Video taken during controlled alcohol testing. Subject with well-documented true blood alcohol concentration between 116 and 120 mgs/100mls blows 303 mgs/100 mls as a result of mouth alcohol bias. Mouth alcohol detection algorithm fails to flag phenomenon as "Invalid Sample". Instrument is an Intoxilyzer 5000C.

The mouth alcohol detection system is looking for a pattern in the blow, whereby the BAC concentration rises quickly and then falls. However, breathing patterns during a subject test on an approved instrument aren't always the same. There are occasions when the presentation of some alcohol in the blood combined with some alcohol in the mouth or stomach, with a long gentle blow is a presentation not contemplated by the design of the instrument. The instrument accepts the sample as a valid sample, notwithstanding the presence of mouth alcohol.

Mouth alcohol detection systems are far from foolproof. Protocols followed by legal systems and operators must therefore require:

1. 15 to 20 minute observation/deprivation period before each subject test to watch for alcohol or food placed in the mouth, burping, belching, vomit
2, two tests with good agreement 15 minutes apart
3. never use a screening device (PBT or ASD) to attempt to obtain a quantitative result (ASDs and PBTs usually don't have mouth alcohol detection systems)

Intoxilyzer 5000 in need of maintenance. Second start. Instrument crashes again during start-up diagnostic RAM Check 2 error. Operator solves problem by tilting, shaking, banging, wiggling wires. Instrument resumes diagnostic and prints "Diagnostic OK". Instrument enters idle loop ready for breath test or other keyboard entry. Intoxilyzer® is a registered trademark of CMI, Inc. The Intoxilyzer® 5000C is an "approved instrument" in Canada.

This video shows that an unreliable approved instrument, in urgent need of maintenance, can pass Diagnostics checks. An approved instrument may display or print "Diagnostics OK" and yet be very unreliable. It seems that even though an instrument crashes during start-up, if you turn it off and on again, or do a warm reboot, you may be able to get it "working". It may still be unreliable as to ethyl alcohol quantitative analysis but it "works". There are serious problems with the idea that the function of an approved instrument is to "work" i.e. get from point A to point B. Crowns sometimes use the analogy of an automobile - it's not perfect but it gets from point A to point B. See for example section 320.34(3) of the Criminal Code of Canada:

"(3) The application shall be in writing and set out detailed particulars of the information that the accused seeks to have disclosed and the likely relevance of that information to determining whether the approved instrument was in proper working order. A copy of the application shall be given to the prosecutor at least 30 days before the day on which the application is to be heard."

If an instrument that cannot reliably quantitate ethyl alcohol passes its Diagnostics, is that proof that it "works" or is "working"? What does "proper working order" mean? Perhaps there is a prima facie appearance of "proper working order", after tilting, shaking, banging, and wiggling wires.

"Reliability" is a concept used in measurement science. It combines known accuracy, precision, and specificity, with no significant drift OVER TIME, together with known calibration, calibration intervals, qualification of operators, and following protocols set by the manufacturer and a program with supervision by a metrological authority.

Intoxilyzer 5000 in need of maintenance. Intrument crashes during start-up diagnostics check following warm up "Not Ready". Operator attempts pressing Green "Start" button. Red "Power" button eventually turns instrument off. Pressing red "Power" button do

Intoxilyzer 5000 in need of maintenance. Following two prior crashes but completion of start-up diagnostic, operator enters Esc Esc D to conduct stand-alone diagnostic check. Instrument crashes again at RAM CHECK 2 leaving test card in printer. Operator s

Intoxilyzer 5000 in need of maintenance. 100 alcohol standard at 34 C. Exposure to mixture of room temperature 50 alcohol standard and GOJO containing limonene bubbled into breath tube during air blanks plus GOJO on hands near flask air entry. After initial AMBIENT FAIL ACACA cal. checks now reading as low as 62 mg/100mls or even lower if INTERFERENT flag at 24 mgs/100mls. 0 has been raised by ethanol and limonene during air blanks far above 20 mg/100 mls threshold. In the process GOJO/limonene has either contaminated sample chamber, tubing, simulator or 100 alcohol standard or instrument's interferent add/subtract circuitry has been affected ultimately result in in 145 cal. check and 144 cal. check when room air used as ambient. Quaere: does contamination of an instrument and accesory equipment by an ambient interferent necessitate a change to the alcohol standard? Is it therefore necessary that the police log and the Crown disclose every "INTERFERENT" error (cards and COBRA/ADAMS) since the last solution change and prior to the subject test so that the defence will have full notice that the alcohol standard may have been unreliable? Intoxilyzer® is a registered trademark of CMI, Inc. The Intoxilyzer® 5000C is an "approved instrument" in Canada.

Intoxilyzer 5000 in need of maintenance. Experiments with GOJO hand cleaner containing limonene partially dissolved in cold 50 mg/100mls alcohol standard bubbled into breath tube during some air blanks. Effects on cal. checks with 200 alcohol standard. Sometimes but not always "Interferent" flag. Cal. checks far below and far above acceptable. Simulator/tubing/Intoxilyzer reliability compromised because cal. checks no longer reliable. Intoxilyzer® is a registered trademark of CMI, Inc. The Intoxilyzer® 5000C is an "approved instrument" in Canada.

Intoxilyzer 5000 in need of maintenance. 100 alcohol standard at 34C. Note high cal. checks on ACACA and ACABA sequences. Calibration check system (sample chamber, interferent detect/subtract/add circuitry, tubing, simulator, or standard) has been compromised by previous experiments bubbIing mixture of cold 50 mg/100mls standard and GOJO containing limonene. Quaere: Once there has been any indication of an "Interferent" should the alcohol standard be changed? Should any "Interferent" flag since last alcohol standard change be disclosed to the defence since the alcohol standard may not be reliable? Intoxilyzer® is a registered trademark of CMI, Inc. The Intoxilyzer® 5000C is an "approved instrument" in Canada.

Intoxilyzer 5000 in need of maintenance. 100 alcohol standard at 34 C. Prior exposure to mixture of room temperature 50 alcohol standard and GOJO containing limonene bubbled into breath tube during air blanks plus GOJO on hands near flask air entry. ACACA cal. checks now reading as high as 134 mg/100mls. GOJO/limonene has either contaminated sample chamber, tubing, simulator or 100 alcohol standard or instrument's interferent add/subtract circuitry has been affected. Quaere: does contamination of an instrument and accesory equipment by an ambient interferent necessitate a change to the alcohol standard? Is it therefore necessary that the police log and the Crown disclose every "INTERFERENT" error (cards and COBRA/ADAMS) since the last solution change and prior to the subject test so that the defence will have full notice that the alcohol standard may have been unreliable?

This experiment shows that RFI in the breath room can change BAC indications on the display of an evidentiary breath instrument without triggering an RFI Interference error message. The "faraday cage" steel case of the instrument did not prevent the RFI from penetrating the instrument.

Intoxilyzer 5000 in need of maintenance. Instrument in ABA (Using EscEsc B) and ACABA mode. Print inhibit on. 100 alcohol standard at 34.0C . RFI applied at various positions at back of instrument during B in sequence. RFI from two Cobra walkie-talkies Channel 3 00 used, one transmitting, another receiving, so you can hear when "Talk" is pressed. Intoxilyzer® is a registered trademark of CMI, Inc. The Intoxilyzer® 5000C is an "approved instrument" in Canada.

This is another experiment that shows that RFI in the breath room can change BAC indications on the display of an evidentiary breath instrument without triggering an RFI Interference error message. The "faraday cage" steel case of the instrument did not prevent the RFI from penetrating the instrument.

Intoxilyzer 5000 in need of maintenance. Instrument in ABA (Using EscEsc B) and ACABA mode. Print inhibit on. 100 alcohol standard at 34.0C . RFI applied at various positions at back of instrument during B in sequence. RFI from two Cobra walkie-talkies Channel 3 00 used, one transmitting, another receiving, so you can hear when "Talk" is pressed. Intoxilyzer® is a registered trademark of CMI, Inc. The Intoxilyzer® 5000C is an "approved instrument" in Canada.

Please note that this instrument twice erroneously displayed an "INVALID SAMPLE" (mouth alcohol detection) error message as a result of RFI but did not display an "RFI interference" error message. On other occasions, the instrument erroneously displayed a BAC analysis with no "RFI Interference" error message.

Subject with zero BAC. Intoxilyzer 5000 in need of maintenance. Instrument in ABA (Using EscEsc B) and ACABA mode. Print inhibit on. 100 alcohol standard at 34.0C . RFI applied at various positions at back of instrument during B in sequence. RFI from two Cobra walkie-talkies Channel 3 00 used, one transmitting, another receiving, so you can hear when "Talk" is pressed. Intoxilyzer® is a registered trademark of CMI, Inc. The Intoxilyzer® 5000C is an "approved instrument" in Canada.

Experiment to explore the effects of a hand cleaner such as GOJO containing limonene in the ambient air respecting reliability of the alcohol standard. 100 SAS in simulator at 34.0C. Operator runs ACACA mode. Cal. check drops from 103 to 81, 96, 85 when hand cleaner in air near breath tube on prior air blank and then returns to 111, 105. Note that 111 is outside of acceptable range. It appears that ambient hand cleaner in the breath room can reduce an apparent cal. check by 20 mg/100mls. Although the breath test result itself may be reduced, the reliability of the simulator/alcohol standard has been compromised and so the breath test itself is not reliable. If the instrument is, for example, reading 11 mgs/100 mls too high and the problem is not with the simulator (note the 111 cal. check), then this inaccuracy may be masked. Cal. checks could be in the normal 90 to 110 mgs/100mls range only because the cal. check reliability was compromised by the ambient limonene and the instrument's response thereto.

Main simulator has 100 SAS at 34.0C. This video is a sequel to two longer videos. Experiment using cold extra simulators at 40 mgs/100mls and 100 mgs/100mls gradually warmed to simulate slow changes in ambient ethanol in room air. Instrument does not trigger "Ambient Fail" because the change is gradual. The results of this experiment show why it is very important for the operator to control breath room ambient conditions and pay close attention to any indications of bizarre ambient conditions. Those indications may include strange cal. checks, ambient fail messages, and others in the weeks before and after the subject tests. In order to make full answer and defence respecting the issue of instrument reliability the defence needs full disclosure of conditions that affect the SAS reliability including expiry, stability, temperature, AND AMBIENT CONDITIONS IN THE BREATH ROOM. Please note that the need for such full disclosure is even more important if ambient conditions potentially include ethanol PLUS INTERFERENTS such as limonene.

This is an example of RFI causing erroneous BAC analysis indications on an evidentiary test instrument without triggering an "RFI Interference" error message. Please notice the "purge fail" phenomenon when the instrument cannot clear the 029 result (zero has been reset too low by the RFI).

Intoxilyzer 5000 in need of maintenance. Intoxilyzer® is a registered trademark of CMI, Inc. Instrument in ACACACACACA mode. Print inhibit on. No simulator, no source of ethanol or interferent. RFI from two Cobra walkie-talkies Channel 3 00 used, one transmitting, another receiving, so you can hear when "Talk" is pressed. The Intoxilyzer® 5000C is an "approved instrument" in Canada.

This experiment shows that RFI can cause many different erroneous error messages and BAC results without triggering the correct error message which should be "RFI interference". Note the "purge fail" when the air blank won't clear.

Intoxilyzer 5000 in need of maintenance. Instrument in ACACACACACA mode. Print inhibit on. No simulator, no source of ethanol or interferent. RFI from two Cobra walkie-talkies Channel 3 00 used, one transmitting, another receiving, so you can hear when "Talk" is pressed. Intoxilyzer® is a registered trademark of CMI, Inc. The Intoxilyzer® 5000C is an "approved instrument" in Canada.

This is another example of zero, as set by the instrument, being compromised by RFI. There is no "RFI Interference" error message and yet there are cal. check results (with no source of ethyl alcohol). Note the "purge fail" caused by zero being set at about -.024 to -.029. A zero air blank therefore reads as 024-029.

Intoxilyzer 5000 in need of maintenance. Intoxilyzer® is a registered trademark of CMI, Inc. Instrument in ACACACACACA mode. Print inhibit on. No simulator, no source of ethanol or interferent. RFI from two Cobra walkie-talkies Channel 3 00 used, one transmitting, another receiving, so you can hear when "Talk" is pressed. The Intoxilyzer® 5000C is an "approved instrument" in Canada. RFI during air blank and internal standards generates phantom ethanol during cal. check.

Intoxilyzer 5000 in need of maintenance. Keyboard completely detached. However, operator runs ACABA subject test that appears normal and produces Intoxilyzer Test Record. Intoxilyzer� is a registered trademark of CMI, Inc. The Intoxilyzer� 5000C is an "ap

Intoxilyzer 5000 in need of maintenance. Instrument runs start-up diagnostics following Not Ready warm-up and passes start-up diagnostics but shows date as "25/45/08". Keyboard or keyboard controller board is defective and detached as can be seen from han

Intoxilyzer 5000 in need of maintenance. Keyboard or keyboard controller or battery defective and so keyboard detached. Instrument, however, runs 5 acceptable stand-alone cal. checks indicating no errors. Operator using ACA dip switch since keyboard detac

Intoxilyzer 5000 in need of maintenance. Operator cannot correct date error using keyboard so uses dip switches at side of instrument to reset date. Instrument returns to idle loop ready for cal. check or subject test. Intoxilyzer� is a registered tradema

Intoxilyzer 5000 in need of maintenance. Following "Processor Errors" crash, re-start , stand-alone varying cal. checks, operator adjusts dip switch to put instrument back into ACABA mode. Instrument is ready for breath test. Excellent air blank 000, cal.

Intoxilyzer 5000 in need of maintenance is re-started using Red start button following crash during prior start-up 0004. Instrument runs through start-up diagnostics and indicates "Diagnostics OK" notwithstanding immediately prior crash and high/drifting

Intoxilyzer 5000 in need of maintenance. After Processor Error crash 0004 and re-start 0005, cal. checks (stand-alone "Start" test button after changing dip switch) show instrument and/or simulator are still unstable with very different cal. checks and on

Intoxilyzer 5000 in need of maintenance. Follows processor errors crash, re-start, and bad cal. check 0008. Operator now runs further stand-alone cal. checks obtaining 96 and 103, two results within 90 to 110 range. Operator now feels its is safe to bring

Intoxilyzer 5000 in need of maintenance. After "Not Ready" warm up period instrument crashes at "Processor Check" indicating "Processor Errors" and immediately goes into DVM mode indicating high and fluctuating DVM values. Sample chamber is probably conta

Intoxilyzer 5000 in need of maintenance. Operator runs two groups of stand-alone cal. checks with no simulator attached. Cal. checks at .00. However, instrument prints test cards notwithstanding no simulator, no alcohol standard, no temperature maintenanc

Intoxilyzer 5000 in need of maintenance. Following 3 crashes, two rectified by shaking and banging machine, operator enters Esc Esc B breath test sequence. Subject test completed producing Intoxilyzer Test Record as evidentiary "proof" that instrument is

Intoxilyzer 5000 in need of maintenance is re-started using Red start button following crash during prior start-up 0004. Instrument runs through start-up diagnostics and indicates "Diagnostics OK" notwithstanding immediately prior crash and high/drifting

Attorney Justin J. McShane of the McShane Law Firm demonstrates how to create a false positive on an Intoximeters Alco-Sensor FST (in Canada we call them ASDs) using ordinary household bread.

Subject test .034 with no alcohol in subject's body as a result of Nokia mobile phone call received. Note cal. check .096 and clear diagnostics test printed.

Experiments respecting effects of radio frequency interference on Intoxilyzer 5000 64 series upgraded to 66 series. Intoxilyzer in need of maintenance. Nokia 3500 cell phone purchased at Walmart June 2009 used to make calls during various Intoxilyzer 5000 sequences. ACABA mode, 100 alcohol standard. When RFI applied during first Air Blank often results in "Ambient Failed". When applied during first or second flying wings Internal Checks often results in "Unstable Reference". When applied during subject test sequence often results in high readings. Breath subject is alcohol free at all relevant times. Intoxilyzer® is a registered trademark of CMI, Inc. The Intoxilyzer® 5000C is an "approved instrument" in Canada.

In this particular example, subject blows .530 at 2:03 in video with no alcohol in subject's body. See also .253 at 4:19.

Experiments respecting effects of radio frequency interference on Intoxilyzer 5000 64 series upgraded to 66 series. Intoxilyzer in need of maintenance. Nokia 3500 cell phone purchased at Walmart June 2009 used to make calls during various Intoxilyzer 5000 sequences. ACABA mode, 100 alcohol standard. When RFI applied during first Air Blank often results in "Ambient Failed". When applied during first or second flying wings Internal Checks often results in "Unstable Reference". When applied during subject test sequence often results in high readings. Breath subject is alcohol free at all relevant times. Intoxilyzer® is a registered trademark of CMI, Inc. The Intoxilyzer® 5000C is an "approved instrument" in Canada.

Subject with no alcohol blowing .700, i.e. beyond the range possible on the evidentiary test instrument. A "Range Exceeded" error message results. There was no RFI error message.

Mississauga Toronto Ontario lawyer Stephen R. Biss speaks about the effect of cell phones on the reliability of the evidentiary breath instruments used in Ontario Canada. Radio frequency interference from a transmitting Nokia 3500 cell phone causes an apparent blood alcohol concentration beyond 700 mg/100mls on an Intoxilyzer 5000 64 series product number 564 CA model similar to the Intoxilyzer 5000C used peviously by police in Ontario Canada. Police in Ontario now use the Intoxilyzer 8000C.

The Intoxilyzer 5000C evidentiary breath testing instrument used in Ontario, Canada is an Intoxilyzer 5000 product number 564. The 64 series model uses technology from the 1980's. The 5000 64 series was replaced in the United States in the 1990's by the 6

Intoxilyzer 5000 64 series mouth alcohol

Defence lawyer, Stephen Biss, explains the impact of interferents such as hand cleaners on the Intoxilyzer 5000C approved instrument in Canada.

False "Ambient Fail" error message generated by RFI.

Intoxilyzer 5000 in need of maintenance. When subjected to cell phone radio frequency interference during the first air blank, this particular instrument, in need of maintenance, interpreted RFI as ambient alcohol in the room air.

This is an example of what is supposed to happen: An "Inhibited RFI" error message is triggered when RFI applied. The Instrument is an Intoxilyzer 5000EN.

The Intoxilyzer 5000C is an "approved instrument" in Canada. The Intoxilyzer 5000EN is two generations beyond the Intoxilyzer 5000C.

This is an example of a false Error Message. The instrument indicates "Interferent" i.e. a chemical interferent when the cause is radio frequency interference (RFI). Error messages are not always reliable to identify the correct error.

ACACA calibration checks using 100 solution, first cal. check 102, second cal. check indicates "Interferent" then 106 when exposed to cell phone RFI

An "Invalid Sample" error message can be triggered by Cell Phone RFI. Error messages are not always reliable to indicate the true cause of the error.

Radio frequency interference can cause a "Range Exceeded" error message on an approved instrument.

Range exceeded error caused by RFI

The Intoxilyzer® 5000C is an "approved instrument" in Canada. TheIntoxilyzer® 5000EN is two generations beyond the Intoxilyzer® 5000C.

The Intoxilyzer 5000C is an "approved instrument" in Canada. The Intoxilyzer 5000EN is two generations beyond the Intoxilyzer 5000C.

The Intoxilyzer® 5000C is an approved instrument in Canada. The IIntoxilyzer® 5000 66 series is one generation beyond the 5000C which is a 64 series. Instrument subjected to cell phone transmission during cal. check and during subject test results in BAC o

The Intoxilyzer® 5000C is an "approved instrument" in Canada. The Intoxilyzer® 5000EN is two generations beyond the Intoxilyzer® 5000C. Instrument accepts sample and generates test card showintg a BAC of 12 mg/100mls from subject with 0 mg/100mls BAC. The

The Intoxilyzer® 5000C is an "approved instrument" in Canada. Cell phone transmission triggers Unstable reference flag.

Do cell phones that are receiving only, not transmitting, cause radio frequency interference to scientific instruments such as evidentiary breath testing equipment? An Intoxilyzer 5000 in need of maintenance is used to conduct an ACABA breath test using a

Does receiving a text message in a breath room interfere with calibration checks conducted by a breath alcohol technician before and during evidentiary breath testing? Nokia 3500 cell phone receives text messages during calibration checks. Using 100 solut

If a text message is received by an idle cell phone in a breath room, could it affect Intoxilyzer results? This video suggests that there needs to be more study on this issue since clearly in this video a subject with 000 BAC managed to blow 122 mg/100mls

Details of Intoxilyzer 8000C Esc Esc D sequence