Servicio de Información Comunitario sobre Investigación y Desarrollo - CORDIS

Sampling and sample reduction

Methods for sampling and sample reduction that are of importance with regard to testing methods of fuel properties in order to ensure that the required fuel properties are met (e.g. methods for reducing samples for sample preparation for physical-mechanical and chemical tests). The potential impact of this issue on the biofuel markets is considerable.

The experimental work has allowed the relative bias of different sampling methods and the influence on sampling variability of different increment sizes to be assessed. This work has also allowed the number of sampling increments required to give a satisfactory level of sampling variability to be estimated, for the biofuels investigated (i.e. woody-fuels such as GROT, pellets and sawdust as well as straw bales). On the basis of these experiments it can be concluded that none of the methods used in the experiments gave disastrous results. Thus, no method can be ruled out from practical use and should be included in the CEN standard.

For sampling in sum it is revealed: with exception of particle size distribution, no evidence of a relative bias between the methods of sampling and testing can be revealed for testing moisture, ash and chlorine content of the fuels investigated.

Sampling from tipped lorry-loads is not biased relative to sampling from a conveyor for the methods recommended for GROT and sawdust.

Based on the work on GROT it can be concluded that for moisture and ash, the results on the relative bias do not provide any reason for preferring sampling from the conveyor over sampling from the heap. Different from that, particle size distribution shows relative bias implying the necessity for the CEN standard to define one preferred method.

For moisture and ash of sawdust analysis the results concerning the relative bias of the two methods show the same as for GROT. For particle size distribution, the results imply a relative bias between the two methods in case of smaller increment sizes (0.2 litres or 1 litre). Thus, if sampling from the stopped conveyor will be the reference method, then sampling from the heap is acceptable provided that sufficiently large increments are taken.

The results obtained for pellets in the investigation of relative bias indicate that the moisture and ash content vary from lorry-load to lorry-load. However, both are not affected by the handling, what is true for the particle size distribution.

For straw bales, the experiment results have shown that sampling with the hook is not biased relative to sampling with the coring machine. Taking five increments per bale will results a relative sampling error of about 10%. Furthermore, doing one determination per sample will give a relative error of test results (i.e. repeatability) of about 5%. Both relative errors might be acceptable for routine tests. Due to the effect of straw bales position, when using either the hook or the coring machine, for sampling straw bales should be turned on their side. In addition, increments should be taking from both sides, and from the full depth of the bale.

Preferred methods for sample reduction (i.e. those gave least variations between sub-samples) in each of the experiments are outlined. Some general conclusions may be drawn tentatively from this. "Riffle" is the preferred method to use with the coarser materials (GROT and pellets), if a rotary divider is not available. If available, a rotary divider is the preferred method for determinations of moisture and particle size distribution on pellets. An explanation should be sought for the poor performance of the rotary divider in comparison with the riffle when determining ash. ¿Coning¿ is the preferred method with sawdust, but the riffle performed nearly as well. The special method used to sub-sample straw, handful sampling (on straw coarse cut prior to reduction), is the preferred method for determination of liberated or partially liberated properties of straw, but not for determination of moisture.

In general is was revealed, when a sample is taken for the purpose of determination of moisture-content, sample reduction should be avoided, since the materials dried noticeably during the reduction process and the result of the test will be affected.

However, it is important that technicians should regularly and routinely check the achieved repeatability with whatever sample reduction methods applied for.

Finally, it is suggested that there is further need to:

- Extend the work on sampling other solid biofuels in order to cover the breadth of materials to be found throughout Europe (e.g. wood chips, bark, reed canary grass, olive waste and briquettes),

- Develop a guideline on how users of CEN standards can ensure reliable results of sampling and sample reduction procedures as well as can decide the frequency of sampling.

Información relacionada

Reported by

Porthiddy Farm West, Berea
SA62 6DR Haverford-west, Pembroke-shire
United Kingdom
Síganos en: RSS Facebook Twitter YouTube Gestionado por la Oficina de Publicaciones de la UE Arriba