15-thing every Extractive Metallurgist should know about Hydrocyclone

Either it could be a separation of fines from the coarser material or separation of oil from sand, Hydrocyclone is proven and versatile equipment in the mineral, mining, and petroleum industry. Due to the cost and energy effectiveness with a low maintenance requirement, Hydrocyclones are ubiquitously in the industries since 1894. Using high centrifugal force, ores can be separated, classify, or filter from the solid and liquid component of feed. In the conventional Hydrocyclon, separation of material is based on particle size. The heavier particle will be situated towards the wall, and consecutively it reports to the underflow. In the case of the small size, they will be more tends to report in overflow through the opposite direction of spiral flow, which formed due to the pressure gap from outward to center.

In Hydrocyclone, we can say with confidence that the partition efficiency or separation efficiency at any particle size could not be ideal(100%). One reason is the error in the designing of Hydrocyclone. It designed to handle specific pressure and to maintain steady cut-size at a particular point. However, this may not be the case if the error in calculating the relation between machine parameter and particle separation (no machine can be ideal) calculated through correlation formulas, which are generally empirically formulated. Hence, It is highly unlikely to get perfect separation. Secondly, while operation, we consider specific gravity over the feed is homogenously distributed (If we do not believe this, this will not only be for size separation) and size/radius of a particle playing a role indifference of centrifugal force. Nevertheless, density also contributes a role in the process, and hence, higher specific gravity particles with small sizes can rest towards walls and, consequently, report to the underflow. Thirdly, fines will always follow the water, and that will be the cause of inefficiency in separation. Given the fact, the underflow will still contain some water, and hence there will be fines.

While operating Hydrocyclone in the mineral processing plant, metallurgists, or the process engineer should know the following 14 things that are useful for analyzing the performance of Hydrocyclone.

1. Ideally, the Hydrocyclone is designed to make a perfect separation between coarse particles and fine particles without employing any moving part. However, separation occurs over a size range

2. Cyclone Efficiency: 

• The partition or performance curve is the method of representing the cyclone efficiency.
• The curve relates the weight fraction or percentage of each size fraction found in the feed, which reports to the underflow.
• The the sharpness of the separation depends on the slope of the central section of the partition curve. More vertical is the curve higher is efficiency. 

3. Therioticly, the separation point is the size at which a particle has a 50-50 percent chance to report to the overflow or underflow. Consider the example; if the pressure is higher than the target pressure, the d50 will be finer than the desired. The pressure should remain constant throughout the process To determine the performance of the Hydrocyclone, the following variable needs to be measure.

• The particle size distribution in the three process stream. (Feed, overflow, and underflow)
• The slurry density in the three process stream
• The the mass flow rate in the three process stream
• The pressure drop, measured at the feed inlet

4. A fraction of fine particles always report to the underflow; hence, experimentally observed partition curves do not asymptote to zero but to a minimum called the bypass. This can be interpreted as a fraction of all particles in the feed bypassing classification and reporting directly to the underflow stream. Short-circuiting of feed material to the overflow stream may cause the partition curve not to reach a partition value of 1 (100%): this is not common. The effect o bypass on classification performance is taken into account by correcting the partition value:

5. It is found that bypass equals the water recovery from the feed to the underflow, although there are no fundamental reasons why this happens. (We can see bypass amount in the figure: 1 down below. The uncorrected graph above the x-axis is the amount of bypass)

6. Mass balance

• In the case of Hydrocyclone, there is only one inlet and two outlets. Hence the grade of the inlet is equal to combine grade of the outlet material. Hydrocyclone is designed to operate on the principle of size-wise separation. Therefore, grade enhancement will not be achievable. 
• The use of an efficient and general method to solve the mass balance of a network of Hydrocyclones connected in any configuration has demonstrated in the computation method (HYDRAS) 

7. Change in flow-rate shows a change in pressure, and this caused by upstream equipment failure. 

8. Feed density also affects the cut size of the Hydrocyclone. As the density of feed increases, coarser cut-size will be. 

9. The cone angle plays a vital role in setting the cut point of a Hydrocyclone. Generally, for fine cuts with a Hydrocyclone, a 10-12 degree angle cone angle is used. To make a coarser cut, the angle is increased as needed. Twenty-degree cones are conventional in a lot of aggregate applications, but 40 degrees and even 60 degrees have been utilized. Not seen as much, the installed angle of the Hydrocyclone will also change the cut point. From the vertical plane, laying the Hydrocyclone down will cause it to make a coarser cut. The cone angle must be taken into consideration so the Hydrocyclone is not tilted too far. However, a more than 75-degree tilt can almost double the cut size.

10. The fish hook effect: In the research community, there is a controversial view regarding fish hook effect. Some researchers observed that this effect is the result of an error in calculation. On the other hand, a different opinion is that this effect is formed from physical forces. It means that the recovery of relatively finer particle sizes is initially higher than that coarser particle sizes in underflow up to critical particle size, and after that recovery increase with an increase in particle size. This happened when wt% is in between 20-30% 

11. Corrected partition curve has an 'S' shape curve (Fig:2)or can be described as a sigmoid curve. Which can be defined as a two-parameter function. The two parameters determine the cut size and the sharpness of separation. The fishhook partition curve(Fig:3) can be modeled using the sum of a corrected partition curve and an inverted partition curve multiplied by a bypass fraction. 

 

12. Since the classified stream must reverse its direction near the spigot, during which some coarse particles are carried upward to the vortex finder. The extent of this misclassification depends on the spigot size, while the classification depends on the centrifugal force field in the cylindrical and conical sections. Thus, for "excellent." classification, the allowable concentration of solids in the feed is limited by the spigot size. Another stream that contributes to misclassification is short-circuiting flow. 

13. In the case of calculating corrected efficiency 'Ec' curves, there are three models: Lynch model, Plitt-Reid model, Harris model. While applying these mentioned models to the industrial equipment, the following things should be taken into consideration.

• The Plitt - Reid function cannot satisfactorily explain all the different shapes of the reduced efficiency curves observed in practice, such as complex ores and coal, etc. Its specific single parameter functional form restricts its applicability. 
• That Rao's approach had been very successful in industrial practice.
• The Lynch approach which can be generalized as Ec(d) = f (d / d50c, a 1 , a 2 ..... i ) does, however, represent all the invariant (or approximately so) reduced efficiency curves, which have been reported over the last five decades
• Both the Plitt - Reid function and the Whiten function, W(a), are satisfactory to describe the most efficient curves observed in practice. However, the latter is reported to be more accurate, although the former is more straightforward to use. 
• Most of the observed variation in classification index, m, reported by Plitt, is attributable to 'normal' sampling and analysis errors. The geometric mean between the extreme values of the classification index minimizes the errors of prediction and is a constant value for the system. 

14. In a conventional way of separation, solid/solid through single-stage Hydrocyclone may not be practical due to excessive load on centrifugal force. To improvise the separation efficiency where density differential is low, then the multiple stages of classification can be used. When targeting low fines in the underflow, it may be necessary to have multiple stages of Hydrocyclones if there is a significant amount of fines in the feed. This is especially true when there is a large amount of the size fraction near the cut point. The network of multi-stage classification can be divided as follows. 

• Multiple product separation networks (e.g., clay processing)
• Complex processing network (e.g., counter-current washing and dewatering)
• Enhanced classification network (e.g., closed-circuit grinding

15. We can predict the product size distribution and solid recovery for changes in feed size distribution. If we assumed that the bypass is equal to the water recovery, the liquid and volumetric balance could also be estimated.