PULSCAR ANALYSIS - P0500 (VEHICLE SPEED SENSOR "A" MALFUNCTION) AND SPEED-SIGNAL FAULTS IN NHTSA'S PUBLIC FILES PulsCar analysis of NHTSA public data, October 2026. License: CC BY 4.0 Attribution: "PulsCar analysis of NHTSA public data, October 2026." Published alongside https://www.pulscar.io/blog/p0500-code ------------------------------------------------------------------------------ Prepared October 2026. Source data: NHTSA public flat files (complaints received 1 January 2015 to 4 October 2026, the latest received date in the files; manufacturer communications) and the documents NHTSA publishes (manufacturer communications and Part 573 recall filings), downloaded one at a time from NHTSA. Files in this folder: the sorting rules file (the sorting rules, the rules used to settle disagreements and the pool history); p0500-3913-complaints-read.csv (one row per complaint); p0500-owner-reported-amounts.csv and p0500-amounts-excluded.csv; p0500-documents.csv; p0500-complaints-not-sorted.txt; COUNTS.txt. 1. Complaints (3,913) First search (2,256): every complaint whose narrative names P0500 or a sister code (P0501, P0502, P0503) by any spelling we searched (a letter O for zero, a space, hyphen or period after the P, the suffix 00, or a bare number after "code", "DTC", "error", "CEL", "coded" or "fault"); or names a vehicle speed sensor or a transmission, output, input or turbine speed sensor in owners' words together with a word of failure; or names the speedometer together with a word of failure and a context word (check engine, MIL, code, cruise, shift, transmission, ABS, traction, limp, reduced power); or names the wheel-speed side (wheel speed sensor, ABS sensor, tone ring, reluctor, exciter ring) together with a word of failure and one of speedometer, cruise, shift, check engine, MIL, code, limp. The searches are deliberately strict: the loose version (any speedometer or speed-sensor mention) returns about 4,900 narratives, most of them crash accounts, brake events or lamp complaints that say nothing about a speed signal. Second search (completeness search, 1,657 added): a separate search of the whole file for what the first one missed, run after the first batch was sorted. It proposed 3,549 candidates: 1,357 were each read on the full narrative and added (speedometer failures with no context word, mostly instrument-cluster blackouts; the Ford 6R80 output-shaft speed sensor and molded lead-frame family; misspellings such as SPEDOMETER and SPEED CENSOR; other speed-signal codes P0715-P0723, P1500, P2159, P1722 and U1412; wheel-speed sensors tied to limp mode, cruise loss, a wrong speed or shifting). 2,192 were found by three broad searches (speedometer plus a failure word 907; speedometer with looser failure wording 564; Ford, Lincoln and Mercury lead-frame complaints 721) and were not read one by one: a random draw of 100 from each group (fixed in advance, seed 20261021; 300 in all) was added and sorted, and the other 1,892 were not sorted. found_by in the complaints file says which search found each complaint and, for the draws, which group. Sorting: the sorting rules (published in full in the sorting rules file) were written, dated and fingerprinted before any complaint was sorted; an addendum for the second batch was fingerprinted before that batch was sorted. Every complaint was sorted twice, independently, by the same written rules, each time from the rules alone, without sight of the other sorting (columns sorting1 and sorting2), from an excerpt of the narrative around the matching words (the whole narrative when nothing matched). One category (S speed sensor; A wheel-speed or ABS side; W wiring or connector; C instrument cluster, computer or software; T transmission or driveline internals; O another cause; K checked and cleared; U no diagnosis or repair stated; N not a speed-signal event on this vehicle) and five yes/no flags (speedo, cruise, shift, recall, returned). Agreement of the two sortings over all 3,913: category 3,618 (92.5%); speedometer flag 3,876 (99.1%); cruise flag 3,902 (99.7%); shift flag 3,834 (98.0%); recall flag 3,820 (97.6%); returned flag 3,833 (98.0%). Settling: every complaint on which the two sortings differed on any field (480 complaints) was read and settled by one of the written rules R1 to R54 (listed in the sorting rules file); the rule applied is in the rule_applied column. Later full-text checks: the sorters had seen an excerpt, not always the whole narrative, so two further checks were made on full texts under the same written rules. (a) The recall, returned, shift and cruise flags were checked on the full text of 1,398 complaints whose wording pointed to such an event: 570 flag changes. (b) Every complaint first sorted U (no diagnosis or repair stated; 2,213 complaints) was read again on its full text for a cause, a shop finding or a repair stated later in the narrative: 255 changed category (rule R200 was added for one kind of false match). A third consistency step (rule R300) moved 59 complaints in which a dealer or mechanic named the Ford molded lead frame as the cause from wiring or transmission to speed sensor, because they had been sorted before the lead-frame rule was written. 736 complaints in all were changed by these checks; their resolution says "changed in full-text check" and rule_applied names the rule. Result: 3,384 relevant, 529 not relevant (N). Counts are counts of complaints, not cars. A complaint is evidence that an owner wrote to NHTSA; nothing here is a failure rate. Re-sort of a random sample: 120 complaints drawn at random were sorted again, independently, on their full texts, and compared with the published sorting. The category differed on 20 of 120 (16.7%); the flags differed on: recall 6, shift 1, speedometer 3, returned 3, cruise 3. Of the 20 differences, 11 were complaints published as U (no diagnosis or repair stated) that the re-sort placed in a named cause, K or N, and 3 went the other way, to U. This is a known limit of the category counts: shares by category are approximate, and the U share is the least certain. Not counted: the 1,892 complaints of the three broad groups that were not drawn (p0500-complaints-not-sorted.txt) are in no total. In the random draws, 89 of 100 complaints of the speedometer group, 73 of 100 of the looser speedometer group and 81 of 100 of the Ford lead-frame group were relevant, so the published counts understate the whole and the unsorted complaints should not be read as a different kind of complaint. Motorcycles and heavy trucks are sorted like all other vehicles. Complaints that only mention a speedometer in a crash, a sudden acceleration or a brake event, or an ABS or traction lamp with no speed-signal effect, are category N. Ford: 1,288 of the 3,384 relevant complaints are Ford (38.1%) and 937 are the Ford F-150 (27.7%). The Ford share is large because of one family: the molded lead frame that carries the output speed sensor in Ford's 6R80 automatic transmission (F-150, Expedition, Mustang, Navigator and others) - 384 of the 483 speed-sensor (S) complaints are Ford - and because the second search looked for that family by name. It describes the complaints, not how common the fault is across makes. 2. Documents (356 candidate documents, 356 files read, 320 distinct texts, 250 families) Sources: NHTSA's index of manufacturer communications and the documents NHTSA publishes. First search: index entries whose summary names P0500-P0503, or a vehicle speed sensor or signal, a transmission, output, input or turbine speed sensor, the speedometer or a wheel speed sensor together with a word of fault, repair or code: 166 entries kept (15 name a code in the summary). Second search: 190 more documents found by a separate search of the index (1,745 index entries not already listed were checked and about 700 summaries read): bulletins on transmission speed-sensor codes P0715-P0723, Ford 6R80 lead-frame campaigns and recalls, vehicle speed sensors on heavy trucks, speedometer faults, output-shaft tone rings, and wheel speed signals with a check engine light. Download: all documents were downloaded from NHTSA's own site, one at a time. 2 of the 356 candidate documents have no PDF on NHTSA (10057366 and 10190567); the other 354 came as 356 files (two documents come as two files each; the second file is listed as the document number followed by -2). Exact copies (the same text under another number) account for 36 of the 356 files, leaving 320 distinct texts. Reading: the files fall into 250 families of near-identical documents (revisions, copies and look-alikes). One document of each family was read in full; the other files of a family were compared with it line by line, or are exact copies. Papers that are scans or images (27 files) were read from images of their pages; for them, the codes and the relevance were read off the page images. Every verbatim quotation taken from a document was checked against the text of the named document. The document list says, for each file, what was done (column read_in_full), whether its own text names P0500, P0501, P0502 or P0503 (names_p0500_to_p0503_in_own_text; for papers read from images, what the images print; empty where there is no PDF), and its relevance (DIRECT: about the target fault; LISTED: a target code is one of many in a list; RELATED: a companion code or the same family of fault; NOT RELEVANT; motorcycle or heavy truck: sorted apart, with a one-line note). number, make and date in the document list are as NHTSA's index gives them (the index date is often not the date printed on the paper, and the index expands years); vehicle years, engines, build dates and coverage terms are as printed in the documents. Coverage status "as of September 2026" is the authors' computation from printed terms. duplicate_of marks exact copies. pdf_url is the address on NHTSA's site; it is empty where NHTSA gave none or the address was not looked up. 3. Owner-reported amounts (267 figures in 241 complaints; 468 candidate figures looked at, 201 excluded) One written inclusion rule (fixed before the figures were decided): a dollar figure is included when the owner states it as a price, charge, bill or quote for diagnosing, repairing or replacing a vehicle speed sensor (transmission output or input speed sensor, lead frame, a valve body that carries it), a wheel-speed sensor, tone ring, hub or bearing assembly, the instrument cluster, the wiring or connector of the speed signal, the transmission or computer when the narrative ties it to the speed-signal fault, or a diagnostic fee asked for this speed-signal event, on that vehicle. A figure for a whole transmission replacement or rebuild is included only in the group transmission, and only when the narrative ties the transmission work to the speed-signal fault. Excluded: mileage written as "NNk"; vehicle value, payments, trade-ins or refund requests; repairs of other systems; totals that mix this fault with other work unless the narrative attributes the figure to the speed-signal part; restatements of a figure already counted in the same complaint; hypothetical figures and other owners' costs. Status comes from the owner's own words (paid, billed, quoted, offered, not stated). Every excluded figure is listed with its reason in p0500-amounts-excluded.csv. Figures are what owners wrote, not audited invoices. Known gap: figures were collected when written with a dollar sign, 'k' or the word dollars. A later look at 24 plain-number figures with cents (for example 3700.00) in the relevant complaints found 12 that would have counted (10 complaints). They are not in the file. Added, they would move the median of cluster figures from $950 to $900, transmission from $6,000 to $6,500, wiring from $650 to $675 and wheel-speed from $1,943 to $1,953, and leave the other group medians unchanged. 4. Limits The complaint file holds the cases where an owner wrote to NHTSA, in the owner's words, with no verification; owners whose trouble ended are mostly not in it. The shares reported (for example how many speed-sensor replacements were followed by a return of the problem) describe the complaints, not all cars. About 58% of the relevant complaints (1,958) state no diagnosis or repair (category U): owners often write before, or without, a shop finding. The category counts carry the uncertainty stated above (the re-sort of 120 complaints differed on 20). The document set holds what NHTSA's index lists; a maker paper that is not in the index is not in it, and 2 candidate documents could not be read because NHTSA has no PDF for them. The speedometer, the cruise control and the shifting of an automatic transmission also depend on the instrument cluster, the ABS wheel-speed sensors and the transmission itself; the study follows the speed signal wherever the complaint puts the cause, and the category says where.