OPINION OF THE COURT
HUBERT L. GRIMES, County Judge.THIS MATTER came before the Court on Defendant’s Motion to Suppress the blood alcohol test submitted in the instant case. The Court makes the following findings of fact.
Defendant was arrested and charged with the offense of Driving Under the Influence of Alcoholic Beverages on or about October 27, 1989. The defendant was arrested by a member of the Ormond Beach Police Department and subsequently taken to the Volusia County *83Sheriffs Operation Center and requested to submit to a blood alcohol test. Defendant consented to the test and the readings resulting from the test were as follows: Test 1 — .193, Test 2 — .163, Test 3 — .205.
Defendant argues to the Court that the blood alcohol test, as indicated above, does not substantially comport with the regulatory requirements as promulgated by the Florida Department of Health and Rehabilitative Services in that HRS Rule 10(d)42.0211 requires that two tests are acceptable if within a plus or minus .02%, if not then a third test should be given. Defense further argues that the results as indicated from the Intoxilyzer 5000 reading should be suppressed in that they do not fall within the plus or minus .02% rule as indicated by HRS and further that such readings must be consecutive in nature. After an initial hearing on the matter on September 10, 1990 the Court granted the defendant’s Motion to Suppress the blood alcohol test as it appears to the Court that the breath test samples were not within the acceptable range of plus or minus .02% and they were not consecutive as it appears a reasonable interpretation of the rule required.
The State requested a rehearing on the Court’s granting of the defendant’s Motion to Suppress based upon the fact that the State discovered additional evidence of a document prepared by the arresting officer at the time of the test that “the defendant did not blow until the tone stopped”. Therefore a rehearing was conducted before the Court on November 27, 1990 and additional testimony was offered on January 22, 1991.
The hearing on January 22, primarily addressed the question of whether the test results accurately reflected the defendant’s blood alcohol level and more specifically, whether the second test was in fact a valid sample even though there was no apparent indication on the face of document that registered it as an insufficient sample.
At the hearing, Deputy Hyashi of the Volusia County Sheriff Department testified that there are several reasons why the machine would have accepted the second sample as a valid sample, even though the arresting officer indicated that the defendant stopped blowing before the tone stopped. The testimony of the state’s witness indicated that the machine was dependent upon three primary criteria, to wit; a) pressure, b) time, and c) slope satisfaction. To meet the pressure requirement, the subject must blow with enough force (approximately two inches of pressure) for a minimum of four to six seconds, time. The sample within the instrument must come to a stable reading. The slope factor is met when the blood alcohol level registered by the instrument every .6 tenths of a second changes by less than .03. It was *84further testified by Deputy Hyashi that if any of these three criteria are not met, the machine will print “insufficient sample” on the print out. In the instant case, no such “insufficient sample” was noted on the print out. The only indications that there may have been an invalid sample stem from the comment made by the arresting officer on the operational check list Form 1615 that “on second sample, subject stopped blowing before tone stopped”.
Thus the issue presented to this Court is whether samples as provided are valid and admissible even though the face of the intoxilyzer print out does not show samples as being consecutively within the .02 range and neither is there a machine indication of an “insufficient sample”.
In examining the print out, it logically follows that for there to be a .03% discrepancy between the first reading and the second reading and a .042% reading between the second and third readings, it would appear that the machine must have been operating improperly. This conclusion is buttressed by the testimony of the State’s expert witness, Dr. Rarick. The instrument is designed to provide valid test results without the necessity of human factor being involved. The instrument is apparently programmed that if one of the three criteria are not met it would normally make a note of that by printing the notation “insufficient sample”. The interesting and crucial point in the instant case appears to be that all of the three basic criteria for a sufficient sample were properly met by the defendant’s blowing into the machine, yet the machine did not pick up the defendant’s failure to blow for the entire time period as a basis for an insufficient sample.
Law enforcement, the judiciary and our citizens have come to place great reliance upon the ability of breath test instruments to accurately perform and report their respective findings in each case. The legislature has even relied upon these devices to distinguish penalties imposed upon persons convicted of DUI in our state. For example the presumption of impairment arises if there is more than a .10% B.A.C. reading. The penalties are further enhanced if there is a .20% reading or above. These distinctions are drawn as a large result based upon our reliance on the integrity of the instruments which measure the alcohol content of human blood.
Faced now with a question of whether the results obtained in the instant case are reliable there are two lines of departure. One is that the results are reliable, but for the defendant “inviting error” and tricking the machine. The other is the instrument’s failure to detect *85that Defendant “stopped blowing before the tone stopped” and hence thereafter not labeling it as an insufficient sample shows that it was not properly working in that it failed to catch the “defendant’s trickery”.
This Court rejects both conclusions as stated. However, as stated, both by the state’s experts Rarick & Hyashi, “something must be amiss if the results are that far out of alignment”. The state’s argument of invited error must be rejected for two reasons — 1) the concept only applies in courtroom proceedings and 2) Defendant cooperated by giving three tests. To credit Defendant with being able to intentionally thwart a scientific instrument by the way he blows is directly counter to logic and credits Defendant with a great deal of human wisdom, more than this Court is prepared to find.
Because of the “high tech” nature of the Intoxilyzer 5000 and the fact that it is designed to be free of the needs for human interference, the failure of the machine to note sample two (2) as an insufficient sample generally means that the Court would accept it as a valid sample for admissibility. The fact that sample two (2) is a .03 differential more than the first test, therefore, requires that a third test be conducted. The third test was thereafter a .205 or .042 more than the immediately preceding test. When there is such a substantial deviation in the test results, it follows that “something must be wrong with the machine”. Yet, this same machine which is programmed by the manufacturer to show insufficient sample if one of the three prongs is not properly attained, failed to do in the instant case and therefore the test, in order to be accepted, must be supplemented by the human factor. This minefield is exactly why courts have a duty to determine the admissibility of evidence in a criminal prosecution.
The citizens of our state place great weight upon the capability of breath test machines. Yet the errors, either in the machine not picking up the fact that Defendant did not blow for the requisite time period and was therefore insufficient, or the fact that a sufficient sample was given to the machine and the chemical analysis of the sample created such an aberration that the results cannot be relied upon. Adopting either approach results in the same conclusion. The test results in the instant case are not legally rehable and must be stricken in this cause.
The state will not be allowed to utilize the test results in prosecution of the defendant but may proceed on the theory of impairment if it so elects.
DONE AND ORDERED IN CHAMBERS THIS 19th DAY OF MARCH, 1991.