Coal to biomass (wood pellets) mill conversion

To convert a pulverised fuel (suspended) boiler into one that burns biomass, one modification required, is on the coal mills/pulverisers. Existing attempts, such as using an existing coal classifier with blocked off areas for increased throughput, seems to not be the solution. With biomass, the flakes (broken pellets) will react differently to the air due to a flake having a very low density, therefore conventional classifiers will over classify the flakes causing a reduced throughput to the boiler. With dynamic classifiers, the rotating cage can easily over classify the particles reducing throughput. Opening the vanes of a conventional classifier, or reducing the rotating cage speed of a dynamic classifier, will have a negative impact on the biomass flake distribution to the burners.

Coal Milling Projects (CMP) assisted Drax Power Station (UK) to convert their conventional Babcock 10E vertical spindle mills into biomass equivalents. CMP managed to design a biomass classifier that avoided the problems as explained above. By means of a calibrated biomass flake, and with the help of  STAR-CCM+®, the classifier managed to reduce system resistances, increase the existing primary air (PA) fan capacity, give maximum throughput of biomass flakes, have no throughput of biomass pellets (i.e. maximum classification), have no biomass material rejecting into mill reject boxes, provide an even distribution to the boiler burners, and cost a fraction of any of the conventional technologies available.

The presentation will give an insight into the CFD methodology followed, using STAR-CCM+, to design a biomass classifier meeting all the requirements. The correlation of the site tests with the CFD results will be discussed. Furthermore, the use of STAR-CCM+ in the milling industry will be discussed and how it plays an integral part in discovering better designs faster. 

Corniel Zwaan
Session Time Slot(s): 
07/03/2017 - 2:25pm-07/03/2017 - 2:50pm
Room II
Presentation Image: 
Session Track: