RFI (Radio Frequency Interference) and Breathalyzers

radio frequency interferenceThe effect of radio frequency interference (RFI) or electromagnetic interference (EMI) on the results of breath tests conducted by the police seems to be a perennial issue in criminal courts. I first encountered this issue when it was raised in the early 1980s in relation to the Breathalyzer 900A. Doug Lucas (who was my director at the Centre of Forensic Sciences at the time) wrote an excellent chapter on the Breathalyzer including the RFI issue as part of a book called “Breathalyzer Law in Canada”, (Carswell Publishers). Unfortunately, this book is not readily available to the forensic community so I will cite extensively from Doug’s chapter as I address this blog update.

“Those who do not learn from history are doomed to repeat it”
George Santayana (1863 -1952)

As we have now already covered, the Breathalyzer was originally invented by Robert F. Borkenstein in the early 1950s (see previous blog). The basic instrument measured alcohol by determining the color change in glass ampoules containing a potassium dichromate/sulphuric acid solution, by comparing the voltage generated by photocells when the breath technician activates a light source and observes the needle on a null meter (WOA30101-2).

In the early 1980s a different model of the Breathalyzer was manufactured which contained an electronic 90 second timer and an amplifier to boost the signal originating from the photocells.  This was the “Breathalyzer Model 900A”.  By adding an amplifier, the Breathalyzer unknowingly became susceptible to RFI from police radios. The amplifier would convert the RFI (of the proper wavelength and strength) into a voltage and cause the needle of the null meter to move.  If the radio transmission occurred exactly when the breath technician activated the light source to measure a drinking driver’s BrAC, it could possibly add to the result. The older Breathalyzer model 900 contained no amplifier and was not susceptible to RFI.

In Canada (and now in most jurisdictions) the breath test procedure of a drinking driver includes a 15-20 minute (observation/deprivation/arrest) period, blank tests, an external calibration test and two separate consistent breath tests. It was, in fact, by using this procedure that breath technicians in Ontario observed the effect of RFI in those few locations where the transmitter antennae was located near the Breathalyzer room. To ensure that RFI presented no problem, these locations had policies of no radio transmissions during breath tests and/or relocated the Breathalyzer to a more shielded room.

RFI became an issue when the manufacturer of the Breathalyzer 900A (Smith and Wesson) issued a general advisory on September 10th, 1982 that read:

“the possibility exists, although unlikely, for higher or lower normal test results”

At the Centre of Forensic Sciences in Toronto, we tried to deliberately fool the breath technicians by having a handheld radio transmitter next to a Breathalyzer 900A and activated it only and exactly when the technician measured the breath test results.  But, even under these extreme conditions, the technician always detected the interference.

Smith and Wesson on December 1983 issued a revised statement about RFI in Canadian operations:

“..the probability of radio frequency interference causing inaccurate subject test readings that are still in agreement is, for all practical purposes, non-existent.   Even assuming for a moment that this might happen, it would likely be detectable by a qualified breath test technician who performs these tests using Canada’s protocol.  With this in mind, it would appear that the test procedures as detailed in the Smith and Wesson Customer Advisory, dated September 10th, 1982 would be, at best, of limited value in view of the extremely rigid breath testing protocol used in Canada”

Although RFI was not a significant risk for programs with external calibration checks and duplicate breath tests, there was speculation that there would be a much greater potential risk in breath testing procedures in which only one breath test was conducted as RFI would have to occur only once, rather than multiple times. In response, a “technical fix” was developed by the manufacturer of the model 900A by issuing an RFI kit to retroactively attach to the instrument to prevent the amplifier from being affected by radio waves.

Today, virtually all modern breath alcohol testing instruments contain amplifiers and electronic circuitry that could generate a false reading.  However, now generally all are shielded against RFI and may even contain an RFI detector. Pat Harding’s classic study of the accuracy of the Intoxilyzer 5000 in police use found that there was no evidence that falsely elevated BrAC test results were obtained due to electromagnetic interference (WOA30406).

What we have learned from this historical review of RFI and breath testing instruments is that a proper breath testing procedure provides a much greater safeguard against false high breath test results than any technology. And as shown in the 1980’s, a rigorous breath testing protocol will exclude the possibility of RFI even in instruments like the Breathalyzer Model 900A that initially did not include any shielding or an RFI detector.


RFI (Radio Frequency Interference) and Breathalyzers — 5 Comments

  1. Pingback: Duplicate Breath Tests: A Necessary Scientific Safeguard

  2. Bob, I am glad my blog is proving useful to you. You can be sure I will continue to write on topics relevant to the field of forensic alcohol toxicology. Feel free to join the WOA community as well through the sign-up form (its free), as I provide more extensive updates and papers to my members on a monthly basis. Thanks for your comments…they are always appreciated.


  3. I hope you continue this series. It is useful for those who do not work exclusively or exhaustuvely in the field of DWI defense.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>