Subscribe Us

header ads

Surface coatings to increase the tool wear life

 

Surface coatings to increase the tool wear life

Producing intricate shapes quickly with excellent precision and good surface quality at a low cost is one of the advantages of die casting. But, maintaining this advantage is difficult. A die casting die can cost more than the machine that operates the die. Die casting is hard on dies and die components and pulling the dies and pins for repair and/or to polish them is costly.

TD COATING

To help maintain the advantages of die casting and production up time, die life needs to be extended by reducing repairs. A tool treatment process called thermal diffusion (TD) can help die casters reduce maintenance and repairs and extend core and core pin life.

The TD process prolongs die life by decreasing the damage caused by thermal and mechanical cyclic loadings and the reaction with active cast materials that severely damage dies and die components. It reduces the soldering, corrosion, erosion, wash-out, wear and heat checking from the loadings and reactions with other materials. It also provides excellent peel and adhesion strength. Die life is significantly extended and repairs and polishing are reduced.

The process also helps protect aluminum die casting dies which are especially vulnerable to die casting problems.

One of the reasons it works effectively in die cast operations is found in the process itself. During the treatment process, the substrate surface of the core and core pins is modified, creating a metallurgically bonded non-porous vanadium carbide (VC) layer.

The VC layer is infused into and onto the dies and die components (Fig. 1 & 2). This VC layer is .0001- to .0006-inch thick and has a hardness of 3200 Vickers at room temperature.

Even under the high working temperatures used in die casting, the VC layer retains its hardness level. And, once the die is cooled to room temperature, the original hardness of VC is restored to previous levels.

For example, at 800° Centigrade, the hardness of the VC layer drops to approximately 900 Vickers, which is still above the hardness of other treated surfaces. As the VC layer cools to room temperature, it returns to near its original hardness of 3200 Vickers. Its ability to maintain its hardness under high temperature applications allows it to perform well in casting operations.

The process also is effective in die cast operations because the hardness of the VC layer is higher than abrasives such as aluminum oxide, which is found in hone stones and sandpaper used for polishing cores and pins. This allows any aluminum that mechanically sticks to a pin to be easily removed without significant damage or dimensional change to the pin. Polishing time and the resultant wear on the pin are greatly reduced.

 

What is TD?

Developed in Japan in 1969 at the Toyota Central Research and Development Laboratories, Inc. by Dr. Tohru Arai, the TD process is a high-temperature tool treatment. During the process a nonporous metallurgically bonded vanadium carbide (VC) layer is diffused into and onto tooling substrate. The VC layer is .0001- to .0006-inch thick and has a hardness of 3,200 to 3,800 on the Vickers hardness scale. The VC layer retains its hardness under high temperature working conditions like die casting.

Because the VC layer is diffused into and onto the tool substrate, it significantly reduces wear, soldering and corrosion and provides superior peel strength and adhesion strength. It extends the life and performance of core pins, dies, punches and tooling by 5 to 50 times and more.

The process is effective with air-hardening cold and hot working die steels such as A2, D2 and H13, high speed steels and cemented carbides. The steels should have a .2 percent or greater carbon content.

In Japan, the treatment process has been shown to be especially effective with conventional hot working die steels such as H13, a tool steel typically used to make die cores and core pins (see Fig. 3). Damages caused by die casting were significantly reduced. The results from European die casters tooling treatment have been detailed in several European publications as well as Die Casting Engineer.2

By using treated tooling, the Japanese also realized other benefits. They reduced interruptions in the production process, helping to eliminate the reconditioning of the casting condition. This reduced cast scrap and energy usage.

The surface quality and dimensional accuracy of cast products were also improved. Using the treatment process meant less sticking of cast materials on pins. The improved surface quality and dimensional stability of the cored surfaces also allowed thread tools to last longer.

 

Benefits from Using TD

  • Improves pin life 10 times or more by reducing corrosion and erosion damage which can cause pin breakage
    during cast ejection.
  • Reduces polishing work which can damage pin surface.
  • Increases casting shots several to ten times or more in comparison to polishing to remove stuck aluminum.
  • Drastically reduces cleaning operations to remove pickup.

TD COATING
Decreases die repair work and soldering since it reduces the tendency of aluminum to mechanically attach to die surface. When soldering occurs, adhesion can easily be removed by polishing with a scotch-brite pad or emery paper without taking die sets out of production.

Production Evaluation of Coatings and Surface Treatments for Die Casting Dies

 Interaction between the die surface and cast metal leads to both mechanical and chemical degradation of the surface. The mechanical degradation is caused by the high speed impingement by the melt particles of the die surface near gates. This type of die wear is called washout. The chemical affinity of the cast metal atoms for the die surface leads to the formation of intermetallic layers at the interface. This phenomenon is called soldering. Production experience has shown that oxide layers and other protective layers on the die surface successfully retard the onset of soldering and washout of these surfaces. There are many surface engineering techniques and a wide range of coatings available in the market which are potential candidates for soldering and washout prevention. The quest is to identify suitable candidates to combat the harsh die casting conditions. As part of this quest, production evaluation campaigns are being held at die casting companies with varying matrix of die wear problems. These production campaigns will help pinpoint the best surface engineering technique and composition for maximum die life increases in a predetermined production campaign. The first production campaign was held at Pace Industries in Monroe City, Mo. In this campaign, the use of coatings (chromium carbide, chromium nitride and vanadium carbide) significantly reduced the soldering tendency and the soldering-related downtime was totally eliminated for the production run of 180,000 shots. The results of this campaign were reported in an earlier paper published in the 1997 NADCA Transactions. This paper contains details of the second campaign that was run at Premier Tool in September-December 1998.

Premier Tool & Die Cast Corporation in Bemen Springs, MI, was selected as the ß-test site for the production evaluation campaign presented in this paper. The objective of the production evaluation campaign at Premier was to evaluate promising coating compositions and application techniques for their efficacy in preventing the occurrence of soldering on the core pins in the two 16 cavity dies for suspension mounts. This campaign was designed to involve both surface treatments and surface coatings.

Production Evaluation/Process Conditions

Premier offered a challenging testing environment in its high volume 16 cavity die. The die was used to produce automobile suspension mounts. The problems associated with the 16 cavity die were not solely soldering-related. There were issues regarding cooling, gating and load bearing capacity of the ejector pins too. Parts were sticking to the ejector half. The cores were snapping at the head. Soldering on the cores was severe especially near the parting line. The cores were removed 1-2 times every shift for polishing. 1-4 cores broke per run. The die had to be removed every 5,000 shots for die maintenance.

Premier addressed some of the above problems by incorporating design changes. The dimensions of the core were modified to eliminate the breaking of the head. Heating and cooling lines were provided within each core to reduce the occurrence of potential hot spots that would promote soldering. The dimensions and position of the ejector pins were modified to facilitate ejection.

Coating/surface treatment selection criteria


Premier, in its 16 cavity die, offered many potential sites for severe soldering and washout problems. The cores on the cover half of Die 2 did not possess a cooling line and were highly prone to soldering. The geometry of the core pins was complicated with non-uniform dimensions that made efficient heat transfer tough. Hence an effective coating or surface treatment candidate, for combating the conditions at Premier, needed to possess the following properties:

  • Sound adhesion to the substrate
  • Sufficient hardness and toughness
  • Low chemical affinity for and low solubility in molten aluminum
  • Superior oxidation resistance
  • Good thermal and impact shock resistance
  • High thermal conductivity to dissipate heat from the interface quickly
  • Compatible Coefficient of thermal expansion with the
    substrate

A potential surface engineering technique should be able to deposit the coating uniformly in spite of the complexity of the geometry. This surface engineering technique could either be a coating deposition process or surface treatment process.

  • If material with desired properties is added to the surface, then the process is called a Coating Deposition Process
  • If the chemistry and/or micro-structure of the substrate of the base material is altered, then the process is called a Surface Treatment Process

Physical Vapor Deposition (PVD) coating

 

Physical Vapor Deposition (low deposition temperature) and Thermo-reactive diffusion process (high temperature deposition process) were identified as potential coating techniques. Nitriding and Carburizing were identified as potential surface treatment techniques. A duplex treatment was also included in the design to compare the performance of a surface textured PVD coating and a smooth PVD coating. The micro texturing of the substrate was provided by shot micro-peening. Three coatings, CrNx, CrxCy and BxC, were the selected physical vapor deposition (PVD) coatings and VC was the selected thermo-reactive diffusion (TRD) coating. CrNx (PVD) on shot peened substrate was the selected duplex treatment. Ferritic Nitro carburizing and ion nitriding were the selected surface treatments. Table 1 includes relevant information about these selected candidates.

 

PVD COATING

Physical Vapor Deposition (PVD) processes deposit coatings on a substrate atomistically, i.e. atom by atom. The material to be deposited is transported in the form of vapor, either through a plasma or vacuum, to the substrate on which the vapor condenses. The source for the vapor could either be thermal or non-thermal. These processes can deposit both single elements and compounds as coatings. The thickness of the coatings can vary from a couple of nanometers to a few millimeters. All PVD processes are line of sight processes. Usually, PVD coatings possess columnar structure which is not as good as equiaxed structure in combating liquid metal corrosion. The columnar grains provide a pathway for the molten alloy to diffuse through. Physical Vapor Deposition is classified into three types: evaporation, sputter deposition and ion plating. The production campaign had candidate coatings applied by either evaporation process or by sputter deposition.

  • Arc Evaporation: Vacuum evaporation takes places at gas pressure ranges of 10-5 Torr to 10-9 Torr. The coating material is in an electrically neutral state and is expelled from the surface of the source at thermal energies typically from 0.1 to 0.3 eV The substrate is preheated to elevated temperatures (200 to 1600°C) for dense and equiaxed grain morphology.
  •  
    1. CrNx: This coating was provided by Multi-Arc Inc which utilized the arc evaporation process to deposit the coating. In this process, a vapor plasma is generated by striking an arc between the solid cathode (target) and the arc source. The arc melts a small area (10 micrometers) of the cathode surface generating metal droplets (Cr), ions and large volume of free electrons. This vapor is highly ionized (up to 80%) and arrives at the substrate with high energies (50 eV). The substrate temperatures are in the range 200 to 550°C. Nitrogen gas is inducted in the vacuum chamber to create nitrides. Process times are of the order of 4 hours for a coating thickness of 6 microns.
  • Sputter Deposition: In this process, the substrate is deposited with particles vaporized from a surface, which is called the sputtering target. It is a non-thermal vaporization process where the coating material is dislodged from the surface of the target by momentum transfer from energetic particles which bombard the surface. The substrate is positioned in front of the target so as to intercept the flux of sputtered atoms. Sputter deposition can be performed in a vacuum or low pressure gas (<5 mTorr). Sputter deposition can also be deposited at higher gas pressures (5-30 mTorr).
    1. CrxCy: This coating was provided by Balzers Tool Coating Inc., which applied it by the e-beam sputtering process. In the e-beam evaporation process, the surface of the workpiece is bombarded with noble gas ions in order to remove contaminants and to sputter off some substrate material. These substrate atoms then condense with the coating element (Cr) which is then evaporated in the second stage. Reactive gas (carbon) is then introduced into the chamber, which combines with the chromium ions on the surface of the workpiece to form hard CrxCy coatings.
    2. BxC: This coating was provided by Diamond Black Inc. It was applied by magnetron sputtering process, which is performed at low temperature (250°F). The coating was vacuum sputtered to a thickness of 0.00008" or 2 microns at 250°F.

Thermo-Reactive Diffusion Process

 

In the TRD process, the carbon and the nitrogen in the steel substrate diffuse into a deposited layer with carbide-forming or nitride-forming elements such as vanadium, niobium, tantalum, chromium, molybdenum or tungsten. The diffused carbon or nitrogen reacts with the carbide and nitride forming elements in the deposited coating so as to form a dense and metallurgically bonded carbide or nitride coating at the substrate surface. The possibility of distortion is present with this high temperature process. Dimensional changes due to the high phase transformations in the heat treatment of the base steel and the formation of the carbide layer are a good possibility. The coatings formed have a fine and nonporous composition. Though the diffusion layer is thin, it is very dense and shares a sound metallurgical bond with the substrate.

VC: This coating was supplied by Arvin TD using the TRD process. The high temperature salt bath TRD process was performed in a molten borax bath at 850 to 1050°C (1560 to 1920°F). Immersion time ranges from 0.5 to 10 hours to obtain an optimum carbide layer thickness of 7 -10µm.

 

Duplex Treatment

 

Metalife+ CrNx; The duplex coating was formed by first shot peening (micro-shot) the substrate and then coating it with CrNx (Arc Evaporation PVD). The CrNx coating was applied by the arc evaporation technique by Multi-Arc. The shot peening treatment was supplied by Badger Metal. Badger Metal Technologies applied a patented micro-peening treatment (Metalife) on the core pins and the DME pins. The treatments for die casting are categorized by "T" processes (T10, T21, T41, T61 and the newer T71). The surface of the pin is impacted with special media, a temporary plastic flow of the metal (with penetration depths of 0.010 to 0.015 in. on 44 to 48 HRC surface) results in generation of compressive residual stresses inside the peened surface layer.

 

Surface Treatments

  1. Ferritic Nitrocarburizing: Ferritic Nitrocarburizing processes are thermochemical processes, which involve the simultaneous diffusion of both nitrogen and carbon to the surface of ferrous materials at temperatures completely in the ferriric phase field. The primary objective of such treatment is to improve the anti-scuffing characteristics of ferrous engineering components by producing a "compound layer" on the surface, which has good tribological properties. A single phase epsilon carbonitride compound layer is produced supported by nitrogen rich subsurface diffusion zone. Dynamic Metal Treating Inc. uses fluid bed (salt bath) ferriric-nitrocarburizing at (600°F to 1000°F) and steam blueing. The ferriric carburizing process takes between 4 to 12 hours. Steam blueing is achieved by sealing in a tempering furnace at 350°C to 370°C and introducing steam. This process creates a blue black surface finish due to the formation of a tight blue oxide layer. A surface hardness higher than 70 HRc is achieved. The growth in dimensions, due to the process are, of the order of 0.0001 to 0.0002" per side. Typical compound zone depths of 0.0005 to 0.001" and case depths of 0.005" are achieved. Dynablue 10B was used in the tests.
  2. Ion Nitriding: Ion nitriding is an extension of the conventional nitriding process using the plasma-discharge physics. In vacuum, high-voltage electrical energy is used to form plasma through which nitrogen ions are accelerated to impinge on the work piece. This ion bombardment not only cleans the work piece surface but also heats up the surface and provides active nitrogen for nitriding. Two different companies supplied the ion nitriding treatments for the campaign at Premier.

a) Ultralow: Ultralow process, applied by Advanced Heat Treat, is basically ion nitriding process with treating conditions optimized to reduce the white gamma prime compound zone on nitride surface. The nitriding parameters were adjusted to yield an average surface hardness of 90 to 94 HR15N on the core pins for Premier Tool and Die Cast and 90 to 91 on the DMG pins. The case depths are maintained from 0.006 to 0.008 in.

b) Ion Wear: lon Wear treatment was provided by Sun Steel. This process involves a combination of ion nitriding at 400 to 565°C and steam treating (oxidation). This diffusion treatment creates a multi-layer composed of complex oxy-carbo-nitrides up to 0.0004" (10 micrometer) and case depths up to 0.025" (0.6 mm). For the Premier test the treatments had a surface hardness of 800-850 HV1, a compound layer of 0.0002 in (5 microns) thick and case depth of 0.005 to 0.0009 in (75 to 125 microns)

Diffusion Carbide Coating for Distortion Control

Carbide coating by TRD (Thermo-Reactive Deposition and Diffusion) can ensure very high adhesion strength of the carbide layers onto substrates due to the nature of the carbide formation mechanism. Carbide coatings made by this method have very good resistance to erosion and corrosion against molten cast metals.

Therefore, carbide coating by TRD has large potential to prevent failures of mold components used in die casting and other casting methods and to provide tremendous benefits to the die casting industry. In fact, many field test results recently obtained are accelerating the wide application of the process in the American die casting industry as in Japan, where the process has been successfully used since the mid-1970s.

TRD coating is usually carried out at temperatures similar to hardening temperatures for steels; for example, 1877°F (1025°C) for H-13. Steel substrates are quench-hardened during cooling from the coating temperatures and followed by tempering. Therefore, distortion may occur. Overcoming distortion was the key point for successful application of the process. Working surfaces of the molds must meet required dimensional specifications on completion of the coating operation since the coated molds are put into use without a finishing process. Carbide layers formed in the TRD process are too thin to be polished off for dimension control.

Distortion in application of high temperature coating processes such as TRD and CVD is often overstated by recalling the distortion in ordinary hardening of steels. People often expect the degree of distortion to be similar to that in ordinary hardening of die molds based on past experience. This thinking is not correct. Ordinary hardening is done in most cases on the premise that the articles are to be ground to the finish size with grinding allowance large enough to get the finish size. This premise lessens the incentive to heat treaters to minimize distortion. Equipment and procedures are well organized and extra care is usually taken in the coating application to minimize distortion. Furthermore, ordinary hardening is applied to annealed steels while TRD coating, in most cases, is applied to already hardened steels. This brings about a good effect on dimensional change due to a smaller difference in microstructure change before and after TRD processing. Thus, the coating has been successfully applied to very close tolerance tools with tolerances of some µm in diameter. Following is an explanation of causes and countermeasures for distortion and some examples of results obtained on die casting molds.

Criteria of distortion

Dimension (size) change: Symmetric change and shape of tools remains unchanged ("a change of size without a change of shape"). This includes size difference between the edges and in the center of the faces.

Deformation: Non-symmetric and shape is changed ("a change of shape and size").

Most tools change shapes to barrel or spool, more or less, depending on the shape and size of molds, and the coating condition. This is usually symmetric and stays in the category of dimension change.

Deformation becomes evident by curvature of long axes of slender tools and non-symmetric-shaped tools. Another typical one is out-of-roundness of ring-shaped tools.

Dimension of molds after TRD coating

Dimensions of molds will be changed by the following two reasons:

  • Buildup of carbide on substrate
  • Dimensional change of substrate material before and after the coating operation.

The dimension of tools after the coating, therefore, can be shown in relation to the initial dimension:

Da = Db + 2 Tc + D
Da ; Dimension after coating
Db ; Dimension before coating
Tc ; Thickness of coatings
D ; Dimension change of substrate

Carbide layer thickness, Tc, is usually selected from 0.0002 to 0.0005 in. (5 to 12 µm) in die casting application regardless of shape, size and kind of mold components to be coated. Tc can usually be small enough to not be a concern. D can be much larger than Tc in the case of large cores, large core pins, etc., depending on various factors and should be a more serious concern than Tc. Under-sizing or over-sizing to the targeted final dimension can compensate for the possible dimensional changes. Therefore, scattering of Db, Tc, and D should be of more serious concern rather than their individual values.

 

Causes of the dimension change and related factors

 

Tc is controlled mainly by TRD bath temperature, immersion time in the bath, and bath control and has a very minor effect. Scattering of chemical composition and microstructure of substrate steels can change the thickness of carbide layers through scatter of carbon content in the matrix phase (austenite) at the coating temperatures. However, this problem is not serious in H-type steels widely used in die casting since they contain relatively small carbon and alloying elements. Therefore, scattering of Tc can usually be out of consideration in die casting applications.

D and its scattering can be affected by a number of factors related to the shape and size, substrate materials, heat treating condition, and coating condition. Volumetric change of the substrate by change of microstructures of the substrates is the primary reason for D. That is to say, from pearlite to martensite and retained austenite, (in the case TRD on the un-hardened molds), change in amount of austenite (in the case of TRD on hardened substrates), and tempering of martensite. However, D is not isotropic.

The size and shape of molds highly affects the difference in temperature between the surface and core during hearing or cooling which produces thermal stress and transformation stress. The barrel- and spool- shaped changes are determined not only by the values of these stresses but also the sequence in time by which the transformation stress is added to the thermal stress. Temperature differences and stresses will also be induced by change of thickness within a mold. Larger size, higher temperature, and larger rate of heating and cooling will make D more considerable.

The banded structure in forged and rolled steels, which is large amounts of carbide particles in a line parallel to the rolling direction, make the expansion and construction heterogeneous. This carbide alignment usually leads to larger size movement in length than in diameter, thickness and width. The banded structure is determined not by the size of steel stocks but by steel making processing, including ingot size, forging ratio, and rolling direction etc.

Size change of tools can occur even at room temperature by transformation of "retained austenite" (high temperature phase of steel) to "martensite" (the phase caused by quench-hardening of steel). Except for tools that need very strict dimensional tolerance, this change of size is usually negligible.

In some cases, some factors are very serious and others are small and negligible.

 Causes of deformation and related factors

 

Distortion is formed by the following reasons:

  • Non-uniform heating and cooling during high temperature coating and heat treatment
  • Non-uniform micro-structure in substrate material

Residual stress produced during tool making, such as machining and grinding.

Creep by gravity during high temperature coating and heat treatment.

Most heating and quenching equipment is not likely to make uniform heating and cooling within tools. The complicated shape accelerates this tendency. Not enough spacing between molds, or molds and basket components also results in differences of the heating and cooling rate. The heterogeneous microstructure in forged and rolled steels gives rise to different transformation temperatures and the time lag in transformation within a mold leading to deformation. Improper machining conditions bring about residual stress in molds, and relieving of the stress at high temperature can cause deformation, if heating rate is high. Steels have very small yielding stress at high temperature. The loading of self-gravity easily results in deformation such as warpage of long slender cools loaded in the baskets in such a way that the end of tools arc placed on the basket bottom, or disk shaped tools placed on a plate with bend, for example.

 

How to get TRD coated molds with minimum distortion

 Following are some countermeasures to ensure minimized distortion:

  • Select good type of substrate materials
  • Select premium quality materials
  • Cut out molds from steel stocks in consideration of rolling direction
  • Machine and grind under proper condition to minimize residual stress
  • Make molds with tight dimensional control
  • Apply under- or over-sizing on targeted size
  • Harden molds under precisely controlled conditions
  • Apply TRD coating under precisely controlled conditions
  • Control movement and deformation by selection of proper tempering conditions
  • Apply deformation correction after TRD coating
  • Design mold with consideration of finish grinding on non-working surface to ensure extremely tight tolerance.

As mentioned before, the H-type steels are basically good materials. It is, however, highly recommendable to use steel stock produced by the same steel maker since there is some scattering in quality, difference in smelting method, forging ratio, and scattering of chemical composition etc., between steel makers and even between steel stock made by the same steel maker.

Bending 

Bending of pins was measured on two long pins. A pin 7 in. (180mm) long with 0.4 in. (10 mm) shank diameter and long tapered portion bent about 0.003 in. (0.08 mm). Another pin 7 in. (180 mm) long with 0.6 in. (15mm) shank diameter and short taped working area bent 0.0016 to 0.0024 in. (0.04 to 0.06 mm). These bending were successfully corrected by pressing.

 

Out-of-Roundness

 

Deterioration of out-of-roundness was evaluated in several examples. Out-of-roundness of the un-hardened round core specimens increased only 0.00015 - 0.00047 in. (4 - 12 µm) in both ID and OD. Those of hardened round core specimens showed only small increases: 0.00004 - 0.0004 in. (1-10), and 0.00008 - 0.00075 in. (2 - 19 µm).

 

Range of Diameter and Length

 

The relation between the range (maximum — minimum) of diameter and length were investigated, in the core pins before coating and those after the coating. The size range can increase by max. 0.00067 in. (17 µm) in diameter and max, 0.0018 in (40 µm) in length. Making the molds with minimized scattering of dimension should be the first step to satisfy the tight dimensional tolerance.

 

Application to die casting molds

 

As for TRD application, benefits similar to those obtained in Japan were reported in the USA two years ago. The application in American die casting is now being accelerated and a large number of pins and cores have been put into production reporting satisfactory results. Application on various types of pins including squeeze pins ranging in size from very small pins such as 0.157 in. (4 mm) in shank diameter to large pins as 1.57 in. (40 mm) in shank diameter and 19.7 in. (500 mm) long. Some types of core inserts with very complicated shapes, and sprue cores with 6 in. (150 mm) in diameter have been successfully used. Substrate steels are usually H-13 and other H type steels. Modified high speed steels are rarely used.

 

Conclusive Summary

 

It seems to be commonly believed that high temperature coating processes cannot be applied to dimensionally tight products. This is not necessarily true. Punch makers in Japan have been making standard type punches for metal stamping with 0.00008 to 0.00012 in. (2 to 3 µm) tolerance in diameter and less than 0.0002 in. (5 µm) in bend for more than twenty years. The results of Example 1 and others suggest that die casting pins with similar tolerances can be made without difficulty by well-considered procedures. It cannot be expected to industrially realize similar tight tolerances for some large cores, especially with complicated shapes. However, the examples of actual applications shown in the previous paragraphs suggest that even relatively large cores can be successfully TRD coated, satisfying the required tolerance.

Unfortunately, a number of factors relate to distortion problems, most of which are out of the die caster's specialty. It should be emphasized that to easily control all these factors, the die caster should keep in close contact with steel makers, heat treaters, tool makers and especially with the coaters if he wants to apply the high temperature coating process to tight tolerances molds. The extra effort will be worthwhile and shown through cost savings.

 Cleaning and coating procedure:

 1)      Cleaning Procedure:

a)      Incoming Inspection of the Tools.

b)      De-greasing using Alkaline Solutions.

c)      Removal of In-organic contaminants using Aluminum Oxide powder and Aluminum Oxide Slurry.

d)      Rinsing the tools with Water.

e)      Drying tools using Dry Air.

 After drying the tools are loaded in the appropriate fixture and taken up for the next step of coating.

2)      Coating Procedure:

a)      Loading the tools with the fixture inside the coating chamber.

b)      Creating vacuum to a level of 10-5 Torr.

c)      In-situ plasma cleaning  of the tools and heating them using radiant heating to 200°C.

d)      Creation of Ti ions by striking arc on the surface of the source.

e)      Bombardment of Ti ions and introduction of N2 gas.

f)       This forms TiN and gets adhered to the Tools.

3)      The tools removed from the coating chamber and are allowed to cool in atmosphere.

4)      Post- coating inspection of the tools is carried out prior to dispatch.

 

References

1.  Ford, Eric, Special Coatings Improve Die and Mold Performance, DIE CASTING ENGINEER, Sept./Oct. 1990, p. 36.

2.  Arai, T., "Research and Applications of Carbide and Nitride Coatings onto Aluminum Die Casting Molds in Japan," NADCA Transactions, T95-102, p. 327.


SOME MORE HELPFUL TOPICS :

Bending Force Calculation

Forming Limit Diagram

Sheet Metal blank size calculation

Lean Manufacturing Terminologies

Brief Comparison of Hydraulic and Mechanical press

TECHNICAL TERMINOLOGY OF WHEEL TIRE

Raw material calculation of Complex Deep drawn Geometry

Modify the product to make it aesthetically better and more cost effective

How creativity results in major cost savings

How concrete analysis resolve warranty issues

Proud of being Granted with your first patent

How hastened decisions affect the business in the long term

How cost reduction is the key factor to capture the market share

How to choose number of Draw process for deep drawn geometry

Essential parts of a draw die

 Everything about patent

 sheet metal processing and latest trends

 Electric vehicle (EV) – Future of Transport

 Surface coatings to increase the tool wear life

 Mechanical Properties of Cast Carbon and Low Alloy Steels

 HYDRAULIC PRESS - BRIEF TERMINOLOGY

 A TO Z TERMINOLOGY OF AUTOMOTIVE CONTROL SYSTEMS

 A to Z Terminology of surface treatment and heat treatment

 Raw material standards


Post a Comment

0 Comments