Florida is not on one breath machine anymore. The state’s breath testing rules were amended effective July 17, 2025 and again effective July 1, 2026, and they now name the Intoxilyzer 9000 alongside the Intoxilyzer 8000, with a separate set of state forms for each.
This page is about what the newer instrument actually does, what its own operator manuals concede, and what independent testing found when someone put one on a bench. Where a number comes from a particular state’s manual rather than from Florida’s, it says so.
Where the Intoxilyzer 9000 Stands in Florida
Chapter 11D-8 of the Florida Administrative Code governs breath testing in this state, and it now runs in parallel for the two instruments. The department inspection rule, the agency inspection rule, and the operational procedures rule each name the 8000 and the 9000 and assign each one its own numbered form.
That matters for a practical reason. Every one of those forms is a record, and every record is a place where the required steps either happened or did not. The rules also still require two breath samples collected within fifteen minutes of each other and agreeing within 0.020 g/210L, and if the first two disagree by more than that, a third sample must be analyzed. A fuller map of the scheme is on the alcohol testing program rules.
How the Intoxilyzer 9000 Works
Like its predecessor, the 9000 identifies alcohol by infrared absorption. A source shines infrared light through a chamber holding your breath, and the instrument measures which wavelengths get absorbed. The pattern of absorption is used to identify ethanol, and the amount of absorption is used to quantify it.
The changes from the 8000 are real. The 9000 uses four filtered wavelengths rather than two, and its source pulses roughly ten times a second rather than four, which means it takes far more individual readings across a single blow. Georgia’s forensic laboratory states that it checked ethanol’s absorption pattern in the nine micron region against the published responses of more than eighty volatile compounds.
One detail is worth holding onto. The instrument sets its zero reference by measuring the detector signal while the chamber is full of ambient room air. The room is part of the measurement, which is why an Ambient Fail is not a trivial message. See also what besides alcohol can set off a breathalyzer.
What the Machine Requires From Your Breath
Georgia’s operator training manual for this instrument lists three acceptance criteria: flow, volume, and the slope of the alcohol level during the blow. The published requirements there are a minimum of roughly 1.1 litres in a single exhalation, a minimum flow rate of 0.15 litres per second, and a three minute window to produce an adequate sample. When flow drops below that rate the instrument stops accepting the sample.
Neither the subject nor the operator sees the alcohol level until the report prints.
These figures come from Georgia, not Florida, and are given here because Florida’s own operational procedures sit in a state form rather than in the rule text. What applied on a particular night is a question for the records in that case. How a person breathes into the machine is not a neutral detail either, and that is covered on how breathing pattern changes a breathalyzer result.
The Intoxilyzer 8000 and the 9000 Compared
| Feature | Intoxilyzer 8000 | Intoxilyzer 9000 |
|---|---|---|
| Infrared source | Heated and cooled wire, pulsing about 4 times per second | Pulsed source at about 10 times per second |
| Filters | Two filter points | Four filters |
| Wavelength region | 3.4 and 9.36 micron | Reported as the 9 micron region, exact wavelengths not published |
| Filter tolerance | Plus or minus 0.5 micron | Not published |
| Readings per sample | Roughly 80 in a ten second blow | Roughly 400 |
| Interface | Keypad | Seven inch colour touchscreen |
| Exhaust port | Can be blocked, which raises the reported result | Shielded by a plate, harder to block |
The fair summary is that the 9000 is a real improvement in optics, in sampling density, and in resistance to tampering, and that it is not demonstrably better at the two failures that matter most in a contested case, which are mouth alcohol and radio interference. A better machine measuring the same assumptions still inherits those assumptions. See how a breathalyzer works and are breathalyzers accurate.
What the Error Messages Mean
The instrument reports specific conditions rather than simply failing, and each message is a fact about what happened that night. This table is drawn from a state operator training manual for the Intoxilyzer 9000.
| Message | What it means |
|---|---|
| Invalid Sample | The instrument saw the alcohol level fall during a single blow, which is the pattern mouth alcohol produces. The test is rejected. |
| Insufficient Sample | You did not reach the required volume, flow rate, or time. Depending on what happened next, this can be treated as a refusal. |
| No 0.020 Agreement | The two breath samples disagreed by more than 0.020 g/210L. Under Florida’s rule a third sample must then be analyzed. |
| Range Exceeded | The reading came in above the instrument’s reportable range. |
| Interferent Detected | The optical bench identified a compound other than ethanol in the sample. |
| RFI Detected | The radio frequency detection circuit sensed interference during the sequence. |
| Ambient Fail or Purge Fail | The room air itself did not read clean, so the instrument could not establish a valid zero. |
| Dry Cal Chk Out of Tolerance | The control test using the dry gas standard fell outside its window. This takes the instrument out of service. |
| ITP Out of Tolerance | An internal standard check fell outside tolerance. |
| Diagnostic Fail | A self-check failed. |
| Incompatible Software Version | The software on the instrument does not match what the program expects. |
An Invalid Sample or an Interferent Detected is the machine documenting its own problem. Those records, and the data behind them, are obtainable. See how to read breathalyzer results and breathalyzer calibration and inspection records.
What the State’s Own Manual Concedes
The most useful material on any breath instrument is rarely written by the defense. It is written by the programs that train the operators, and it is published.
On the twenty minute observation period. Georgia’s manual states that the wait does not require the operator to administer it personally, and does not require the officer to watch the subject continuously for twenty minutes. Florida’s rule requires the officer to reasonably ensure nothing entered the mouth and that the subject did not regurgitate during that window. What “reasonably ensure” looked like in a given case is a factual question. See the 20-minute breathalyzer observation period.
On mouth alcohol. The same manual concedes that the alcohol level will occasionally not drop enough for the instrument to flag a sample as invalid, and that the breath profile printed on the report is not intended to be used to visually identify mouth alcohol. See slope detection and mouth alcohol.
On measurement uncertainty. Georgia prints an uncertainty figure on the ticket, stated as plus or minus 5 percent or 0.005 g/210L, whichever is greater, at a 95 percent confidence level, and states that reaching 99 percent confidence would require plus or minus 7 percent. It also identifies the largest single contributor as the natural variability in how the subject provides the sample.
On breath against blood. That manual states that measured breath will run on average 10 to 15 percent lower than measured blood. That one cuts in the tested person’s favour and is worth knowing for exactly that reason.
What Independent Testing Found
In 2017 a forensic criminalist ran a new Intoxilyzer 9000 through an independent bench review and published the results in three parts. The limits of that work should be stated plainly, because they are real: it was one instrument, roughly sixty tests over about ten hours, the author wrote his own caveat that the sample was not large enough for proper statistical analysis, and the publication is not peer reviewed. It is the most detailed independent look available, and it is not a validation study.
On mouth alcohol, the review reported falsely elevated readings that the instrument did not flag, with the invalid message firing only at much higher levels of contamination. A different manufacturer’s instrument caught the same contamination in every trial.
On radio interference, an active mobile phone was detected only some of the time, and only when held within a couple of inches of the antenna. In one sequence the instrument produced two very different back-to-back readings without indicating any interference. See radio frequency interference.
On interferents, the instrument passed, and that belongs on this page too. Challenged with isopropanol, acetone, methanol, and d-limonene, it correctly identified the interfering chemical every time. Several other compounds were never tested. A page that only reported the findings it liked would not be worth reading.
What to Ask For in a 9000 Case
- The instrument’s registration and its department inspection records, including any inspection following a repair.
- The monthly agency inspection reports and the inspector’s written remarks, which routinely say more than the boxes on the form.
- The full electronic record for the subject test, including any aborted or invalid sequences on that instrument that night.
- The control test results and the dry gas standard or reference solution records.
- The diagnostic records, the stability records, and the software version.
- The operator’s permit, and the agency inspector’s permit.
How that material gets used is on how we challenge a breathalyzer result.
Questions About the Intoxilyzer 9000
Is the Intoxilyzer 9000 used in Florida?
Yes. Florida’s breath testing rules, chapter 11D-8 of the Florida Administrative Code, were amended effective July 17, 2025 and again effective July 1, 2026, and they now name the Intoxilyzer 9000 alongside the Intoxilyzer 8000. Each instrument has its own set of state forms for operational procedures, agency inspection, and department inspection. Whether the agency in a particular case was running a 9000 or an 8000 is a question for the records in that case.
Is the Intoxilyzer 9000 more accurate than the Intoxilyzer 8000?
It is a genuine improvement in some respects and not in others, and the honest answer depends on which failure you are asking about. The optics are better, it takes far more readings during a blow, and independent testing found it correctly flagged every interfering chemical it was challenged with. The same testing found it no better at catching mouth alcohol, and possibly worse at detecting radio interference. A newer machine does not fix the assumptions that every breath test rests on.
How much breath does the Intoxilyzer 9000 require?
Georgia’s operator training manual for the same instrument gives a minimum of roughly 1.1 litres in a single exhalation, a minimum flow rate of 0.15 litres per second, and a three minute window to produce an adequate sample. Below that flow rate the instrument stops accepting the sample. Florida’s own operational procedures are set out in the state’s form for this instrument, and the specific values that applied on the night of a test are worth confirming from the records rather than assumed.
What does Invalid Sample mean on an Intoxilyzer 9000?
It usually means the instrument saw the alcohol level fall during a single breath, which is the signature of alcohol coming from the mouth rather than from deep lung air. The machine rejects the test. The message is useful to a defense because it is the instrument reporting its own detection of a problem, and the surrounding data is discoverable.
Does the Intoxilyzer 9000 catch mouth alcohol?
Sometimes, and its own manuals do not claim otherwise. Georgia’s manual concedes that occasionally the alcohol level will not drop enough for the instrument to flag the sample as invalid, and that the breath profile printed on the report is not intended as a tool for visually identifying mouth alcohol. An independent review in 2017 found falsely elevated readings that the instrument did not flag.
What records exist for an Intoxilyzer 9000 test?
The instrument stores subject tests, control and calibration checks, diagnostic records, and stability records, and it can reprint them. The state program keeps agency inspection reports and department inspection reports on their own forms. Those records are the substance of a breath test defense and they are obtainable.
Related: the main breathalyzer and breath test defense page, how a breathalyzer works, calibration and inspection records, and the 11D-8 rules reference. See also breathalyzer repairs and the Kilburn ruling.
A newer machine is still a machine. The Intoxilyzer 9000 takes more readings and uses better optics than the instrument it replaces, and it still converts breath into an assumed blood figure using a population average. The partition ratio did not change when the hardware did.
This page is general information, not legal advice, and it does not create an attorney-client relationship. Breath testing in Florida is governed by Fla. Stat. 316.1932 and 316.1934 and the Florida Administrative Code chapter 11D-8, which was amended effective July 17, 2025 and July 1, 2026. Specifications attributed to a state operator training manual are that state’s, and the procedures that applied in any particular case are a question for the records in that case. Procedures and rules change, and every case turns on its own facts. Past results do not guarantee a similar outcome.
