Leaf tests

Vacuum filters are usually simulated with a Buchner funnel test or filter leaf test (54). The measured parameters are cake weight, cake moisture, and filtration rate. Retention aids are usually evaluated using the Britt jar test, also called the Dynamic Drainage Jar, which simulates the shear conditions found on the paper machine and predicts performance (55). 

Figure 18-95 also contains a schematic layout of the equipment which is required for all bottom-feed leaf tests. Note that there are no valves in the drainage line between the test leaf and the filtrate receiver, nor between the filtrate receiver and the vacuum pump. 

At the start of the leaf test run, the hose between the test leaf and filtrate receiver should be crimped by hand to bring the filtrate receiver to the operating vacuum level. The use of a valve at this point is not only less convenient but very frequently results in a hydraulic... 

It is difficult to plan a filtration leaf test program until one test has been run. In the case of a bottom-feed test, the first run is normally started with the intention of using a 30-s cake formation time. However, if the filtrate rate is very high, it is usually wise to terminate the run at the end of 15 s. Should the filtrate rate be very low, the initial form period should be extended to at least 1 min. If cake washing is to... 

In any leaf test program there is always a question as to what vacuum level should be used. With very porous materials, a vacuum in the range of 0.1 to 0.3 bar (3 to 9 in Hg) should be used, and, except for thermal-drying apphcations using hot air, the vacuum level should be adjusted to give an air rate in the range of 450 to 900 mVm h (30 to 40 cfm/ft measured at the vacuum. 

Bottom-Feed Procedure The procedure for coUecting data using bottom-feed leaf test techniques is as follows ... 

FILTRATION LEAF TEST DATA SHEET - VACUUM AND PRESSURE... 

Top-Feed Procedure The sequence of operations with a top-feed leaf test is the same as in a bottom-feed test, except that the leaf is not immersed in the sluriy. The best method for transferring the slurry to the top-feed leaf is, of course, a function of the characteristics of the sluriy. If the particles in the sluriy do not settle rapidly, the feed can usually be transferred to the leaf from a beaker. If, however, the particles settle veiy rapidly, it is virtually impossible to pour the slurry out of a beaker satisfactorily. In this case, the best method is to make use of an Erlenmeyer flask, preferably one made of plastic. The slurry is swirled in the flask until it is completely suspended and then abruptly inverted over the leaf. This technique will ensure that all of the sohds are transferred to the leaf. 

When a flocculated feed is added to a filter tank, there is a definite time lag before this material reaches the surface of the filter medium. Since this lag time is not known at the time of testing, a lag time of 8 to 10 minutes should be allowed before starting the first leaf test on a flocculated shiny. Two, or perhaps three, tests can be run before the elapsed time exceeds the probable retention time in the full-scale filter tank. With knowledge of the elapsed time after flocculation and data relating to the rate of degradation, the rates obtained on the leaf test runs can be adjusted to some constant lag time consistent with the anticipated full-sc e design. 

It is advisable to start a constant-pressure filtration test, like a comparable plant operation, at a low pressure, and smoothly increase the pressure to the desired operating level. In such cases, time and filtrate-quantity data shoulci not be taken until the constant operating pressure is reahzed. The value of r calculated from the extrapolated intercept then reflec ts the resistance of both the filter medium and that part of the cake deposited during the pressure-buildup period. When only the total mass of diy cake is measured for the tot cycle time, as is usually true in vacuum leaf tests, at least three runs of different lengths should be made to permit a rehable plot of 0/V against W. If rectification of the resulting three points is dubious, additional runs should be made. 

Leaf Tests A bomb filter is used for small-scale leaf tests to simulate the performance of pressure-leaf (leaf-in-shell) filters. The equipment used is a small [50.8- by 50.8-mm (2- by 2-in)] leaf, covered with appropriate filter medium, suspended in a cell large enough to contain sufficient shiny to form the desired cake (Fig. 18-108). The shiny may be agitated gently, for example, by an air sparger. 

Plate-and-Frame Tests These tests should be conducted if the use of a filter press in the plant is anticipated at least a few confirming tests are advisable after preliminaiy leaf tests, unless the sluriy is veiy rapidly filtering. A laboratory-size filter press consisting of two plates and a single frame may be used. It will permit the observation of solids-settling, cake-pacldug, and washing behavior, which may be quite different for a frame than for a leaf. 

Compression-Permeability Tests Instead of model leaf tests, compression-permeabihty experiments may be substituted with advantage for appreciably compressible sohds. As in the case of constant-rate filtratiou, a single run provides data equivalent to those obtained from a series of constant-pressure runs, but it avoids the data-treatment complexity of constant-rate tests. 

In principle, filter bed permeabilities can be calculated using the Carman-Kozeny equation 2.53. For slurries containing irregular particles, however, cake filtrabilities together with filter medium resistance are determined using the Leaf Test (Figure 4.13). In this technique, a sample of suspended slurry is drawn through a sample test filter leaf at a fixed pressure drop and the transient volumetric flowrate of clear filtrate collected determined. 

A leaf-test was carried out on the new medium with slurry giving the following results ... 

Leaf test - Assume constant Ap Integrated filtration equation of the fonu... 

The proof consists in showing that if leaf test U,p,T) could be done by just adding pushdown stores of the type described in this section, then it could be done by an ordinary program scheme without augmentations, but that is impossible. 

Safety factors for scale up from laboratory leaf tests are difficult to generalize. On the basis of pilot plant work, adjustments of 11-21% are made to plate-and-frame filter areas or rates, and 14-20% to continuous rotary filters, according to Table 1.4. 

Figure 11.8. A filtration leaf test data sheet (Dahlstrom and Silverblatt, 1977).

Filtration leaf test results indicate that the filtration rate of a protein product is 50 dry lbs/(ft2 hr). What size production filter would be required to obtain 100 dry lbs of filter cake per hour ... 

Dorr-Oliver Inc., Filtration Leaf Test Procedures. Stamford, CT Dorr-Oliver Inc., 1972. 

---