The following excerpt about breath testing is taken from a portion of Chapter 5 from The Georgia DUI Defense Trial Practice Manual, © 1996 William C. Head. The reader should understand that the material provided is in no way comprehensive, but outlines certain legal attacks on breath testing. Historically, many breath tests (and some refusals) have been excluded prior to trial on other grounds other than the ones mentioned in the excerpt.

5.1.2.b. Mechanical and Electronic Features of the Intoxilyzer 5000

The primary testing method employed by this machine is infrared spectrophotometry, or "infrared absorption" as it is usually called. This is the method used by most breath testing machines in current use across America. This machine contains a computer, a printer, a panel screen, an LED read-out, an infrared energy source, a sample chamber (where the subject's breath and room air are alternatively analyzed), and the filters mentioned earlier. See the schematic drawing of the Intoxilyzer 5000, in the copy of its patent, Appendix D. Note: the patent shown is the three filter model and no additional patent for the five filter has been filed.

The 5000 has three different "operating modes" that it utilizes in processing a breath sample. Mode A is the "air blank" mode of the machine. Mode B represents the subject's breath testing mode for the machine. Finally, Mode C is the calibration mode. The computer program is custom-programmed by CMI for each state by means of the Z80 microprocessor. This controls the sequence which will be followed for each test.

For example, in a single test state, the program might be set for an "A-B-A" sequence, which means that it will draw an air blank, sample the subject's breath and then draw another air blank. This sequence never does a calibration check, so no "C" mode is included.

Georgia's program is set up as an A-B-A-B-A sequence. Dual testing has been "required" by the state since January 1, 1995. Notably absent from the Georgia microchip program is the calibration check in the Mode C capability of the machine. Georgia (once again) has opted to NOT utilize this available (but more expensive) mode for operating the machine against a constant, wet-bath solution calibration standard. If utilized, the calibration standard of known ethanol concentration would be kept in an adjacent, sealed container with a stirrer, heater and a thermometer so that constant verification of the reliability of the subject's test result could be guaranteed.

Georgia has opted for a plan of periodic maintenance wherein the implied consent area supervisor services the machines quarterly. Georgia has studiously avoided placing even this lax requirement in its GBI Implied Consent Rules. So, in heavy use areas, such as the City of Atlanta, the machine may run 600 to 1000 tests before having its calibration checked against a known alcohol standard, whereas Hahira may have checked only 25 subjects in the same time period. Other states check all machines every 30 days (e.g., Florida), some weekly. Many states check it with every subject test by way of the calibration check.

The protocol for the area supervisor's quarterly visit is for him or her to mix a known quantity of distilled water and a vial of ethanol for use in the wet-bath simulator. No portion of this mixture is retained for later independent analysis or verification of accuracy.

Interestingly, the GBI did not establish any testing protocol whereby the area supervisors would check the machine's capability to accurately detect and report the presence of any common interfering substances until approximately July 1, 1995. Therefore, the GBI and its area supervisors cannot truthfully state that the Intoxilyzer 5000 was capable of "identifying" a problem of interference from any of these chemicals prior to mid-1995.

Ironically, once testing began, only acetone was provided for the quarterly testing. No toluene or acetaldehyde are provided or ever mixed with ethanol in order to determine whether the detectors are functioning properly.

Other states, like Florida, have detailed calibration requirements, including MONTHLY checking of the machine FIVE different times against MIXTURES of various chemicals at varying concentrations, so as to assure that the machine can do what it is represented to be able to do.

QUERY: If Georgia puts these machines out in the field and never verifies that the devices can detect and report common interfering substances, are we any better off than when we dealt with a Taguchi-less Intoximeter 3000?

Following the GBI's ostrich mentality, it seems that the agency has adopted this approach to thwart the General Assembly's legislative intent in passing O.C.G.A. §§ 40-5-67.1 (g)(2)(F) and 40-6-392 (a)(1).

Beyond the problems associated with no simultaneous calibration check, the program for Georgia's version of the Intoxilyzer 5000 is incapable (during a routine test of a subject) of running a complete self-diagnostics check UNLESS it is turned off completely and then restarted. The diagnostics check is important because it searches for bad components in the machine, in order to avoid the chance of running a test on a malfunctioning device. The old Intoximeter 3000 could run a diagnostics check while "ON," but, as currently programmed, this new Intoxilyzer 5000, series 768GA machine cannot. This important safeguard was not specified by the GBI for the Z80 microprocessor chip when the order for machines was placed.

The first officers trained on the new machine were told not to turn the machine "on" and "off," although later, some classes have been told about the importance of this step. Again, there is no written protocol or regulation to REQUIRE all operators to do this. It is usually not done. Moreover, since the machine (as programmed) takes 21 minutes and 14 seconds to go from the "off" position to be "ready" for testing the subject, few departments will ever add this step to their testing routine.

Whenever a diagnostics check is not done, the operator cannot truthfully tell that the machine is running properly, because it has not been run through its self-examination diagnostics cycle to verify this fact. In busy locations, officers are not turning it off at all.

Expect operators of the Intoxilyzer 5000 to know nothing about the significance of this due to their cursory training without the benefit of written materials. You must subpoena the area supervisor or a defense expert to come to court to explain that this step of turning the machine "off" is the only way to assure that "all parts are in working order," as mandated by O.C.G.A. § 40-6-392 (a)(1)(A). For a list of the current area supervisors, see Form 4.09.

The copies of breath testing cards shown in Form 5.01 indicate the difference between a test run with diagnostics (Fig. 1) and WITHOUT (Fig. 2). Note: Fig. 1 is a test run by the GBI in checking the machine, not a standard test for a subject. Officers in busy locations are sure to ignore this step until they lose many cases on this point, and then it will be made a requirement.

In Forms 5.02 through 5.10, several other CMI printout cards are shown. These were obtained from Mr. James Panter of the GBI pursuant to an Open Records Act request on behalf of the Georgia Association of Criminal Defense Lawyers. The various "reports" that the machine, as programmed, is capable of rendering offer insight into attacks upon the machine.