Castool Community

In 1996 I wrote an article called, “Who are the Superextruders?”  Who are the extruders  whose average rate of productivity clearly exceeds that of almost all others who are making similar product on similar presses?

The answer, of course, was those few extruders for whom having the correct temperatures all the time, every time, was almost an obsession.

That article received more favourable attention than anything I had previously written, and its catch phrase, “Temperature, Temperature,  Temperature!” is still often quoted today.

In my travels , visiting extrusion plants in all parts of the world, the variance in productivity between plants is quite apparent.. For example, am average extruder with an 8” press extruding 6063, produces about 3500 net lb per hour (1587 kg per hour) whereas a Superextruder usually makes about 5300 lb/hr (2404kg/hr), over 50%more.

The principal factor in successful extrusion is temperature establishment and management, and the maintenance of consistent temperature conditions.  Ideally we preheat the die uniformly to the billet operating temperature which for 6XXX commodity extrusion will be 860ºF(460C).  That is, an 860ºF(460C) billet temperature will allow extrusion without undue acceleration delay.  We should also be extruding in the preferred extrusion ratio range of around 40 – 80.

Some extruders choose to use higher billet temperatures or higher die temperatures to start up, either because they have inadequate temperature management, or they are trying to extrude at too high an extrusion ratio, or both.

In a perfect world, to achieve near perfect temperature management and consistency, it could be argued that in every instance we should preheat bolsters to avoid the initial chill when a fresh die ring assembly is transferred from a correctly preheated environment and placed next to a cold bolster.  Usually, however, we can get away with it for most dies.  An exception is with dies known to be hard to push, such as some difficult hollow dies.  In this instance, it is preferable to preheat a bolster to around 500ºF(226C) to reduce any heat loss and avoid problems pushing, and also the need to preheat initial billets any higher than 860ºF(460C).

A billet preheat temperature of greater than 860ºF(460C) should only be considered if, when in steady state extrusion, there is an acceleration delay of more than 3 – 4 secs.  After starting at 860ºF(460C), the billet temperature should be adjusted to allow breakthrough at peak pressure without significant acceleration delay.  The job can then be run at a speed that generates the ideal platen exit temperature for the alloy being used. For example 1050 – 1100ºF (476 – 500C) for 6063 alloy.  These practices should allow ram speeds of no less than 26 ipm for any 6063 alloy extrusion.

Of course, all of this assumes the following:  Reliable billet preheat and a capacity to achieve 860ºF (460C) at the required billets/hr throughput.  Reliable die preheat (such as Castool single cell die heaters.).  Adequate control of temperature losses in the die stack (Offset container heating – Castool QR technology).  The die design is correct for the target ram speed.

This finally brings us to another aspect of thermal management – quenching.

If we can control the process temperatures to create acceptable product at a ram speed of 26 ipm or more, we must be able to satisfactorily cool the extrusion to achieve both the required mechanical properties and also the required profile, without unacceptable distortion.

Yes, there still are Superextruders who can do it all.

Worst Practices Checklist

• Press is not precisely aligned.
• Die is not uniformly and adequately preheated.
• Die is too strong.
• Die bearings are badly oxidized.
• Billet is poorly cut, surface is not clean.
• Billet is insufficiently taper heated.
• Dummy block is no longer contracting.
• Container is overheating.
• Container liner exit temperature is vertically inconsistent.
• Too much Dag is being used.

Much has been written about Best Practices for the extruder. A number of major multi-plant extruders already have Best Practices manuals. These are usually very detailed, and are meant to ensure that all their facilities anticipate anything that may prevent 100% quality and maximum productivity. The obverse to this is Worst Practices List. This includes common but avoidable problems in production system between the billet and the die. From this preliminary list, an extruder may identify some of the areas in his process that can be improved. Few extruders can honestly claim to have none of these problems.

It’s Not Just the Die

The die is the heart of the extrusion process and, until fairly recently, it was the main focus of the extruder’s attention. Now however, many die makers can provide dies that will make perfect product from the first push but only if the alloy is at optimum temperature for maximum speed as it enters a properly preheated die. The prime focus of extruder is now on improving the efficiency of his production process.

No attempt has been made to prioritize these problem practices since their real importance and frequency is impossible to quantify. For the extruder who is sincerely committed to ongoing improvement, concentrating on the basic purpose and function of each component involved in managing the temperature of billet, and utilizing the state-of-art technology currently available, is a certain formula for immediate improvement in both productivity and profit.

In discussing the function and effect of different parts of the extrusion process for the purpose of improving efficiency, it is advisable to avoid evaluating any part individually, without regard for its interaction with other components. Maximum productivity can only be achieved if all parts of process work together as a coordinated interactive system.

Overcoming Worst Practices

Press is not precisely aligned: Press alignment should always be the first item on any list of extrusion practices. Good extrusion depends on all components of the press being physically in precise alignment, and the die being mounted exactly in center of the container. If this is not done, good extrusion is impossible. Regular inspection at operating temperatures is essential, with emphasis always on pre-venting rather than correcting misalignment.

Die is not uniformly and adequately preheated: The die is usually designed to already be completely at operating temperature when the first push begins. If it isn’t, a perfect profile is usually impossible until one or two billets are wasted in heating the die. The answer to this problem is single-cell die oven. This will bring the die quickly and uniformly to operating temperature. To avoid the initial capital expense of complete battery of single-cell ovens, dies may be held at a moderate temperature for some time in a traditional chest oven, then the necessary heating quickly completed in a single-cell oven when the die is needed.

An extruder today should be able to assume that his die will produce good product immediately, and concentrate on optimizing his production process.

The die is too strong: Anything that prevents the die from creating good product at maximum speed and with minimum scrap is counterproductive. Unfortunately the die maker usually does not have the luxury of making a perfect die for perfect operating conditions. In real life he must provide a die that is best suited for its anticipated actual use.

If the die maker knows that die will not likely be uniformly at operating temperature before the first push, he must make it strong enough to withstand the resulting high breakthrough pressure. Press speed can then never be maximized. A strong die is a slow die.

If the die maker knows that the die will be uniformly at operating temperature before the first push, the break through pressure may be reduced by 30-40%. A lower breakthrough pressure allows cooler billet temperatures, and thus greater press speed.

The die corrector used to modify dies primarily to bring the profile to the required tolerances. The integrity of profile can now usually be taken for granted. The die corrector’s prime function now is to provide feedback on temperatures and breakthrough pressures to help the die maker to provide more productive dies.

Very high breakthrough pressure, for example, can bend die, and cause the core to deflect and distort the profile, once this danger has been understood and included in the design equation, large, thin, complex shapes that were previously thought impossible to efficiently extrude now become viable.

Die bearings are badly oxidized: When a die is held too long at or near operating temperature in chest oven, the bearings will oxidize. A satisfactory finish cannot then be obtained on the extruded product. The solution to this problem is, of course, the rapid and controlled heating of single-cell die ovens.

Billet is poorly cut, and surface is not clean: To avoid the air temperature and blisters from poorly sheared and two part billets, logs can now be precisely cut with an in-line narrow-cut saw, then welded together before being automatically cut into billets. When the end of the current log is detected, a new log from the magazine is positioned in the cutting line. The logs are then locked firmly in place, and their ends welded together. The joined logs then pass through the cutting and loading process as if no weld existed.

Billet should always be kept clean, because the skin may be inadvertently carried into the product. Scrap will inevitably result.

Billet is insufficiently taper heated: The friction of the die bearings causes heat to be increasingly generated in the alloy as it is pushed through the die. In order to achieve isothermal extrusion, that is, to allow the alloy to pass through the dir at its maximum operating temperature and speed at all times, the billet must be initially heated to a temperature that reduces from front to back in order to compensate for this heat of friction.

Taper heating the billet can best be achieved by electrical induction heating. To combine the economy of gas heating with accuracy and repeatability of induction heating, billets may be first preheated to a base temperature in a gas-fired oven. The hot billets are then transferred to an auxiliary induction billet heater where multiple separately controlled heating zones are programmed to quickly and accurately provide the necessary taper heating. Once the taper heating program for any shape has been confirmed by both calculation and experience, it can again be successfully used for even a single billet. If the temperature of the billet is not adequately tapered before extrusion begins, maximum press speed is impossible.

Dummy block is no longer contracting: For the dummy block to work properly, a thin film of alloy must remain between the block and the container liner at all times during the extrusion process. Its thickness should be uniform. With a soft alloy, the clearance that creates this film be only about 0.006 in. If the clearance is more, the alloy will penetrate the gap in the first push. If much less, this essential film of alloy will be stripped from the liner. Stripping the film of aluminum off the liner results in scrap due to blisters, and also to inferior alloy being carried into the extrusion, instead of being discarded in the butt.

An effective dummy block must expand quickly under load. It must separate cleanly from the billet at the end of the stroke, then contract immediately and return through the container without stripping the film of alloy from the liner. The measure of real value of a dummy block is its ability to continue to contact fully after an unusually large number of pushes, before it takes permanent ser and no longer contracts. The operating lives of contemporary dummy blocks can vary widely.

Container is overheating: If the container temperature sensors and heating elements are not close to the liner, overheating can easily occur, Heating elements, unless properly controlled, can reach temperatures of 1 ,300° – 1 ,400°F. The container mantle is usually of 4340 steel, which may begin to temper and soften at 1,000°F. If the mantle softens, bellying of the liner will likely occur. This will allow a buildup of impurities from the billet skin that will eventually end up in extrusion. Scrap will result.

Container liner exit temperature is vertically inconsistent: At the die end of the container, the temperature of the top of the liner is usually considerably higher than at the bottom. This is caused primarily by the heat rising inside the container housing. The result is that the alloy entering the die at the top is less viscous than at the bottom, and therefore flows at a greater velocity through the upper apertures,

A rule of thumb is that every 10°F difference between the top and bottom of the liner will cause 1% difference in the runout length. On a long table, unless the upper doe apertures are choked, major problems will occur, especially when using a puller. The solution is for the container thermal control system to have top and bottom as well as axial temperature control zones.

Too much Dag is being used: In the billet delivery system, the final factor is the introduction of lubrication. Ideally, the dummy block would pass smoothly through the container, and at the end of the stroke, the butt would fall off. Unfortunately this doesn’t always happen.

Too much lubrication has always been anathema to extruders. The old saying used to be, “Use no lubrication, then wipe off any surplus.” We have learned much about extrusion since then, and much about necessity and the effective use of lubrication.

At the end of each extrusion cycle, the fixed dummy block must separate instantly and cleanly from the butt, without pulling the extruded section from the die and also without breaking the mandrel or stud in the dummy block. Sticking can be a serious problem. It is essential, therefore that both the dummy block and the billet are properly lubricated to provide immediate and effortless separation.

Effective lubrication ensures instant and clean separation of dummy block from the butt. It also ensures clean butt release from the shear. It keeps the container seal face clean and free of alloy, and reduces scrap due to blisters.

Powder or liquid boron nitride, developed specifically for light metal extrusion, is today universally considered to be the ultimate lubricant.

Conclusion

Over the years, the buying practice of most extruders has gradually evolved. At one time relationship buying was common practice. The extruder’s supplier very often became a personal friend. He then usually had an opportunity to meet and competing supplier’s price. Next, almost all tooling was treated as commodity, and price took precedence over all else. Now, the astute extruder is making every effort to measure the real value of his purchases. He understands the importance of the interaction between some components, the value of undivided responsibility whenever possible, and the need for a detailed and tight specification to ensure that competing suppliers will provide products of at least equal value.

The successful extruder’s focus now is on improving profit by improving productivity. We can no longer afford any poor practices.

Key Temperatures

Tempering Temp. of  Common Materials

H-13       46-48 HRC            585°C (1,085°F)
4340       34-38 HRC            540°C (1,000°F)

Max. Billet Temp. 485° to 500°C (900° to 930°F)

Max. Exit Temp. 570°C (1,085°F)

Alloy Melt Pt.

6063       600° to 650°C (1,110° to 1,200°F)
2024       500° to 640°C (930° to 1,185°F)
7075       475° to 649°C (885° to 1,185°F)

Heating Elements

Not Controlled 700° to 760°C (1,300° to 1,400°F)

Torch (Open Flame)

Not Controlled 3,000°C (5,000°F) Plus

Editor’s Note: After reading the most recent issue of Light Metal Age which included an illustration of a press operator with an open torch near an extrusion die, Paul Robbins sent this quick guide to key temperatures. We are including this key as part of Robbins’ instructive extrusion practices article to encourage better practices in the extrusion industry.

Castool’s Robotic Die Expediter schedules dies, accurately and safely heats dies, and compliments existing press practices.

The goal is to have the temperature of the die the same as the billet during the first push. Only then can we accurately predict break through pressure, and optimize die design.

Dies lose 5 degrees C every minute in air, and 10 degrees C every minute in the die slide. Break through pressure is increased by 1% for every 5 degrees C the die is below billet temperature.

The RDX monitors the die temperature from the moment the die is removed from the cradle by the press operator, until the ram pushes the first billet.

In extrusion, as in most other industries, there are some theories that seem logical and therefore often remain unquestioned.  There is also some misinformation that calls for debunking.

A useful approach is, “If it sounds too good to be true, it probably isn’t.”

The following facts have been proven, and can be documented.

Regarding the Heating of  Dies

Fantasy: Several years ago, a misguided but enthusiastic supplier actively promoted the use of microwave to quickly heat extrusion dies from within.  He even claimed to reduce energy costs by using only the part of the infra-red spectrum that heats H13 steel.

Fact: A microwave oven generates heat by dielectric heating.  It uses microwave radiation to heat water or other polarized molecules.  This obviously does not include H13 tool steel.

Fact: Only the surface of the die is heated by radiation, the interior of the die is heated by conduction.  Bringing the die completely and uniformly to the required steady state operating temperature can only be accelerated by increasing its surface temperature, and allowing it time to stabilize.

Fact: There are two main limiting factors in heating dies.  The first is the danger of softening the die by heating it beyond its annealing temperature.   For H13 steel, this is 1085ºF.  The second factor is the danger of dissipating the nitride.  This can begin at about 1000ºF.

Fact: The time required to heat any die can be accurately calculated from the following factors – surface area, mass, material specific conductivity, rate of energy input in KwHrs, rate of heat loss from the heating chamber.

Fact: The rate of heating depends primarily on the conductivity of the steel used.

Assurance

All the facts shown have been proven to be true, under standard operating conditions.

I will welcome any questions or comments.

Note: This is a 3 part blog, part 2 tomorrow and part 3 Wed.

In extrusion, as in most other industries, there are some theories that seem logical and therefore often remain unquestioned.  There is also some misinformation that calls for debunking.

A useful approach is, “If it sounds too good to be true, it probably isn’t.”

The following facts have been proven, and can be documented.

Regarding Container Temperature Control

Fantasy: The increase in temperature caused by the friction of the alloy passing through the container liner is responsible for overheating and annealing the mantle, and the eventual bellying of the liner.

Fact: Container mantles cannot overheat solely from heat inflow from the billet – even at a billet throughput of 3 tons per hour on an 8” press.

Fact: Container mantles overheat and anneal, and liners belly, due to inadequate temperature control systems.

Fact: Isothermal extrusion is best achieved by taper heating the billet before it is placed in the container.  Some speed control will also assist in maintaining constant exit temperature.

Fact: Mainly due to temperature build-up within the container mantle/housing, the alloy exiting the die at the top is usually hotter than at the bottom.  This is particularly true of large containers. Since the viscosity of the aluminum alloy is reduced as its temperature increases, allowing it to flow more easily, any increase in temperature will be directly reflected in variable runout lengths with multi-hole dies, or shape issues with high vertical aspect profiles.

Fact: Accurately predicting and measuring the relationship between the viscosity of aluminum alloy and its temperature is difficult.

Fact: A practical rule of thumb used by some major extruders, is that a 5-10ºF increase in billet temperature may result in at least 1% increase in runout.  If the difference in temperature between two vertically separated apertures in a die is 40ºF, the difference in runout could therefore be more than 4%.  On a 200 ft. runout for example, an 8 ft. difference in length can certainly cause problems for the extruder.

Fact: Container temperature control cannot replace taper heating the billet for isothermal extrusion.  However, if zoned vertically as well as axially, an effective thermal control system can eliminate any bottom to top increase in temperature in the die, and consequently eliminate an unacceptable increase in runout.

Fact: As it is often the most costly component of an extrusion production system, the operating life of a container and liner is important to any extruder.  Containers almost never fail because of inadequate material, nor because of excessively high operating temperatures, containers usually fail due to overheating during preheating, as a result of inadequate temperature control systems.

Assurance

All the facts shown have been proven to be true, under standard operating conditions.

I will welcome any questions or comments.

Note: This is a 3 part blog, part 3 tomorrow.

In extrusion, as in most other industries, there are some theories that seem logical and therefore often remain unquestioned.  There is also some misinformation that calls for debunking.

A useful approach is, “If it sounds too good to be true, it probably isn’t.”

The following facts have been proven, and can be documented.

Regarding Dummy Blocks

Fantasy: When cleaning the container prior to changing alloys, if the cleanout block removes no aluminum from the liner, it indicates that the dummy block is operating effectively.

Fact: Wrong! If the cleanout block removes no aluminum it indicates that either the dummy block is not expanding and contracting properly, or that the relationship between the container, dummy block and clean-out block diameters is not correct.

Fact: One of the functions of the expanding dummy block is to expand quickly under load, and to maintain a secure seal with container wall leaving only a thin film of alloy on the liner.  The front of the shell of the dummy block should reach full expansion just before the billet is upset, so that almost all of the trapped air will be evacuated.

Fact: Expanding the dummy block is not difficult.  Ensuring that at the end of the stroke it will immediately contract to its original diameter can be a problem.

Fact: As long as the stress on the lip of the dummy block shell remains within its elastic range, it will continue to expand and contract satisfactorily.  If, however, the stress exceeds the yield strength of the material, the dummy block may no longer contract.  This will result in it stripping the skin of alloy that coats the liner and seals the dummy block.

Fact:t If no alloy is left on the liner, there is no longer a secure seal.  Entrapped air will result in scrap due to blisters.  Inferior alloy from the skin of the billet will be carried into the product instead of being discarded with the butt.  There will also be excessive wear on both the dummy block and the container.

Fact: Excessive stress can be avoided by precisely limiting the extent to which the tapered mandrel enters and expands the dummy block.

Assurance

All the facts shown have been proven to be true, under standard operating conditions.

I will welcome any questions or comments.

The most valuable knowledge an extruder can have is a thorough and accurate understanding of the extrusion process.  This statement may appear patently obvious, but many extruders still consider both die design and the extrusion production process to be an art as well as a science.  This is no longer true.

There was once a time when the extrusion of aluminum was perhaps almost as much an art as it was a science.  Usually the die designer learned his trade from his own experience.  This was an empirical process.  What worked for him before was likely to work again.  He didn’t really understand or use the laws of physics that govern the flow of metal through the die.  His success depended on his experience, his talent, and on the close cooperation of the extruder’s die corrector.

Those days have gone forever.

Extrusion is basically a simple process.  If the die is properly designed, and impeccable alignment, thermal as well as physical, maintained throughout the production process, good product should result from every billet.  This doesn’t always happen.  But it no longer depends on the personal skills of the die designer, the die corrector and the press operator.  Over the years, light metal extrusion technology has improved to an extent that now, with computer-assisted die design and die cutting, the extruder should be able to get good product from the first billet every time.  If he doesn’t, and his alignment is correct, he may be well advised to consider another die maker.

I have been asked a few times this week about container to die alignment.

Container to die alignment should be controlled to be within +/-2.5mm.  This can best be checked and monitored by not shearing the butt from the die on the last billet of any order.  After allowing the die and butt to cool to room temperature the distance from the edge of the die to the edge of the butt can be measured in each of the 12, 3, 6 and 9 o’clock positions.  Knowing the top of the die when it is located in the press, misalignment and direction of misalignment can be easily reported to maintenance department for correction if out of tolerance.  This should be done on both die slide pockets, and at a frequency depending on the extent of the problem.  Initially it can be done daily for each die slide pocket. Establish SPC control charts for each die slide pocket.
This is an important practice to overcome poor performance from dies, and avoid correcting dies that don’t need correction.

For any extruder, the three basic rules of productivity are:

First, fill the container as much as possible.

Second, empty the container as fast as possible.

Third, repeat rules one and two as often as possible.

In other words, productivity depends on container utilization,

ram-speed, and contact time.

If the Extrusion Ratio is in the ideal range of 40 – 60, it should be possible to utilize most of the container length most of the time.

Cut length at the saw will restrict using full length billets most times, but you should be able to operate within 75% at all times.

An opportunity exists for most presses to increase the container length, usually by about 3 inches (75 – 100 mm).  There’s usually enough available daylight to do this.

And remember that it’s always possible to improve your productivity

We continue to be asked questions about shear blades. The good news is that extruders are starting to understand the importance of a good die shear. We have categorized shears blades into 3 categories based on butt thickness.

A sharp knife-edge blade works best for soft alloys such as 1XXX and 3XXX
alloys.  The tendency is to run to very short butts (because of minimal billet shell zone there is no need to leave a longer butt) and the butt tends to naturally curl with a simple knife-edge blade.  There is therefore no need for a radius to encourage curling – in fact because of the stickiness of these alloys it is arguably best to avoid prolonged contact with the blade.

Hard alloys like 2XXX and 7XXX need to run to longer butts (typically 15% of the upset billet length).  Because of this the butt will never curl and it is necessary to simply shear the butt from the die entry.  A sharp blade doesn’t promote a straightforward shear, and therefore typical blade designs for these hard alloys are more of a blocky shear design.  But do not encompass all 7XXX into this category.

The leaner 7XXX alloys such as 7003, 7005, 7020 fall into the same group as the harder 6XXX alloys such as 6061, 
are run to shorter butt lengths typical of 6061 and therefore can use the scoop type design.  Indirect extrusion of hard 2XXX and 7XXX alloys should be employing much shorter butt lengths than for direct extrusion, and in this instance it makes good sense to use the typical scoop type blade.

So in essence, the most effective blade design tends to more depend upon the butt length used rather than the alloy itself (of course, the alloy dictates the butt length).  It’s worth noting that some plants employ rather long butt lengths to some critical 6061 alloy products (e.g. automotive), and do tend to use the blockier blade design to ensure a good shear.  However having said the alloy itself is not instrumental in driving the optimum blade design, in the case of really soft 1XXX and 3XXX alloys, alloy stickiness does come into reckoning and it’s best to keep away from the curved scoop type blades.