The 6th meeting of CEN/TC 264/WG 2 took place online from the 6th to the 8th of April. This meeting was organized after the cancellation of the meeting in Delft, The Netherlands due to Covid-19 impact. There were three days altogether from 9:00 to 13:00 dealing with the comments sent by the experts of the member states to the group. Now the draft is one step closer to be ready.
Many decisions were taken during this meeting. Many comments were just accepted, others were accepted but with modifications and a few of the comments were rejected. An important part of the meeting dealt with the calculation of uncertainty, but there were also other important issues. Let's try so summarize a bit 3 days of meetings.
There were many interesting discussions on the first day of the meeting, but I would say that the most relevant one came after a suggestion of a German group to change how the uncertainty is dealt in the draft. In Germany, the measurement uncertainty values of each testing laboratory and each applied standard reference method are checked during the accreditation process by the Regional Environmental Ministries in 2 ways:
- Measurement uncertainty of the labs. Learn a bit about this here.
- Sampling uncertainty by the legal obligation to participate at stack emission proficiency tests at the emission simulation apparatus (ESA) in Kassel. Read a bit more about this here.
Important to note that the procedure of calculation of a measurement uncertainty in Germany differs from the method used in the new EN 13725. Also, by using paired measurements it is possible to calculate random error but there are more difficulties to calculate bias. That is, it is difficult to calculate if the expected value of the measurement results differs from the "true value".
To have an idea of the complexity of the calculation of bias it is important to note that the use of n-butanol as European Reference Odour Mass has some limitations. That is why a Secondary Reference Odour Mass has been defined in this standard. Also, there is a procedure to calculate a "floating bias", expressed as intermediate precision due to changes in panel composition, which is one of the most important sources of uncertainty in the overall uncertainty estimation.
| Measurement repeatability | Intermediate precision | Reproducibility |
|---|---|---|
|
|
|
In another order of things, there was also an interesting discussion on the correction of drift through time in olfactometers and on considering the correction of this drift in the dilutions setups in order to set a maximum allowed drift.
Also, as you probably know, there are many countries around like Germany, Netherlands or Chile that have two separate standards, one dealing with olfactometric analysis and a second standard usually dealing with sampling. In a previous meeting, splitting the existing document into two standards was suggested. Finally, this idea was taken under consideration in this meeting, and a final decision was made on not to split the standard and to leave it as it is now.
The list of changes made in the standard during these last meeting is quite extensive and it is difficult to comment about them all in just a couple of paragraphs. If you are attending the next NOSE, there will probably be a presentation by Mr. van Harreveld on the new features of the coming EN 13725. Also, we plan to write a bit more on uncertainty one of these days.
If you find this article interesting, you might also be interested in these articles:
