MIKELL, Judge.

Following a bench trial, the trial court found Mehdi Sultan guilty of driving under the influence of alcohol, driving with an unlawful alcohol concentration, and exceeding the maximum speed limit. Sultan appeals, claiming that the trial court erred in denying his motion to suppress because the arresting officer did not perform the horizontal gaze nystagmus (HGN) test in accordance with law enforcement guidelines, because the officer lacked probable cause to arrest Sultan, and because the officer's use of a passive alcohol sensor flashlight violated the Fourth Amendment's prohibition against unreasonable searches and seizures. For the reasons that follow, we find no reversible error and affirm.

[T]he trial court's application of the law to the facts is subject to de novo review if the facts are stipulated, or if the critical facts do not depend on the testimony of witnesses who are subject to cross-examination. However, a trial court's ruling on a motion to suppress frequently involves a mixed question of fact and law. When the outcome of a motion to suppress depends on the credibility of the witnesses or on disputed facts, and the trial court has not committed an error of law, the court's ruling will not be disturbed on appeal. As a reviewing court, we must accept the factual and credibility determinations and inferences drawn by the trier of fact, even if we disagree with them, as long as there is evidence in the record to support the trial court's findings.

So viewed, the evidence shows that in the early morning hours of February 26, 2006, a City of Duluth police officer observed Sultan's car traveling at 60 mph in a 45 mph zone. The officer radioed this information to a second police officer, who stopped Sultan in a shopping center parking lot. The second officer advised Sultan that he had stopped him for speeding. As was his custom, the officer placed his passive alcohol sensor flashlight just inside the driver's side window, and the sensor reported a high positive indication for alcohol inside the vehicle. The officer could also smell alcohol emitting from the vehicle, although he could not recall if the odor was coming from Sultan's person. The officer asked Sultan to step out of the car, and asked Sultan if he had been drinking. Sultan admitted he had been drinking "a little." The officer also noticed that Sultan's speech was slurred and that Sultan had "red watery, glossy eyes."

The officer asked Sultan to perform several field sobriety tests. First, the officer performed the HGN test. According to the officer's initial testimony, Sultan exhibited six of six clues on the test, indicating that he was less safe to drive. The officer subsequently admitted, while viewing the videotape of the examination, that "I see what I did wrong." The officer testified that he probably did not perform the equal tracking portion of the HGN test, which was used to qualify the subject to take the remainder of the test. The officer also testified, among other things, as to the requirements for performing the test for nystagmus at maximum deviation, including that the stimulus be held for five seconds at maximum deviation on each side of the head. The officer maintained that he did hold the stimulus for four to five seconds, although Sultan argued that, in view of the video evidence, the stimulus was not held for more than two seconds.

The officer next had Sultan perform the walk and turn test and the one-leg stand test. According to the officer, any more than two clues on these evaluations indicate that someone is less safe to drive. Sultan exhibited three clues on the one-leg stand test, including swaying, raising his arms more than six inches, and placing his foot down. The officer initially testified that Sultan demonstrated two notable clues on the walk and turn test by taking eight steps rather than the nine steps that he was instructed to take, and by missing heel to toe on several steps. The officer changed his testimony upon reviewing the videotape of the examination and admitted that Sultan took nine steps as instructed.

As the last part of the field examination, the officer asked Sultan to take a portable breath test, which was positive for the presence of alcohol. Based on Sultan's performance on the field evaluations as well as physical manifestations of impairment, which included slurred speech and red, watery, and glossy eyes, the officer concluded that Sultan was under the influence of alcohol to the extent that he was less safe to drive, and he placed Sultan under arrest.

1. Sultan claims that the trial court erred in denying his motion to suppress because the HGN test was not performed in accordance with law enforcement guidelines. We agree that under the circumstances the trial court should have suppressed the HGN test, but conclude that Sultan suffered no harm as a result.

Evidence of a defendant's performance on an HGN test is considered to be evidence based on a scientific principle or technique. "[S]uch evidence is admissible upon a showing by the party offering the evidence that (1) the general scientific principles and techniques involved are valid and capable of producing reliable results, and (2) the person performing the test substantially performed the scientific procedures in an acceptable manner." Sultan acknowledges that the state satisfied the first prong of the test. "[A] trial court may judicially notice, without receiving evidence, that the standardized HGN test has been established with verifiable certainty." Sultan contends, however, that the state failed to satisfy the second prong required for the HGN test's admission, and the trial court therefore erred in failing to exclude the test in its entirety.

The trial court found that the officer failed to conduct the equal tracking portion of the HGN test and that the officer improperly performed the portion of the test intended to check for nystagmus at maximum deviation. The test for nystagmus at maximum deviation accounted for only two of the six possible clues indicating impairment, and the trial court did not find that the officer improperly performed the portion of the test accounting for the other four clues. However, according to the officer's testimony, the equal tracking portion of the test was intended to qualify the subject. The officer indicated that the qualification of the subject was necessary for the remainder of the HGN test to be valid. It follows that because the officer failed to qualify Sultan as a suitable subject of an HGN test, and in view of the officer's testimony as to importance of that qualification, no portion of the HGN test could be relied upon as evidence of Sultan's impairment.

In view of the foregoing, the officer substantially departed from the principles and procedures that were the basis for the HGN test's reliability, and the HGN test should have been excluded. Nevertheless, "[i]t is fundamental that harm as well as error must be shown for reversal." In its order denying Sultan's motion to suppress, the trial court expressly found that it "did not credit the results of the HGN evaluation." Furthermore, during the bench trial the trial court noted for the record that "the court is not considering the HGN as credible at all. . . . That is . . . [a] scientific test[] . . ., and it does have to be done in a certain way or it's just not credible. So the court does disregard the HGN." Thus, the record demonstrates that the HGN test evidence did not contribute to the verdict of the trier of fact, and it follows that Sultan was not harmed by the trial court's denial of his motion to exclude the test.

2. Sultan also claims the trial court erred in denying his motion to suppress because the officer did not have probable cause to arrest him for driving under the influence of alcohol. We disagree.

The test of probable cause requires merely a probability — less than a certainty but more than a mere suspicion or possibility. To arrest a suspect for driving under the influence, an officer need only have knowledge or reasonably trustworthy information sufficient to authorize a prudent person to believe that the suspect was actually in physical control of a moving vehicle, while under the influence of alcohol to a degree which renders him incapable of driving safely.

Evidence shows that an experienced officer observed Sultan with slurred speech and red, watery eyes, that Sultan's breath tested positive for the presence of alcohol, that Sultan admitted to drinking "a little" alcohol, and that the officer smelled the odor of alcohol coming from Sultan's vehicle. Sultan also failed the one-leg stand field sobriety evaluation. Accordingly, the trial court was authorized to conclude that the officer had probable cause to arrest Sultan for driving under the influence of alcohol.

3. Sultan further contends that the trial court erred in denying his motion to suppress because the officer's use of a passive alcohol sensor flashlight violated the Fourth Amendment's prohibition against unreasonable searches and seizures. Sultan claims that he had a constitutionally protected expectation of privacy inside his vehicle, and that the officer violated that privacy interest by placing the intake valve of the passive alcohol sensor inside the driver's side window. Pretermitting whether Sultan's argument has merit, the officer's use of the passive alcohol sensor was not introduced into evidence at trial and thus Sultan fails to show he was harmed by a failure to suppress the evidence directly gathered by the sensor. "Even if there had been error in denying the motion to suppress, . . . [i]t would not have been reversible error until and unless evidence seized during an illegal search was tendered and admitted against the defendant, though properly and timely objected to."

Furthermore, the use of the sensor, even if unlawful, did not taint the remainder of the arresting officer's investigation. As a rule, "the indirect fruits of an illegal search or arrest should be suppressed when they bear a significantly close relationship to the underlying illegality." According to the arresting officer, the passive alcohol sensor draws in the ambient air within the passenger compartment of a motor vehicle and "detect[s] an odor commonly associated with an alcoholic beverage coming from the vehicle." The officer testified that, independently of the sensor, he noted the odor of alcohol coming from the vehicle, and there was no testimony showing that the officer's decision to pursue his investigation was dependent upon a positive reading from his sensor. Thus, even assuming that the officer violated Sultan's privacy interests when he placed his instrument inside the driver's window, the evidence supporting Sultan's conviction for driving under the influence was not the fruit of that search. In other words, it cannot be said that "the evidence to which instant objection is made has been come at by exploitation of that illegality." Accordingly, the trial court did not err in denying Sultan's motion to suppress.

Judgment affirmed. Johnson, P. J., and Phipps, J., concur.

MIKELL, Judge.

DECIDED JANUARY 30, 2008.