Most extrusion defects that are directly traceable to pugging operations occur due to insufficient mixing of bodies.   The typical extruder was not designed to do major delamination of clay minerals or mixing of bodies.  Mechanical feed mechanisms are designed to chop feed materials and direct them into the extruder where they are compressed to feed through the die.  That is all.  To expect pugging/extruder systems to do a lot of mixing/blending/delaminating of bodies, or to put the final touches on mixing which wasn't fully completed prior to this step, is to invite problems.

Other extrusion defects which are traceable to body formulations are not discussed here.  Contrary to opinions I have heard, dilatancy is not a result of poor mixing -- dilatancy is usually the result of distributions that are too narrow (i.e., a major percentage of body particles are exactly the same size) and/or extrusion rates that are too fast.  Some bodies are simply dilatant, and the fact that extrusion problems occur is not the result of the extruder, but of the body.  Some bodies are dilatant when sheared at high rates, but they are okay (shear-thinning) when sheared at lower rates.  Such problems are extruder-related, but mostly due to poor particle size distributions and/or chemistry in the body.  Neither of these cases are discussed further in this article.

Mixing

Some mixing occurs in extruders and delaminators simply because bodies are sheared as they are pushed forward through the augers (or by the piston) and through expansion/restriction chambers towards the die.  But the intensities of mixing by pug mills and extrusion augers are insufficient to allow unmixed raw materials and water to be fed into the pug mill to produce well-mixed, homogeneous bodies by the time they exit the production die.  Even less mixing occurs in piston extruders.  Pug mills are designed to chop filter press blanks and other feed materials so they can flow into the auger(s).  Although this looks like mixing, the primary function of extruders is extruding -- not mixing.

Use mixers for mixing, extruders for extruding, and delaminators for delaminating.  Yes -- each of these devices will perform some of the other functions, but they typically won't be highly efficient when performing the other functions.  If you want to mix and delaminate and extrude, then put all three devices in series (i.e., mixer feeds delaminator which feeds extruder.)

Defects occur in extruded product when vast amounts of mixing need to be performed during the pugging/extruding operation.  For example, if the feed materials are too dry, handing the pug mill operator a water hose to achieve proper viscosities assumes the pugging/extruding operation will perform adequate mixing.  This will frequently produce high solids chunks of body with lower solids regions between them.  It then depends on the size of the die openings relative to the size of the high solids chunks whether or not extrusion surface and internal defects will appear.  If the die sizes are large relative to the size of the high solids chunks, extrusion defects may be minimal.  The fact that extrusion defects are minimal, however, does not eliminate other problems caused by regions of high and low solids contents in the product such as differential shrinkages, cracks, and low strengths.  If die openings are small relative to the high solids chunks, the body going through the die may alternate between zones of high and low solids contents, which can cause extrusion problems.  

Normal Extrusion Augers

Extrusion augers are designed to push bodies towards the die and create high pressures as they do so.  Most production augers which accomplish this are relatively long, continuous augers.  In this configuration, augers designed primarily for compaction do not perform much mixing.

Such augers are designed to force the bodies towards the die, while increasing pressures as the body moves down the extruder.  Some shear takes place near the surfaces of the auger and at the barrel wall, but cutting and shearing are not primary functions in this type of system.

Delaminator Augers

When delamination of the body is a design function of the device, the auger shaft will consist of many split augers and the barrel of the extruder may contain many stator sections.  Delamination takes place as the body is constantly cut by the split augers and sheared past the stators.  Such designs cut, shear, cut, shear, cut, shear, etc., and all of this occurs at relatively high pressures (but not as high as can be achieved in a normal extruder with continuous auger.)  Mixing occurs in these devices due to the fact that they are designed to produce delaminating conditions.  But once again, mixing is not the primary design function -- delaminating is.

What is delaminating?  When clay/kaolin bodies contain many large book stacks of kaolinite particles, kaolinite plates are not free to move as individuals.  Delaminators help the rheological/processing properties of such bodies.  Why are clays and kaolins present in the body in the first place?  In most cases, clays and kaolins are present because they provide yield stresses and shear-thinning flow properties typical of clays and kaolins.  Large book stacks of kaolinite particles, however, minimize the plastic property contributions of these raw materials.  To get the most plastic properties for the money, clays and kaolins should be well delaminated before use.

Outwardly, delaminators look like normal extruders, but they have very different internals.  Split augers and stators repeatedly cut and shear bodies at low shear rates and high pressures as they move through the device.  Slightly higher rpms, which may enter the dilatant regime, can also help this process.  Although this process performs some mixing, the low shear/high pressure environment optimizes delamination -- not necessarily mixing.  That is, the goal is not to distribute kaolinite platelets homogeneously throughout the body, but to delaminate the books stacks and free each platelet to travel as an individual.

De-Airing Chambers

Some de-airing chambers in vacuum extruders are poorly designed.  They allow chunks of body to remain in the vacuum chamber for extended periods, so they eventually dry out, break off, and fall back into the batch.  Dry chunks then function as rocks in otherwise uniform plastic forming bodies.  These may not cause extrusion problems, but if shrinkage is complete in these dry chunks, differential shrinkage and cracks can occur within the production bodies.  Once again, the size of the die openings relative to the sizes of these chunks can cause problems.  Dry chunks like these may not cause any problems during the extrusion of large wares (like bricks.)  But dry chunks may cause severe problems in the extrusion of small, fine, and/or intricate shapes (like catalytic converter substrates, for instance.)

Some de-airing chambers are poorly sealed so air leaks in and de-airing operations are not efficient.  Some operators have attempted to solve this problem by increasing the size of the vacuum pump rather than by simply fixing the seal.

Why use vacuum extrusion in the first place?  The goal is to pull all of the air out of the body so the body can be compressed by the main extrusion auger without hindrance from internal air bubbles.  Any compressed air bubbles that remain in the extruded product can expand as the body exits the die. 

To de-air properly, the requirement is that vacuum chambers, pumps, and seals are ALL functioning well.  Any body buildup in the vacuum chamber is a direct signal that something is wrong with the extrusion process and that extrusion defects can be expected.

Remnant Cut Planes in Column

Some systems use a variety of devices between the end of the extrusion auger and the die.  Egg beaters, expansion/contraction chambers, etc., have been used.  And then of course, mandrels used to extrude hollow shapes must be mounted inside the extruder, ahead of the die.  Spider legs in mandrel mounting rings cut through the body, as do some of the devices such as the egg-beaters mentioned above.  Two questions need be asked here:  (1)  Will any remnant cut planes remain by the time the body reaches the die?  (2)  Has the body been sheared sufficiently ahead of the die to heal any remnant cut planes?

The end of the extrusion auger can leave a spiral cut that may extend through the die and into the extruded column.  This frequently shows itself as an "S"-crack in the dried extrusion.  As the body flows over the end of the auger, a spiraling cut-plane is formed.  Unless the body is forced to shear somewhere between the end of the auger and the die so this cut-plane can knit and heal, a weak, remnant spiral plane can extend into the extruded column.  Cuts like this almost always close under the pressure, but that does not mean that they knit or heal.  Without shearing action to knit and heal such cuts, weak remnant cut planes passing through the die can later crack.  This frequently occurs in large extruded blanks that are dried for later turning on lathes.

Sometimes mechanical devices (such as the 'egg beaters' mentioned above) are installed in the final chamber between the auger and the die to perform some final mixing, homogenization, and/or delamination and disrupt and heal the cut left by the auger.  But these too cut through the body and leave their own cut-planes.

Expansion/contraction chambers between augers and dies that cause bodies to shear are the best way to knit and heal remnant cut-planes.  This can take the form of a large chamber following the auger, which then constricts the flow through a small opening which then again opens into a large chamber before once again constricting to feed the die.  The shear imposed as body flows through a small opening between the two large expansion chambers can be sufficient to knit, heal, and eliminate remnant cut-planes. 

Remnant Shrinkage Inhomogeneities

Most pre-die processing inhomogeneities will show as shrinkage variations in extruded products.  Clay bodies are known to have 'memories'.  That is, they remember what has happened to them in earlier processing.  Forcing a clay body through an oval constriction prior to the final round die may produce a round extrusion, but when the drying and firing operations are completed, the extruded shape may no longer be round, but once again show an oval shape consistent with the earlier oval constriction.  When extruded products do not retain their desired shapes, problems can frequently be traced to flow patterns ahead of the die.

Dilatancy and Finite Element Predictions

Some finite element analysis (FEA) software packages are used to predict extrusion, injection molding, and general flow behaviors of high viscosity bodies.  Most such FEA packages, however, unless recently modified, were designed to predict flows of homogeneous plastic materials which DO NOT CONTAIN ANY PARTICLES.  Most such models assume highly viscous, but nevertheless simple fluid properties. 

Why is this important?  Without suspended particles, dilatancy does not occur.  So most FEA programs used to predict extrusion behaviors DO NOT take suspended particles into account -- and they therefore cannot predict dilatant behaviors or associated processing problems.  This is important because all systems which contain particles can exhibit dilatancy.  Many don't, but they all can.  How high are body solids contents?  How high are production shear stresses?  And most importantly, how high are production shear rates?  When software packages ignore the presence of suspended particles and the types of rheology suspended particles produce, dilatant behaviors and dilatancy-caused production problems will never be predicted.

Dilatancy causes many production problems, so it is important to know the conditions under which it can occur.  When software packages ignore the presence of particles, they cannot predict such problems.  Process conditions and rates that are exactly similar to those that allow successful injection molding and/or extrusion of plastics may be disastrous for the injection molding and/or extrusion of ceramic bodies. 

The shape and size of flow channels in the ware also affect these phenomena.  Plastic may flow well through small openings to produce fine, delicate shapes, but bodies containing suspended particles can exhibit dilatant blockages in these same shapes.  Once a channel is blocked, open volumes downstream of the blockage will not be filled -- which means the part will end up in the scrap bucket.

This author is the first to sing the praises of computer models, but only those models that include all applicable phenomena.  Sometimes, phenomena are not included because they aren't considered to be important,  Some phenomena aren't included because they are unknown.  Particle/particle collisions are known to occur.  They are important in the processing of particulate suspensions so they should be included in any predictive models.  Computer models which ignore particles and dilatant effects (which includes the vast majority of models) are not completely applicable to most ceramic forming bodies.  Their predictions should be very accurate for plastic extrusion and injection molding systems.  When conditions are right (that is, when flow rates are relatively low), their predictions will be very accurate for high solids particulate forming bodies as well.  But as flow rates, shear rates, speeds of processing, or solids contents increase, or flow channel sizes decrease, their predictions can be totally off base for high solids particulate forming bodies.

Summary

Extruders are designed to feed body into the auger system, de-air in vacuum extruders, compress body in the main auger, and force it through production dies.  Just as some people believe ball mills can function as mixers, many believe extruders also function well as mixers.  Ball mills are called "mills" because they perform milling -- otherwise, they should be known as "ball mixers."  Extruders are called "extruders" because they perform extrusion -- otherwise, they should be called "mixers."  Certainly, each of these types of equipment performs some mixing, but that is not their primary purpose.  To expect them to perform excellent mixing as they perform their design function is wishful thinking at best.

Materials fed into pug mills and extruders should be well-mixed at the desired solids content.  Neither mixing nor adjustment of solids contents are efficiently performed when high solids forming bodies are fed into pug mill/extruder systems.

De-airing chambers in vacuum extruders should be well-maintained, unclogged, and free of leaky seals. 

Cuts created within extrusion bodies between the auger and the die should be well-sheared to knit and heal the cuts.  Otherwise, cracks can form at remnant cut planes carried through to extruded wares.

Cross-sections of expansion/contraction chambers between auger and die should be round (rather than square or oval) to minimize flow inhomogeneities that will cause shape deformations in extruded products.

Finally, when finite element packages are used to predict flow behaviors of ceramic bodies, they should take particle/particle collisions into account.  Otherwise, they may work well for gentle flow conditions (low solids contents, low shear rates, etc.), but they will not properly predict the onset of dilatancy in the more intense conditions which typify the production of high solids ceramic forming bodies.

Miscellany

Suggested topics for future issues of this E-zine .... Please continue to send your ideas or questions for future topics.  Thanks.  Until next time ...

Thanks for signing up to receive Ceramic Processing E-zine
To unsubscribe, click the "Remove Me" link below.

Copyright © 2005  Dennis R Dinger

103 Augusta Rd, Clemson, SC 29631   (864) 654-5731

consulting@DingerCeramics.com

www.DingerCeramics.com

All Rights Reserved.

All Rights Reserved.