We have a two-chamber ball mill, for one quality we put 60 per cent cold clinker and synthetic gypsum (total H₂O 25-30) in the mill and we add 40 per cent already milled blast furnace slag (4400 Blaine at 0.32O) in the separator or in the online mixer. We suffer from soft lumps in the finished product silo and of decreasing two days results after aging for four to six weeks in the silo (winter period). Where is the origin of the problem? Humidity of the gypsum (but we do not suffer it in the pure Portland quality), humidity of the slag? grinding? Any experience?

I do not have specific experience of your lump formation problem, however it must be connected with amount of heat generated in the mill during the grinding process and the amount of water that must be dried from the gypsum and slag. You need to develop a heat balance model of the mill that allows you to test different scenarios with different rates of addition of the materials and points of addition of the slag. This will allow you to identify the limits of the drying capabilities in the process. You may be able to get round the problem by a process adjustment or you may need to add heat to the system.

We have a two-chamber ball mill, for one quality we put 60 per cent cold clinker and synthetic gypsum (total H₂O 25-30) in the mill and we add 40 per cent already milled blast furnace slag (4400 Blaine at 0.32O) in the separator or in the online mixer. We suffer from soft lumps in the finished product silo and of decreasing two days results after aging for four to six weeks in the silo (winter period). Where is the origin of the problem? Humidity of the gypsum (but we do not suffer it in the pure Portland quality), humidity of the slag? grinding? Any experience?

I would have thought it unlikely that the slag is the cause of the problem. Do you adjust gypsum addition level to the clinker/slag combination to take account of the 40 per cent slag addition? As the synthetic gypsum contains a high free water content of up to 11 per cent (assuming all gypsum as .2H₂O) any reduction in the gypsum addition may help avoid lump formation. No details have been provided of the clinker chemistry but if there is scope to increase clinker free lime level this can often act as a desiccant to remove small amounts of excess moisture from cement stored in silos and avoid air setting/lump formation.