Subscription Required

There have been a number of articles written some while ago including by Lintec and Fobas which deal with the question of repeatability and reproducibility of results. We should recall that the purpose of the tests is now different in that based on analysis of the MARPOL sample, some one could conceivably go to jail. With just a single measurement those repeatability and reproducibility errors, for a fairly difficult measurement such as sulphru which lacks the precision of some other tests, can have important consequences. As a result and against a lot of opposition, the MEPC has ammended the procedure (not the test, the procedure) which is now defined in MEPC.1/Circ.614 but which doesn't come into force for some time yet and apparently there is some conflict with the EU over this. For fuel quality testing many properties are determined by a single measurement. For example, I have witnessed viscosity measurements in refineries where they simly run a single sample through a single capillary viscometer (ASTM D445) but in other refineries they will run the tes with four or more viscometers and discard the highest and lowest values and average the rest. Which is right? If it were just for operational reasons then a single test result might be sufficient but if were at risk of going to jail then I would want to see more rigorus testing performed that would help minimise at least the repeatability errors. The sad thing si that because no one goes to jail over a density reading the same approach to legal measurements has been applied to density as it has to sulphur. Does it matter? It might.

How can this be possible: Products that are bought and sold within the ISO 8217:1996/2005 specifications, which are worldwide used and accepted as an standard, with all applicable testmethods as described in the ISO specifications. IMO has verified that ISO 8754 is applicable for the test method of the sulphur content, but they do NOT accept an result that would be fully within those mentioned test methods (up to 1.58% m/m) This is not in conformity of the agreed ISO 8217 regulations and can to my knowledge only been changed if this would be amended in the ISO 8217 international standard. It is very strange that IMO's MEPC can just alterate or not use an international worldwide accepted standard and make their own interpretations about the fact that a result according them is out of the specifications, but according the ISO methods is within the requested specifications. I wonder what the outcome will be if a case like this will go to court !!!

I agree with you Martin and I find it strange and I have heard of cases where borderline figures of 1.57%S have been disputed. The maximum figure of 1.58%S was arrived at based on the ISO 4259 assertion that a fuel's true value of any parameter can never established exactly and hence the Repeatability And Reproducibility principle. The IMO would have to come to the party before some serious can of worms come out of the legal can!