Stephen BissOIML R126 is the International Standard for Evidential Breath Analyzers
Updated: Aug 8, 2022
Interesting things about R 126:
It exists and yet Canadian Judges and lawyers don't know about it, notwithstanding the fact that it is an international standard and Canada is a member state of OIML.
It is not referred to as a reference or a footnote in the most important publications by the Alcohol Test Committee of the Canadian Society of Forensic Science.
It has been referred to occasionally in the past to justify the Canadian evidentiary breath testing.
OIML R 126 uses specific terminology that is the same vocabulary consistent with other OIML, BIPM, and ISO scientific documents and vocabularies.
It connects well with Canada's Weights and Measures Act, the Fairness at the Pumps Act, the Constitution Act 1867, and Magna Carta.
But we do something different in Canadian evidentiary breath testing and law.
Unlike the Fairness at the Pumps Act, our evidentiary breath testing system and law do not connect with important concepts of measurement science that are recognized in OIML R 126.
"verification of a measuring instrument"
"initial verification"
"verification after repair"
"mandatory periodic verification"
"legal metrological control"
"metrological authority"
"metrological supervision"
"maximum permissible error (MPE)"
Note: OIML R 126 was updated in 2021.
Purpose:
To obtain admissions that:
there exists an international legal measurement standard for evidential breath analyzers
the International Standard for Evidential Breath Analyzers is OIML R126
that OIML R126 is the international standard referred to by Brian Hodgson in his paper
that Canada is a member state of OIML
that Canadian scientists have played a role in the development of this international standard
that the international standard does not state that one control test data point is adequate to check the reliability of evidential breath analyzers
to introduce the concept of MPE or "maximum permissible error" at low, medium, and high data points, such as 50, 100, 300 mg/100mls
to consider the different applications of MPE at each of: type approval, initial verification of the specific instrument, verification after repair, all contrasted with at time of subject tests.
Q. Nothing at your fingertips. All right. Now, Brian Hodgson, in his paper (see footnote 1) refers to an international standard called O-I-M-L R-1-2-6 in relationship to evidential breath testing. A. Okay. Thank you. MR. BISS: And Your Honour, I’m going to pass you up a copy. My friend’s – oh, [the local] Police and the Crown will have copies of these already. I think I have one for Madam Reporter as well. Q. Just in terms of the O-I-M-L, as I understand it, the O-I-M-L is an organisation with the name that’s on the front page of this document. Have you heard of this document – this – this organisation before? A. Yes. Q. And it exists as a result of a convention – an international convention from 1955 and specifically, in terms of member states of that convention, the member states are listed at Tab 9. A. I’ll have to take your word for it. Q. And you – all, right, and you’ll agree with me that on – at Tab 9, one, two, three, on the fourth page down it’s indicated that Canada is a member state. The representative for Canada is Mr. Allan Johnson for Measurement Canada. A. One moment please. Q. Four pages down in Tab 9.
A. Yes. Q. All right. Now, there is a document contained in this bundle called O-I-M-L R-1-2-6. It’s at Tab
Number 7, the 2012 version. And will you agree with me that page 6, that the scope of the recommendation, top of page 6 item number 1. “This recommendation applies to quantitative breath alcohol analysers that render a measurement result of alcohol concentration in exhaled human breath for the purpose of establishing compliance with national policy for fighting against alcohol abuse.” That scope would appear to apply – the instruments that are being described are the same types of instruments that we are dealing with all of the time, we call approved instruments. A. Yes, the next paragraph down says, “These types of quantitative breath alcohol analysers are referred to by some national authorities as evidential breath alcohol analysers and serve to provide the principle means by which a definitive alcohol measurement is obtained. Yes. Q. That’s – would seem to fit out concept of an approved instrument. A. Yes. Q. And the scope of this document does not include devices that provide a preliminary result or don’t quantitatively indicate a measurement result. In other words, something like a screening device. A. I’d have to read the whole document to make that assessment, but... Q. Just take a look... A. ...I could.... Q. ...at the third paragraph under scope, page six. A. Correct, yes.
Q. All right. Now in terms of Canadian connection with this document, I’m just going to pass up a list of committee members, of O-I-M-L, and specifically,
within O-I-M-L, individuals that are listed include someone called – well, I can see the email address, is Benny dot Wong at C-S-F-S dot C-A. Do you know somebody by the name of Benny Wong? A. He works for the R.C.M.P. in Ottawa. Q. Right. And another person that’s mentioned on this committee is Felix Como, a company called A-C-S Alcohol Countermeasure Systems. A. Yes. Q. You are familiar with that individual? A. I am not, but I am familiar with the company. They produce alcohol standard solutions that are often used in simulators for checking the calibration of the devices. Q. I think that was the same company that the Crown attorney was cross-examining Mr. Kupferschmidt about doing consulting work for them on one of the earlier days in this proceeding. A. That's correct. I believe he certifies their alcohol standard solutions prior to it being released... Q. And then there’s another reference... A. ...to us to be tested. Q. ...to a – someone by the name of Carl Cotton, of Measurement Canada. A. Yes. Q. Do you know that person? A. I do not. Q. And then there’s another person, Mr. Robert L’Agile (ph). A. Yes. Q. Do you know that individual? A. I do.
Q. Somebody who you work with every day at the Centre of Forensic Sciences? A. Yes, I do. Q. I see. Now, going back to Exhibit 12, Exhibit 12 in this matter, which are the document disclosed with respect to this particular Intoxilyzer, and the certificate of calibration that we talked about before at page 7 of Exhibit 12, the certificate of calibration for the Intoxilyzer 8000 certified by C.M.I. Inc., makes reference to an O-I-M-L standard. O-I-M-L R-1-2-6. A. Um-hmm. Q. It specifically makes reference to an edition, 1998. A. Yes. Q. Whereas the book that I’m just showing you, and we just referred to the tab we referred to, is the 2012 version. I just want to make sure that you see that we have in this document.... A. 2013 edition. Q. Actually, you’ve just – hold on for a moment, and Tab 7 – Tab 7 is 2012. A. Yes. Q. Tab 6 is the 1998 version. A. Yes.
Q. All right. Now, again, going back to this practice of only checking at one data point, only checking at 100 milligrams per 100 mils. I want to suggest to you that in the international literature, the practical problem of an instrument during time of use in deviating beyond certain limits, in the international literature they call that the maximum permissible error. The M-P-E. A. I don’t know that definition. I’m sure
that’s something that you’re reading? Q. All right. If I can just... A. Yeah. Q. ...refer you please, in O-I-M-L R-1-2-6 page 10 there’s an indication there of – at paragraph 5.2, a paragraph called Maximum Permissible Error. A. Sorry, what.... Q. Maximum Permissible Errors. A. For which section? Which tab? Q. Ah, this is in Tab Number 7. A. Oh, Tab 7, sorry. Q. Yes. A. Five point.... Q. Five point two and specifically five point two point one, on page ten. A. Maximum Permissible Errors for type approval and initial verification and verification after repair.
Q. Yes. I just wanted to ask you, it’s clear from this document that in this international standard the maximum permissible error is used not just with respect to the initial evaluation of an instrument, but also with respect to occasions when the instrument is verified initially, and where there has been a verification after repair. A. I would need to read this paragraph. Q. If you just might read then, paragraph 5.2.1. A. All right. If you can repeat your question then? Q. So, under 5.2.1 it’s pretty clear that the O-I-M-L has defined a maximum permissible error for type approval, which I gather is something similar to an evaluation
of an instrument initially, to initial verification of an individual instrument, and verification after the repair of an instrument. A. That’s what that appears to be yes. Q. And the maximum permissible error is stated as a maximum but is not limited to just 100 milligrams per 100 mils. A. I don’t see where you’re reading that. Q. Well, I see that it indicates that the maximum permissible error, positive or negative, is .02, so that is – is that – that’s plus or minus 2 milligrams per 100 mils. A. Yes. Q. Or five percent of the referent value of the mass concentration. So, for example, if we were dealing with 40 solution then it would be 2 milligrams per 100 mils, which is the same as what it is in milligrams per litre, if we use it as a percentage. For 50 solution, it would be the same thing in the sense that it’s plus or minus 0.02 milligrams per litre or 0.02 milligrams per litre calculated as a percentage. Whereas with 100 solution, it is a maximum of 2 milligrams per litre or 5 – I’m sorry, milligrams per 100 mils or 5 milligrams per 100 mils if you do it by way of percentage, under the maximums. A. Yes. Q. Now what’s interesting there is, the maximum permissible error is described across the range, the measuring range of the instrument, as described in paragraph 5.1. A. So, the range they’re giving at the top is 9 to 200 milligrams of alcohol in 100 millilitres of blood. Q. Yes.
A. So the breath alcohol analyser shall be capable of measuring all mass concentrations in the range of 0 to 2 – 0.00 milligrams per litre to at least 2.0 milligrams per litre, which is 0 to 200 milligrams of alcohol in 100 millilitres of blood. Q. Yes, A. However, in the measuring mode, the breath alcohol analyser may indicate 0.00 milligrams per litre for mass concentrations equal to or small than a given value defined under the responsibility of the national authorities, such a masking function shall be able to be cancelled in maintenance mode. Q. All right, so we know that with respect to the Intoxilyzer 8000C as with the Intoxilyzer 5000C the first approximately 6 milligrams per 100 mils are masked? A. No. If the result is six, as – that is shown as zero. It doesn’t mask it for actual results that are larger than that. Q. I see. All right, but anyway, my point is, if you read 5.2 and 5.2.1 maximum permissible errors in this document in the international literature standard of what’s to be expected of the performance of these instruments, is the maximum permissible errors are defined as a percentage or a number of milligrams per 100 mils but it applies across the range. It’s not just at 100 milligrams per 100 mils. A. It seems to be – that’s your interpretation of it. It doesn’t actually specify the number of samples that have to be used to check that result, in that – just giving you a concentration range that the result should be between 0 and 200 milligrams of alcohol in 100 millilitres of blood, doesn’t say that there should be one at 50, one at 100 and one at 150.
Q. I see, but if one.... A. Additionally, there’s also 5.2.2 which is maximum permissible errors for breath alcohol analysers in service. Q. Yes. A. Right? Which would probably be more applicable to what’s being used and what’s being done in the field as opposed to what’s being done in a laboratory or being done during verification like by the Alcohol Test Committee. Q. All right. A. The parameters are...
[Note: There was an error by the cross-examiner in the original chart put before the witness that showed incorrect calculations of MPE. This was subsequently corrected after Mr. Kupferschmidt noticed the error while reading the transcript. The error stemmed from the difference between BrAC and BAC. The corrected chart is shown here.]
MR. BISS: Q. Just to understand those numbers, you can tell me if my math is right. So, for in service, those are the maximum permissible errors. If we were looking at 40 standard, 50 standard, 100 standard or 300 standard. Those are the maximum numbers in the international literature. A. At first glance, they appear to be – your math appears to be correct. Q. All right, so if – using the international standard, then when we’re dealing with 40 solution we should
be looking at a maximum permissible error of – from the perspective of the international literature, we should be looking at a maximum permissible error of plus or minus 3 milligrams per 100 mils. Right? A. That is correct, based on these values, yes. Q. And same with 50 and at 100 we should be looking at a maximum of plus or minus 7 milligrams per 100 mils. Right? A. That is correct. Q. But my point is that the concept of maximum permissible error in the international literature seems to permit that maximum permissible error to vary across the measuring range of the instrument. A. Well, the range here is from 0 to 200. You’ve actually gone up to 300. Q. Yes. All right, well, if we do it with 200 then it would be a similar calculation. A. Yes, it would. Q. But my point is that in terms of international literature it seems that when looking at maximum permissible error for an instrument in service the international literature expectation is a whole lot tighter than plus or minus 10 milligrams per 100 mils, for an instrument in service. A. Correct. Q. And not only that, but it is a number that needs to be considered across the measuring range of the instrument. Not just at one value of 100 milligrams per 100 mils. A. Again, I don’t see where it says that in this document.
Q. Well.... A. It doesn’t specify that there should be more than one check of the concentration range or the linearity in this document.
This cross-examination continues under the heading:
International Std OIML R126 Mandates Maximum Permissible Error blog entry of January 16, 2018
1. Hodgson, Brian article "The Validity of Evidential Breath Alcohol Testing"