Jeff Gwirtz: Determining How Aggressively to Grind

The topic of mill stock granulation profile and its graphical and mathematical evaluation for mill control purposes is important. The value of representative sample collection, proper sieve testing careful stock observation, mathematical analysis, and graphical representation is critical in both control and design.

This article presents a comparison between several commercial first break sifter configurations using various 1BK roll gap settings and their potential impact on downstream milling systems.

John Winfield presented a model break granulation curve concept in “Break Release/Extraction Survey – 1988,” Technical Bulletins – Association of Operative Millers, 1989, pages 5,415-5,422. In his paper, model break granulation curves were shown at various release settings. The curves for sizes larger than the target break release appeared to be estimates and were parallel for various releases.

This area of break granulation would be essential for the estimation of coarse and fine break separation for subsequent second break grinding considerations. To estimate the slope of the curve beyond the break release selected by Winfield, first break grinding runs of hard red winter (HRW) wheat tempered to 16% moisture were conducted at various roll gap settings. The roll gap setting in inches and metric measurement units are provided in Table 1 below. The six granulation profiles (A-F) resulting from the various roll gap settings are provided in Figure 1 on p. 31.

Roll gap measurement is dependent on the individual making the measurement with a feeler gauge. Individual preferences to the feeling of resistance as the gauge is inserted or pulled from the gap can vary from person to person.

Moreover, when adjusting rolls, a better estimate of average gap can be obtained if the rolls are rotated into different positions and the gap measured across the roll grinding line. While it may not be possible in most cases, it will reveal that rolls are not perfect cylinders, and high spots will be observed where the rolls are more closely set and tight than low spots, indicating the gap is wide. Judgment on roll gap will vary from person to person and may be monitored best with break release, provided samples are collected and sieved properly.

Granulation Curves

The six granulation curves A-F shown in Figure 1 on p. 31 are very similar to the model granulation curves proposed by Winfield. Best fit lines originating from the point-of-break release achieved using a 1041-µm screen to each respective endpoint would provide useful mathematical approximations. Graphical analysis of reported data is adequate for most purposes as shown in Table 2 on p. 32.

Table 2 shows sifter cuts made on the first break sifter in three different commercial mills. For each mill, various first break sifter fractions (screen μm) may be dispatched as flour or additional grinding, sifting, and purification as indicated by the flow diagram. The percentages of incoming stock to first break for the various granulation curves with release through a 1041-µm screen shown are provided, along with the average, minimum, maximum, and range for the fraction identified. Granulations B and F closely represent the lowest and highest first break release, respectively, identified in the Winfield survey.

Roll gap measurement is dependent on the individual making the measurement with a feeler gauge. Individual preferences to the feeling of resistance as the gauge is inserted or pulled from the gap can vary from person to person.

The three mills take differing approaches to first break grinding and stock separation. When comparing more typical releases for first break, one might look to granulation curves C and D for example. As we make the comparisons moving from C to D granulation for the A-Mill, the increase in load from the second cut in A- and B-Mill and the third cut in C-Mill increases by 34.8%, 15.1%, and 22.7%, respectively. The third cut in A- and B-Mill and the fourth cut in C-Mill increase by 35.7%, 12.9%, and 8.7%. The fourth cut in A- and B-Mill and the fifth cut in C-Mill increase by 29.8%, 59%, and 59.6% respectively. However, flour production at first break in hard wheat milling is variable depending primarily on kernel hardness, as well as roll corrugation sharpness. Changes in granulation or break release can dramatically impact both the quantity and quality of stock moving downstream.

Behavioral Differences

Every mill design is different, resulting in differences in product stream flow rates and quality characteristics. Once the equipment is installed, the difference in design becomes somewhat of a fixed constraint that must be balanced to achieve optimal operational efficiency and product quality. The flow and equipment sizing are based on milling engineer design and operational preference and experience, as individuals may view the milling process differently from one another.

During startup, the behavior of the wheat itself requires some modification to roll and purifier settings (if employed), in addition to sifter and purifier screen changes. Such changes may be based on the mill designer’s or startup miller’s judgment, who may view the problem encountered differently resulting in varyingly different solutions. It is strongly recommended that flow and operational changes be documented during mill startup, as well as during planned or unplanned wheat mix changes.

In many cases, the first and second break are combined together with one grind step following the other, with conveying and sifter capacity designed to handle the mass and volume of ground stock.

Once control is established and the mill balanced, it may run smoothly until the next wheat mix change or crop year. Finding the balance is important to maintaining production efficiency and product quality uniformity for the installed mill flow optimization.

Consider how changing first break grinding from that characterized by curve C to that shown by curve B. As the rolls open up, the energy consumption at first break would be decreased. Perhaps the stock becomes cooler and moisture loss decreased at first break. Will the roll drive motor be operating efficiently with less grinding pressure? The ground first break stock becomes denser, particle size increases, and aerodynamic properties altered, requiring increased air flow and energy for conveying. Pneumatic air system adjustments may be required, as stock quality difference demands more or less energy to convey efficiently. Coarser stock entering the sifter may increase sieve cloth wear; however, the finer clothing may last longer due to less product flow. Is this an efficient use of available sifter space (depth) and surface? Sifter separation efficiency may increase as will likely the increased impact of bare bolting. There will likely be less load to the purifier(s) from first break. Will this increase moisture loss? Will there be enough product to adjust the purifier properly and make the desired size and ash shift?

Anticipate screen and air changes at purification to accommodate changing load and particle size. As the top scalp moves to the next grinding step, will there be adequate grinding and sifting capacity to handle the increased load? In many cases, the first and second break are combined together with one grind step following the other, with conveying and sifter capacity designed to handle the mass and volume of ground stock. Where the first and second break operate separately with intermediate sifting, the destination of separated products may be the same from both first and second break sifters. However, conveying and sifting may be limited to handling product within a specific mass, volume, and aerodynamic characteristic.

As the percentage of retained stock increases, sifting efficiency has decreased. Some carryover is reasonable and is a guard against bare bolting, which is especially critical for flour clothing.

The coarse, dense material of granulation B likely would move quickly across today’s higher-speed sifters with small throw impacting sifter efficiency when compared to granulation C. Sifting efficiency is measured by determining the presence of potential through stock in the overtail of a given sifter section.

Many mill engineering firms provide designs that have data collection capability on mill behavior such as feeder roll engagement or disengagement frequency and/or speed changes, as well as roll grinding energy consumption.

In Mill B, for example, a sampling of tail overstock from the 475-µm screen would be test-sifted on the same screen to determine the percentage of material in the sample that should have passed through the 475-µm screen. As the percentage of retained stock increases, sifting efficiency has decreased. Some carryover is reasonable and is a guard against bare bolting, which is especially critical for flour clothing. Excessive carryover may indicate sieve blinding as a result of poor cleaning action. In either case, mill balance and product quality are compromised.

What Automation Requires

Winfield stated, “The complete and scientific automation of the milling process should incorporate data as to the load distributions to the various passages and machinery in the mill flow.”

To an extent, we have seen this happen with automation packages capable of returning a mill to specific settings based on prior experience with a given wheat mix. The ability of rolls to respond to variations in load to the roll keeps the mill running, but is it truly optimal relative to mill balance, efficiency, and optimal uniform product? Many mill engineering firms provide designs that have data collection capability on mill behavior such as feeder roll engagement or disengagement frequency and/or speed changes, as well as roll grinding energy consumption.

Unfortunately, however, the data may not be collected or analyzed to improve the process and future designs taking into account machine and wheat behavior. The data collection capability is there. The data needs to be mined and analyzed. Until then, it is an opportunity missed!

Achieving existing mill flow balance and keeping the mill in operation no longer may provide the competitive edge.

Such data combined with data on the behavior of mill streams, including load and quality, would present an opportunity to further improve both efficiency and quality of our milling processes. Achieving existing mill flow balance and keeping the mill in operation no longer may provide the competitive edge. We need to do more work to create understanding of mill design and balance, if we are to do more with less.

Dr. Jeff Gwirtz is CEO of JAG Services, Inc., an international consulting company in Lawrence, KS; 785-341-2371 or jeff@jagsi.com. He also is adjunct professor in the Department of Grain Science and Industry at Kansas State University, Manhattan.

From Second Quarter 2021 MILLING JOURNAL