https://www.bulksolids.nl/phenomena_in_silos.html

BSE bulk solids engineering Tel.: +31 53 434 45 66

Print: 11 Oct 2024 23:01
logo BSE bulk solids engineering

Phenomena in silos

Pressure build-up, product breakage, silo quaking, segregation

This paper discusses phenomena that can cause problems when storing bulk materials in silos. Including solution directions and some practical cases.
See also the paper on bulk technology with the basics of silo design.

C O N T E N T S
 

Pressures in silos

Product damage due to normal, shearing and impact forces

Increasing capacities or shorter lead times sometimes require larger raw material inventories. For companies that store their raw materials in silos, this means investing in higher or wider silos. In some cases, however, this leads to problems. This is because in a larger silo, the bulk product experiences greater pressure. This changes its properties, and usually not for the better.
The stored product may break (degradation) or clump together ( consolidation), hampering the outflow from the silo or causing problems for the further progress of the process. High pressure can further give rise to plant shocks, malfunctioning vibrating floors (i.e. poor dosing) and even collapsing hoppers. To lower the pressure in the silo, the main focus should be on the wall of the silo.

Product damage

In a production process where particles are made, e.g. animal feed pellets, it is unavoidable that some of these products break already during the process. Besides leading to production losses and product depreciation, this can also adversely affect the processes themselves.
This product damage can occur at several places during the production process or transport. In this story , we look at causes when storing in silos, and ways to limit the damage.

Product breakage

During storage, the product in a silo undergoes normal and shear forces, and possibly impact forces when the silo is filled. The normal and impact forces can lead to breakage or fragmentation of the particles. The shear forces cause surface damage by forming cracks or breaking out small pieces.
Consequences may include:

  • depreciation of the product;
  • generation of dust, leading to contamination and risk of dust explosions;
  • change in flow properties, and therefore possible flow problems;
  • increased risk of segregation ;
  • problems in subsequent processes due to changes in density, permeability, etc.

Damage that can occur due to impact during silo filling is not considered further here. Limiting the drop height or fitting suitable chutes reduces those problems.

Calculation of silo pressure

Cover sheet DIN-EN 1991-4, Silo Loads
EN 1991-4, to calculate silo loads

We describe the pressure occurring in silos using the Janssen formulas. Although this method will not accurately reflect the actual occurring pressure, the outcome does give a good indication of the pressure curve and the factors that determine pressure. For comparing pressure levels, this is not an a problem because the pressures are always calculated with the same relative inaccuracy.
The standard Janssen formula for vertical pressure (sigmaZ) in the cylinder is used in common standards (e.g. EN 1991-4, the replacement for the long-used DIN 1055, part 6). This is used to calculate pressure at a given depth (z), i.e. for a given filling level.

The equation is a function of
  • the ratio area to circumference of the cylinder (the vertical part);
  • the wall friction coefficient (mu_W = tan(phiW));
  • the ratio between vertical and horizontal silo pressure, indicated by a factor (lambda or K)
  • the bulk density of the product (gamma = rho*g)

The horizontal pressure (sigmaH) can be calculated by multiplying the vertical pressure by the pressure ratio (lambda or K). The horizontal pressure in the cylinder is the same as the wall pressure, the pressure perpendicular to the wall (sigmaH = sigmaW = K * sigmaZ). The shear stress at the wall is the wall pressure multiplied by the wall friction coefficient (tau_W = sigmaW * mu_W).

Influences on silo pressure

The pressure gradient therefore depends on the wall friction, but also on the ratio of surface area to circumference of the silo. One can deduce that high pressure can occur in relatively high or wide silos and in silos with low wall friction. The latter may be caused by the deposition of grease on wall, which is quite common in silos containing animal feed, for example soybean meal. The standard menttions this phenomenon.
Low wall friction also occurs when the bulk material has polished the silo wall. This requires mass flow in the silo. Furthermore, the bulk material must be relatively hard, but a soft product may also contain a hard contaminant (e.g. sand). The combination can eventually cause mirror-smooth walls.
Another cause of undesirably high pressures in the bulk material is that the silo is used for other raw materials than the silo was designed for. Changing the properties of a raw material (e.g. bulk density) can also be a cause.

Silo pressure reduction options

Of the factors that determine pressure according to the Janssen formulas, bulk density and pressure ratio cannot be changed much, if at all. Reducing the filling height is trivial; we leave these out of consideration. Remaining are the wall friction and the area - circumference ratio.

In cases where high pressures have undesirable consequences, one will strive to reduce pressure. Various methods are used in practice, almost all of which affect the wall (friction).
Incidentally, it should be remembered that these methods may be a solution for a particular case but are unlikely to work when a different product is stored in the same silo. After all, wall friction is product-specific.

Options for reducing pressure in silos

Some options for pressure relief are:
  1. Placing rings,
  2. Applying a rim of mesh,
  3. Installing one or more dividing walls,
  4. Increasing the wall friction; i.e. making the wall rougher (profiling or coating),
  5. Installing inserts in the silo,
  6. Applying intermediate cones.

1) Placing rings

Placing rings in the vertical part of the silo is structurally simple. However, correct sizing is difficult. A ring that is too large can give rise (locally) to core flow, as product remains on the rings. With perishable product, this is a problem.

2) Applying rims of mesh

A related method is to insert strips of mesh into the cylinder. These strips are usually attached at the bottom edge. The same disadvantages apply here as with rings. An additional disadvantage is the construction itself. Due to high loads, the mesh itself may crack, or be pulled away from the wall altogether.

3) Installing dividing walls

Janssen's formula is a function of the ratio of surface area to circumference of the cylinder. Lowering this ratio will, like increasing wall friction, lead to a reduction in silo pressure. In fact, the ratio of surface area to circumference in a silo is nothing more than the storage volume of the cylinder relative to the amount of silo wall with which the bulk material comes into contact.
For a horizontal cross-section of a circular cylinder, it is easy to deduce that the ratio of surface area to circumference is equal to the diameter divided by four. This is also true for square silos. Reducing the silo diameter D thus leads to lower silo pressure at the same filling height, unlike for liquids.
So for a newly built silo, the obvious choice when silo pressure is a problem is to opt for several smaller cells.

silo pressures at various cell dimensions
Silo pressure with and without partition(s)

For existing silos, one option is to increase the wall surface by installing one or more intermediate walls. This reduces the diameter and increases the wall area. The effect of this measure is easy to calculate. Installing intermediate walls has the added advantage of allowing different types of products to be stored. Disadvantages lie mainly on the structural level. The wall must be solid and the silo often has to be reinforced at the attachment points. This leads to expensive constructions. A disadvantage with regard to flow is that the opening becomes eccentric. See the situation with feed pellets for an elaboration of silo partitions in practise.

4) Increasing the wall friction

Increasing the wall friction coefficient leads to a substantial and almost linear reduction in silo pressure at the bottom of the cylinder. At the top of the silo, the influence is smaller because the pressure is built up . An example of this pressure reduction is given in the figure, where the pressure profile is shown for a fill level (Z), a silo diameter of 3 m, a bulk density of 1000 kg/m3 and a pressure ratio of K=0.4.

Silo pressure at different wall friction
Silo pressure profiles for different values of the wall friction
It is assumed that the wall friction coefficient is increased from 0.3 to 0.6. The figure clearly shows that the value of the vertical pressure decreases significantly especially deeper in the silo.

It is assumed that the wall friction coefficient is increased from 0.3 to 0.6. The figure clearly shows that the value of the vertical pressure decreases significantly especially deeper in the silo. In the given example, it means that with the rougher wall, the same pressure is reached only at about 10 metres filling height as with the smooth wall after about four metres. The value of the wall pressure also decreases in the same proportion.
The value of the shear stress at the wall becomes substantially higher at the top of the silo due to the increase in wall friction. Important here is the summed shear stress along the wall: this is the load carried by the wall. For cylinders, this is the load that can cause buckling.
It is logical that this load increases: after all, the decrease in vertical pressure on the product is caused by the column being supported more by the wall.

Unfortunately, the possibility of reducing silo pressure by increasing wall friction is rather limited in practice. This often requires some kind of coating or cladding that provides the required friction increase, is suitable for the product concerned and is also sufficiently wear-resistant.
In addition, the rougher wall itself may give rise to product wear, although the lower wall pressures are again beneficial in this regard. Furthermore, a less smooth wall is more likely to lead to fouling of the silo. Finally, the increase in wall friction is limited by the value of the product's internal friction. If the walls are too rough, the flow will no longer flow flow along the wall but along a boundary layer of the product itself, where the wall friction coefficient is equal to the tangent of the internal friction (mu_W = tan(phiI)).

Hang-up in silo with tie bars
Hang-up in silo with tie bars

5) Installing inserts

Hang-up in silo with tie bars Inserts are flow obstructions. They can be formed by inclined plates, cones, funnels, chains, horizontal bars, etc. From practical experience, they are known to cause large pressure drop. The main advantage of inserts is that they can often be relatively small. However, here also lies the biggest disadvantage: it cannot be predicted how big the pressure drop and thus the load on the insert will be. The influence on the flow pattern is also poorly predictable. Flow problems may occur, or the product may get stuck at altitude (hang-ups). Restarting stalled flow can then be very difficult, the very reason being the built-in structures. Installing one or more so-called intermediate cones in the cylinder is then a better alternative.

silo pressure decrease by intermediate cone
Decrease of vertical silo pressure due to an intermediate cone

6) Using intermediate cones

With intermediate cones, the same effect occurs as with the actual cone or hopper of the silo. The pressure in the bulk material decreases, while the load on the wall increases. The pressure drop and the structural load are easily predictable. See the figure, which shows the pressure profile in a cylinder with an intermediate cone. As the pressure drop is greatest at the top of the hopper, intermediate cones of limited height already result in a significant pressure drop. In practice, an intermediate cone protruding only 3% of the diameter of the cylinder from the wall can already reduce the pressure by about half.
An advantage of using intermediate cones is also that mass flow is guaranteed. In the case of mass flow, there is normally no mutual movement between the particles when lowering into the cylinder. This can lead to settling and clumping. When using intermediate cones, the product mass at the location of the intermediate cone is set in motion again. This negates consolidation effects.

See the following sections for some real-life case studies on pressure reduction.

Partitions for pressure reduction in silo

Limiting pressure in a storage silo due to breakage of feed grains

In the contribution above, we discussed some theoretical possibilities to reduce pressures in storage silos. This is to reduce damage to the stored product, e.g. grain breakage, or powder sticking together. This article discusses two practical examples.

Livestock feed pellets in large silos

At a feed mill, animal feed pellets are produced, which are then dried and stored before being transported to customers in bulk or bagged. In the past, storage took place in relatively small storage silos. Due to production expansion, larger silos are now used. This had the disadvantage of significantly more damage to the pellets. Since the type of flow in the large and small silos was the same, mass flow in both cases, it was concluded that the cause of the increased damage was due to the higher pressure in the silos. The question then naturally arises: can anything be done about it?

Starting points and possible solutions

Starting points and possible solutions The small silo cells used previously were rectangular, with a diameter of 1.5 x 1 m, and a height of 8 m. The outlet hopper was sufficiently steep for mass flow. The pellets had with the following properties: wall friction relative to the cylinder wall (mu_w) = 0.48; bulk weight (gamma) = 8 kN/m3.
Based on this, with a chosen value for the stress ratio (lambda or K) of 0.4, the standard Janssen formula can be used to calculate the maximum occurring vertical pressure at fully filled silo can be calculated: sigma_z = 12.4 kN/m2. This maximum pressure did not lead to serious product damage in the past.

veevoederpellets
Veevoederpellets; gepelleteerde/geëxtrudeerde deeltjes

The new silo cells were also rectangular, but now with a diameter of 2.0 x 2.8 m, and a height of 10 m. Mass flow also occurred in these silos. The vertical pressure occurring in these cells when the silo was fully filled was: sigma_z = 23.4 kN/m2, almost twice as high as before. As too many pellets broke, the pressure therefore had to be reduced. The high pressure turned out to be mainly caused by the larger cross-section. Indeed, at a filling height of 2.3 m, the pressure was already higher than at a filling height of 8 m in the old silo. Filling the silo less high with that was not an option, and undesirable anyway because of the large unused volume.

Installation of partition walls

Initially, it was investigated whether pressure could be sufficiently reduced by incorporating intermediate walls, and how high these walls should be. The first situation investigated was the incorporation of a single wall, dividing the cell into two equal parts (see figure, option B). This already provides a considerable reduction of pressures in the silo, because the magnitude of pressure corresponds to the ratio of surface area divided by circumference. Nevertheless, for the current cell, the pressure reduction did not appear to be sufficient.

silo pressure using partitions
Silo pressure using partitioning walls

Therefore, the incorporation of two mutually perpendicular intermediate walls was investigated. In the event that these walls were made as high as the silo, the pressure would be well below the permissible pressure (see figure, pressure curve A). This means that the partitions need not be as high as the silo. A modified version of the Janssen formula can be used to calculate how high the wall would have to be to arrive at exactly the desired pressure at the bottom of the silo. In this particular case, partitions with a height of 8 m proved sufficient.
NB: This means that even above the partition wall the pressure is already close to maximum (which occurs at 2.3 m filling of the silo without partitions). From this, we can deduce that in the cell covering a quarter of the silo, the pressure is still barely increasing: the end value has been reached. In practice, this means that the cell of 1.0 x 1.4 m may be "infinitely" high: the pressure will not increase further.

Considerations for partition walls

For the application of partitions to reduce silo pressures, the following comments are important:

  • Further raising the partitions above the point where a sufficient reduction is achieved is usually not useful. Further pressure reduction is not achieved.
  • Installation of intermediate walls has major structural consequences. A larger part of the total silo load is absorbed higher in the silo walls. The structure must be checked for this and modified if necessary.
  • Because the intermediate walls are loaded on both sides, it seems plausible that these walls need to absorb less horizontally. Because of the possibility of non-symmetrical flows: one cell may empty while the other(s) are stationary, the building standards recommend calculating the partitions as normal silo walls.
  • Sometimes flow lags in one or more cells, or stops altogether. As long as product continues to flow out, this is not noticeable. However, time consolidation can then cause these cells to become completely blocked due to bridging.
  • The vertical pressures considered here are usually not the highest pressures in a silo. At the site of the transition to the hopper, peak pressures can occur that are considerably higher. However, these peak pressures occur very locally and, moreover, these peak pressures are also reduced to the same extent as the vertical pressures

Installation of intermediate cones

The silo pressures keep increasing in the vertical section from top to bottom. In the hopper, the pressures decrease again; the hopper walls carry very much of the total weight. Especially the upper part of the hopper takes a lot of pressure. This effect can be used by incorporating one or more limited-height hoppers at different heights in the cylinder.

silo pressure with intermediate cones
Silo pressure profile with and without intermediate cones
The incorporation of these so-called intermediate cones was also investigated for the feed pellet problem. It was calculated that three intermediate cones only 30 cm high were needed to keep the pressure below the permissible level, see the figure. This large pressure drop through an intermediate cone means that the structure is heavily loaded at that spot. This vertical load must be absorbed by the wall or by reinforcements (vertical ribs) to be absorbed.
For the present situation, this was not attractive, and so the solution with partitions was chosen.

Practical verification

To check whether the chosen method of reducing pressure would actually lead to less product damage in practice, several tests were carried out. In one of the new cells, two mutually perpendicular partitions with a height of 7.5 m were fitted. This modified cell and the adjacent unmodified cell were then alternately filled with layers of pellets. Such that a similar product was stored in both cells. During filling, a number of samples were taken from the filling stream. After filling, both silo cells were emptied and samples were taken from the product flow at set intervals.

From all these samples, the average length of the pellets was determined by counting and measuring as a comparison measure for the occurrence of the degree of product damage. The average length of the material before the silo was 16.8 mm. After storage in the silo cell without partitions, this length had decreased to about 12.5 mm. At the cell with partitions, the length had reduced to about 14.4 mm. Thus, although, as also expected, damage to the product still occurs, a significant improvement did occur by incorporating the partitions.
During the tests, both used cells were immediately emptied completely so that part of the storage product was not under the high pressures. In practice, the cells are often not always totally emptied but regularly refilled. As a result, more material does come under the higher pressures. In that case, the improvement due to the partitions will be even greater.

Pressure reduction using intermediate cones

Avoiding shock in silos containing milk powder

Problematic loading

At a company where milk powder is loaded in bulk trucks, they had extended the finished product silos to increase their capacity. However, the problem after this modification was that the silos now shocked when emptied to such an extent that weighing was disrupted. Indeed, sometimes such large shocks occurred that the silo appeared to be tons lighter. The loading was then stopped by the software, as if the target weight had been reached. Restarting took a lot of time. To get the desired quantity into the bulk truck, loading had to be done "very carefully", which also took a lot of extra time.
Shocks were most severe when the silos were full, but not only because of the greater mass in the silo: apparently, shock behaviour was also negatively influenced by the degree of filling. Interestingly, the occurrence of shocks was related to the history of a batch. For instance, in these round silos, shocks occurred above an 18-tonne fill when the bunker had been full (50 tonnes). If the bunker was filled with no more than 30 tonnes, no shocks occurred at all.
There were two types of silos, with smaller and larger diameters. With the same filling in mass, the smaller cells shocked less.

Shocking, quaking silos

A literature review revealed that silo quaking can have several causes. One well-known one is the occurrence of slip/stick (see box). Temporary bridges can also occur as a result of varying wall friction, which can cause severe shocks when collapsing. Another cause of shocks is the reinforcement or venting behaviour of the product combined with silo geometry. Also the change from mass to core flow in the silo, irregular outflow due to, for example, an asymmetric opening or a malfunctioning discharge mechanism can create shocks. See causes of silo quaking.
In all cases, higher pressure in the silo gives rise to more and bigger shocks. So even if one does not know (exactly) the cause of the quaking: sufficient pressure reduction will bring the solution.

Stick-slip behaviour

Milk powders and whey powders are known to exhibit stick-slip behaviour. One speaks of stick-slip when a product has markedly different static and dynamic wall friction. When measuring the wall friction of milk powder, this is clearly evident. When the sample is stationary, the wall friction is high (stick). Once it starts moving, the friction becomes lower so that the sample shifts quickly (slip). The product then comes to a standstill again and the effect repeats itself.
Even without measurement, it is already clear from most products that they will exhibit stick-slip behaviour. It can be felt when the product is in a plastic bag, for example. If it is squeezed, one feels that deformation is jerky. This is often accompanied by noise: the product 'squeaks' while deforming. See registration stick-slip.

Installation of intermediate cones

In a silo, the pressures in the vertical part increase from top to bottom. In the hopper, the pressures decrease again; the hopper walls now carry much of the total weight. It can be experienced in practice: when the outlet opening is very small, the pressures are also very low: the product can be stopped by hand, even when there is a 10-metre 'column' of product in the silo.

Round silo with intermediate cones
Schematic representation of a circular silo and the pressure profile with intermediate cones

The pressure drop is greatest at the top of the hopper. The solution with intermediate cones takes advantage of this. See figure above. With an intermediate cone only 25 cm high, the vertical pressure can already be reduced by a factor of 2.

Design of intermediate cones

The proper functioning of intermediate cones depends on the design. Here, the most important axes are determining the allowable pressure, and measuring the wall friction and internal friction.
For a given silo geometry, the wall friction is the most important factor for the pressure occurring in the silo (broken line in the figure above). Furthermore, the maximum allowable pressure must of course be determined. In existing situations, this is relatively easy, by establishing the filling level above which problems occur. In other cases, measurements of the product properties (e.g. reinforcement) combined with the operating conditions can indicate what the maximum allowable pressure is. Furthermore, the pressure drop in a cone or intermediate cone depends on the internal friction of the product. This can be determined with the Jenike shear cell. If there is reason to do so, flow and bridging behaviour of the product can be studied.
Using arc theory (Jenike, Enstad, Benink), the pressure drop when applying intermediate cones can be calculated based on the product properties. With this, the dimensions of the intermediate cones can be determined exactly. The more intermediate cones are used, the smaller the total summed height and thus the material consumption becomes. However, the assembly costs increase. The structural engineer can then choose the optimum number of intermediate cones for a specific situation.

Intermediate cones in practice

silo with intermediate cones (seen towards top)
Intermediate cones in practice: very small relative to silo diameter

In the situation with milk powder, one intermediate cone could reduce the pressure sufficiently, but then it had to be quite high, compared to application of two or more cones. Moreover, the fewer intermediate cones (for the same maximum pressure level), the higher the pressure drop and thus the local load.
However, more intermediate cones means more work and is thus less cost-effective. For these reasons, three intermediate cones were chosen: this did not require additional reinforcement of the cylinder wall. The photo (taken through the outlet opening towards the top of the silo, when the silo is horizontal) shows that the three intermediate cones are relatively small.

Predictable pressure reduction

Avoiding high product pressure is especially important for products that can cause silo quaking. A well-known indicator of this is stick-slip behaviour, which occurs, for example, with milk powders and whey powders. Pressure should also be limited for fragile pellets, such as pressed or extruded pellets, to prevent breakage. With sticky and choppy products, higher pressure is detrimental because problems worsen, often more than proportionally. Even with large mass-flow silos or silos with asymmetric flow, one must be very mindful of the adverse effects of high pressure.
Application of intermediate cones for pressure reduction has several important advantages. The biggest advantage is the predictability of pressure reduction. This allows the number and size of the intermediate cones to be determined exactly. If large silos are built for fragile products, for example, it is advisable to include intermediate cones during construction. The costs are then considerably lower than if they have to be fitted into an existing silo.

Quaking, vibrating and honking silos

Causes and mechanisms

From humming to honking and quaking

Silos for storing powders or granules can generally be designed to do their job properly. But sometimes curious and often troublesome things occur. Some silos, in combination with certain types of bulk material, are found to produce a somewhat humming sound ('the silo sings'), or worse, a periodic occurring sound as if a truck were honking (so-called silo honking).
Another phenomenon is a continuous vibration of the silo (silo vibrations) in which other parts of the structure may also vibrate. The worst phenomenon involves the occurrence of sometimes very heavy shocks, silo quaking. These are then felt throughout the silo building and even in its surroundings and can cause damage to the structure.

Case: Quaking silo containing milk powder

Silo quaking can have several causes. One well-known one is the occurrence of slip/stick. As a result of varying wall friction, temporary bridges can form, which can cause severe shocks when collapsing. Another cause for shocks is the reinforcement or venting behaviour of the product combined with the silo geometry. Also the change from mass to core flow in the silo, irregular outflow due to, for example, an asymmetric opening or a malfunctioning discharge mechanism can create shocks.
In all cases, higher pressure in the silo gives rise to more and larger shocks. In most cases, the shocks decrease when the level in the silo drops. There is then a level at which shocks no longer occur. Interestingly, the occurrence of shocks may be related to the history of a batch.
For example, in a round silo of milk powder (diameter 3.7 m, height 11 m), shocks occurred with a filling of only 18 tonnes when the bunker had been full (50 tonnes). If the bunker was filled with no more than 30 tonnes, no shocks occurred at all. See pressure decrease with intermediate cones.

Flow pattern

In a silo, for a given product, depending on the hopper angle and wall friction, mass flow or core flow will occur. With mass flow, all the bulk material is in motion and flow takes place along the wall, whereas with core flow the bulk material flows (partially) within itself and stagnant zones will exist.

Mass flow First-in, First-out Mass flow
Mass flow; First-in, First-out (FiFo)
Core flow, funnel flow, shaft building Kernstroming, schachtvorming
Core flow, funnel flow, shaft building

In both flow types, irregularities during flow can lead to shocks. Although there are various causes, the mechanism is basically the same. There are or there will be stagnant areas in the bulk material that suddenly do start moving and then come to an equally sudden stop again.
Depending on the internal friction, elasticity and damping of the product, this leads to shocks in the silo. The strength of the shocks and the shock frequency is largely determined by the space given to the product to slide off and the amount of product involved.

Causes of vibrations and shocks

The phenomena referred to here should not be confused with the sudden collapse of semi-stable vaults or bridges as can occur in silo storage of cohesive materials, and which often causes significant damage immediately. We are talking about more or less periodic phenomena that occur with usually free-flowing products that are quite hard and have not too high internal friction. Well-known examples are maize, cement clinker, coal, iron ore, and various hard plastic granules.

From practical experience, several situations are known that can lead to silo quaking:

  1. stick-slip, internally or along the silo wall,
  2. In silos where flow is at the boundary between mass and core flow.
  3. Slipping of stagnant zones in core flow silos,
  4. Due to uneven drawdown by feeders,
  5. In mass flow silos, due to dilation and compaction of the bulk material.

1) Stick-slip

Stick-slip can occur when a relatively large difference exists between static and dynamic friction. This can occur with both internal friction, and wall friction. In this case, the bulk material does not shear or slide along a wall evenly. But it starts moving with difficulty, then shoots through and is slowed down again, a constantly repeating mechanism. This creates a jerky pattern of movement that can lead to vibrations or shocks when interacting with the wall, and depending on the stiffness of the structure.
In which types of bulk material stick-slip occurs is difficult to predict, but when measuring internal friction and wall friction, it will be detected quickly. See the image below for a schematic representation of wall friction measurement with and without stick-slip.

stick-slip at wall friction
Registration of wall friction measurement, with and without stick-slip

The occurrence of stick-slip also often turns out to be pressure-dependent: at higher pressure, the effect is (much) greater. This means that when the pressure is reduced, the degree of stick-slip decreases and sometimes even disappears completely.

2) Reversal of mass and core flow

Alternation of mass and core flow can cause silo quaking
Alternation of mass and core flow can cause silo quaking

In a storage situation where the flow pattern in the silo and product is in the transition region between mass and core flow, a situation as outlined in the figure may occur.
Due to initial filling, the bulk material in the hopper may be quite compacted. Because the hopper is just not steep enough for mass flow, a flow channel will form in the bulk material itself. This flow channel is usually wider at the top than at the bottom; a funnel of product forms. Stationary areas exist around the flowing part. Because the bulk material dilates during flowing (in other words, it becomes less dense, as space is required for movement), the horizontal pressure in the flow channel is reduced resulting in less support for the stagnant zone.
Moreover, there will be friction between the flowing and the stationary bulk material, "pulling" on the stationary mass. This may create a situation where the stationary area, or portions of it, suddenly shoot down the wall.
This causes the bulk material in the flow channel to be compressed again and the slipping areas slowed down where a shock may occur. Then the whole thing starts all over again causing periodic jerking.

3) Shifting mass in core flow silos

varying core flow
Shocks may occur due to varying core flow

A fairly steep flow channel can be formed in a core flow silo, as sketched in the figure opposite. This flow channel is normally always refilled as bulk material flows in from the top of the stagnant zones.
Again, the bulk material in the flow channel will dilate, creating less horizontal support for the stagnant areas. This can lead to the shooting of sheaves of product past slip surfaces in the bulk material itself. The sudden braking of this bulk material in the flow channel again may lead to a shock in which the bulk material in the channel is also compacted again and the process starts again.

 

4) In the case of malfunctioning feeders

screw feeder uneven discharge
Shocks due to uneven extraction

In properly functioning mass flow silos, all the bulk material is in motion and no stagnant zones exist. However, it regularly occurs in practice that a feeder installed below the discharge opening disrupts this ideal pattern. This is the case, for example, when a poorly designed discharge screw, vibrating feeder or discharge conveyor extracts the bulk material only over part of the discharge opening.
This creates an internal flow channel and stagnant zones in the hopper, which can lead to shocks in the same way as outlined above.

 

5) Shocks due to dilation during flow

dilatation at flowing powder
X-ray image of dilatation in bulk material when flowing

Even in silos in which mass flow occurs, shocks can still occur. Because often the entire mass in the silo is involved, these shocks can be very large. The phenomenon can be explained as follows. In the vertical part of the silo, with sufficient filling degree, the bulk material sinks as a more or less solid block, the so-called plug-flow. In the hopper, movement must take place in the bulk product itself, which means that dilatation of the product must occur. This dilation starts at the outlet opening and extends upwards in the hopper. As a result, the bulk product in the cylinder, the plug, will experience less support from the dilated part in the hopper. As a result, it suddenly shoots down a bit after which it is slowed down again by the bulk material in the hopper, resulting in a shock. Also, the bulk material in the hopper is compacted again where the process starts again and another shock may occur. The size of the shock here is determined by the bulk material properties, by the space given to the bulk material to sink and the amount of mass in the plug involved.

Combination of causes

The above phenomena sometimes also occur in combination. For instance, in core flow silos where the upper part of the product still sinks like a plug, the same shocks can occur as in mass flow silos. The occurrence of stick-slip can also sometimes initiate or amplify shock behaviour. The following will discuss in more detail options that exist to prevent or solve the problems described here.

Possible solutions for shocking silos

Prevention is better than cure

The contribution above described how and why vibrations, shocks and other undesirable phenomena can occur in a silo. We will now elaborate a little further on possible consequences of these problems and, more importantly, how they can be prevented or solved.

Problems caused by shaking silos

In the case of silo singing or light vibrations, the consequences are usually not major. At most, it is irritating to operating personnel in the silo building and immediately around it. The same applies to 'honking', although this has occasionally led to complaints of neighbourly noise and the need to adjust usage patterns. It becomes more serious when the vibrations are such that parts of the structure vibrate along with it so that damage can occur. In the case of silo quaking, the matter certainly needs to be considered seriously. It not only leads to an undesirable situation for the operating personnel, but can also cause disruption of measuring equipment, or make dosing from the silo irregular because the dispensed product does not have a constant density. Severe impacts may even damage the silo or its supporting structure. It is notable that standards prescribing the calculation of silo loads pay little attention to this potential load increase.

In the design phase of the silo

Already during the design phase, the issues mentioned here should be addressed. It should be avoided that the design results just on the boundary of mass and core flow, as this often gives rise to shocks due to unstable dead zones. It is better to sit well within the mass flow area, or if this is not possible or desired, to opt clearly for core flow . When using wedge-shaped hoppers with elongated discharge openings, the chosen extraction equipment should be designed to extract bulk material along the entire length of the opening. This is to avoid dead zones above part of the opening.
Finally, if possible, so-called stick-slip flow of the material along the silo wall should be avoided. This can give rise to vibrations, but can also initiate or amplify heavier shocks. The possible stick-slip behaviour can usually be easily detected with a standard shear test. This involves sliding the material across a test plate under various loads, measuring the shear force required as a function of displacement. With stick-slip, large variations in the measured shear force occur, as shown in the figure below. If significant stick-slip is found to occur with a particular combination of bulk material and silo wall, a different wall material or wall coating should be used if possible.

stick-slip at wall friction
Registration of wall friction measurement, with and without stick-slip

Modifying existing situations

In the design phase, the occurrence of shocks and vibrations can to some extent be investigated and then avoided. But the occurrence of shocks also remains unpredictable to a certain extent. It may also turn out that the bulk material does behave differently from what was assumed in the design phase, or because completely different products are stored in the silo. The question is then how to solve the problem of shock or vibration. The contribution above described that there may be several causes underlying the problem. First step towards a solution is therefore to identify which causes play a role in the situation at hand. On that basis, a solution can be sought.
The following briefly outlines solutions for various causes.

Stick-slip vibrations

When stick-slip flow occurs in a silo, it is usually evidenced by slight vibrations felt at the cylinder wall. This need not be a problem unless the vibrations interfere with weighing or measurement, for example. Vibrations can also cause shocks at higher fill rates. To resolve this, it is necessary to determine whether stick-slip behaviour of bulk material occurs along the wall or internally. This can be determined with a shear test, for example with the Jenike shear cell. .
If there is stick-slip along the wall, an attempt can be made to improve flow behaviour with a suitable wall covering or coating. If there is only internal stick-slip, there is little that can be done about it. However, it is practically almost always the case that at lower pressures stick-slip behaviour occurs to a much lesser extent, both along a wall and internally. A solution is therefore usually to reduce the pressure level in the silo by suitable measures to suppress the stick-slip effect. See silo pressure reduction options..

Oplossen van schokken bij een grensontwerp

core flow to mass flow
Obtaining reliable mass flow

When a fairly regular pattern of not too severe shocks occurs, it may be due to a boundary design, where the flow oscillates somewhat between mass and core flow. This can be checked by measuring the wall friction, and determining whether it, in combination with the funnel angle, is at the boundary area. If a boundary situation exists in a given case, as indicated in the figure as point A in axial-symmetric flow (left graph), an attempt should be made to arrive at true mass flow. Several possibilities exist for this.
The often simplest method is to reduce the wall friction by finding a suitable coating or cladding of the cone wall, where the situation ends up e.g. at point B, with good mass flow. Unfortunately, it does not always prove possible to find a suitable and also wear-resistant cladding. A second possibility is to make the cone steeper, so that e.g. point C is reached. Mass flow will then also occur, but it is a drastic solution that is usually not easy to realise in an existing situation.

Approximating plane-symmetric flow
Approximating plane-symmetric flow in an axi-symmetric situation

In situation with round or square hoppers and thus axially-symmetric flow, there is another possibility: switching to plane-symmetric flow. This type of flow requires a less steep hopper angle to get mass flow, see point A in the graph of the plane-symmetric situation.
Adjusting the hopper is best, but often too drastic in practice. With a displacement cone in the hopper (see sketch of cross-section of a circular silo), plane flow can be approached. The incorporation of the cone creates an elongated discharge opening at the bottom around the cone, resulting in a semi-planar flow profile. The disadvantage of incorporating a displacement cone is that the location and proper shape of the cone is rather critical to get optimal results. In addition, the cone must be quite large, and a support structure is required that will usually hinder the flow around the displacement cone. Design guidelines for this can be found in the literature.
Finally, there is still the possibility of not changing the silo geometry itself, but using vibrators on the hopper or blowing in air to improve the flow pattern, to get more bulk material moving. However, the effects of these measures are often not that great, and usually only work locally.

Resolving shocks at core flow

If core flow is found to occur, initial attempts can be made to obtain mass flow in one of the ways described above in situation where flow is at the boundary. However, in practical situations with core flow, the funnel angle is not nearly steep enough, or the wall friction is so high, that there are no easily practical solutions. One can still try to improve the flow behavior by vibrators on the funnel or blowing in air. But for real core flow, the chance of success is small because the flow is relatively far from the wall. However, when real shocks occur, it is sometimes possible to use air cannons to move the stationary zones at regular intervals. This does not completely prevent the shocks but will make them more regular and less violent.

Resolving shocks at mass flow

Vibrations and shocks can also occur in a siutation with mass flow. These can be caused by a malfunctioning drawdown mechanism in which unstable areas above the outlet are created. The system should be checked for this and the extraction mechanism adjusted. However, shock problems can occur even with properly functioning discharge devices. Usually, with smaller filling of the silo, the intensity of the shocks appears to become less. In that case, it may be useful to reduce the pressure on the hopper. See silo pressure reduction options..