Waylay weight woes

Production managers, quality control managers and maintenance engineers alike have encountered weigh cells that never worked properly. Why? Why have so many switched from a supposedly high accuracy load cell solution to old fashioned level probes or sight glasses? This article touches on the worst culprits for problems and how to avoid them.

By Ryan Titmas, product manager, Sartorius North America

Share Print Related RSS
Page 2 of 3 1 | 2 | 3 View on one page

Figure 2 shows the best type of installation. Notice how the load cells are placed in the middle of the beams. Each cell is situated between two frames. This frame network ensures the load cells will move together, creating a more repeatable weighing environment.

Figure 2. Constructing a rigid frame is critical for repeatable weighing.

Figure 2. Constructing a rigid frame is critical for repeatable weighing.

In Figure 2 the load cells are mounted on top of a bottom frame assembly that provides a rigid foundation. The weight of the vessel is being transferred to the load cells and then to the bottom frame. This transfer of force occurs with minimal deflection. Although the load cells can be mounted directly to the floor without the bottom frame, the floor must allow little deflection and settling. The load cells are measuring deflection down to thousands of a millimeter, so if your foundation deflects, you aren’t going to achieve acceptable accuracy. Most architectural designs allow a floor deflection that would be unacceptable for weigh cell operation.

Figure 3 illustrates another typical design problem found in tank weighing. It shows load cells installed on the end of very long tank legs. In this example, errors are due to the legs moving in different directions. Measurements become unrepeatable as extraneous deflection causes variability in movement.

Figure 3. Shortening the span between the tank base and the weigh cells reduces the bending moment causing poor repeatability and inaccuracy.

Figure 3. Shortening the span between the tank base and the weigh cells reduces the bending moment causing poor repeatability and inaccuracy.

The second drawing in Figure 3 illustrates a good load cells’ installation; the legs from the tank are shorter, minimizing separate leg movement, and the load cell has a top and bottom frame to help induce a more uniform loading pattern. These figures show a common error; each weighing applications is different.

Rigid tank construction with a solid foundation, top/bottom frame, short legs — got it. What other tank design factors must be considered? One of the most overlooked factors is the vessel’s peripheral connections and accessories, such as pipes, mixers, ladders, walkways, etc. Each of these external fixtures influences the weighing system. This is another reason why design of a process and the vessel should be included when considering load cells.

Pipes entering and exiting a vessel are always exerting forces on it. Under stable, ambient conditions, e.g., constant temperature, pipes and nozzles don’t create serious problems for weigh cells. We all know that industrial applications are far from “ambient.” Thermal expansion and movement will adversely affect weighing results. To better visualize this effect simply stand on your bathroom scale and take your weight reading (don’t worry I don’t like what my scale reads either!) Next weigh yourself by gently pushing on the wall, and then once more by pulling on a nearby towel rack. The results are different aren’t they? This type of action/reaction force is similar on vessel weighing applications where pipes are present.

Figure 4 illustrates the possible forces and displacements of piping: a rigid pipe can lift some weight off the vessel and cause a displacement (dP). A force Fp along the pipe can push the vessel to the side and exert forces downward or vise versa.

Figure 4. The pipe force (Fp) added to the vessel appears as a “false” low weight at the weigh cells.

Figure 4. The pipe force (Fp) added to the vessel appears as a “false” low weight at the weigh cells.

By carefully considering pipe supports and nozzle design, it may be possible to reduce pipe problems. A simple way to dampen the effects of pipes on weighing measurements is to have all pipe connections enter from the side. If possible, place pipe hangers as far away from the tank as possible to allow for flexing as the tank moves up and down during loading. Rollers may help but they will eventually fail and cause problems. Some movement can always be expected. Flexible pipe connections and bellows may be the best solution. These connectors should lessen some of the influence of pipe forces on the vessel and moderate pipe fatigue. However, ensure the flexible connections are truly flexible. A 1-in. flex joint on a 4-in. pipe probably isn’t effective against either pipe fatigue or weigh cell interference.

Installation

Once you ensure that your tank is  properly designed and all pipe connections are flexible, the next step is to select the correct weighing system. Although shear beams and bending beams can be used, I suggest compression-type load cells. Unlike shear-beam or bending beam, compression load cells are designed to eliminate the introduction of non-vertical loads. This is accomplished by carefully engineering a curved top and bottom. These rounded components allow the cells to “right” themselves during side loading by a principle of restoring forces (Figure 5).

Figure 5. Heat shield protects a 20,000 kg compression load cell and mounting hardware.

Figure 5. Heat shield protects a 20,000 kg compression load cell and mounting hardware.

Compression cells, like those pictured in Figure 6, are designed to shift during vessel expansion and are constrained mechanically to eliminate the influences of other non-vertical forces.

Figure 6. Compression weigh cells are generally the choice for resisting influence of non-vertical forces.

Page 2 of 3 1 | 2 | 3 View on one page
Share Print Reprints Permissions

What are your comments?

You cannot post comments until you have logged in. Login Here.

Comments

No one has commented on this page yet.

RSS feed for comments on this page | RSS feed for all comments