Hi, I have a question as well. I am reviewing past calibration records of a particular instrument.

One of the calibration date shows 03 March, 2006 on the cert. The next cert date shows 19 September, 2006. The recall interval is 6 mos.

My question is that isn't the next date should have shown 03 September, 2006 instead of the 19th of September?

Thank you for your input.

Regards,

Tungsten

Hi, I have a question as well. I am reviewing past calibration records of a particular instrument.

One of the calibration date shows 03 March, 2006 on the cert. The next cert date shows 19 September, 2006. The recall interval is 6 mos.

My question is that isn't the next date should have shown 03 September, 2006 instead of the 19th of September?

Thank you for your input.

Regards,

Tungsten

What are the requirements? If the documentation says that recall has to happen in the month that the instrument comes due, there's no problem. If the documentation says that the thing must be recalled on or before the end date, then there might be a minor issue. What's the statute of limitations? :D Why worry about something that happened nearly three years ago?

Hi, Thanks for you all for the prompt reply. I have no cal procedure available. This particular equipment is a LION Intoxilizer 6000 IRL EBA (Evidential Breath Analizer) currently used by the Limerick (Ireland) Garda (Police) for evidential breath tests to determine the amount of alcohol in the breath (micro grammes in 100 ml of breath).

My job is to dig up the "dirt", and prove it to the court that there are (several) holes in the calibration intervals and the subject equipment was used from time to time in "out of calibration", otherwords it was ovedrue.

Regards and many thanks.

Tungsten

What are the requirements? If the documentation says that recall has to happen in the month that the instrument comes due, there's no problem. If the documentation says that the thing must be recalled on or before the end date, then there might be a minor issue. What's the statute of limitations? :D Why worry about something that happened nearly three years ago?

Thanks Jim for your input. Yeah, why should we worry...about something that happened almost 3 years ago. I wish it was that easy. I am involved in a court case where my job is to find holes whether the instrument (a breath analyzer) used for breath tests used during out of calibration condition.

Regards,

Tungsten :tg:

My job is to dig up the "dirt", and prove it to the court that there are (several) holes in the calibration intervals and the subject equipment was used from time to time in "out of calibration", otherwords it was overdue.


A weak defense might come from if that device - wherever it was used - ever failed a calibration. If so, how long was the calibration period? Because that would provide evidence that the period was potentially too short.

However, that can be easily countered if they can show that the device passed the previous calibration prior to the event in question, and a calibration after the event. That infers that it was in calibration throughout the period, whether it was late or not, leaving a need for another defense - such as operator training, documentation, etc.

Late calibration is only a risk that later calibration may find the device out of calibration, causing all data back to the previous calibration to be suspect. It is not evidence of out of calibration. A jury might not be able to follow that logic, though.

Of course, this is not legal advice - but it is a good place to start...:tg:

Hi, and thanks very much for your reply. So, otherwords if I stopped by the cops on 19th and they took a breath test with a device which was due for calibration on the 12th of the same month, there is no, or little argument from the point of legality?

Hi, and thanks very much for your reply. So, otherwords if I stopped by the cops on 19th and they took a breath test with a device which was due for calibration on the 12th of the same month, there is no, or little argument from the point of legality?

IF the calibration on the 25th showed the device in calibration, no - there is little to stand on. (Again - not legal advice - but good gage advice)

This is getting interesting...:). Actually, the next calibration did show the equipment in calibration with a note on the service record that the No. 1 Ethanol cylinder was replaced (the record did not say whether it was faulty, empty or there was a leak). The purpose of the ethanol cylinder (there are two altogether) in the unit is to self-calibrate the device prior to the breath test, and post self-calibrate again at the end of the test. It called the "blank" and it should show zero value.

The other thing is what looks interesting that the test was done at 14.5C, .5C BELOW the OIML International Atmospheric Recommendation on Evidential Breath Analyzers. The guide clearly states that for non-portable EBA's the rated operating conditions must be between 15 deg.C and 35 deg.C.

0.5C does not make much of the difference on the value of the outcome, but in the point of legality (I think...) could be challenged.

Back to the cal intervals, I am wondering why there is calibration intervals at all? Few days here and there out of calibration, seems like does not matter.

Any thoughts would be appreciated. - Thank you!:read:

This is getting interesting...:). Actually, the next calibration did show the equipment in calibration with a note on the service record that the No. 1 Ethanol cylinder was replaced (the record did not say whether it was faulty, empty or there was a leak). The purpose of the ethanol cylinder (there are two altogether) in the unit is to self-calibrate the device prior to the breath test, and post self-calibrate again at the end of the test. It called the "blank" and it should show zero value.

If it passed before and after the calibration, then replacement of the "zero standard" should be simply preventive maintenance - nothing there. If they do not do before and after changing the 'zero standard' (lessons learned here for you folks that do not ask for that data from your cal labs), then they have no idea if it was accurate or not before the change, and could be an issue.

The other thing is what looks interesting that the test was done at 14.5C, .5C BELOW the OIML International Atmospheric Recommendation on Evidential Breath Analyzers. The guide clearly states that for non-portable EBA's the rated operating conditions must be between 15 deg.C and 35 deg.C.

0.5C does not make much of the difference on the value of the outcome, but in the point of legality (I think...) could be challenged.

Never a good thing to not follow the standard, but, as you noted, it may be scientifically challenged by showing that the error was statistically inconsequential to the reading (% error, if you will, would not be large enough to bring the original data back to legal range).


Back to the cal intervals, I am wondering why there is calibration intervals at all? Few days here and there out of calibration, seems like does not matter.

Any thoughts would be appreciated. - Thank you!:read:

Calibration intervals are designed to mitigate the risk that the upcoming calibration may find the device out of calibration. It is not evidence of out of calibration, nor is calibrating within the interval a guarantee that the next calibration will pass. If it does not - even if done within the calibration interval - all readings back to the previous good calibration are suspect.

(Again - not legal advice - but good gage advice)

Something that could be questioned is the number of times a gage is utilized between calibrations. I have extend calibration intervals for those gages that only see occasional use where those that see continual use are calibrated more often. It becomes more of a count for the number of uses vs. a time period, for my use, a bit more accurate for maintaining calibration of a precision gage.

Something that could be questioned is the number of times a gage is utilized between calibrations. I have extend calibration intervals for those gages that only see occasional use where those that see continual use are calibrated more often. It becomes more of a count for the number of uses vs. a time period, for my use, a bit more accurate for maintaining calibration of a precision gage.

This IS exactly what I was thinking. The number of time the device was used in between calibrations. For a breath analyzer it should NOT be more than a thousand times. Frankly, in Ireland, this can be more. Far more.:tg: Thanks very much for your input!:applause:

This IS exactly what I was thinking. The number of time the device was used in between calibrations. For a breath analyzer it should NOT be more than a thousand times. Frankly, in Ireland, this can be more. Far more.:tg: Thanks very much for your input!:applause:

Yes, but again, that is just a manner of developing calibration intervals. It it was used a bazillion times between calibrations - or just once -and both calibrations passed, it is still all good.

Yes, but again, that is just a manner of developing calibration intervals. It it was used a bazillion times between calibrations - or just once -and both calibrations passed, it is still all good.

I think if there is a solid trend in using the instrument significantly MORE times than the recommended, but the cal intervals (6 mos) were NOT tightened, I'd definitely question accuracy. Don't you think?:(

Yes, but again, that is just a manner of developing calibration intervals. It it was used a bazillion times between calibrations - or just once -and both calibrations passed, it is still all good.

Yes and no. There is a specific guide for EBS' that the unit can not be used MORE than a thousand times in between calibrations. Again, this is all legality vs. practicality. (Unfortunately) I don't have much room for science to support the case. All I can grab into is LEGALITY. :thanks: for your input!

I think if there is a solid trend in using the instrument significantly MORE times than the recommended, but the cal intervals (6 mos) were NOT tightened, I'd definitely question accuracy. Don't you think?:(

You can always question it - but there really is nothing to support a problem existed by exceeding "recommendations" if the device passed the calibration each time. It's not quite the same thing as conceptually as running your car 20,000 miles between oil changes.

Again, this is all legality vs. practicality. (Unfortunately) I don't have much room for science to support the case. All I can grab into is LEGALITY.

True. Funny how they can be divergent worlds.:tg:

You can always question it - but there really is nothing to support a problem existed by exceeding "recommendations" if the device passed the calibration each time. It's not quite the same thing as conceptually as running your car 20,000 miles between oil changes.

True, and you're right! However there is one other thing. The manual of the LION Intoxilizer 6000IRL states that the machine produces an "ambient fail" reading if unable to conduct a test due to temperature, but this did not happened. Question: if the unit set to produce acceptable readings and print between 15 and 35 degrees C, how come it STILL printed a result at 14.5? :confused: Thanks for your inputs.:agree:

Hi, and thanks very much for your reply. So, otherwords if I stopped by the cops on 19th and they took a breath test with a device which was due for calibration on the 12th of the same month, there is no, or little argument from the point of legality?

You're in Ireland, so I have no idea what the laws might be. In the US, assuming evidence is admissible to begin with, the idea is to convince a jury that the calibration status of a device used in a particular instance is questionable--reasonable doubt.

True, and you're right! However there is one other thing. The manual of the LION Intoxilizer 6000IRL states that the machine produces an "ambient fail" reading if unable to conduct a test due to temperature, but this did not happened. Question: if the unit set to produce acceptable readings and print between 15 and 35 degrees C, how come it STILL printed a result at 14.5? :confused: Thanks for your inputs.:agree:

Now THAT is a valid question!

Calibration is simply verifying the instrument is operating like it is intended. So either the instrument has been working like it should, or not.

The instrument has a tolerance, and there is error with testing methods. That tolerance and error may/ may not be a factor in testing.

Say.. you are testing Blood Alcohol content, and it's important if the actual BAC is .08 or .1. If the instrument is only accurate to within .05, say, then you may have a case. If it is accurate to like .01 or better, I don't think you would get too far with that.

Now, if this instruments has a "calibration" program that must be ran prior to each use, that is different.

I would think you would be interested in determining the repeatability of the instrument. How reliable is this thing?

Bottom line: This instrument is suitable to be making the claims during testing, or it is not. Your burden is to prove in court that the instrument is/ was not suitable.

The instrument is only part of the equation. How about the operator (both contribute to errors)? Maybe it is easier to 'cast doubt' on the 'competency' of the operator if you had a hard time finding fault with the instrument.

The instrument is only part of the equation. How about the operator (both contribute to errors)? Maybe it is easier to 'cast doubt' on the 'competency' of the operator if you had a hard time finding fault with the instrument.

Hi Harry, Thanks very much for your input, good thinkin'!

I have a letter in front of me regarding the competency of the Garda Training Manual on the EBA from the Medical Bureau of Road Safety (University College of Dublin) the official body in Ireland who approves/calibrates the device. (The Intoxiliser is NOT certified by INAB. It is approved, supplied and tested (calibrated) by the Bureau pursuant to Section 6 of the Road Traffic Act, 1994. The MBRS is accredited to ISO1725 standards by INAB)

In the letter the Buerau notes that "There is nowhere in the course notes that indicate to the operator that they should not proceed with an Evidential Breath Test if the values are outside the temperature and relative humidity mentioned".

There is sufficient evidence that the test was performed at 14.5C outside of the capacity of the instrument.

Competent training? NOT!:nope:

Calibration is simply verifying the instrument is operating like it is intended. So either the instrument has been working like it should, or not.

The instrument has a tolerance, and there is error with testing methods. That tolerance and error may/ may not be a factor in testing.

Say.. you are testing Blood Alcohol content, and it's important if the actual BAC is .08 or .1. If the instrument is only accurate to within .05, say, then you may have a case. If it is accurate to like .01 or better, I don't think you would get too far with that.

Now, if this instruments has a "calibration" program that must be ran prior to each use, that is different.

I would think you would be interested in determining the repeatability of the instrument. How reliable is this thing?

Bottom line: This instrument is suitable to be making the claims during testing, or it is not. Your burden is to prove in court that the instrument is/ was not suitable.


Brad, thank you for your input. I have no data for repeatability of the unit. The accuracy of the device is 2 digits, acceptable. At this time all I've got is the subject's actual readout results.

The cycle of a test is as follows:

Blank... 000 (making sure there is no residual alcohol in the system (e.g. from a previous test. The value given by the machine MUST be zero)

Simulator Check 1... 035 (this is the first self-cal cycle PRIOR to the first blow of the subject. 035 used as a "standard" is the maximum allowed BrAC in Ireland (035ugm in 100ml of deep lung breath))

Blank... 000

Breath Specimen 1... nnn (the first blow of the subject)

Blank...000 (making sure that there is no alcohol remaining in the chamber. Preparation for the second blow)

Breath Specimen 2... nnn (the second blow of the subject)

Blank... 000 (making sure that no alcohol remains in the system)

Simulator Check 2...035 (self-cal, final)

To answer your question whether the instrument is suitable and/or reliable to make reliable claims? It is NOT. EBA's are widely criticized in forensic science regarding their accuracy and repeatability. Their operating principle is based on Henry's Law, which states that there is a fixed ratio exists for a fixed ratio of concentration of alcohol in water at a fixed temperature, therefore the ratio of alcohol in water is fixed. True. In theory, there is a constant ratio exists between the concentration of alcohol in the blood and the concentration of alcohol in the alveolar air of the lungs.

Now, here comes the interesting part, which I think I could challenge in court.

The average ration among human beings is generally accepted at 1:2100, that is, the same weight of alcohol will be contained in one cubic centimeter of pulmonary blood, as in 2.1 liters of deep lung breath. and most of the EBA's are set to this ratio.

But this is just an AVERAGE!

The actual ratio of any given individual can vary from 1:1500 to 1:3000, with a resultant error in ANY breath analysis arising from the 1:2100 ratio.

The subject device, the LION Intoxilizer 6000IRL is set to a blood/breath partition of 1:2300, which is near to the ratio of young, healthy 30 years old individual (1:2280). The subject is much, much older, with a chronic lung disease. It has been proven that both increasing age and lung disease can cause dramatically increased blood/breath partition ratios.

In this case the subject was tested with a device which is set to an average, a "one-size-fits-all". I'd call this the "convicted by a printout".

Fortunately, there are only few country remains which uses EBA's. It has been determined that such tests scientifically not convincing enough to demonstrate as "inherently reliable".

Consequently, the legal challenge mounted on behalf of the plaintiff should be considered in this light.

There is sufficient evidence that the test was performed at 14.5C outside of the capacity of the instrument.


Ok, let's accept that as a fact. Which is why you need to determine how much of an effect that would have on the indicated measurements. That is one estimated value of the entire uncertainty of this measurement process. I certainly don't mean to bore you with a bunch of "Calibration lingo", but for your situation, I do think it is important. Which is why I quote Harry below:

The instrument is only part of the equation. How about the operator (both contribute to errors)? Maybe it is easier to 'cast doubt' on the 'competency' of the operator if you had a hard time finding fault with the instrument.

Precisely. hence, the entire measurement process needs to be taken into account. As you stated the accuracy of the instrument is two digits (would that be +/-.01?), but that is only part of this picture.

You seem to have one decent line of defense about the variance due to using averages. Maybe if you could develop a total uncertainty, that would give you a second defense.

You're in Ireland, so I have no idea what the laws might be. In the US, assuming evidence is admissible to begin with, the idea is to convince a jury that the calibration status of a device used in a particular instance is questionable--reasonable doubt.

Thanks very much for your input Jim. Ireland is pretty much the same jury system as in the States. I don't think I'll have much of a problem to convince the jury.

The judge is a different story.

Actually, I'm from the States (Orange County, CA), working in Ireland for the last two and a half years. :mg:

Ok, let's accept that as a fact. Which is why you need to determine how much of an effect that would have on the indicated measurements. That is one estimated value of the entire uncertainty of this measurement process. I certainly don't mean to bore you with a bunch of "Calibration lingo", but for your situation, I do think it is important. Which is why I quote Harry below:



Precisely. hence, the entire measurement process needs to be taken into account. As you stated the accuracy of the instrument is two digits (would that be +/-.01?), but that is only part of this picture.

You seem to have one decent line of defense about the variance due to using averages. Maybe if you could develop a total uncertainty, that would give you a second defense.

Hi Brad, thanks kindly for your input. There is a generous 17.5% uncertainty budget in place in Ireland which would cover pretty much all areas of relevant uncertainty.

The other fact is that in reality a 0.5 degrees C difference in temp would not make much of a difference in the outcome of the result.

Here, I am only questioning legality, compliance, effectiveness of operator training, and the reliability of a device which is set to an average of an obsolete system exists only in a few countries in the world.

Luckily I don't have to deal much with reliability to convince the jury, simply because how reliable any test can be if the convicting device is set to an average, uses an average blood/breath partition that is, outside of its stated operating temperature range?