system pressurization, visual inspection, photographic documentation of any leaks found, and a written report with repair recommendations. The fee is typically a separate paid service, credited toward approved repairs — $50 on repairs of $500+, scaling to a full waiver on bills over $2,000. If you authorize the repair, Some shops charge $80-$120 for smoke testing alone but don't include the upstream and downstream diagnostic work — we include both because they're both required for an accurate answer.
Typically 30-45 minutes shop time, sometimes longer for very small leaks (P0456) or for vehicles with hard-to-access service ports. The actual smoke pressurization is fast — usually under 5 minutes to find an obvious leak. The setup work (sealing the system, accessing service ports, scan tool commands to close vent valves) and the documentation work afterward are what take the time. Plan for an hour at the shop including check-in.
No, when performed correctly with proper equipment. The pressure used is 0.5-1.0 PSI, which is below the design pressure of any factory seal or gasket. The smoke fluid is mineral oil-based, non-toxic, and non-flammable at the temperatures and pressures used. The risks come from improperly performed tests — using shop air without a regulator (high pressure damages seals) or testing systems that aren't designed for it. We use OEM-approved smoke fluid (like UltraTrace UV) and proper pressure regulation on every test.
It will find virtually every leak that's large enough to set a code — meaning leaks at or above the 0.010"-0.020" range. Some intermittent leaks that only occur under specific conditions (extreme heat, high vibration, full fuel tank) may not show during a static smoke test. For these, we sometimes need to perform the test under more representative conditions (warm engine, vehicle at operating temperature) or run the test with the engine running for vacuum/intake leak diagnostics. About 95% of leak-related EVAP codes resolve with a single shop-temperature smoke test; the remainder need additional diagnostic time.
Either (1) the new gas cap doesn't seal properly — common with aftermarket caps that don't match OEM dimensions, or (2) there's a second leak somewhere else that the original tech didn't find because they assumed the cap was the only issue. Both are textbook smoke-test scenarios. We re-run the smoke test, find the actual remaining leak, and quote the right repair. Call (530) 826-4275 with your codes and we'll let you know what to expect.
In some cases, yes. Smoke testing the cooling system can reveal compression leaks into the cooling jacket — smoke escaping from the radiator cap when pressure cycles, or smoke visible in coolant during pressurization. This is a specialty application of smoke testing rather than the typical use case, but it can be diagnostic on suspected internal head gasket failures. More commonly, head gasket diagnosis uses combustion gas testing of the coolant, compression testing, and leakdown testing. Smoke testing is one tool in the toolbox here, not the primary one.
Generally no. EVs don't have an EVAP system (no fuel tank), no intake vacuum system in the conventional sense, no exhaust system, and no PCV system. The systems that smoke testing diagnoses on gas vehicles largely don't exist on EVs. The exception is the high-voltage battery cooling system on some EV applications, where smoke testing can occasionally help diagnose coolant leak sources. But for routine EV diagnostic needs, smoke testing isn't typically required. Detailed coverage on our EV repair page.
If your car failed smog because of an active EVAP, intake, or exhaust leak code, then fixing the leak (and clearing the code, plus completing a drive cycle to reset the readiness monitor) will resolve the smog failure. Most EVAP-related smog failures clear within 50-100 miles of driving over 2-3 days after the repair. We do smog check service in addition to smoke testing, so we can find the leak, fix it, run a drive cycle, and certify the smog all at one shop. Saves a trip.
Depends on the codes. If your scanner shows EVAP codes (P0440-P0457), lean fuel trim codes (P0171, P0174), boost codes (P0299), or any code that suggests a leak rather than a sensor or electronic failure, smoke testing is likely needed. If the codes point to an electronic failure, sensor failure, or transmission issue, smoke testing won't help — you need different diagnostics. When you call, give us the codes and we'll tell you which path makes sense for your specific situation.
Trusted by Chico drivers since 2007. Written estimates.
Or call: (530) 826-4275
