Jun 10, 2022

# How to effectively establish static head while sizing centrifugal pumps in mining applications

• Article
• pumps
• centrifugal pumps
• piping

In the course of our design activities, we often come across pumping system calculations. One of the parameters that needs to be established for proper pump sizing is the system’s static head. This deliberation frequently arises when one specific type of piping configuration is considered. We often see it in in-plant pumping systems, specifically in slurry applications. This blog article will attempt to demystify the issue and provide some practical recommendations on how to properly establish static head value.

1. Inspired by a recent project performed by BBA experts, this second part is a follow-up to a blog article published in 2021.

2. ## The mandate

The mandate included pumping slurry to elevated process equipment with a top-feed arrangement. The piping layout needed to be Type 2 (Figure 1), i.e., a system with a piping section at an elevation higher than the discharge point.

3. The basic piping layout option that was chosen appears in the 3D model screen shot (Figure 2).

4. The slurry also needed to be distributed evenly along the elongated feed box to properly supply the process. Therefore, the feed pipe was split twice symmetrically to provide four equally spaced downcomers reporting to the feed box.

For similar slurry piping arrangements, a four to six diameter straight section is usually designed right after the first tee to facilitate material homogenization in the vertical section of the manifold. Tees are preferred, instead of an elbow, as they create additional turbulence, which facilitates material mixing. In this case, with an ND=250 mm pipe, a 4D = 1 m was chosen for the first vertical section. As such, the elevation of the highest point of the feed pipe, from the discharge end of the downcomers (not submerged, so that SH2 in Figure 1 would measure to the discharge pipe end elevation), would measure 3.0 m (DH in Figure 1). With respect to slurry SG=1.2, we can expect a vacuum of 3.6 m of water column (i.e., 0.36 atm vacuum) in this pipe, if an air/vacuum release (AVR) valve were not installed in the top section.

An alternative schematic was also considered, represented in Figure 3, by splitting the feed pipe upstream and below the equipment operating platform to feed the feed box from two sides, sharing the feed flow between two feed pipes.

5. This routing allows the use of smaller pipes (ND=200). In this case, each of them was split only once into two downcomers, while reporting slurry exactly at the same locations to the feed box, as it was in the basic concept. With this concept, said elevation difference (DH in Figure 1) will also be reduced down to 2.0 m, so the resulting vacuum will measure 2.4 m of water column or 0.24 atm vacuum, should it be decided to keep the top pipe without an AVR valve.

It is obvious that air will accumulate much faster in the basic concept since the vacuum there is higher. Naturally, the smaller the vacuum in the top pipe, the better off the system is with a higher chance of uninterrupted pumping.

If there were no AVR valves installed, the pump would need to be able to charge the said high pipe (a siphon) during startup. As discussed in Part 1 of this blog article, in this situation, it is necessary to look up the pump curve, which is naturally flat for a centrifugal slurry pump. This poses a risk for proper pump operation because the piping has a gooseneck.

The proposed pump has a curve, as submitted by the vendor and shown in Figure 4.

7. This clearly shows that the difference between the shutoff (SO) head and the total dynamic head (TDH) at the operating point stands at only 2.0 m of slurry column.

To charge the siphon, the pump must be pumping at least 15% or 20% of the design flow rate (refer to the discussion in Part 1 of this blog article), in which case the corresponding TDH will be at about 35 m, thus the said difference goes further down to 1.7 m, which is clearly below DH=3.0 m in the basic option and barely passes DH=2.0 m in the alternative layout.

This proves that if the pump were sized for the system static head, measured between the downcomer discharge elevation and slurry level in the pump box (SH2), the pump would not be able to start pumping, provided it had a fixed speed pump motor, or if it were not designed to override motor speed rated at 100% with a variable speed motor option.

To ensure stability and problem-free pumping operation, it is obvious that AVR valves must be installed on the top pipe. In which case, the system static head is always measured between the centreline of the feed pipe top section and the level of slurry in the pump box (SH1).

Once the AVR valve is included in the concept, if the top pipe was lowered by 1 m, as in the alternative piping layout, pump energy consumption would also be reduced and would be an additional benefit to this routing. A rough estimate of energy savings, in this case, was calculated as follows (with respect to 1 m of static head and some friction loss reductions).

8. ## Conclusions

This case study corroborates the following conclusions drawn in Part 1 of this blog article:

1. In the case of the Type 2 piping arrangement, it is always prudent to install an AVR valve on the top pipe.

2. The pumping system static head must be measured as the difference in elevations of the centreline of the highest section of the feed pipe and that of the slurry level in the pump box (SH1).

3. Lowering the top of the gooseneck will provide for some energy savings, resulting in potentially tangible reduction in operational costs.