Running Centrifuges in Series

There are those who tout the practice of series centrifugation as a beneficial procedure. This is not true and is the result of misunderstanding the reason for centrifuging weighted fluids. As has been stated, the objective in centrifuging weighted drilling fluids is the removal of colloidal and near-colloidal solids, not the separation of drilled solids.

As the process is described, the first stage recovers the barite, while the second rejects the drilled solids and returns clean fluid to the system. Inasmuch as the centrifuge cannot separate drilled solids from barite, and the second stage cannot separate the finest drilled solids and barite particles from the fluid, the description is clearly inaccurate. At the first stage, the underflow, consisting of the larger solids particles (both barite and low-gravity solids), is returned to the system. The overflow, that is, the liquid and finer solids (both barite and low gravity), is then routed to another centrifuge that is operated at higher g force and makes a finer cut.
At this stage, the underflow is discarded and the overflow, containing the finest and most damaging solids, is returned to the circulating system. Assume for illustrative purposes that the D50 cut points on barite are 8 at the first stage and 4 at the second. The net effect is the removal of most of the barite between 4 and 8, and most of the low-gravity solids between 6 and 12. The barite that is removed is in a perfectly acceptable size range, and no benefit is derived from its removal. While the removal of low-gravity solids before they become small enough to become troublesome is beneficial, the benefit cannot be expected to justify the loss of desirable-sized barite.
There is an additional factor to consider in deciding whether or not to utilize this questionable practice. Unless the mud density is being reduced, the desirably sized barite that is disposed of in the first stage has to be replaced. As much as 30% of API-quality fresh barite can be particles smaller than 6 and much of this can be expected to be colloidal. Replacement of the discarded 4–8 material with fresh barite, which can include as much as 15–20% colloidal particles, increases the concentration of colloids, thereby exacerbating rheological problems and increasing the need for dilution of the active mud system.
Another consideration is that while low-gravity solids tend to be more troublesome than barite, solids problems in weighted fluids, which tend to contain much more barite than low-gravity solids, are frequently caused by excessive concentrations of colloidal and ultra-fine barite. This process returns these problem solids to the drilling fluid.
The validity of this analysis is not dependent on the assumed cut points. Whatever they may be, the result is the same. The solids that are removed are those between the cut points of the two stages, and the finest—most damaging—solids are returned to the mud system. Respected experts have been counseling against running centrifuges in series for decades. (See George Ormsby’s chapter, ‘‘Drilling Fluids Solids Removal,’’ in Preston Moore’s Drilling Practices, for an example.) They point out that the process does not, and cannot, work as described. Some go a step further and point out that the process reduces drilling-fluid quality and is harmful rather than beneficial. There are many situations in which the use of multiple centrifuges is clearly economically attractive; however, they are added for increased capacity and must be operated in parallel, not in series.