99 posts
Blending Silo Modelling
I would like to know if there is any way to predict the time that a certain feed from VRM to blending silo will come out as a particular Kiln feed?
Know the answer to this question? Join the community and register for a free guest account to post a reply.
537 posts
Re: Blending Silo Modelling
norew:
I would like to know if there is any way to predict the time that a certain feed from VRM to blending silo will come out as a particular Kiln feed?
Norew,
Predicting the time that a change of raw meal chemistry will come out of a blending silo depends on many, sometimes inter-related factors such as volume of the silo, degree of filling in the silo, silo extraction pattern, duration and volume of each extraction, degree/efficiency of silo meal aeration, dimensions of the silo (including those of the blending cone), and physical characteristics of the raw meal... to mention only a few.
Many of these things are unable to be measured, therefore theoretically predicting such an event is practically impossible.
However, you can use a tracer element present in the raw meal to estimate the silo blending behavior empirically. This tracer needs to be something that generally stays very constant for at least twice the capacity of the silo, then step-changes in magnitude for an equal amount of time. The magnitude of the change in tracer concentration must be as far above the sensitivity of the analytical test for that element as possible.
The minor elements (MgO, K2O, Na2O or SO3) can be used for this purpose, if they change significantly and often enough and the analysis method is sensitive enough to detect clear step-changes in concentration.
By regularly analysing the chemistry of the feed into and out of the silo (eg hourly) you can measure the degree of blending at any time and the total time taken for the change to occur.
eg If you look at the attached picture at the top of my post you will see a graphical example of this technique. The white line is 2-hourly raw mill %MgO and the yellow line is the 2-hourly blending silo %MgO. The time taken for the change to occur completely (or in part) can be measured directly from this trend graph.
Flourescent dyes can also be used but the above method is simpler and easier, especially if you already have regular step changes in one or more trace elements and regularly test both raw meal and kiln feed chemistry.
Regards,
Ted.
99 posts
537 posts
Re: Blending Silo Modelling
Norew,
What do you mean by "manual"?
Regards,
Ted.