Used measurement methods and practices

4.4.1

Manual sampling is still very common practice since more than half of the respondents answered using manual sampling in their primary quality measurement as Table 5 represents.

Further observation between different power categories revealed that the percentage of manual sampling systems was a little bit higher in the lowest power categories. However, the increase in automation cannot be clearly proved when power plant size is bigger. The approximately same amount of half automated systems was found in each power category, and the only fully automated sampling system was found from the smallest power category.

Table 5. Different sampling methods in each power category.

Sampling method

Number & percentage of power plants included in each category 101 - 200 MW 201 - 300 MW 301 - 400 MW over 400 MW Total

Manual 5 63 % 4 57 % 2 50 % 2 50 % 13 57 %

Half automated 2 25 % 3 43 % 2 50 % 2 50 % 9 39 %

Full automated 1 13 % 0 0 % 0 0 % 0 0 % 1 4 %

Total 8 100 % 7 100 % 4 100 % 4 100 % 23 100 %

The division of responsibility in the sampling process was one of the interesting research questions since bias issue is unavoidable in the manual sampling. The survey reveals that very often driver who represents supplier is the one who collects samples as Figure 25 represents the share among respondents who use manual sampling.

Figure 25. Responsible personnel for sampling when manual sampling was the used method.

Using the power plant’s own personnel to take samples is a way to reduce supplier affected bias. So, in theory, this practice can compensate the lack of automated sampling system. The prevalence of these methods was compared with intention to analyze how the bias issue is mostly managed at the power plants. Table 6 reveals the percentage of automated sampling systems and the percentages of power plants where own personnel either completely responsible for sampling or makes regular recheck. A notable finding is that automated sampling was more than double more common solution to reduce biased samples in comparison to using power plants’ own personnel to carry out sampling. Automation is, of course, an understandable direction of development because biased sampling could become a concern on the suppliers’ side if power plant operators carried out the whole sampling and measurement process unilaterally.

80 % 13 %

7 %

Driver /Supplier's representative Power plant personnel

Both take own samples or regular recheck

Table 6. Responsible personnel to carry out sampling for quality analysis.

Who collects the samples Number of respondents

Typically used quality analyses included measuring moisture content, heating value, ash content and chip size. Also candling for foreign objects was found in use at a couple of power plants. Elemental analysis was carried out at two power plants 3 – 4 times per year, and one of the respondents told monitoring foreign metal objects weekly through the evaluation of findings in magnet collection. Table 7 represents frequencies of the commonly used measures.

Table 7. Percentages of the power plants carrying out commonly used quality analyses in each frequency category.

Traditional gravimetric measurement was utilized at all surveyed power plants. However, two respondents reported using the gravimetric measurement only as a secondary method. In these cases, the primary measurement was either an X-ray or microwave-based system. X-ray and

microwave systems were also the only non-destructive measurement methods that were found among respondents. These methods were applied either in test usage or operational usage as Figure 26 represents. One of the power plants applied both non-destructive methods; X-ray in operative usage and microwave in test usage.

Figure 26. Non-destructive measurement methods utilized at the surveyed power plants.

4.4.3

The heating value was confirmed being the main basis for fuel pricing at all the surveyed power plants as it was expected based on the literature review. In addition to that, more than half of them told applying sanction pricing if the quality undercuts specific limits. Measured qualities that influence fuel pricing were the heating value (all 23 power plants), moisture content (21 power plants) and ash content (5 power plants).

Discussions also revealed that bad quality fuel is sometimes not accepted in any price even if it still has positive heating value. Negative effects of bad quality become too detrimental, for example, when there is no possibility to balance high moisture content fuel with a fuel that has better heating value. Burning very low-quality fuel is not only harmful to the boiler and other

4 %

parts of the power plant but might also increase flue gas emissions above the regulated limits.

Another reason for banning a specific fuel or supplier could be harmful foreign objects among the fuel. For example, big stones and metal objects might broke conveyers, causing huge costs and production losses.

4.4.4

According to literature, the quality of forest chips can be improved if the entrepreneurs are informed about the quality issues. Often quality of the biofuel depends on the subjective opinions and experience of the entrepreneur and the driver of the chippers and forestry machines. Precise training and feedback systems have given good results to improve practices and the quality of wood fuels. (Ikonen et al. 2013, 6)

Despite the benefits that a good feedback system could provide, only four power plants (17 % in the survey as Table 8 represents) reported the results of the fuel analysis straightaway after delivery. Largest number (61 %) of the surveyed power plants, however, provided load specific reports afterwards. This reporting does not give the supplier the possibility to make quick changes but helps at least to get understanding which loads were the worst and what kind of quality factors should be improved in the long term. A notable group of power plants (22%) did not provide any specific information about the fuel analysis. This means that supplier can make only a rough conclusion of the fuel quality based on the paid price and terms of the purchase contract. In this case, the supplier does not get any specific information which quality factors to improve and how notable the differences are between the delivered loads.

Table 8. Level of feedback provided to the suppliers.

Frequency Percent

straightaway after delivery 4 17 %

Total 23 100 %

Comprehensive and timely provided quality reports could enhance supplier’s possibilities to learn and provide better fuel in the future. If the driver of the machines gets notices about the quality changes during the same day when the chipping happened, there is a much greater opportunity to learn compared to the case if reports are delivered a week or even a month later.

For example, the machine drivers could quickly learn that which wood batches were not dry enough or how the rocks ended among the fuel when the drivers get reports while they can still remember what kind of feedstock was used and in which kind of conditions it was processed.

A brilliant idea could be, for example, a mobile application that would tell the chipping machine driver the quality of the delivered fuel loads instantly at the same moment when the loads are getting delivered to the power plant and analyzed there.