When I work on a DUI case in Michigan with a Datamaster result, I always look at the different times in the police reports and evidence.  

1 - I look for the first police contact time
2 - I look for what test an arrest was made
3 - I look for what time a PBT was done
4 - I look for what time chemical test rights were read
5 - I look for start of the observation time
6 - I look for the times that the Datamaster tests were done - there's usually two of them

The first police contact is the first moment that we can be assured that my client is not drinking, and therefore we can start the clock on time between last drink and the Datamaster.  It's very likely more time than this has gone by, but that's a case by case issue.  

Let's say, the traffic stop happened at 2 am, and the first Datamaster test is at 3:03 am.  With about an hour gone by, the blood level is either at it's maximum level from drinks right before the traffic stop, or on the decline as alcohol was at its peak when the stop was made.  If the prosecutor can't provide the timing on the last drinks, a jury is left to wonder, and a good defense is ready to be made about reasonable doubt as to that level.  

The Datamaster machine looks at your BAC level at the police station, the number will always be higher or lower than the number when you were driving; jury instructions are provided that tell a jury that can infer the number is the same, but common sense, and a good defense will point out that it lacks common sense, and probably should not be followed. 

To be clear, the "second" - refusing the the test altogether would subject you to the Implied Consent law in Michigan.

​If you decide to take the Datamaster and avoid the Implied consent issue kicking in, there is nothing in the law that says you need to blow a second time as a confirming sample.  A second test only confirms the accuracy of the first test, and it could be easier to challenge the Datamaster accuracy with only one result vs two results.  

In Michigan, police agencies now use the Datamaster DMT for breath testing, which has replaced the BAC Datamaster. The new Datamaster uses infrared spectroscopy, which captures a sample of breath from the client. This sample may contain alcohol, which absorb infrared energy.

A known sample of infrared light is sent through the chamber, and if any ethanol is present, it absorbs that infrared light. The remaining light is measured by a detector; this “loss” of energy is then measured and you get an alcohol reading. This measurement is only as accurate as the device and the procedure allows.

Before reaching this point, the machine goes through a number of steps including purging where the internal plumbing is cleansed of any residual substances from prior samples. If it is clear, there will be a .00 reading displayed. The machine will then do it’s ambient zeroing, which an interference filter is inserted in the infrared path to check for contaminants in the ambient air sample. These filters are used to separate ethanol from potential interference such as acetone.


The machine then does a blank test and a .000 will appear on the display if no contaminants are detected. The final step is the internal standard check where a quartz filter is dropped into the infrared light source path. The filter absorbs infrared energy which is a known amount, determined during the calibration of the machine.

The infrared detector receives a reduced amount of infrared energy that was transmitted by the infrared source. The reduced intensity is evaluated by the microprocessor in the Datamaster, and a predetermined result is achieved. This step evaluates the machine’s electronic system, signal analysis, and proceeding system to determine that everything is working properly.

While all of these precautions sound impressive, they mean nothing if the machine doesn’t alert the operator that some of these components that do the internal checks are not working properly.

Datamaster operators will testify under oath that there is a hole in the assumption that things are working properly, because they are not trained on anything other than pressing buttons, and rely on the machine itself to tell them that is working properly. This is like going to the doctor for a medical opinion, then coming back two minutes later for a second opinion, there is no true checks and balance system, because the machine checks itself.

The Datamaster can be impacted by many factors, and another one of these would be various substances in the test-takers body. Ethyl grain alcohol is what is against the law, and what the Datamaster is looking for, but it’s very possible that substances such as ethyl ether, isopropyl alcohol, methyl alcohol, paraldehyde, acetone or trichloroethylene could lead to a misleading result.

Datamaster operators are not properly trained or knowledgeable to detect or factor in the presence of stimulants, depressants, tranquilizers or hallucinogens, all of which could impact the test result. In DUI cases in Michigan, perception is everything. Even if a client didn’t use any of these, it doesn’t matter. If the Datamaster operator acknowledges that these could impact the result, and they actually don’t know of their status for my client, then this uncertainty should be enough to discredit the Datamaster result. The standard of proof is beyond a reasonable doubt, and that lack of certainty is a good reason to doubt, and acquit.

Every person comes to the Datamaster machine with different characteristics; there are 100’s of differences between me and you, and one of these is our body temperature. Not everyone has the same “normal” temperature, and people may be suffering from higher or lower levels due to both internal and external factors. Essentially, the higher your body temperature, the higher the breath alcohol result. I’ve yet to come across a case where the police officer takes my client’s temperature. This is a problem for the test result.

The machine assumes that each person stepping up to the machine has a 98.6 fahrenheit reading. Consider that scientific studies have said that one degree in variance, could cause a 7 percent variance in the test result. What if my client has the flu, and has a 100.6 temperature? That could artificially increase the result by 21 percent, and could be the difference of winning or losing a trial.

In drunk driving cases, there will always be a time where a cop or witness observe the Defendant driving a car. This moment in time establishes when someone was driving, which can then be evaluated with the time that the chemical test was performed.


Recent case law eliminated the “reasonable period of time” rule in Michigan, and said the above scenario is enough to determine how much weight to give the result rather than to have it suppressed via pretrial motion by a judge. If 60 minutes pass between the first point of observed operation or 3 hours, that’s for a jury to decide, and not a judge, which has lead to more trials.


There is a remaining exception to the elimination of the reasonable period of time, which is the single-car accident, because the operation time cannot be determined, because there is no eye-witness to the driving. If someone hits a deer, and there is nobody around to determine the time, then we have no idea when the driver last operated. Unless the Defendant provides this information then there is no way to know, and courts have said under this scenario it is possible for a judge to determine that such an extensive or unknown time delay could make the test result irrelevant, and suppressible before getting in front of a jury at trial. If my client was involved in a single-car accident, and the last moment of operation cannot be determined then we may move for suppression of the results due to a lack of reasonable time period.


This technique puts the burden on the prosecutor to show the test is relevant by establishing operation. If they can’t do that then the tests are simply not reliable; my client could have been passed out in his car for three hours before someone else gets eyes on my client.

In Michigan the Datamaster machine must be maintained and tested properly in order for it to work properly. The machine determines the fate of tens of thousands of people in Michigan each year, and the machine can’t be treated like your car.

Sticking with the car analogy, law enforcement must always rotate their tires, get scheduled oil changes and always be operating within the most stringent guidelines; essentially getting the maximum gas mileage and performance at all times. If there is a lapse in these guidelines then the test is not reliable, and should be suppressed.

In Michigan, the rules say a class II, IIIA, IIIB or IVB operator must verify every machine on weekly basis using a vapor solution provided by Michigan State Police or using a compressed alcohol gas device that delivers exactly .080 grams of alcohol per 210 liters of vapor. Based on this deliverance, the machine MUST record a result of .076 to .084. Recent case law has ruled that the weekly compliance rule will not automatically suppress the result, but rather it will factor into the weight of the evidence for a judge or jury to determine.

When I get a case, I order the Datamaster logs, and see how the machine was testing before and after my client’s test was performed. Courts have ruled that attack of the logs is permissible and the other test results are admissible as a factor for a judge or jury to consider.
Along with the weekly test, the Datamaster needs to be inspected and verified for accuracy and cleanliness every 120 days by a class IVB operator or a representative from the manufacturer.
If the rules were not followed, or the machine is providing a number outside of the range then it should be attacked. You have two swings at this issue; pre-trial motion requesting that the judge suppress the results, and/or at trial where you can challenge the weight of the results, but they’d be admitted for a judge or jury to hear.

These techniques may require an expert to educate a judge or jury on what weight if any should be attributed to the results. In addition to the overall weight, it may be possible to use the results to one’s advantage.

If I am working on a case where my client recorded a 0.08 or even 0.09 on the Datamaster, and after looking at the logs, the machine is “testing high” at 0.084, we know the machine is turning the 0.08 into a higher number, which helps me argue that it’s possible, if not likely that my client actually could have blown a 0.076, but was only elevated by the margin of error allowed by the machine.

Finally, another method I like to point out to a jury is the machine being tested with a 0.08 solution. That’s probably a good idea for a case where my client blew a 0.08, as the jury knows that number could actually be 0.076 or 0.084, but if my client allegedly blew a 0.18, the machine was not tested for accuracy for what my client actually blew. This is like hitting the popcorn button on your microwave for everything. Would you hit the popcorn button to reheat iced cold chicken? No, and that’s how we attack the test result.

In Michigan, police officers use the Datamaster machine to test breath samples; this device is electronic and subject to the same interference as other devices that generate electrical power. One kind of interference that is possible is radio frequency interference or RFI.

Although the machine apparently has a built in sensor, which is designed to seek and identify this sort of interference, there is no guideline or information about how sensitive the sensor actually is during the tests. Just like a metal detector; some may pick up the metal on your belt, and some may not, but we all know there is metal being undetected. The same holds true for radio interference; we know there are radios in and around a police station, potentially even right next to the device if the officer has one; if the sensor doesn’t go off, does that actually mean something? It’s my position that there is always the possibility of some radio interference, even if the sensor does not pick it up.

In my experience there are at least five types of radio interference that must be detected, many which may not be picked up with a low sensitivity sensor.
It’s imperative to test for radio interference from mobile transmitters from patrol cars, ambulances and fire trucks, even news trucks driving around and private citizens using them. It must test for military and private security radio, base radio communication from the police station itself, portable transmitters from handheld police radios and a scenario where all of them are acting in concert.

It’s not a stretch to have a situation where the police station is surrounded by private citizens transmitting radio signals, military operations being conducted in the airspace, each cop having his own radio on his person, multiple police cars, and emergency vehicles transmitting and the police station itself (where the test is being conducted) sending out various signals to those same cars and officers.

Many prosecutors, police officers and judges roll their eyes at radio interference, and I really don’t quite understand why? It’s likely that the attorney presenting it moved on too quick without having the witness acknowledge all the possibilities, and the fact that we have no idea how sensitive the device actually is or isn’t.

Another problem area, which goes unchallenged is the machine only conducts a test at the very beginning; some of these tests take 10 to 15 minutes. Radio waves come and go, especially at a busy police station, and to simply ignore further tests is absurd. Some states have acknowledged this gap by placing shields on the device, but Michigan has not done so, which is an excellent issue to put in front of a jury to challenge the test results. In my view, until a shield is adopted by Michigan then no Datamaster test is actually fully reliable; how can a jury believe beyond a reasonable doubt that the test was accurate when a neighboring state doubted their same tests, and added the extra precaution.