New machine grading standard…

The results of the (second) CEN Enquiry for prEN 14081-2 (the standard that governs machine grading settings) are out.  Approval was by 85% of the required 55% of National Members approving – but just 66.7% of the required 65% by weighted population (see previous post for more on CEN voting maths…and Brexit).  This means we will soon be getting some quite big changes to how machine grading settings are calculated – but not changes as huge as in previous drafts of this standard).

Germany, Sweden and Italy voted no – three countries with a big interest (and knowledge) of timber grading – so what are the big issues?

Let’s first consider why the standard needed updating.  It was hard to follow, did not make use of the potential of modern grading machines, and needed updating to be compatible with recent changes in other standards (especially EN14081-1 and EN384).  More importantly – it was not complete.  You could not use it, alone, to make grading settings – CEN TC124 WG2 TG1 had a big list of additional checks and clarifications that had been developed as result of experience and knowledge gathered in grading settings projects for more species, bigger datasets and wider growth areas.

What are the changes?

Requirements for strength grading machines haven’t really changed, but the way settings are calculated and verified has.  The procedures are less prescriptive, and they are (hopefully) simpler to follow.

For development of settings for machine controlled systems – the overall checks are more or less as they are currently (according to the current standard and TG1 decision list). The cost matrix is retained in the form of (another) machine verification check – but doesn’t ask for anything that didn’t need to be done before (*incorrect, see below).  Indeed it is less important in that it doesn’t need to be passed for all grade combinations.

Modifications are made to what can be a settings area.  We were already bending the rules before (necessary for large countries like Russia) and these changes mostly make sense.  The country check procedure in the TG1 decision list has been added (with slight modification of stiffness requirement) – it’s not a perfect solution, but it is certainly better than the current standard.

New fixed settings are added for longitudinal resonance machines.

A much modified version of output control is included.  This was in response to evidence that the current system is not able to adapt quickly enough to shifts in resource quality – but retaining it as an option because it is used by a small number of sawmills in Europe.

And a new process called “adaptive settings” is included that using IP records to adapt settings. The idea makes a huge amount of sense as many grading machines can measure properties with good accuracy and hold long records of data as they grade.

So what are the main comments received?

Many comments relate to editorial matters – simple formatting, spelling and English language issues … but also more substantial comments on the understandability of the standard.  It is not just grading specialists that need to be able to follow this standard – and of course if it is hard to understand in English it will also be hard to translate (correctly) into the other CEN languages.  There are also a few causes where the clauses actually say the opposite of what they are supposed to mean.

It is asked why the standard needed to make changes to procedures that are already in use – that there is no justification given for why changes are needed – and that there has been no comprehensive study on the impacts of the changes (and the associated costs to industry).

There is objection to the possibility that some existing settings would need to be verified or recalculated (see below).  There are a few settings in long-standing use that are based on sampling that would not be allowed under the current system (current standard and country check from the TG1 decision list).

There are also comments on the specific requirements of some clauses and procedures – including ones that seem to have loopholes allowing checks to be passed by making clever choices – rather than by fulfilling the intended requirement.  It would be hard to write a standard that cannot be cheated by someone with a mind to do so – but it is more important that the reason for requirements are clear (since we hope that anyone doing these calculations does what to ensure the graded timber are properly safe)

There are also questions about whether the procedure for adaptive settings works – and is suitable for implementation.

The comments will be reviewed, and responded to, at the next meeting of CEN TC124 WG, which will take place 17-18 October.

 

The text concerning existing settings:

“Since the previous version of this European Standard (EN 14081-2:2012), grading settings work, and research data, have provided more information about the variation in wood properties. Several new rules were created by CEN/TC124/WG2 to update the procedures and ensure safety of grading – particularly of settings covering many countries, and are referenced in the guidance paper (see Annex A). This new version of the standard updates the procedures according to the guidance paper.

Machine control setting that have previously been approved will need to be reviewed to ensure that the basis of the settings complies with the requirements of this European Standard where the settings area covers more than one country or standardized area. The requirements on previously approved machine control settings are:
– if the settings area covers a single country or standardized area, the settings continue to be valid (no change);
– if the settings area covers more than one country or standardized area, and item 15 in the decision list of CEN/TC124/WG2/TG1 (the ‘country check’) was applied and passed, the settings continue to be valid (no change). This is the case for all settings approved October 2012 and later;
– if the settings area covers more than one country or standardized area, and item 15 in the decision list of CEN/TC124/WG2/TG1 (the ‘country check’) was not applied and/or not passed, the settings will continue to be valid for a period of three years following the date of publication of this European Standard. To remain valid after this period it will be required that the combined country check in this European Standard is shown to be passed. It may be required to remove countries from the settings area to comply with the requirements (if there was no sampling). If the combined country check is failed, settings should be recalculated according to this European Standard.”

* above it is said that the cost matrix was the same procedure as the current one. This is not right as the prEN applied the 0.2 limit to both the “incorrectly upgraded” timber and the “incorrectly downgraded” timber (instead of just the “incorrectly upgraded” pieces as now).  This actually makes the check impossible to pass!