'Why' & 'what factors are responsible' are just about repetition.

Now, it would be handy to learn about the nature of the fumes, not gases, condition & location of any separators, cross section detail of the flues; are these balloon flues or cylindrical (12ft)?

If saltation velocity is relevant then either your pipes are too big or you have some mighty big & ineffective fans.

On an old plant there is nothing unusual about deposition. In the old days we used balloon flues with cleaning access; or if you were really rich a screw conveyor was located at the root of the balloon flue.

Steelworks are hot, dirty, smelly places & full of accretions of one sort or another. Your problem is housekeeping, not wondering about the saltation velocities of particles about which there is no given information.

If you have a separator upstream; overhaul it.etc etc. Go through the upstream equipment first before you start questioning the suitability of a simple pipe which, in your own description has served a long & useful life.

Another point, if the particles are abrasive then the deposit wasn't such a bad thing after all; was it? Maybe that's why the pipe was so old. You only found the deposits when you opened up the line; but if they were 'severe' then the flow conditions would have altered accordingly before the refit. Did they, or don't you know?

Do some investigating & come back with history & numbers. ■

To continue:

I bet that if you can find the structural calculations for the pipe rack you will see that there is load condition for dust loads inside & outside the pipe. There'd better be! ■