Fossil Fired Boiler Fly Ash Erosion vs. Flue Gas Velocity

[MechEng1995]

B&W's "Steam" offers information on the relationship between fly ash erosion of boiler tubes and flue gas velocity. They state than the metal loss on convection pass tubes is proportional to the total ash quantity passing through the boiler and is an exponential function of flue gas velocty. This is obviously intuitive, However, they offer no additional help on what the exponentional function is or how to determine it. Anyone have any ideas?

Thanks.

 

[metengr]

Most likely you would need to develop your own correlation with actual plant operating data. The information mentioned by B&W is for general application purposes and is based on several variables, one of which is the type of coal that is used in the boiler.

 

[davefitz]

Meteng is correct. It is approximately to the 3.5 power of gas velocity, and is also a function of the ash erosiveness and also of a configuration factor near the impact points. Hi silica ashes are problematic, but other ash constituents also need to be considered.

Some techniques to reduce the effect of ash erosion, besides lowering velocity, include anti-laning baffles ( near tube bends at HRA front and rear walls), using inline tube bundles, providing a furnace "nose" and other furnace geometry details to reuce the gas flow unbalance ( to be determined by modeling via numerical models).

To reduce gas velocity , operating at lowest practical excess air helps, and that implies mills are properly maintained for correct fineness and the burner to burner coal flow unbalance must be minmimized.

 

[davefitz]

For a recent research report by Mitsui Babcok on this subject, see:

http://www.dti.gov.uk/energy/coal/cfft/cct/pub/r241.pdf?nourl=

http://www.dti.gov.uk/publications/pdflink/&pubpdfdload=04/701

or google: erosion particle impact

 

[davefitz]

The above referenced paper provides useful correlations of erosion of mill components by coal and also tube erosion by fly ash.

The paper confirms the 3.5 exponent, if one recognizes the plots are nomralized as (mg metal loss / kg erodient hitting surface). The rate of erodent hitting the tube surface is proportional to velocity times (lb ash/ lb fluegas), so the 2.5 exponent would be increased to 3.5 if teh erosion rate is not normalized.

Ash particles that are smaller than 25 um do not contribute to erosion of tubes, since they follow the flue gas streamlines , and due to the "law of the wall" such small ash particle will not impact the wall surface except due to brownian motion.

Only ash particles that are harder than steel will cause erosion , and the paper demonstrtes that silica and pyrites that are larger than 25 um diameter are the only significant contributors to ash erosion.