Case Studies



Case Study 1: COG Reduction

Aximill eliminated the pre-drying process and double handling of the material, resulting in an overall COG (Cost Of Goods) reduction of 32 percent.

zeolite

Zeolite, 2 to 3 Micron, 0.5% Moisture

A client of Aximill was processing zeolite with numerous machines to achieve an average particle size of 2 to 3 micron. This involved drying the product, ball-milling it, passing it through a classifier, and returning the oversize particles back to the ball-mill, as a batch process.

Aximill provided a full turn key solution, as a continuous process from start to finish, processing raw zeolite with a moisture content of 18 percent, and an approximate size of 6mm, to a finished size of 2 to 3 micron, and a 0.5 percent moisture content in a single pass. The breakdown was 98 percent under 5 micron, and 80 percent under 3 micron. No further classification was necessary.



Case Study 2: Tahini and Almond Oil

Aximill versatility.

As a result of a customer enquiry, the staff at Aximill set up a special trial to produce sesame seed oil and tahini. The client supplied several kilograms of hulled, toasted sesame seeds, and his main interest was in finding a more efficient way of extracting oil from the seed. Based on Aximill's experience with almond oil and almond flour, the mill was set-up with a closed loop to run a nitrogen atmosphere (refer to photo of setup - note: blue tube at base of receiver, for nitrogen purge).

Experimental Setup

Experimental closed-Loop Aximill Setup

The Aximill is fitted with a rotary valve on the intake to feed the product, and the machine and receiver are purged with a cubic metre of food-grade nitrogen. While running, nitrogen continues to be fed into the system at 2 to 3 litres per minute, to maintain system pressure and prevent the rotary valve from introducing air. In order to bypass the sesame seed flour stage, and oily paste, a high initial rotor speed was used. The initial results produced a coarse oil with particulates, samples of which were isolated for filtering as a source of oil. Throughput was comparable to almond oil (click to link).

The rotor speed was increased in stages, producing a finer and finer oil, and ultimately resulted in a fine oil with particulates well under 5 micron, mostly under 2 micron. We had produced an incredibly fine tahini. The taste-test was outstanding, very very smooth. After 4 months of standing, a 100 mL sample jar contained only a thin (5 mm) layer of oil on the surface of the settled product, easily re-blended with a quick shake!



Case Study 3: Milling Versatility

Aximill 250

Aximill 250

Rather than purchasing multiple machines, it takes only one Aximill.

An organic grain farmer approached Aximill, with the desire to mill his own grains and legumes. He brought a range of samples, including chickpeas and millet, and was extremely impressed with the quality of the flour, the low temperature processing, and the kilos/hour attained using the Aximill. The price was a sticking point, compared to some of the competitive machines available.

Some months later he returned, after thoroughly researching the market and discovering that he had to buy three different machines to mill his range of products. Rather than purchasing all three machines, he could instead do everything with one Aximill.



Case Study 4: Energy Savings: Aximill Vs. Jet Mill

65% energy saving, with Aximill.

In order to ascertain the energy savings of using Aximill technology as opposed to conventional milling, we set up a trial which was supervised by the chief engineer of a talc quarry. In due course, several tonnes of raw talc arrived in bulker bags, and divided into 500 kg trial batches. The material had been crushed and passed through a 20 mm minus sieve, in order to ensure a level playing field.

The current method used by the talc quarry is as follows: The talc is quarried, passed through a crusher, and passed through a 20 mm minus sieve. The 20 mm minus material is then forwarded through a coal-fired rotary drier, and into a large roller mill. The material from the roller mill is sieved to achieve a dry less than 80 micron material, with the over-size passed back into the roller mill. The under 80 micron material is then fed into a jet mill, which further reduces the size to less than 30 micron. The material is classified and the over-size passed back to the jet mill. The air for the jet mill is supplied by a 250 kW compressor. There is an option to run the jet mill at 400 kg/hr, to achieve the less than 30 micron, or to run at 500 kg/hr and classify the material with the over-size returned to the jet mill, depending on output requirements.

The Aximill 1000 is capable of taking the talc at 20 mm minus. The first batch was used to optimize the setup of our 1 metre Aximill by running 100 kg trial lots to achieve the correct particle size, versus throughput. Samples were kept from each subsequent batch before and after milling, to test for moisture loss and residual moisture. A number of batches were run through under normal production conditions, and timed. Checking the particle size after the first batch, it was discovered that we had 10 percent over-size (above the 30 micron specification). Simply by raising the operating revs of the machine, the next batch came in at 2 percent over-size without classification.

TABLE 1 - Test Results of Milling Talc with the Aximill

Test # Original Size Moisture (pre-mill) Milled Size Moisture (milled) kg/hr Hz / RPM Classification Cone Blades Cooling Temperature / Humidity
1 < 20 mm 6 to 14.5 % < 35 µm 0.17 % 612 50 / 3900 15 mm / 45 deg 24 mm Tungsten Non 10.6 deg C / 96 %
2 < 20 mm 6 to 14.5 % < 30 µm 0.28 % 600 52 / 4060 15 mm / 45 deg 24 mm Tungsten Non 10.6 deg C / 96 %
3 < 20 mm 6 to 14.5 % < 30 µm 0.19 % 607 52 / 4060 15 mm / 45 deg 24 mm Tungsten Non 11.5 deg C / 85 %

SUMMARY

The Aximill 1000 was able to process and dry the talc at 600 kg/hr, within the specification for the material. The drive motor of the Aximill 1000 in this instance was 75 kW, and average ampere draw during the trial was approximately 100 amperes. Ancillary equipment consisted of: a fractional kilowatt motor for the elevator feeding the mill; a 12 kW side-channel blower servicing the product filter; and an additional fractional kilowatt motor, which drives the rotary valve under the product filter. By comparison, the talc quarry jet mill uses a 250 kW compressor, and a classifier. This translates to a 65 percent energy saving, plus this does not include the rotary drier, the roller mill, the sieve and all the ancillary motors used by the standard plant which are now made redundant by the Aximill.

At the request of the supervising engineer, the vanes on the rotor (which generate the airflow for the machine) were weighed at the beginning of the trial because of wear concerns. After two tonnes of material had passed through the machine, the vanes were re-weighed. There was no visible or measurable wear on the tungsten-carbide faces of the vanes, but some loss of stock around the bolt holes on the sides of the vanes. Experience has shown that this is normal breaking-in wear, which progresses no further with subsequent use.



Case Study 5: Sizing Trial

Particle size consistency, despite disparity of product.

Recently, Aximill undertook a trial to determine consistency of particle size when milling different products on identical machine settings. The machine was set up to run at a motor speed of 45 hertz, coupled with the other controlling parameters (airflow, temperature, control ring, etc), with the expectation of achieving 10 to 20 micron particle size. Four different products were trialled: Gumbi Gumbi, banana stem fibre, black chickpea, and sweet potato. The banana stem fibre and sweet potato had been pre-dried and pre-milled using a Fritsch mill. The Gumbi Gumbi consisted of leaves, twigs, and short sections of stems up to 10 mm diameter. The chickpeas were whole and dried. The sample sizes were all under 200 grams, which has the drawback of producing over-sized particles when the product feed is stopped. If there is insufficient product in the machine and can't collide with itself, it simply goes around and around until the machine stops. Subsequently, the larger the sample, the smaller the contamination with oversize. The oversize is apparent when looking at the bell curve (Figure 1).

The products were run through one at a time, with the machine cleaned between runs. The samples were carefully collected, labelled, and packaged for analysis and sizing (using a computerised laser sizer). The results can be seen in Figure 1, which has the various bell curves super-imposed over one another. The results were gratifying. A much greater variation was expected, particularly with the Gumbi Gumbi. Wood fibre tends to be a lot tougher than fibre in legumes and banana stems. The bell curve on the right is put in as a comparison, showing sweet potato after passing through a Fritsch mill. A closer inspection of Figure 1 will show two pink graphs which represent two samples of chickpea. One sample of chickpea was run at a reduced speed of 40 Hz, the other parameters remaining the same. Interestingly, the peak of the curve is lower indicating a lower percentage of fines (70% reduced to 60%), but the bottom of the graph shows an increase in larger particles (in the upper range of the graph). What is of interest, is that the width of the bell curve is unaffected, ie. the spread of size is consistent in all cases, as a result of the specific controlling parameters used (listed earlier).

trial

Figure 1: Bell Curve Comparison Between Different Products



Case Study 6: Throughput Trial

Aximill demonstrates ultrafine, dry milling, without use of classifier.

In 2016, a 110 kilowatt Aximill was built specifically for the agricultural industry. The requirement was for 500 kg per hour, at sub 5 micron, continuous, and at ambient temperatures up to 48 degrees C. Ancillary equipment included a large refrigerated chiller to provide cooling water for the rotor chamber, bottom bearing, and oil, and also included was a microprocessor to maintain constant load on the machine by controlling the feed rate.

The setup of the plant is of standard design with a bulk loading hopper from which the material (in this case a fertiliser of sub 1 mm granules) is elevated to a smaller receiving hopper fitted with level sensors. At the base of the hopper a rotary valve controlled by the microprocessor feeds the material into the Aximill intake. In a single pass through the Aximill, the material is reduced to sub 5 micron, and via two outlet tubes passes into two cyclones. The product exits the base of the cyclones via rotary valves and is then elevated to a receiving hopper. The air and dust (fine fertiliser) from the top of the cyclones passes through a filtration system. The accumulated ultra fine fertiliser passes through a rotary valve at the base of the filter and is elevated to join the rest of the material in the receiving hopper from which it is metered into bulker bags. The filtered air (25 cubic metres per minute) is vented through the roof into the atmosphere.

The Aximill easily reached its design criteria. At first run, just over 500 kg/hr was achieved, with average particle size around 2.5 micron. Without the use of an external classifier, this is an unprecedented result. By turning up the air speed to the desired level, the throughput instantly jumped to just under 600 kg/hr with the particle size unchanged. At this throughput, power level was approximately 83 percent of maximum, and further optimisation of airflow and setup of the mill will make it possible to achieve higher throughputs.

There is growing interest in the micronisation of agricultural products such as gypsum, lime, phosphates, diatomaceous earth, and like products, because of better plant uptake and/or absorption, and lower losses to groundwater and the environment, through the use of drip irrigation and targeted spraying. Sub 5 micron particles can be absorbed by plant leaves and lead to quicker results than conventional fertilisation.

Aximill1000

Aximill 1000 with 110kW motor

Page last updated: 14 September 2016