Vertical Auger Based Powder Filling Machines: Principles, Variations and Applications
Powder filling machines come in a wide variety of sizes, shapes, filling principles and degrees of technical sophistication in response to the ever widening diversity of powder filling applications around. Here we look at vertical auger fillers which represent by volume 95% of powder filling machines sold, the remaining 5% comprising; vacuum fillers, cup fillers and gravity fillers.
The principle of the vertical auger filling machine is very simple. The basic filling head comprises five primary components:
1. Hopper containing the powder to be dosed (removed).
2. Auger or screw rotating vertically through the hopper.
3. Funnel or tube outlet to hopper, through which the lower parallel part of the auger doses.
4. Auger drive.
5. Agitation blade to assist powder feed into the flights of the auger (with drive).
The lower parallel flights of the auger within the funnel are machined to a constant pitch so that each pitch has a precise volume. The auger drive rotates the auger at constant speed to produce a continuous dosing flow. The agitation blade, generally separately driven and controlled, rotates in the opposite direction to the auger, de-aerating and homogenising the powder, and breaking any bridge which, in non-free flowing powders, tends to form. The agitation blade extends right down to the throat of the funnel preventing rat-holing and cavitation and ensuring that the flights are fully filled. The start/stop signals to the auger drive control the number of revolutions.
If an auger with flights of perfectly even pitch, evenly filled with powder of constant bulk density and even particle distribution, rotates a precise number of revolutions, theoretically there would be zero deviation in dose from one cycle to another. In the majority of applications, where these conditions more-or-less apply, and where any deviation is not life threatening or prohibitively expensive, volumetric filling is entirely appropriate. Typically, accuracies in the range of ± 0.5% to 1% can be expected.
Where such factors as variations in bulk density, particle size/distribution, high cost or criticality of dose demand closer tolerances, weigh-fill technology and alternative drives can be employed. Before we look at these alternatives, we should first examine the standard auger/funnel “tooling” types for the most common powder groups.
Non-Free Flowing Powders e.g: Talc
The diagram on the right shows shows the set-up described above with the auger flush with the bottom of the funnel and the agitation blade extending into the throat of the funnel. The auger featured shows larger diameter “overflights” in the hopper section. This assists the flow of an otherwise reluctant powder down to the funnel throat where the agitation blade can feed the flights of the smaller diameter parallel section. In most cases the agitation blade control will be set to “with fill” mode, stopping between fills to avoid overworking and compacting the powder. If the powder was highly aerated coming into the hopper, it might be necessary to switch to “continuous agitation” mode to assist removal of the air. A larger hopper volume will aid de-aeration and homogenisation by allowing greater residence time for the powder within the hopper. If the powder were somewhat better flowing but not completely free flowing, a slightly different approach would be required.
This may be a good point to introduce a simple (but not completely foolproof) test for evaluating the flow characteristics of powders. If when you insert a pencil into the top of a container of powder and then remove it, the hole immediately collapses in on itself, then the powder would be deemed to be free flowing. If, however, a perfect impression of the pencil remains, then it is non-free flowing. Partial collapse might indicate an intermediate powder.
Intermediate Powders e.g: instant coffee powder
Often such a powder might be damaged if large diameter compressive auger overflights are used in the hopper section, so an auger with parallel flights would be selected. Furthermore, an intermediate powder might tend to dribble after the auger has stopped. In such cases, it is necessary to create a slight back pressure and retain the powder in the funnel by the addition of a “drip washer” to the bottom of the auger or crosswires/gauzes to the bottom of the funnel. The correct solution varies with each powder and tends to be more of a “black art” than a science.
Free Flowing Powders e.g: salt
When free-flowing powders are being filled, a method of cutting off the powder supply is required. Usually, the best method of achieving this objective is to use a “spinner disc” . The spinner disc is a saucer shaped disc attached to the end of the protruding auger, designed to retain the product flow. All powders have an angle of repose, the angle at which a collapsing heap of powder will come to rest rather than continue to spread outwards: the angle (top enclosed angle) for small smooth spherical beads would be much greater than, for instance, granulated sugar. The gap between the bottom of the funnel and the spinner disc is determined by the powder dosing rate, i.e; the gap needed to permit unrestricted product flow. The spinner disc is sized to prevent the powder trickling over the rim of the disc once the auger has stopped, taking into account this gap and the powder's angle of repose. As the spinner disc throws the powder outwards, a simple collection cone funnels it down to an appropriate size. The spinner disc approach achieves a perfectly clean fill without expensive shut-off valves and complicated pneumatic/moving parts.
The size of the auger and funnel is determined by a number of factors; fill weight, dosing rate, accuracy and container neck opening. The smaller the auger/funnel, the slower the dosing rate but greater the accuracy. Conversely a larger auger delivers more quickly but less accurately. Where a wide range of fill weights are required, more than one auger/funnel set may be required to achieve the required combination of speed and accuracy. As a rule of thumb, a ratio of 5:1 for each set may be considered appropriate. Accordingly, if a range of weights from 10g to 1 kilo+ is required, three tooling sets would probably suffice; one small set for weights of 10g to 50g, a second for 50g to 250g and a large set for 250g to 1.25 kilo.
Drives
A choice of three drive configurations are available, to suit the application (and budget). Servo drive offers the best performance, with high power/torque characteristics and precision stopping position. Directly coupled inverter drives or clutch/brake offer lower cost alternatives.
Volumetric Filling
When filling volumetrically the auger stop signal is controlled from a shaft encoder monitoring the number of auger revolutions, or a timer.
The choice of system is dependent upon the speed required, batch size/changeover times and budget available. When filling semi-automatically, the upper speed limit is dictated by the time taken to manually handle the container. If one second, normally the minimum, is allowed for picking up the container, presenting it to the filling nozzle and putting it down again, a one second fill time will restrict the output to 30 fills per minute. Others factors will limit the output further, for example:
- Larger fill weight
- Slow auger rpm required due to a tendency to damage the product or generate dust.
- Small container neck opening restricting the auger diameter.
- Container handling difficulties, bags/sacks etc.
When filling rigid containers automatically the same restrictions apply. The speed may be further limited by the need to lift or vibrate the container at the point of fill. A simple pneumatic lift is used to provide a clean neck-entry fill, with neck location for narrow-neck containers. A hydraulic system provides the constant rate of descent required for bottom-up filling, and greater resistance (power) when the powder need compacting/compressing to fit the full fill weight into the container. Vibration may be required during the fill to settle granular products.
Oral suspension antibiotic filling into glass/plastic bottles, a common pharmaceutical macro-dosing application (see left) typically runs at between 15 and 30 containers per minute on a single head in-line system. Doubling the number of filling heads has the effect of virtually doubling the output. Containers are indexed through two-at-a-time, each head filling 100% into alternate containers.
In the left image above, we see a twin head machine filling dietry supplement powders into plastic cans at 40-50 cpm. Further increases can be made by going to a four head system (see centre photograph above); here filling freeze-dried and agglomerated coffee, with wormscroll container transport. The most frequently chosen solution is the continuous motion rotary machine (above right) with either intermittent or continuous filling. Containers are fed into a rotary turret possessing the requisite number of transfer funnels.
Any of the volumetric systems above can be linked to a downstream checkweigher or weigh cell. This checkweigher/weigh cell can perform three important functions; it can provide weight data, reject filled containers for under/over weights and/or monitor trends in powder bulk density. Such trends, generally caused by stratification of fines and larger granules during bulk storage/handling, cause fill weights to increase/decrease pro-rata. This data can be fed back to the filling head to compensate by increasing/de-creasing the dose to suit. In the case of intermittent filling this is done by increasing/decreasing the number of revolutions. When filling continuously it is achieved by adjusting the auger rpm.
When filling internally-taken medicines into glass bottles, the variation in one bottle weight to another is often greater than the upper and lower reject limits for the fill. If 100% weight validation is required, it is necessary to weigh the empty and filled bottles, and deduct the tare (empty) weight from the gross (filled) weight to arrive at the nett weight of fill. The use of free standing checkweighers introduces the problem of maintaining control of the containers during the weighing and filling processes and therefore of referencing the gross weight of each bottle with the tare weight of that same bottle. At slower production rates this potential problem is addressed by choosing an indexing rotary system whereby the tare and gross weigh stations are incorporated into the rotary turret, registration being mechanically guaranteed by the turret starwheel system.
With such systems, outlet is restricted by the speed at which the bottles can be indexed and reaction/settling time of the weigh cells, although multiple filling heads and weigh cells can be incorporated to double/treble/quadruple the output. Where higher speeds still are required, a mechanically registered intermittent tare and gross load cell system integrated into the infeed and outfeed starwheel assembly of a continuous motion rotary turret provides the ideal solution.
Weigh Filling
A gravimetric filling system may be selected for a number of different reasons:
- Inconsistency of tare weight (as discussed above)
- Criticality of fill weight and need to document/validate same
- Inconsistency of particle distribution and therefore volumetric accuracy
- Cost of give-away
The simplest weigh fill option is filling directly onto a weigh cell.
In a semi-automatic configuration, weigh cell can be used in a variety of ways:
Single speed/single shot
Least expensive but the speed can be compromised by the need to use an auger small enough to provide the required accuracy, or the accuracy compromised by an auger large enough to meet the required speed.
Bulk and dribble
The auger can be run at high speed for the bulk fill and slow speed for the dribble top-up. Much improved performance but the accuracy can still be compromised by reaction/settling time of the weigh cell.
Bulk, predict and top-up
Again, high speed bulk filling but the auger is stopped just short of the target weight, a static weight taken, and the top-up calculated and converted into a volumetric dose. Although the top-up is volumetric, if the volumetric accuracy (perhaps 2%) is applied to the top-up weight (typically 10% of the target weight), an overall accuracy of ±0.2% can be expected.
Weigh cells can be incorporated into automatic filling in different ways to different purposes.
In a single head system, a single weigh cell placed directly under the filling head can be used in exactly the same way as when filling semi-automatically; single shot, bulk and dribble, bulk predict and top-up, can tare weigh the empty container and provide weight data as required. If the reaction/settling time of the weigh cell prohibitively restricts the production output, moving the weigh cell downstream and filling volumetrically with feedback (as described earlier) might have to be considered, with upstream tare weigh cell as required.
Alternatively, two heads could be used, but instead of the two heads dosing onto two live weigh cells to double the output, a bulk and top-up system would be preferred.
This provides the optimum of speed and accuracy; the volumetric first head bulk filling quickly with a large auger, the second head making the gravimetric top-up with a more accurate small auger. Furthermore, the cost of one weigh cell is saved.
For even greater speed and accuracy, the weigh cell can be moved between the two filling heads turning it into a bulk fill, predict forward and top-up system with intermediate weighing.
This offers the speed of large auger bulk fill together with the accuracy of filling the top-up with a smaller auger. In this case the weigh cell would feed back to the bulk filling head to maintain the proportion of the bulk fill, feed forward to the top-up head and predict the top-up dose needed, but could not provide final weight data. If required, a second weigh cell downstream of the top-up head could feed back to the top-up head and forward to any reject station. If the containers vary in their empty (tare) weight, tare weigh station could be added upstream of the bulk filling head.
