Fountain of Innovations New Coatings ALL4®: AlCrTiN4® nACRo4®

TiXCo4®

CROMVIc3®

TiB2

Lifetime Guarantee for CEC Cathodes

New Ultra Fast Decoating System

Basis: Coating Equipment of the 11 Series

www.platit.com - IMTS Chicago W-1680 - AMB Stuttgart 1Z140 - JIMTOF Tokio E2003

COATING

Rotating PVD Cathodes with Lifetime Guarantee or

Current New Developments in the π-Technology PLATIT AG, Selzach, Switzerland

Figure 1: Series 11 Standard PLATIT coating equipment with rotating cathodes.

π111, π 211, π311 & π 411 – the magical „11“ is the distinguishing mark of the π-system series from PLATIT with rotating cathode technology (Figure 1 [1]). The „11“ connects PLATIT not just with its native city of Solothurn [2]. PVD systems with rotating cathodes have been on the market for 11 years. In the meantime, there are 196 π systems in 38 countries (status as of the beginning of June 2014). Thanks to the large number of systems used in a wide variety of industries, a broad base of experience has been accumulated. The technology established itself in the high-tech area, which of course was only possible due to constant enhancement. This article summarizes the goals of the new developments that will be presented at the AMB 2014 in Stuttgart. 2

●

Werkzeug Technik 141

●

25 August 2014

COATING

1. New coatings thanks to flexibility (ALL4®, CROMVIc3®, TiB2) What does flexibility mean? Flexibility is the versatility and the possibility of being able to make a choice and change quickly between the many options. The π systems cover the entire spectrum of current PVD coatings [1]: – Conventional mono and gradient coatings, – Multi and nanolayers, – Nanocomposites and oxide coatings, – DLC films as well as, Figure 3: Tool life comparison with high performance coatings at hobbing.

– Triple and quad coatings are possible. The absolutely unique flexibility characteristic of the technology is the programmable stoichiometry of the coatings. The largely non-alloyed cylindrical tube targets are arranged very close to one another. At the same time, the targets are very close to the parts to be coated. This has the advantage that not just the vacuum chamber, but especially the whole system can be constructed very compactly. The height of the target is always longer than the height of the coatable area in the chamber. The composition of the coating can be generated with user programming along the en-

3

●

Werkzeug Technik 141

●

25 August 2014

Figure 2: Coating structure of the QuadCoating ALL4® (AlCrTiN4®).

tire coating construction. As an example, the new quad coating ALL4® (AlCrTiN4 [3] (Figure 2). – begins with a CrTiN adhesion layer, – creates the transition with good cohesion between adhesion and core layer with a AlCrTiN gradient, – builds up in the middle an elastic core as multilayer with Al/CrN, – continues the hard coating part with a AlCrTiN top coating and – can complete the composition with the top coating CrCN as lubrication. And all of this without expensive alloyed targets with 3 non-alloyed

COATING

Figure 4: Comparison of the built up edges at aluminum cutting. SEM and EDX after 283 m tool life. Material: 3.4365 AlZnMgCu1,5 - Tool: Torus end mill Ø12mm – r=2.5mm – z=2 - vc=377 m/min – ae=5mm – ap=6mm – fz=0.2 mm/rev.

cathodes, Ti, Al and Cr. To accelerate the process, it is possible to use an AlCr cathode in the center as a booster.

– abrasive wear resistance (tungsten from carbide is hardly visible through the coating).

ALL4® is an all-rounder, a universal coating, is used for milling, hobbing (Figure 3 [4]), fine punching, and also for forming.

The CROMVIc3®- (DLC3: ta:C) -coating was deposited with the new coating system π 211 whose development is not yet complete. It achieves the smallest built up edge. But the EDX "sees" too much tungsten, which also indicates insufficient abrasion resistance, e.g. due to too-low coating thickness (~0.5 μm). The basis of the new system is a new, patented "straight-on" ARC filter [5], with the help of which the droplets can be reduced significantly [1]

Depending on the required coatings and the requirements posed of them, deposition can be realized with – ARC- (LARC®: Lateral Rotating Cathodes and CERC ®: Central Rotating Cathode), – PECVD- (Plasma Enhanced Chemical Vapor Deposition) and

2. Productive decoating The decoating of cutting tools for regrinding is an important precondition for their recoating with high quality. "Productive decoating" sounds paradoxical. Decoating does not produce any new product but removes an old, used coating so that the new coating can bond well on the reground tool and achieve high performance. Productive decoating means first and foremost speed. It may not "hold up" the production of the new coating. The current decoating processes are slow. They require several hours or even

– Magnetron Sputtering (SCiL®: Sputtered Coating induced by LGD®) technologies. The new TiB 2-coating is deposited with the latter technology from a specially-segmented central SCiL cathode (Figure 4). This coating is used first and foremost for aluminum cutting. For soft aluminum cutting, the edge build-up is critical. The pictures of the SEM and EDX analysis show that the new TiB2coating demonstrates very good behavior both with respect to

Figure 5: Decoating unit PLATIT-CT.

– edge build up (little aluminum on the edge) as well as with respect to 4

●

Werkzeug Technik 141

●

25 August 2014

COATING

days. The new PLATIT-CT decoating system (Figure 5 [6]) reduces the time to less than one hour. It will be presented in a publication that will appear at the same time as this article [7].

3. LGD® – new etching procedure for optimal coating adhesion

Figure 6: Principle of the LGD® etching.

ble, which prevents the formation of additional hollow plasmas.. Figure 6 shows an extreme but clear example. The slightly corroded gripper was placed in a tool sleeve and etched

Without having to install additional equipment (anodes, cathodes, filaments, ion sources, etc.), thanks to a new patented process, the rotating cathodes ensure optimal coating adhesion. The new LGD ® process (Lateral Rotating Cathode Glow Discharge) generates a highly-efficient argon etching [8]. During LGD etching, a cathode is ignited in the direction of the chamber door while a second, adjacent cathode is switched as anode, see Figure 6. The electron flow between cathodes 1 and 2 generates a plasma with high ion density. Conditioned by the arrangement in the vacuum chamber, the plasma is fed through the carousel with the substrates. The plasma stream is not just efficient, but also very sta5

●

Werkzeug Technik 141

●

25 August 2014

with LGD. On the free area, the rust could be removed, the surface became shiny. Of course the areas covered by the sleeve remained rusty.

Figure 7: Optimized (extremely short) process flow with rotating cathodes of the π411 system for depositing of a complex QuadCoating4®.

COATING

Figure 8: Comparison of the target costs with spot and rotating cathodes.

Thanks to the easy parameterization of the flexible LGD etching, this process can be used efficiently and to great advantage both for a wide variety of shaft tools as well as for substrates with complex surfaces such as forming tools. Indispensable is the technology for coating hobbing [4]

4. Low target costs due to π-cathodes If simple monolayer coatings are deposited, then all cathode spaces can be occupied with the same targets. For the depositing, all cathodes are always run at full power and they achieve the highest productivity. Of course today, monolayer coatings have been replaced by much more complicated coating structures. This is precisely where the PLATIT systems, with their wide variety of cathodes, can exploit their full potential. In comparison to this, a system with common spot cathodes would have to have at least 3 different rows of cathodes to deposit the quad coating described above, which cuts productivity by two-

thirds. With the rotating cathodes of the π411 system, door-to-door cycle times of less than 4 hours can be realized (Figure 7). And this for complex coatings. Are these complex coatings even feasible with conventional cathodes?

least fifty nACo4®-like coating batches without changing. (Of course under the same conditions, like thickness, loading, rotation etc.)

portant share of the variable costs. The π-cathodes provide a much larger target area than the conventional flat cathodes:

In the recent years, PLATIT has established cathode exchange centers (CEC= Cathode Exchange Centers [1], [9]) around the world (Figure 9). Thanks to the cathode exchange centers, the user must bear no storage costs for cathodes and targets. With the help of the centers, the user always receives a completely-prepared cathode within a short time. That is:

5. Cathode change centers The target costs account for an im- allow lifetime guarantee

–

π -times larger than a planar target with the same width and length,

– 17 times greater than a spot target with the same diameter. Unfortunately, a comparison of the target costs per batch or per tool between rotating and common cathodes is not possible, since the system manufacturers do not disclose the life span of the spot cathodes. PLATIT conducted a long series of experiments on the rotating cathodes and collected data from customers around the world. Figure 8 shows an example from this data collection. In order to be more costeffective, a conventional spot cathode set would have to be able to deposit at 6

●

Werkzeug Technik 141

●

25 August 2014

– with only first-class, tested target materials, – whereby all wear parts are replaced, – the cathode is fully set and "burned in", – the cathode is tested and certified in a 24-hour check under vacuum. Each cathode is delivered with a new guarantee. That is, the user has a lifetime guarantee for the cathodes.

COATING

Summary The technology of the rotating PVD cathodes has established itself in the cutting industry over the course of the last 11 years. The advantages are impressive when it comes to the quantity and quality: – Very great flexibility • Wide variety of coatings, • Quick and easy rearranging to production of different coatings, • Own coating development possibilities, – Homogemeous coating thickness distribution, – Extremely good coating adhesion thanks to the new LGD process, – High productivity thanks to extremely short process times, especially for complex high-performance coatings, besonders für komplexe Hochleistungsschichten, – Low variable costs (such as target costs) and – The unparalleled lifetime guarantee for cathodes.

References [1] PLATIT compendium 53rd Edition, 2014, Selzach, Switzerland. [2] Cselle, T., u.a.: Series 11 facilities with four coating generations Werkzeug-Technik, Boulogne, 8/2013.

[3] Morstein, M., u.a.: Roads to tougher nanostructured coatings for cutting at Intermediate temperatures ICMCTF, B6-5, San Diego, USA, April, 2014. [4] Lümkemann, A., u.a.: A new generation of PVD coatings for high performance gear hobbing Conference THE “A” COATINGS, Thessaloniki, Greece, Oct/2014. [5]

πsCOAT ® : Filtered cathodic arc deposition apparatus and method European Patent, Munich, EP2602 354 A1, Dez/2011.

[6] Process for decoating ceramic hard coating of steel and hard metal substrates European Patent, Den Haag, PCT/EP2014/055376. [7] Wittel, u.a.: Productive decoating of PVD coatings Schweizer Präzisionstechnik 2014, p.41-43. [8] Glow discharge apparatus and method with lateral rotating cathodes European Patent, Munich, PCT/EP2012/058092, 2012. [9] www.platit.com New PLATIT website, 2014. (14214-??)

Figure 9: Process of the cathode exchange with the help of the PLATIT CEC centers around the world. (CEC= Cathode Exchange Center)

7

●

Werkzeug Technik 141

●

25 August 2014

Coating Generations and their Structures 1. Generation CT=1.84µm

2. Generation

Monoblock Structure Without Adhesion Layer The monoblock structure without adhesion layer can be produced by the fastest, most economical process. All targets are the same and run during the whole deposition process.

Conventional Structures With Adhesion Layer

Monoblock

Gradient (G)

Multilayer (ML) Period > 20 nm

CT=1.92µm

CT=2.5µm

CT=2.52µm

Especially at high aluminum content the monoblock coating should be started with adhesion layer (e.g. TiN or CrN).

At gradient structure the ratio of hard components (e.g. cubic AlN) will be continuously increased obtaining the highest hardness on the top of the coating.

Multilayer structures have higher toughness at lower hardness than comparable monoblock coatings. The "sandwich" structure absorbs the cracks by the sublayers.

3. Generation: TripleCoatings3®

Nanolayer (NL) Period < 20 nm

Nanocomposite (NC)

5 nm Nanolayer is the conventional structure for the so called Nanocoatings. It is a finer version of multilayers with a period of < 20 nm.

4. Generation:

At depositing Nanocomposites the hard nanocrystalline grains (TiAlN or AlCrN) become embedded in an amorphous SiNMatrix.

Coatings4®

Nanocomposite top layer

Nanocomposite top layer

Monoblock or gradient core layer

Multilayer core layer Gradient core layer

Adhesion layer Adhesion layer CT=2.72µm

CT=2.32µm

Preprint from

  Conference, 

hessaloniki, Greece, Oct.1‐3, 2014     

 

A NEW GENERATION OF PVD COATINGS FOR HIGH-PERFORMANCE GEAR HOBBING A. Lümkemann², M. Beutner1, M. Morstein², M. Köchig1, M. Wengler1 T. Cselle², B. Karpuschewski1 1.

Institute of Manufacturing Technology and Quality Management (IFQ), Otto von Guericke University, D-39106 Magdeburg, Germany

2.

PLATIT AG, CH-2545 Selzach, Switzerland ABSTRACT The aim of the paper was the development of a new high-performance coating tailored for hobs in dry cutting of gears. A range of AlCrN-based coatings was deposited in an industrial rotating arc cathodes PVD unit. These coatings were varied both in terms of their nanostructure and the chemical composition of their top layer. To examine the performance of the different coatings, cutting trials were carried out using the well-established fly cutter analogy test. Significant performance differences between the chemically and structurally modified coatings were found. Doping the coating top layer with titanium in combination with an optimized nanolayer structure lead to a substantial increase of tool life by about 50 % compared to AlCrN. KEYWORDS:

1

gear hobbing, fly cutting test, coating development, PVD, AlCrSiN, AlCrTiN

INTRODUCTION

In the current globalized trading situation it is mandatory for production companies to stay competitive. Advantages can be realized by reducing costs or delivery times, which requires an optimized manufacturing process. One example for this mechanism is the gear manufacturing industry. In 2005, the German turnover concerning gears amounted to 14 billion US-$ /1/. According to the Freedonia Group, there was a worldwide demand for gears of about 170 billion US-$ in 2013, 80% of which are accounted for by the automotive industry /2/. This demonstrates the enormous potential which can be harvested by gear manufacturing companies. To manufacture automotive gears, the applied standard process chain consists of hobbing the steel blanks, followed by heat treatment and finish grinding. Generally, hobbing is the dominating green manufacturing technique for external spur gears. This manufacturing technique is characterized by an interrupted cut. This kind of cutting process is demanding special tool resistance against impacts and rapid changes in temperature. The performance of the hobbing process benefits both from an increased tool life and more productive cutting parameters. This can be achieved by using improved substrate materials and coatings. Starting in the 1970ies, a continuous development of coatings took place, starting from the early TiN and TiCN thin films, followed by the successful broadband of TiAlN coatings (late eighties) towards recent AlCr-based coatings. Connected to the development there is a trend towards application tailored tool micro geometry and coating composition and structure /3, 4,5/. Today, more than 100 different coating systems are available /6/ which is an evidence for the importance of coatings. Depending on the coating elements and its structure, different properties regarding wear resistance can be achieved. Due to the type of substrate material mainly used for hobbing (HSS) and the required tool geometry, “Physical Vapor Deposition” (PVD) coatings dominate this field. For the experimental trials a fly cutting analogy test, which is well established in research, was utilised at the IFQ. The fly cutter generation, measuring and the test execution was performed at IFQ test facilities. The pre-treatment of the tools, the coating development and the post-

1

Preprint from

  Conference, 

hessaloniki, Greece, Oct.1‐3, 2014     

 

treatment of the coatings was accomplished by PLATIT R&D. The tests are based on a tool geometry and on cutting parameters that largely correspond to production settings of a major Japanese transmission manufacturer. 2

FLY CUTTING – AN ANALOGY TEST FOR GEAR HOBBING

The fly cutting analogy test was developed in the 1970ies by the WZL of the RWTH Aachen /7/. Increasing research in gear hobbing prompted this improvement because it would be very time and material consuming to perform research using complete hobs. By using a single tooth of a hob, the newly introduced fly cutting analogy test is able to generate gears, equal to those machined by a real hobbing process.

Figure 1:

Principle of the fly cutting test.

The kinematics of the analogy test (Figure 1) reproduces the wear behaviour of a shifted tooth of a hob. The fly cutter continuously moves with the tangential feed rate ft, following the shift axis of a real hob. After one full tangential motion, the fly cutter is discretely moved towards the next axial position with an offset corresponding to the axial feed fa (Figure 1b). This procedure repeats until the whole width of the wheel has been geared. Even though the machining time for a single gear is increased compared to the real hobbing process (15 to 60 times longer) a statement concerning wear behaviour is already possible after a few milled gears. Moreover, fly teeth can be examined much easier than a full hob tool by methods such as SEM/EDX analysis or optical profilometry. 3

MOUNTING DEVICE – BLASTING AND COATING

A real hob serves as source to gain the trial tools (fly cutters). These fly cutters are cut out from the hob by wire erosion machining. Extracting rows of teeth is the first step in this procedure, Figure 2b,c. The fragmentation of these rows into single fly cutter teeth is again done by wire erosion, Figure 2d.

Figure 2:

Generation of teeth for fly cutting test.

2

Preprint from

  Conference, 

hessaloniki, Greece, Oct.1‐3, 2014     

 

A mounting device like the mentioned one is essential for coating and post-treatment, due to the specific behaviour of a real hob during such processes. While the hob is rotating inside the PVD coating chamber the teeth of one flute are covering the teeth of the following flute, which is comparable to shadowing, Figure 3. This shadow induces a reduction of the local coating thickness, which has an enormous influence on the tool life, as described in 4.4. Stand-alone fly cutting teeth do not reproduce this important characteristic, which results in a severe difference in both wear mechanism and tool life, compared to a real hob.

Figure 3:

Shadowing effect during hob coating.

Similar effects can be suppressed in the pre- and post-treatment. During micro blasting, the cutting edges of free-standing single teeth would be exposed to increased erosion. Therefore, the mounting device is inevitable to carry out the analogy test as close to the real process as possible. 4

IMPROVING THE HOBBING PERFORMANCE

The schematic in Figure 4 shows the main factors influencing the tool performance. In this article we are focussing on the surface quality of the tools, the coating composition and structure. All other parameters influencing the performance were kept constant. Hobbing Conditions

Figure 4: 4.1

Hob Geometry (Makro)

Hob Geometry (Micro)

Dedicated Coating

Gear Geometry

Modulus

Surface Quality

Coating Type & Structure

Cutting Parameter

Number of Starts

Pre- & PostTreatment

Coating Thickness

Project plan illustrating the investigated performance parameters.

TEST SETUP

A Liebherr LC180 hobbing machine has been used for the wear studies. Maintaining constant settings (cutting parameters, gear geometry) and gear blanks taken from a single batch, comparable trial conditions can be preserved. Table 1 is summarizing these settings.

3

  Conference, 

Preprint from

workpiece mn z2 da β2 b

Table 1:

: : : : :

hessaloniki, Greece, Oct.1‐3, 2014     

 

process

2,31 mm 48 130 mm 24,5° 29 mm 20MnCr5 (147HB)

vc fa hcu,max T

: : : :

150 m/min 1,69 mm/rev 0,40 mm 7,30 mm dry climb hobbing

Parameters used in fly cutting test.

As implied by the title of this paper, a high-performance approach in hobbing is aimed at. In this case “high performance” refers to high productivity, where a the minimized cutting time can be achieved by using either high cutting speeds, axial feeds, numbers of starts or small tool diameters, respectively. In the actual industrial gear production case, the number of starts was used as the main variable to gain a higher productivity. The advantages of an increase in the number of starts and the comparison to the other control variables will be discussed in the next chapter. 4.2

HOB GEOMETRY (MACRO GEOMETRY)

The tests were carried out using a tool actually used in industrial production of a Japanese transmission manufacturer. Table 2 shows the major facts regarding tool profile and hob geometry. Two of them diverge from the norm: On one hand that is the small pressure angle to create an integer overlap ration and on the other hand that is the large number of starts. TOOL mn da z0 α p Table 2:

: : : : :

2,31 mm 90 mm 5 14,5° 14 PM-HSS

Test tool geometry.

As mentioned in the previous chapter an increase in productivity can be achieved by using hobs with a large number of starts. When increasing this number from one to five, also the revolution speed of the workpiece increases fivefold. That way the tools’ axial speed increases keeping the revolution speed of the hob (cutting speed) at the same level. One effect of larger numbers of starts is a decrease of generating positions, causing an increase in chip thickness (hcu,max) and in generating cut deviations (δy). Despite having five starts in the applied case, the deviations in the generating cut do not exceed δy = 4,0 µm, thus still maintaining a surface quality matching well with the following production steps. The maximum chip thickness is of more importance, because it represents the load on the fly cutter. Using a CAD-based interpenetration calculation, a duplication of hcu,max altering the number of starts from two to five could be noticed, Figure 5a. To illustrate the benefit in productivity, the cutting speed and the maximum chip thickness depending on the number of starts while having a constant machining time are plotted in Figure 5b. To achieve the same productivity as a hob with five starts, a hob with only two starts needs to be used with a cutting speed of vc = 375 m/min. Especially for old hobbing machines or for PM-

4

Preprint from

  Conference, 

hessaloniki, Greece, Oct.1‐3, 2014     

 

HSS-hobs this cutting speed is exceedingly high. For those applications, an increase of the number of starts is a more feasible solution.

Figure 5: 4.3

Increasing number of starts and its effect on selected cutting parameters.

HOB GEOMETRY (MICRO GEOMETRY)

A ground, sharp cutting edge exhibits a series of micro-defects that can negatively impact the tool life. The idea of edge preparation is to stabilize the cutting edge, in other words to reduce the failure probability and to prolong the tool life. Edge pre-treatment technologies... ... reduce edge chipping ... flatten out surface defects and grinding grooves ... produce a "defined" micro geometry (edge radius, symmetry of rounding) In the same way as the right choice of the hard coating, the edge treatment method must be adapted to its application. Rech et al. /8/ did show the important role of edge rounding on the tool life for gear hobbing tools. For comparison reasons all tested tools in that contribution had the same way of edge preparation and post polishing. So the influence of different micro geometries on the fly hobbing performance could be minimized in this study. For edge preparation, all teeth were wet blasted using the above introduced fly teeth holder, again in order to have the same shadowing effects as for a real hob. The cutting edge radius for all tested teeth was kept in the range of 17 µm to 20 µm, measured before coating at the head of the teeth1. The symmetry of rounding /9/ was kept in the range of k = 0.8 - 1.0 for all teeth being tested. However, the optimization of the micro geometry does already start even before edge rounding, when performing the final grinding step. The following images give an impression of the cutting 1

Measured with a GFM Mikrocad Premium Sensor using standard fit routine.

5

Preprint from

  Conference, 

hessaloniki, Greece, Oct.1‐3, 2014     

 

face of two gear hobbing fly teeth. They were taken at the end of tool tests. Outside the wear region, one can easily identify different kinds of grinding traces on the cutting faces, indicating different grinding qualities. The "original" tooth has big grinding grooves, while the other one shows a nicely ground, much smoother cutting face. Even though the differences in surface quality can easily be seen on the images, the measured surface roughness differs only on a small scale (Figure 6).

Figure 6:

Influence of fly tooth grinding quality on tool life.

The graph on the left side in Figure 6 shows average Rz roughness values measured at the cutting face of original and resharpened fly teeth. On average, the resharpened cutting faces do have a roughness Rz = 1.9 µm, compared with Rz = 2.5 µm for the initial cutting faces showing the grinding traces. However, the relatively small difference in Rz does have a quite big impact on tool performance. The graph on the right in Figure 6 is correspondingly showing the tool life for original and resharpened tools, all coated in the same cycle by a nanostructured AlCrN PVD coating. In our fly hobbing tests, we found a noticeable gain of total tool life for the precisely resharpened teeth, and even more important, a much lower tool life variation. Therefore only very precisely reground teeth with excellent surface finish were used for the coating development described below. 4.4

DEDICATED COATING DEVELOPMENT 411

The preparation of the hard coatings took place in an industrial PLATIT π Arc PVD coating unit. Due to the simultaneous use of four cylindrical rotating arc cathodes /10/, a very high deposition rate (= short cycle times) together with a very high flexibility of coating composition and –structure can be achieved. The aim of these new multilayer high-performance QuadCoatings4® /11/ is to combine several different coating properties in one high-performance coating system for hobs, i.e.:   

Starting with CrN for best coating adhesion Followed by multilayered tough layers such as AlCrN or AlTiCrN Ending with an hard nanocrystalline wear resistant skin on top

To achieve the full potential of coated high performance tools, the hard coatings need to be adapted for specific applications. A basic requirement when comparing different coatings is the reduction of other potentially overlapping influences. One important parameter to be kept constant for all tested fly cutters is the coating thickness. The following graph gives an impression of how much the tool life might depend on the coating thickness in the case of gear hobbing. A sample test revealed that an increase in coating thickness of 20% (0,7 µm) doubles the tool life. Further investigations on how the coating thickness is influencing the wear behavior were not carried out within this study.

6

Preprint from

Figure 7:

  Conference, 

hessaloniki, Greece, Oct.1‐3, 2014     

 

Influence of coating thickness on gear hobbing tool life.

If not otherwise specified all investigated coatings (including the refererence market coating) have a coating thickness in the range of (4.0 ±0.1) µm, measured at the tooth head. Gerth et al. /12/ reported having much lower coating thickness around the cutting edge than on the cutting and flank face. The reason for their observation is likely to be caused by too sharp cutting edges, that locally concentrate compressive coating stresses, resulting in coating cracking and edge chipping to relax the stress, leaving the cutting edge unprotected /13/. In contrast, the multilayered AlCr-based coatings show a relatively low range of internal stress (4.0 GPa to -4.5 GPa on HSS). Together with the beneficial effect of rounding the cutting edge to R>15µm, the result is the typical coating thickness distribution seen in Figure 8. The decrease in cutting face coating thickness (Figure 8) is caused by the effects explained in chapter 3 (shadowing effect). The highest coating thickness is generally observed directly on the cutting edge as the SEM cross section image below proves.

Figure 8: 4.5

Cross-sectional SEM image of the head region of a AlCrN-ML coated fly tooth.

FLY HOB CUTTING TEST RESULTS

Based on the above-described settings for tool micro geometry preparation and coating thickness, a set of fly cutter teeth was coated by different AlCrN-based coatings. All coatings follow the general structure described in section 4.4, where generally AlCrN was selected as the main center part of the coating and the composition of the microcrystalline toplayer was varied in between AlCrN, AlCrTiN and an AlCrN/SiNx nanocomposite. In part, the thickness and type of the coatings’ internal nanolayer structure was additionally varied. Usually three fly cutter teeth were tested per each coating type and the test results averaged. 4.5.1 TESTED COATING PARAMETERS In this project we focused on aluminium chromium nitride (AlCrN) based coatings. TiAlN or AlTiN coatings were not taken into account, since in comparison with these coatings it is

7

Preprint from

  Conference, 

hessaloniki, Greece, Oct.1‐3, 2014     

 

reported that AlCrN shows a superior wear resistance, higher oxidation resistance and lower thermal conductivity at high temperatures /14/. Therefore AlCrN-based coatings can be considered the optimum choice to protect the underlying HSS against oxidation and softening during dry hobbing even at very high cutting speeds leading to crater wear for TiAlN-coated tools /15/. For highly productive cutting conditions as reported in chapter 4.2, the high chip thickness is leading to crater wear, even using moderate cutting speeds. Massive chips apply more load on the cutting face, causing higher thermal and abrasive impact. Indeed, crater wear was observed during both the analogy test as well as the gear hobbing production in Japan using identical machining parameters. In order to prolong the tool life following coating parameters were varied:  

Nanolayer Structure: Generate coatings with different nanolayer thickness by changing the tool rotation speed Composition: Doping the AlCrN top layer with Si (respectively Ti) to influence the coating nanostructure and hardness

4.5.2 NANOLAYER STRUCTURE Even for chemically identical coatings, their micro- and nanostructure can significantly influence the hob performance. With the help of a multilayer structure, simply controllable by periodic target power adjustments, PVD coatings can be optimized in terms of toughness and hardness by changing the composition of the individual sublayers /16/. Furthermore multilayering seems to significantly delay the occurrence of indentation induced large cracks as can be deduced from higher values of the fracture toughness and the critical load for cracking /17/. It is also reported that in nanostructured (Cr1-xAlx)N, alternating CrN/AlN does prevent crack propagation /18/. The composition and thickness of the single layers inside the multilayer is determined by the evaporation rate of the metallic targets /16/, the carousel rotation speed and the dimensions of the coating chamber. Particularly the use of elemental Cr and Al arc targets, as opposed to alloyed Al-Cr targets, has the potential to create a powerful nanolayered substructure (Figure 9).

Figure 9:

Formation of the nanolayered multilayer structure.

To test the influence of the nanolayered substructure on the wear behaviour we varied the substrate carousel rotation speed, thus the nanolayer thickness for two otherwise identical sample coatings. Two different carousel rotation speeds were chosen: 6 rpm and 9 rpm. The resulting thickness distribution including nanolayer structure was simulated2 for a "dummy load" of Ø = 100 mm cylinders placed in the coating device. The result is a nanolayer thickness for the two rotation speeds in the range of 10 nm for 6 rpm compared with 7 nm for 9 rpm. On real tools the nanolayer thicknesses might be a bit different due to antenna effects resulting in an higher coating thickness on the cutting edges. 2

The principle of that simulation is shown elsewhere /19/. 8

Preprint from

  Conference, 

hessaloniki, Greece, Oct.1‐3, 2014     

 

The following two SEM images were taken at the end of tool life for "standard" AlCrN-ML and secondly for AlCrN-ML with optimized nanolayer structure. The wear on the fly hobbing teeth with cutting conditions described above (0) typically starts on the trailing flank, followed by wear on the leading flank and crater wear on the cutting face. The crater wear appearing on the cutting faces close to the edge does influences the flank wear. This may limit the total life time. It can be clearly seen that with optimized nanolayer structure (Figure 10) the crater wear is reduced even after a longer cutting distance.

Figure 10:

SEM of cutting faces, left: AlCrN-ML-coated tooth at 27 m life time right: optimized AlCrN-ML coated tooth at 31 m life time.

As dry cutting is performed, the area where the chips rub on the cutting face heats up considerably. Thereby the martensitic microstructure of the underlying HSS starts to decompose above 550 °C /13/. Stein et. al /20/ measured a reduction of tool hardness in the range of 100 HV next to the crater wear, supporting the theory of friction-induced surface softening. It is assumed that a better thermal insulation, commonly found for high-Al AlCrNbased coatings, protects the steel and thus delays crack formation and consequent layer damage. An optimized nanostructure would add to that, by retarding coating crack formation. 4.5.3 COATING COMPOSITION As already outlined, two versions of AlCrN-ML were tested alongside two new QuadCoatings4®, featuring toplayers doped with titanium and silicon, respectively. Figure 11 is illustrating the calotest images to show the architecture of both QuadCoatings4®.

Figure 11:

Optical mircographs of calotests two QuadCoatings4®, nACRo4 and AlCrTiN4.

The graph below, Figure 12, is showing the flank wear VB on the trailing flank cutting edge. The tool life criterion was a maximum acceptable flank wear of VB = 130 µm. As reference coating,

9

Preprint from

  Conference, 

hessaloniki, Greece, Oct.1‐3, 2014     

 

we tested an AlCr-based coating from the Japanese market coated by a local hob manufacturer. Evidently, all newly developed coatings exceeded the life time of the reference. The graph furthermore shows that for AlCrN-ML with a slight addition of silicon, the tool life did increase in comparison with the undoped base layer. As shown in 4.5.2, the optimization of the nanolayered structure for AlCrN-ML enhances the wear resistance, too. Furthermore, by doping the optimized AlCrN-ML with a slight amount of Ti, an even further increase of tool life was achievable. In comparison with the initial AlCrN-ML the AlCrTiN4 (= opt. AlCrN-ML + AlCrTiN) increases the tool life by more than 50 %.

Figure 12:

Wear curves of the trailing flank.

The three images on the very left in Figure 13 show the wear on cutting faces after the same cutting length of about 26 m/tooth. The Ti-doped version shows an impressive life time enhancement: In comparison to an equally thick market reference coating, this fly cutter can work twice as long (>46 m/tooth) until a flank wear comparable to the reference tooth (≈24 m/tooth) is reached. All in all, using an optimized micro-/nanostructure and Ti- respectively Si-doped top layers the crater wear that occurs for dry hobbing with HSS single teeth is significantly reduced.

Figure 13: 5

Optical measurement (GFM) of crater wear depth on the cutting face.

CONCLUSIONS

In this study a sophisticated fly cutting analogy test for the optimization of gear hobbing tools was presented. The special mounting device for the fly teeth during edge preparation and coating ensured the important characteristic of shadowing effects and coating thickness distribution.

10

Preprint from

  Conference, 

hessaloniki, Greece, Oct.1‐3, 2014     

 

It is shown, that the grinding / polishing quality significantly affects the process stability in gear hobbing. In order to achieve low tool life variation, one has to pay attention to a precise grinding process with excellent surface finish. By using a hob with a high number of starts it is possible to achieve high productivity. A drawback of this cutting strategy especially in dry cutting is that it creates a major increase in chip thickness, which induces crater wear. This effect could however be counteracted by using dedicated PVD coatings with tailored nanolayer structure and thickness. Different structures and compositions of coatings were tested. The Ti-doped QuadCoating4® AlCrTiN4® (CrN + AlCrN-ML + AlCrTiN), with optimized nanolayer, protects the cutting face best. Much higher hob tool life can be achieved maintaining the same or even less crater wear in comparison to the reference coating or undoped AlCrN-ML. 6 1.

REFERENCES Karpuschewski B., Knoche H.-J., Hipke M.: Gear finishing by abrasive processes, Annals of the CIRP 57 (2008) 2, 621-640 (Keynote paper)

2.

McGuinn J., The global gear industry, Gear technology 05/2011, 33-38

3.

Kohlscheen J., Das Prinzip Lamelierung, Werkstatt und Betrieb, WB 05/2010, 56

4.

Peyerl G., Kroh R., AlCrN-Schicht www.maschinenmarkt.vogel.de, 2011

5.

Lümkemann A., Cselle T., Büchel C. Morstein M.: Schneidkantenpräparation und Beschichtung: Schlüsselrollen beim Optimieren von Hochleistungswerkzeugen, XIII Internationales Oberflächenkolloquium - IMSAS / Kanten und Radien (2012) 71 - 78

6.

Cselle T.: Dedication – Integration – Open Source New Rules in the Coating Industry, Werkzeug Technik 118 (2011)

7.

Sulzer G.: Leistungssteigerung bei der Zylinderradherstellung durch systematische Erfassung der Zerspankinematik, Dr.-Ing. Diss., RWTH Aachen, 1973

8.

Rech J., Schaff M.-J., Hamdi H.: Influence of cutting edge preparation on the wear resistance in high speed dry gear hobbing, Wear 261 (2006) 505 - 512

9.

Holsten S.: Mikroprozesse an der Schneidkante, Schneidkantenpräparation: Ziele, Verfahren und Messmethoden (2009) 12 - 33

erhöht

Standzeit

von

Wälzfräsern

bei

Zahnradfertigung,

10. Jilek M., Cselle T., Holubar P., Blösch P., Morstein M.: Arc-coating process with rotating cathodes, European Patent EP1357577A8 (2003) 11. Cselle T., Morstein M., Lümkemann A.: 2013 – The Year of QuadCoatings, Werkzeug und Technik 130 (2012). 12. Gerth J., Larsson M.,Wiklund U., Riddar, F., Hogmark, S.: On the wear of PVD-coated HSS hobs in dry gear cutting. Wear 266 (2009) 444 - 452 13. Gerth J., Werner M., Larsson M., Wiklund U.: Reproducing wear mechanisms in gear hobbing - Evaluation of a single insert milling test. Wear 267 (2009) 2257 - 2268 14. Kalss W,. Reiter A., Derflinger V., Gey C., Endrino J.L.: Modern coatings in high performance cutting applications, International Journal of Refractory Metals & Hard Materials 24 (2006) 399 - 404 15. Kohlscheen J., Knoche H.-J., Hipke M., Lümkemann A.: Coating development for gear cutting tools. Key Engineering Mat. Vol. 438 (2010) 35 - 40 16. Morstein M., Cselle T., Lümkemann A., Kohlscheen J.: Wälzfräswerkzeug mit einer Beschichtung und Verfahren zur Neubeschichtung eines Wälzfräswerkzeuges. European Patent EP2163 661A1 (2010) 17. Karimi A., Wang Y., Cselle T., Morstein M.: Fracture mechanisms in nanoscale layered hard thin films. Thin Solid Films 420 - 421 (2002) 275 - 280 18. Bobzin K., Lugscheider E., Nickel R., Immich P.: (Cr1-xAlx)N ein Review über ein vielseitig einsetzbares Schichtsystem. Mat.wiss. u. Werkstofftech. 37, Nr. 10 (2006) 833 - 841 19. Cselle T., Coddet O., Galamand C., Holubar P., Jilek M., Jilek J., Lümkemann A., Morstein M.: TripleCoatings3® - New Generation of PVD-Coatings for Cutting Tools. Journal of Machine Manufacturing Vol. XLIX Issue E1 (2009) 19 - 25 20. Stein S., Lechthaler M., Krassnitzer S., Albrecht K., Schindler A., Arndt M.: Gear Hobbing: a contribution to analogy testing and its wear mechanism. Procedia CIRP 1 (2012) 231 - 236

11

Standard Coatings The "parent" coatings determine the application fields of all "children" coatings in the same row. The "children" coatings specify PLATIT's standard coatings, which can be deposited by the machine of the columns. The exponent x (coatingx) describes the generation of the coating.

p 80

PL70 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

Nitride TiN TiCN-grey TiAlN AlTiN CrN CrTiN ZrN AlCrN AlTiCrN nACo nACRo TiXCo Oxi-Nitride nACoX DLC VIc cVIc CROMVIc CROMTIVIc nACVIc

TiN1 TiCN2-grey AlTiN1 CrN1 ZrN1

1®

1

cVIc : TiCN +CBC

1

TiN1 TiCN2-grey TiAlN2-ML AlTiN2 CrN1 CrTiN2-ML ZrN1 AlCrN3-NL

TiN1 TiCN2-grey TiAlN2-ML AlTiN2 CrN1 CrTiN2-ML ZrN2 AlCrN3-NL

nACo2® nACRo2®

nACo2® nACRo2® TiXCo3®

1®

13 14 15 16 17 18 19 20

Nitride TiN TiCN-grey TiAlN AlTiN CrN CrTiN ZrN AlCrN AlTiCrN nACo nACRo TiXCo Oxi-Nitride nACoX DLC VIc cVIc CROMVIc CROMTIVIc nACVIc SCiL TiN-SCiL TiCN-SCiL

TiN1 TiCN2-grey TiAlN2-ML AlTiN2 CrN1 CrTiN2-ML ZrN2 AlCrN3-NL AlTiCrN3 nACo3® nACRo2® TiXCo3®

VIc2®: DLC2 cVIc1®: TiCN1+CBC1 CROMVIc2®: CrN1+DLC2 CROMTIVIc2®: CrTiN2+DLC2 nACVIc2®: nACRO2+DLC2

1

1

cVIc : TiCN +CBC CROMVIc2®: CrN1+DLC2 CROMTIVIc2®: CrTiN2+DLC2 nACVIc2®: nACRO2+DLC2

p 311-ECO 1 2 3 4 5 6 7 8 9 10 11 12

p 111

VIc2®: DLC2 cVIc1®: TiCN1+CBC1 CROMVIc2®: CrN1+DLC2 CROMTIVIc2®: CrTiN2+DLC2 nACVIc2®: nACRO2+DLC2

p 311

p 211

VIc3®: DLC3 cVIc3®: TiC+DLC3 CROMVIc3®: CrC+DLC3

p 411 TiN1 TiCN2-grey TiAlN2-ML AlTiN2

TiN1 TiCN2-grey TiAlN2-ML AlTiN2 CrN1 CrTiN2-ML ZrN2 AlCrN3-NL AlTiCrN3 nACo3® nACRo3® TiXCo3®

AlCrN3-NL AlCrTiN4 = ALL4: nACo4® nACRo4® TiXCo4®

nACoX4®

nACoX4®

CrTiN2-ML

VIc2®: DLC2 cVIc1®: TiCN1+CBC1 CROMVIc2®: CrN1+DLC2 CROMTIVIc2®: CrTiN2+DLC2 nACVIc2®: nACRO2+DLC2

PL1001 TiN1 TiCN2-grey TiAlN2-ML AlTiN2 CrN1 CrTiN2-ML ZrN2 AlCrN2 AlTiCrN3

cVIc1®: TiCN1+CBC1 CROMVIc2®: CrN1+DLC2 CROMTIVIc2®: CrTiN2+DLC2

TiN1-SCiL® TiCN1-SCiL®