The Breath Test Margin of Error Texas Prosecutors Don't Mention
The TL;DR
- Every breath instrument has measurement uncertainty. Texas DPS records calibration drift in
g/210Lvalues that prosecutors typically omit at trial. - 0.08 is the legal threshold — not the test result's true value. A 0.08 reading from an instrument with ±0.005 uncertainty cannot prove the actual BAC was at or above 0.08.
- Maintenance and calibration logs are discoverable under the Michael Morton Act and the Texas Public Information Act — if defense counsel asks correctly.
- The right hearing is Texas Rule of Evidence 702 / Kelly — not federal Daubert. Same standard, different label.
Table of Contents
A Texas Intoxilyzer 9000 reading does not equal an exact blood alcohol concentration. It is a measurement — with a known uncertainty range. When the reading is at or near 0.08, the legal threshold under Texas Penal Code §49.04, that uncertainty is the case.
What the machine actually does
The Intoxilyzer 9000, the breath instrument used by most Texas law-enforcement agencies, measures alcohol in deep-lung breath using infrared spectrometry. The instrument fires a beam of light at the breath sample, measures how much the alcohol molecules absorb at specific wavelengths, and converts that absorption into a number reported as g/210L — grams of alcohol per 210 liters of breath. Texas equates that to grams per 100 mL of blood for purposes of §49.01(2)(B).
Like any scientific instrument, the Intoxilyzer has a measurement uncertainty — a range within which the true value is likely to fall. That uncertainty is documented in the instrument's calibration certificate. The Texas Department of Public Safety knows it. The Breath Alcohol Testing Program documents it. Most jurors never hear it.
Measurement uncertainty is not the same as instrument error
An instrument can be working perfectly within manufacturer specifications and still produce a reading whose true value is some increment above or below what is shown. A 0.08 result on a calibrated instrument with ±0.005 uncertainty is consistent with a true BAC anywhere from 0.075 to 0.085 — the legal threshold sits inside the band, not outside it.
DPS-acknowledged uncertainty
Texas DPS publishes the calibration history of each Intoxilyzer in the field. The certificates list the inspector, the date of the calibration, the reference solution used, the instrument's response, and the calculated uncertainty value for that calibration. Typical uncertainties on properly maintained instruments fall in the range below.
Those numbers are not invented by the defense bar. They are documented by the Breath Alcohol Testing Program in DPS calibration paperwork. They are referenced in scientific literature submitted in Mata v. State1 and the body of work building on it. They are simply not foregrounded at most DWI trials, because the State's case is easier when the number on the slide reads as exact.
What this changes at 0.08
Texas Penal Code §49.04 makes it an offense to operate a motor vehicle in a public place while intoxicated, and §49.01(2) defines intoxicated to include "having an alcohol concentration of 0.08 or more." The legal cliff is at 0.08 exactly — not at 0.085 or 0.09.
Run a 0.08 result through the uncertainty math:
- Reported result: 0.08 g/210L
- Instrument-specific uncertainty (sample): ±0.006 g/210L
- Probable range of the true value: 0.074 to 0.086
That range straddles the legal threshold. The State's burden is proof beyond reasonable doubt. A reading whose true value could be 0.074 — under the legal limit — cannot satisfy that burden by itself. The cross-examination question writes itself: Is it possible, given the instrument-specific uncertainty you just described, that the true alcohol concentration was below 0.08?
The job is not to argue that the instrument is broken. The job is to make sure the jury hears what the instrument's own paperwork says about how exact it is — and how exact it is not.
— Njeri LondonWhat the State must disclose
Two procedural rules govern access to breath-instrument records in Texas DWI defense.
1. Code of Criminal Procedure Article 39.14 (Michael Morton Act)
Article 39.14 requires the State, on request, to produce material evidence in its possession or that the prosecution can obtain. Calibration records, maintenance logs, and operator-certification paperwork for the specific instrument used in a defendant's test are routinely held by DPS and obtainable by the prosecution. They are discoverable.
2. Texas Public Information Act (Government Code Chapter 552)
Defense counsel typically also serves a TPIA request directly on the DPS Breath Alcohol Testing Program for the instrument-specific records. This bypasses any back-and-forth with the prosecutor's office and creates an independent paper trail of what was provided when.
We file both, every time
Our DWI files include a Michael Morton request to the State and a parallel TPIA request to DPS within the first two weeks of intake. The dual track means the calibration records arrive even if the prosecutor sits on the discovery letter, and any inconsistency between what the State produces and what DPS produces becomes its own cross-examination point.
How we attack it in motion practice
The mechanics of challenging a breath result in Texas run through the trial-court pretrial schedule. Three filings carry most of the weight:
- Motion in Limine — to require the State to lay foundation outside the jury's presence before any expert references the breath result as exact.
- Motion to Suppress — on instrument-specific grounds (calibration deviation, operator certification gaps, observation-period violations under 37 Tex. Admin. Code §19.4).
- Texas Rule of Evidence 702 / Kelly hearing — on the science. Texas does not use federal Daubert. The functional standard for novel and applied scientific evidence comes from Kelly v. State2 and Hernandez v. State, requiring that the underlying theory and technique be reliable as applied.
The 702 / Kelly hearing is where the uncertainty data lives most naturally. The proponent of the science (the State) has to establish reliability. Defense cross-examination on that hearing is where calibration drift, sensor contamination, slope-detector cutoffs, and operator-error windows get exposed in front of the trial judge before the jury ever hears a number.
Charged with DWI at or near 0.08?
If the breath number on your case sits near the legal threshold, the State's case has more uncertainty in it than the slide will admit. We will pull your specific instrument's calibration records and run the math.
Frequently Asked
Common follow-up questions on Texas breath-test margin of error and how it shapes a DWI defense.
What is the margin of error on a Texas Intoxilyzer 9000?+
DPS lab calibration records list a typical uncertainty of approximately ±0.005 to 0.010 g/210L for Intoxilyzer 9000 instruments operated within manufacturer specs. Defense lawyers can subpoena the maintenance and calibration logs through the Texas Public Information Act, then cross-reference the instrument-specific uncertainty values for the device that ran the test on a particular case. The exact number varies by instrument, calibration cycle, and ambient conditions.
Can a 0.08 BAC test be defended in Texas?+
Yes. A 0.08 reading sits within the realistic uncertainty band of most calibrated breath instruments, which means a properly framed Rule 702 / Kelly challenge can put the State to its proof on whether the legal threshold was actually exceeded. The Texas Court of Criminal Appeals has acknowledged that breath-test results are not infallible, and Mata v. State (Tex. Crim. App. 2001) is one of several decisions where breath-science evidence was scrutinized.
Does the State have to disclose Intoxilyzer maintenance records?+
Texas Code of Criminal Procedure Article 39.14 (the Michael Morton Act) requires the State to disclose material evidence in its possession or that the prosecution can obtain. Defense counsel typically also serves a Texas Public Information Act request directly on the DPS Breath Alcohol Testing Program for the device-specific maintenance, calibration, and operator-certification records.
How long after a DWI arrest can the breath sample be challenged?+
Defense challenges to breath-test admissibility are timed to the pretrial motion practice schedule the trial court sets, not to the arrest date itself. Filings typically occur after discovery is exchanged and before the case is set for trial. The 15-day ALR hearing window is separate — that is the administrative license challenge, not the criminal-case suppression challenge. Both deadlines matter and they run on different clocks.
What is a Daubert hearing in a Texas DWI case?+
In Texas, scientific-evidence challenges run through Texas Rule of Evidence 702 and the Kelly v. State / Hernandez v. State framework rather than the federal Daubert standard. Functionally similar — the proponent of the science must establish that the underlying theory and technique are reliable as applied. Defense counsel uses this hearing to expose calibration drift, sensor contamination, or operator error in the breath instrument used in a specific case.
References
- Mata v. State, 46 S.W.3d 902 (Tex. Crim. App. 2001) — Texas Court of Criminal Appeals decision on the admissibility of retrograde extrapolation testimony in a DWI prosecution. Available via FindLaw.
- Kelly v. State, 824 S.W.2d 568 (Tex. Crim. App. 1992) — controlling Texas authority on the admissibility of novel scientific evidence under Texas Rule of Evidence 702.
- Texas Penal Code §49.04 — Driving While Intoxicated. https://statutes.capitol.texas.gov/Docs/PE/htm/PE.49.htm
- Texas Code of Criminal Procedure Article 39.14 (Michael Morton Act). https://statutes.capitol.texas.gov/Docs/CR/htm/CR.39.htm
- 37 Texas Administrative Code Chapter 19 — DPS Breath Alcohol Testing Program. DPS Breath Alcohol Testing rules.
Njeri London
Njeri is a trial-trained Texas criminal defense attorney with concentrated work in DWI, drug, and Fourth Amendment motion practice across the nine-county DFW footprint. She is admitted to practice in TXND and TXED federal courts and the U.S. Court of Appeals for the Fifth Circuit.
