PulsCar analysis of NHTSA public data, September 2026 Article: https://www.pulscar.io/blog/car-ac-smells-bad License: CC BY 4.0 (cite "PulsCar analysis of NHTSA public data, September 2026") SOURCE FILES - NHTSA complaint flat files (FLAT_CMPL), all reports received 1 January 2015 to 8 September 2026, de-duplicated by report number (ODINO): 777,040 unique owner complaints. Field dictionary: https://static.nhtsa.gov/odi/ffdd/cmpl/CMPL.txt - NHTSA manufacturer communications files (FLAT_TSBS 2015-2019, 2020-2024, 2025-2026): one row per make/model/year per document; counted as distinct NHTSA document IDs. - NHTSA recall flat file (FLAT_RCL): one row per make/model/year per campaign; counted as distinct campaigns. - NHTSA investigations flat file (FLAT_INV). Files were downloaded in September 2026 and loaded into SQLite with an FTS5 full-text index on the complaint description and on the bulletin summary. Counts were run on 17 September 2026. COMPLAINT QUERIES (FTS5 syntax, run against the complaint description field) 1. Musty AC odor (tightened): ((musty OR mildew OR moldy OR mold OR mildewy) AND ("a/c" OR "ac vents" OR "air conditioner" OR "air conditioning" OR hvac OR vents OR "vent system" OR evaporator) AND (smell OR smells OR smelling OR odor OR odors OR stink OR stinks OR stench)) NOT (sunroof OR rain OR flood OR flooded OR accident OR collision OR "water leak" OR "water leaking" OR steering OR airbag OR "air bag") Result: 170 complaints. The first, looser version of this query (mold/musty + vent/AC words + smell words, no exclusions) returned 238; we read 35 random matches and found about 8 that were rain leaks, crash repairs, VOC off-gassing or unrelated (roughly 23 percent), so we added the exclusion list. Sample reading of the tightened set: 110 random complaints read in full. 84 (76 percent) describe a musty, mildew, vinegar or sour odor from the HVAC system; 10 (9 percent) describe an AC drain or wet-carpet mold problem; 16 (15 percent) are something else (coolant smell, VOC odor, burning smell, rain leaks through body seals, unrelated). Of the 94 relevant reports, 52 describe a treatment that was tried (cabin filter, foam or spray cleaner, dealer disinfecting, drain clearing); 41 of those 52 say the smell returned. No report in the sample describes a permanent fix; owners who are satisfied do not file complaints, so this is a selection effect, not a success rate. 2. Exhaust or fumes smelled in cabin: (exhaust OR fumes) AND (cabin OR "inside the vehicle" OR "inside the car" OR "into the vehicle" OR "in the cabin" OR "passenger compartment") AND (smell OR smells OR odor OR odour) Result: 1,257. Ford Explorer 567 (534 of them model years 2011-2017). Not sample-read as a whole; the Explorer subset is consistent with NHTSA's own investigation file (EA17-002). 3. Rotten egg or sulfur smell: ("rotten egg" OR "rotten eggs" OR sulfur OR sulphur) AND (smell OR smells OR odor OR odour) Result: 353; Ford Explorer 223. We read 35 random matches of the whole set and 30 random matches of the 130 non-Explorer reports. Non-Explorer reading: about 60 percent exhaust or catalytic converter, about 10 percent a battery (overcharging or a hybrid pack), about 15 percent fuel-related, about 15 percent unrelated (a seat-belt pretensioner, an airbag, electrical faults). Keyword flags across the 130 non-Explorer reports: "batter" in 34, "catalytic" in 26, fuel or gasoline words in 30. 4. Burning smell with blower, resistor or under-dash words: ((blower OR resistor OR "glove box" OR glovebox OR "under the dash" OR "under the dashboard" OR "behind the dash") AND (burning OR burnt OR melted OR melting OR smoke OR smoking) AND (smell OR smells OR odor OR odors OR stench)) Result: 204; fire field = Y in 62 (30.4 percent) against 1.82 percent across all 777,040 complaints. We read 30 random matches: 24 describe the HVAC blower motor, its resistor or the wiring and connector under the dash; 6 are engine-bay or other electrical faults whose smell reached the vents. A broader query (burning + vents/blower words, 508 results, 146 fires) was read (35 matches) and found to be about half engine-bay sources; it was not published. 5. Heater core leak or smell: "heater core" AND (leak OR leaking OR leaked OR smell OR odor). Result: 134. A broader sweet/coolant/antifreeze + cabin/vents query returned 246; reading 19 random matches found most were engine-bay coolant leaks smelled outside or in a garage, so the broader count is not published. FIELDS AND EXCLUSIONS - Mileage statistics use the MILES field, excluding blank, 0 and 400,000 or more. - "fire", "crash", "towed" are the Y/N fields as filed by the owner; "injury reports" counts complaints with INJURED greater than 0. - Complaints are self-reported, unverified, and are not failure rates. Counts by make or model reflect fleet size, publicity and owner behavior as much as defects. We rank nothing by reliability. BULLETIN QUERY (tsbs_fts on the NHTSA summary field) (odor OR odour OR smell OR musty OR mildew) AND (evaporator OR "a/c" OR "air conditioning" OR hvac OR "air conditioner") Result: 159 distinct documents, 26 makes (the file spells Mercedes-Benz two ways). Every summary was read and assigned one category; 11 of the documents were also opened as PDFs from static.nhtsa.gov and the category set from the document where the summary was ambiguous (GM 22-NA-045 is a calibration update; MINI M64 02 14 is a filter-position and drainage correction; Mazda 07-004/25a is a coil coating). Categories: evaporator odor - clean, disinfect or coat (62); drain or water ingress (19); hardware - evaporator or filter fit (19); software or control logic (19); not odor-focused - matched the words but concerns compressor valves, idle or transmission flashes (15); refrigerant leak diagnosis (8); manufacturer parts request (5); plastic odor (3); exhaust intrusion (3); coolant odor (2); other (3); fuel odor (1). Documents that describe evaporator odor without using the words in the query are not counted. RECALLS AND INVESTIGATIONS - Blower-motor fire recalls were found with: (defect LIKE '%blower%') AND (consequence LIKE '%fire%'), light vehicles only; 10 campaigns listed in the article, populations from the POTAFF field. - Investigations PE16-008 and EA17-002 (Ford Explorer exhaust odor) from the investigations file; the closing resume was read as a PDF. FILES - ac_smell_complaints_summary.csv: one row per query with counts, fire/crash/tow shares, mileage quartiles and top models, plus the all-complaints baseline. - musty_ac_odor_by_model.csv: the 170 tightened musty complaints by make, model and model year. - exhaust_in_cabin_by_model.csv: the 1,257 exhaust-in-cabin complaints by make, model and model year. - ac_odor_bulletins.csv: the 159 bulletins with NHTSA ID, bulletin number, date, years, models, our category and a 160-character summary excerpt. No VINs, names, cities, dealers or free text beyond short excerpts are included. LIMITATIONS Keyword queries miss owners who describe the smell without the words we searched and catch some who use them for something else; the false-positive rates above are from reading, not from a model. NHTSA's bulletin index attaches some documents to many models they do not cover. Populations in recalls are manufacturer estimates at filing. ------------------------------------------------------------------------------ CASE SENSITIVITY — SECTION ADDED 19 SEPTEMBER 2026 ------------------------------------------------------------------------------ NHTSA stores complaint narratives in ALL CAPS for records filed through roughly 2020 and in mixed case for records filed from 2021 on. A case-SENSITIVE search of the narrative field loses about 92% of the complaints filed after 2020, so this is the single largest way an analysis of this database can go wrong while still looking finished. Every query behind this file is case-insensitive, in two layers: * the complaint set is built with an FTS5 MATCH full-text search, which folds case before matching, so it cannot miss a mixed-case record; and * where the narrative is then tested with patterns, the narrative is upper-cased and whitespace-normalized first, so the patterns compare upper against upper. No query in this work uses GLOB or any case-sensitive comparison. This was not taken on trust. On 19 September 2026 every published folder was re-checked against the same database. Re-run here: the HVAC step-1 search reproduces its published count exactly.