PHS Tailored Products

Press Hardened Steels

• Introduction
• Laser Welded Tailored Blanks (Tailor Welded Blanks)
• Tailor Rolled Blanks
• Patchwork Blanks
• Tailored Properties

Introduction

Some automotive components may require higher stiffness, strength, or energy absorbing capacity in a confined local area. One method to achieve these characteristics is to spot weld separate reinforcement panels to the main component. A strategy for improving energy absorption in high-strength components involves joining a second part made from a more ductile but lower strength material.  Neither of these approaches are ideal in terms of manufacturing efficiency and product/process optimization.

Tailored parts are the term given to those parts that may have zones with different thickness, chemistry, or heat treatment, resulting in a reduced number of components, weight reduction, and/or lower costs.  These goals are achieved through part consolidation and by reducing or in some cases even eliminating joining operations.

In cold stamping operations, tailored parts (tailored products) are typically produced at the incoming coil or blank level, but are typically called “tailored blanks” or the specific process/product produced:

• Laser Welded Tailored Blanks (LWTB, also known as Tailor Welded Blanks, or TWB) or Tailor Welded Coils (TWC – not common in press hardening),
• Tailor Rolled Blanks (TRB) or Tailor Rolled Coils (TRC),
• Tailor Welded Tubes (TWT) or Tailor Rolled Tubes (TRT), or
• Patchwork Blanks,

In press hardened components, a single component may be press hardened such that it has what are known as soft zones, or areas of lower hardness possessing increased ductility. The tailored processing of these multi-strength parts can be achieved by^B-14:

• Controlling the incoming blank temperature, (Tailored Heating, pre-process)
• Controlling the quenching rate, (Tailored Quenching, during process)
• Partially tempering (Tailored Tempering, post-process).

These are typically called tailored parts or tailor processed parts. Lastly it is also possible to combine two techniques, that is, making a tailored part using a tailored blank.

Laser Welded Tailored Blanks (Tailor Welded Blanks)

Laser Welded Tailored Blanks (LWTB) are blanks that are produced by laser butt welding of two or more sub-blanks, as shown in Figure 1. In the industry, the terms Tailor Welded Blanks (TWB) and Laser Welded Blanks (LWB) are also used interchangably.^M-46

Figure 1: Steps of making a press hardened laser welded tailored blank (re-created after Citations B-14 and A-8).

Laser welded tailored blanks consist of sub-blanks with:

• Different thicknesses, allowing for use of thinner sheet steels in areas of the component having less rigorous loading requirements. Using thinner sheets saves weight.
• Different grades, optimizing the energy absorption and intrusion resistance characteristics in each area of the same part (such as an automotive B-pillar, see Figure 6b).
• A combination of both.

Laser welded tailored blanks with the same thickness and grade are used to create blanks having dimensions larger than mill rolling or processing capabilities.

Laser welded tailored blanks offer many paths to weight savings and cost reduction, including:

• Reducing the number of parts in the subassembly, such as the need for reinforcements;
• Reducing the number of required forming tools, welding fixtures, etc.; and
• Improved raw material utilization by sub-blank nesting optimization (see Figure 1 and Figure 10c).

The weld area of press hardened laser welded blanks may not transform to martensite, and therefore may show a significant reduction in the hardness. This can be attributed to weld quality and quenching rate.^{B-47, W-3}

Blanked edge geometry of the sub-blanks (notches, underfillings and weld seam pollution) affects weld quality. Separating the blanking into two operations, rough blanking and precision blanking, may improve blanked edge geometry and the resulting weld quality.^M-46

As seen in Figure 1, Aluminium-Silicon (AS) coated sub-blanks may require a secondary ablation operation. The AS coating is removed (ablated) near the weld edge, typically by using a laser. When this AS coating is not removed and filler wire is not used, the aluminium from the coating may pollute the weld. When welding two PHS1500 sub-blanks together, an aluminium-polluted weld may have significantly lower hardness, as shown in Figure 2. A part made of such a blank will fail at the weld zone, both in quasi-static and dynamic conditions.^E-8

Figure 2: Effect of ablation and filler wire on hardness distribution around the laser weld of equal thickness PHS1500+AS150 (re-created after Citation E-8).

Another common type of laser welded tailored blank is where a press hardening steel (typically PHS1500) is welded to a press quenched steel (PQS 450 or 550). Lower strength and higher ductility should be observed in the PQS. Without ablation, a hardness drop is observed in AS coated welded blank, seen in Figure 3a. In quasi-static tests, fracture was observed in the PQS base metal. In dynamic tests, the part failed at the weld zone. When ablation is applied, a B-pillar with a PQS base absorbs more energy compared to the welded blank without ablation.^E-8   In uncoated and Zn-coated steels, ablation is not required since there is no concern about aluminium pollution in the weld.^{A-68, M-2} Figure 3b shows the hardness distribution in the weld seam of galvanized sub-blanks.

Figure 3: Hardness distribution in PHS-PQS laser welded blanks.  a) AS coated sub-blanks with and without ablation (re-created after Citation E-8); b) Galvanized sub-blanks without ablation (re-created after Citation M-2). Note that the initial thicknesses of sub-blanks are different.

There are two methods of ablation.  Full ablation removes the AS coating and the interdiffusion layer (IDL) in their entirety.  In contrast, partial ablation removes only the AS coating, but the IDL remains intact. Full ablation may result in oxidation and decarburization in the weld seam.^{E-8, W-3}

In addition to weld pollution, the hardness drop in the weld seam could also be caused by the local quenching rate. When a welded blank is made using sub-blanks with different thicknesses, misalignment (Δx in Figure 4a) may lower the quenching rate. Misalignment greater than 2 mm could cause over 30% hardness drop, from approximately 500 HV to less than 350 HV.^B-47 A filler wire with high-C content could reduce the critical cooling rate, as shown in the Figure 4b. In a particular example using a filler wire containing 0.3% C presented in this image, the critical cooling rate was reduced to approximately 13 °C/s. Due to the high-C content, a 20% increase in the weld seam hardness may be possible,^E-8 as indicated in Figure 2.

Figure 4: a) Misalignment of the blank in the die could cause lower quenching rate in the weld seam (re-created after Citation B-47);  b) A high-carbon filler wire may reduce the critical cooling rate (re-created after Citation E-8).

The fourth generation Audi A4 (2008-2016 also known as B8) contained some of the earliest applications of press hardened laser welded tailored blanks. The car had five components made of blanks with tailored properties: tunnel reinforcement, left/right B-pillar reinforcements, and left/right rear rails, as shown in Figure 5. As PQS grades were not commercially available at that time, High-Strength Low-Alloy (HSLA) steels were used for energy absorbing applications. As delivered, HX340LAD + AS, had a minimum 340 MPa yield strength. Press hardened parts and their final mechanical properties are shown in Figure 5.^S-65

Figure 5: PHS applications in Audi A4 (2008-2016). The car had a total of three different components and five parts using laser welded tailored blanks (figure and table re-created using data and images from Citations S-65, D-11, V-21, W-5, and S-13).

In Citation K-25, using a laser welded tailored blank resulted in the highest energy absorbing capacity of a B-pillar reinforcement. In this study, PHS1500 (22MnB5) was laser welded to a HC340LA (uncoated HSLA steel with minimum 340 MPa incoming yield strength). Such a welded blank could absorb 3.3 kJ energy without fracture, whereas a monolithic (same thickness, same hardness all around) PHS1500 failed at 2.3 kJ (see Figure 6). PHS1500 with soft zones (see the Tailored Properties section below) passed a 2.3 kJ test but failed at 3.3 kJ.

Figure 6: Energy absorbing capacity of B-pillars increase significantly with soft zones or laser-welded ductile material (re-created after Citation K-25).

Conventional High-Strength Steels are not designed for hot stamping process. HSLA 340 and 410 MPa grades (minimum yield strength, as delivered) and CMn440 steel (Carbon-manganese alloyed, minimum 440 MPa tensile strength at delivery) may be softer than their as-delivered condition when heated over austenitization temperature and slowly cooled at 15 °C/s cooling rate. Furthermore, if the local cooling rate is over 60 to 80 °C/s, a significant increase in hardness (see Figure 7) and sharp decrease in elongation may be observed.^{D-22, T-27}

Figure 7: Vickers hardness variation of several cold stamping steels after austenitization and at different cooling rates (re-created using data from Citation D-22).

Development of PQS grades started around 2007, targeting consistent mechanical properties over a wide range of cooling rates. Currently, typical laser welded blank applications of PQS450 and PQS550 in the automotive industry include B-pillars, front rails, and rear rails. One such car with LWTB components is the 2^nd generation Volvo XC90 (2014-Present). The car has a total of 152 kg hot stamped parts, with approximately 132 kg of PHS1500 and 20 kg PQS450, comprising 33% and 5% of the BIW (excluding doors and closures), respectively. The XC90 has a total of six hot stamped welded blanks (three left and three right), as seen in Figure 8.^{L-29, L-8}  More details about welded blanks with PQS450 and PQS550 are presented in the Grades with Higher Ductility Section within our article on PQS Grades.

Figure 8: Use of laser welded PQS-PHS grades in the 2nd generation Volvo XC90 (re-created after Citation L-29).

Recently PHS1000 and PHS1200 grades have been developed. The yield and tensile strength of these grades increase with hot stamping, and as such are considered press hardening steels. ^{Y-12, G-30} More details about these grades are presented in the Grades with Higher Ductility Section within our article on PQS Grades. Renault conducted an experimental study in 2021 to replace PHS1500-PQS550 laser welded tailored blanks with those made from a PHS2000-PHS1000 combination. As seen in Figure 9, the new materials can absorb the same amount of energy with less intrusion. At the same level of intrusion, the energy absorbing capacity improves by 30%.^B-62

Figure 9: Stroke vs. energy curves of representative sub-assemblies, emulating B-pillar (re-created after Citation B-62).

Laser welded tailored blanks may also be used to create larger blanks that may not be otherwise possible or economically feasible.^M-4  Door rings represent one such application for hot stamping, as introduced by ArcelorMittal in 2010.^A-17  A prototype door ring was produced in 2012, using four sub-blanks, including one PQS550, as shown Figure 10a. The part measured approximately 1500 mm long and 1250 mm high.^{B-63, T-1}  May 2013 saw the first application of a hot stamped door ring with the introduction of the 3^rd generation Acura MDX, running from 2013-2020. The vehicle used a two sub-blank LWB door ring, both PHS1500, with thicknesses of 1.2 mm and 1.6 mm. Through sub-blank nesting optimization, material utilization was improved to 63%. Details can be seen in Figures 10b and 10c.^M-46

Figure 10: Door rings.  a) one of the earliest concepts from 2010T-1;  b) the first mass produced door ring of the 2013 Acura MDX;  c) sub-blank nesting to improve the material utilization.^M-46

For door ring manufacturing, a higher tonnage press with larger bolster area may be required, as well as a wider furnace and heavier capacity transfer systems. In most hot stamping lines, typically two or four parts are formed and quenched in one stroke (known as 2-out or 4-out) to improve productivity and reduce the total cost per piece. Due to the large size and additional requirements, door ring manufacturing is typically 1-out. However, as the part itself replaces four components (A and B pillars, hinge pillar and rocker reinforcement), it can be as cost effective as a 4-out hot stamping operation.^W-6

Although not common, the Acura TLX (1^st generation 2015-2021) and Hyundai Santa Fe (since 2018, 4^th generation) utilize single piece (not from a welded blank) door rings with 1.4 mm and 1.1 mm thicknesses respectively. ^{B-14, H-4}  The 2^nd generation Acura TLX (2021-present) has the door-ring of the 1^st generation model as a carryover.^L-61

Since its inception, laser welded door rings have been used in several Honda / Acura models. The number of sub-blanks was increased to 4 with the 2^nd generation Honda Ridgeline (2017-present). This was the first door ring application in a pick-up truck.^B-52  The Chrysler Pacifica started production in 2017 with 5 sub-blanks, as shown in Figure 11a, including PQS550 for crash energy absorption.^T-19  The 5^th generation RAM 1500 pick-up truck, which debuted in 2018, has a six sub-blank door ring, as seen in Figure 11b.^R-3  In 2018, Acura RDX became the first car to have inner and outer door rings made of PHS1500 laser welded blanks. As seen in Figures 11c and 11d, five and four sub-blanks were used respectively for the inner and outer door rings, all PHS1500. This design further allowed downgauging and lightweighting.^R-26

Figure 11: Laser welded door ring applications: (a) Chrysler Pacifica (SOP 2017) has five sub-blanks (recreated after Citation T-19); (b) RAM 1500 (SOP 2018) has six sub-blanks (re-created after Citation R-3);  Acura RDX was the first car to have two door rings: (c) inner and (d) outer, both with four sub-blanks of PHS1500 (re-created after R-26).

Currently, the European Standard for Laser Welded Tailored Blanks (LWTB), EN 10359^D-2, covers only LWTBs for cold stamping materials. This standard will be expanded to include press hardened laser welded blanks, with an expected release in 2023.

Tailor Rolled Blanks

Tailor Rolled Blanks (TRB) or variable thickness rolled blanks (VTRB) are produced by a secondary cold rolling of an already cold rolled and possibly coated coil. In this secondary cold rolling, the roll gap is adjusted during the process so that the thickness can be varied (tailored) locally, shown in the left image of Figure 12. TRBs can be an alternative to “same material-different thickness” welded blanks.^B-14 Contrary to an LWB, thickness changes are not abrupt, but instead are continuous. Thus, TRBs do not have stress concentration due to the notch effect. Problems associated with weld quality in welded blanks (pollution, geometry, quenching rate, etc.) do not apply to TRBs since welding the blank is not necessary.^H-7

Tailor rolled blanks are typically named by their thicknesses from head-to-tail, and symmetrical sections with same thicknesses are written once. For example, the B-pillars of previous generation Ford Focus (2011-2018), as shown in the right image of Figure 12, has five thicknesses in nine zones. This blank would be named as: 1.35-2.30-2.10-2.40-2.70. The process starts with a 2.70 mm thick coil, and thickness reductions up to 50% would be completed during the tailor rolling process. The typical slope in the Thickness Transition Zones (TTZ) are 1:100, meaning 1 mm change in thickness would require a 100 mm long TTZ. Different slopes could also be utilized.^{Q-7, H-8}

Figure 12: Left image: Principle of tailor rolling process (re-created after Citation Z-5);  Right image: thickness profile and nesting of a B-pillar in a tailor rolled coil (re-created after Citation Q-7).

The tailor rolling process squeezes and thins any coating, and possibly damages the coating as well.^T-4  For this reason, TRBs are typically used in dry areas. Because of similar reasons, in AS coated TRB applications, AS150 (75 g/m² on each side, Al-Si coating) is preferred instead of thinner coatings such as AS80.

One of the first press hardened TRB applications was the B-pillar reinforcement of the BMW X5 (2^nd generation, 2006-2013). The application saved 4 kg/car, compared to a monolithic press hardened part.^P-1  Other applications include: heel piece of MQB (Modularer Querbaukasten, translating from German to “Modular Transversal Toolkit) platform cars – covering many VW Group cars with transverse engine orientation, since 2012^S-107, front crossmember of MLB Evo (Modularer Längsbaukasten, translating from German to “Modular Longitudinal Matrix) platform cars – covering Audi vehicles with longitudinal engine orientation, since 2015^H-44, and roof crossmember of the 10^th generation Honda Accord (2017-present).^M-7  Many other OEMs use tailor rolled blanks, with a more detailed list presented in Citation B-14.

Figure 13: Several TRB applications in recent vehicles (re-created after Citations H-44, P-1, S-107, and M-7).

Patchwork Blanks

In a patchwork blank, one or more “patch blanks” (reinforcements) are overlapped with a “master blank” and spot welded. The spot-welded blanks are then heated in a furnace and hot stamped as a single piece in one stroke. The final part will have increased thickness in the areas of interest. A patchwork blank may reduce the need for post-forming assemblies of reinforcements, as seen in Figure 14. Since the spot welds are also austenitized and quenched, their hardness distribution is typically better than spot welding after hot stamping, as shown clearly in Figure 15.^{B-20, U-12, N-3}

Figure 14: Master blank and patch geometries of a sample B-pillar: (a) before, and (b) after spot welding, (c) after hot stamping (re-created after Citations B-14 and L-52).

Figure 15: Hardness distribution in a spot weld, comparing when spot welding is done before or after press hardening,  (Re-created after Citations B-14 and U-12)

Patchwork blanks allow for the possibility of reducing the number of forming tools and the associated fixed costs. Stamping and post-process joining costs may be reduced as well, leading to a variable cost reduction.  Depending on how the part is engineered, a weight savings may be achieved.  These benefits come at the expense of the additional blanking operation to create the patch blanks, and the pre-process welding stations.^{U-12, T-42}

Optimizing the initial geometry of the patch blank helps reduce these costs.  One approach is to use a one-step inverse simulation in the early planning / feasibility phase. In this method, the initial outline is estimated based on deformation theory of plasticity, requiring only relatively short CPU-times (in the order of a few minutes using a modern PC), with low accuracy (up to 3 mm deviation is common). A trim optimization method is recommended during the design phase of the patch blank blanking dies. In this method, an incremental solver is used with an initially assumed blank outline. Typically, the result of one-step solution is used for the first iteration. The software then compares the outline of the patch after forming and calculates the differences with the desired geometry. Then the initial geometry is modified accordingly, and another forming simulation is carried out. These iterations continue until the deviation is less than the set tolerances. For example, in a B-pillar patch optimization, ±0.25 mm deviation may be achieved in two to three iterations.^{W-8, Z-12, S-108}

Reducing the number of spot welds also reduces the cost of the patch blank. Minimizing the number of spot welds may also reduce the cycle time in welding stations. In some cases, it may also affect the number of spot-welding stations — thus, the initial fixed cost. However, severe wrinkles may form if using an insufficient number of spot welds. Using finite element analysis may assist in finding the optimum number of spot welds for formability.^A-19  In some cases, although the part could be hot formed with a smaller number of spot welds without any problems, more spot welds are applied for crash performance.^U-12

Some of the earliest patchwork PHS applications were used in the B-pillars of 3^rd generation Volvo V70 (2007-2016) and Fiat 500 (2007-present). In the Volvo V70, a total of 46 spot welds were used to create the patchwork blank. Both blanks were uncoated PHS1500, with a 1.4 mm thick master blank and a 2.0 mm thick patch.^L-53  In the Fiat 500, the master blank was 2 mm thick, supported by a 1 mm thick patch, both AS coated, as seen in Figure 16a.^Z-13  In recent years, patchwork PHS blanks have been used in more car bodies, including but not limited to several parts in the 2^nd generation Volvo XC90 (2014-Present)^L-29, rear rail of the Fiat 500X (2014-Present)^M-45, B-pillar of the Opel Astra K (2015-Present)^K-8, B-pillar of the 2^nd generation Range Rover Evoque^F-1, and several Subaru models.^{U-12, A-73}

In the rear rail of the Fiat 500X, the master blank is laser welded with 1.5 mm PHS1500 and 1.6 mm PQS450 sub-blanks. The patch blank is a 1.5 mm thick PHS1500.^M-45  A similar design with different thicknesses was also used in Fiat Tipo/Egea, as shown in Figure 16b.^B-14  For the Opel Astra, the master blank is a 1.3 mm thick PHS1500 with soft zones (see the Tailored Properties discussion below). The patch is a TRB with 1.00-1.95-1.00 thickness distribution.^K-8

Figure 16: Sample automotive applications of patchwork PHS: B-pillar reinforcements of (a) 2007 Volvo V70 (re-crated after Citation N-4), (b)  2007 Fiat 500 (re-created after Citation Z-13); and (c) rear rail of 2015 Fiat Tipo/Egea (Citation T-43, recreated after Citation B-14).

Jaguar I-PACE is an aluminium-intensive electric SUV making its debut in 2018. In this car, the B-pillar reinforcement is made up of a patchwork blank. Contrary to most earlier applications, the master blank is a PQS450, which could be joined easily to the rest of the body by mechanical joining. The patch is PHS1500, which improves the side impact and roof crush performance.^B-21  In 2018, a global Tier 1 supplier showed the possibility of using PHS2000 master blank and patch for a rear bumper beam.^N-6

Improvements in patchwork blank technology includes the weld type and quality. Conventional resistance spot welding has been used in making patchwork blanks. There are studies on using remote laser welding for this purpose as well. In one study, joining a patchwork blank with approximately 50 welds was completed in 35 seconds using 2.2 kW laser power, and in 23 seconds using 2.8 kW.^L-54  Another study showed that when laser welding is used with AS-coated blanks, weld strength is reduced by approximately 40% compared to uncoated blanks.^G-1

“Overlap patch blanks” are a sub-set of patchwork blanks. As seen in Figure 17a, instead of a master and patch blanks, two (or more) sub-blanks are spot welded over an “overlap region” to create a blank like a laser welded tailored blank. The technology was initially applied in cold stamped components.^P-4   Recently an international tier 1 supplier developed door rings and floor panels made from overlap patch blanks that were press hardened. As seen in Figure 17 b and c, a door ring can be created using 5 sub-blanks, including one PQS (shown in green).^G-3

Figure 17: Overlap patch blanks: (a) schematic of a B-pillar blank (re-created after B-75), (b) door ring concept from outer view, and (c) inner view.^G-3

One of the benefits of using overlap patch blanks is the ability to build up larger welded blanks of Al-Si coated steel without the need to employ ablation technology. The overlapped sub-blanks can be resistance spot welded together, thereby avoiding the risk of aluminum polluting the weld pool that ablation would otherwise mitigate.

Patches can be engineered to increase stiffness in critical locations, and the spot welds provide easy adjustment to both the blank and weld as needed.

Figure 18: Overlap patch blanks created with resistance spot welding eliminates the need to use laser welding and ablation techniques.^S-111

Tailored Properties

Tailored properties is a term used for the technology to make a part with hard and soft zones. Hard zones are nearly 100% martensitic, whereas soft zones have a lower percentage of martensite. This type of part may be called a “multi-strength part”. In Europe, the term “tailored tempering” may be used to denote a part with tailored properties. In this article, tailored tempering describes a part which was press hardened as a whole and later locally softened to modify properties in specific areas.

Soft zones may be used for several reasons:

• To improve crashworthiness: Local areas with higher ductility aid in crash energy absorption. An example B-pillar is shown in Figure 6. This type of usage is very common in B-pillars, front rails, and rear rails, as shown in Figure 8. The first application for this purpose was realized in the B-pillar of the first-generation VW Tiguan (2007-2018).^S-13  The technology is also used in rear rails. Both applications are shown in Figure 19. The technology is also used in rear rails of 10^th generation Honda Civic (2015-present)^C-22, and 10^th generation Honda Accord (2017-present). In this particular application, shown in Figure 20, soft zones were designed such that the rear frame would deform in a pre-defined manner and absorb the crash energy efficiently.^{C-22, M-7, K-52}  Tailored parts are used for improved energy absorption in numerous models from Audi, BMW, Ford, Honda, Mercedes and others.^B-14

Figure 19: Rear rail assembly of Honda Civic (10th gen., 2015-Present): (a) Isometric view of the assembly, (b) bottom view of the frame, during rear crash condition (re-created after Citation K-52). A similar design was also employed in Honda Accord.^M-7

Figure 20: Example uses of soft zones for improved energy absorption: (a) first application was in 1st generation (2007-2018) VW Tiguan’s B-pillars (re-created after Citations V-22 and M-8), (b) a more recent application in 2013 Ford Escape’s rear rails (known as Ford Kuga in EU, sold between 2013 and 2019) (re-created after Citation M-59).

• To improve weld/joint strength: When base metal hardness is over 350 HV, the heat affected zone (HAZ) in the spot weld may be the weakest point of an assembly.^B-20  Several other studies have proven the hardness drop and early fractures around spot weld of fully hardened parts, as summarized in Figure 21. When flanges are induction tempered (see the Tailored Tempering discussion below, a B-pillar assembly may absorb 30% more energy than a fully hardened B-pillar.^{H-61, F-2}  In multi-material mix cars, such as the 2^nd generation Audi Q7 (2015-present), “soft flanges” can be used for mechanical joining the PHS B-pillar reinforcement to aluminium components. Hemming of aluminium, around the PHS, can also be used to join the components.^H-62

Figure 21: When spot welding is done on a soft zone: (a) hardness distribution would not have a soft HAZ, and (b) early fractures at spot welds are not observed (re-created after Citations B-14, H-61, and B-64).

• For secondary bending operations: Tailored tempering (softening areas of interest after a fully hardened press hardening process) may be used in bumper beams, where a secondary bending may be required to form an inner flange.^L-40
• To facilitate trimming/piercing: Although not very common, local soft zones may reduce the force/energy requirements and improve the cutting tool life if hard trimming will be used.^L-55

There are three methods to create the soft zones leading to tailored properties:^B-14

• Tailored heating during austenitization of the blank (typically achieved in the furnace),
• Tailored quenching after austenitization (can be achieved in tempering stations or in the forming die),
• Tailored tempering after fully hardening a part (after the press hardening process).

1)      Tailored Heating (Pre-Process)

In tailored heating, areas of interest (the soft zones) are not fully austenitized. The critical heating temperature has been reported as 750 °C by several researchers. When heated below 750 °C and hot stamped, the part has a tensile strength of approximately 600 MPa and over 15% total elongation. As seen in Figure 22, mechanical properties will stay relatively constant with heating temperatures between 650 and 750 °C. Above this critical heating temperature, hardness (almost directly proportional with tensile strength) may increase significantly.^K-53

Figure 22: Effect of blank heating temperature on hardness and converted tensile strength of PHS1500 (re-created after Citation K-53).

There are several methods to achieve tailored heating. In the direct process, where an undeformed blank is being heated, there were four main methods proposed:

• Using a divided furnace,
• Masking soft zones in furnace,
• Heating by segmented contact plates, and
• Conduction heating with controlled current flow.

A divided roller hearth furnace may have gas or electric heating for the first half of its length, ensuring a uniform temperature distribution during heating. In the second half of the furnace length (soaking zone), there may be several electric heating zones across the furnace width direction that can be set to different temperatures. To simplify the schematic, Figure 23 shows a two-zone divided furnace. In the soaking zone, five-zone furnaces were already available as early as 2011. By 2018, furnaces with 32 zones were industrially used to make parts for several German OEMs.^{H-47, E-12, O-13}

Figure 23: Divided furnace concept (simplified with two zones): (a) temperature setting in the furnace affects the temperature distribution in the soft and hard zones; (b) in the tailored soaking area, up to 32 zones may be realized (re-created after Citations B-14, E-12, and O-13).

As heating of the blank in furnace is mostly achieved by radiation, an insulating mask may reduce the local temperature in the soft zones. Ceramic insulators or machined steel blocks may be used for masking purposes. Areas that are not masked will be heated above the austenitization temperature, whereas the masked areas will be at lower temperatures.^N-3  Figure 24 shows a schematic of the process. In addition to masking duty, the inlay should have enough heat capacity to absorb the heat from the blank. When steel inlays (masks) are used, they should be thicker than the blank to have the necessary heat capacity. Stainless steels could be used to avoid scaling of the steel inlay.^{B-65, B-66}

Figure 24: (a) Using masking for tailored heating (re-created after Citations B-14 and N-3); (b) an example mask and blank from Citation K-54.

Although not commonly used for mass production, it was proven that contact plates may be used in tailored heating. Blanks are isolated from the environment during contact plate heating, significantly reducing oxidation on uncoated blanks. Fraunhofer IWU in Chemnitz, Germany, has developed a lab-scale contact plate heater that can generate soft zones. In the hard zones (those heated over 900 °C), the heating rate may be as high as 300 °C/s. The heater and a sample blank are shown in Figure 25.^{S-109, G-48}

Figure 25: Tailored heating in contact plate heating: (a) right after the heating before the discharge, (b) a tailor heated blank with dimensions and approximate temperatures (re-created after Citation S-109).

Two different strategies were developed to generate tailored heat blanks using conductive heating while ensuring no current passes through the targeted soft zones.^M-60  These approaches are applicable only to rectangular blanks. Researchers in Hanover University improved the technology to heat non-rectangular blanks with tailored temperature distribution. In a sample (non-rectangular) B-pillar blank, temperature was kept at 950 °C in the heated zones and approximately at 700 °C in the soft zones. Significant temperature drops were observed in the proximity of electrodes, resulting in non-uniform heating.^B-67  Neither of the techniques are used in mass production for tailored parts.

In the indirect hot stamping process, the parts are formed prior to heating. Thus, it is not practical to apply any of the earlier strategies to get a cold zone in the part. For such components, soft zones are generated by using machined steel blocks known as absorption masses, which have high heat capacity to absorb the heat from the blank. As seen in Figure 26, correctly sized absorption masses keep the soft zones below 750 °C. When quenched, these areas have approximately 500 MPa tensile strength, over 20% total elongation (A₅₀) and over 150° bending angle according to VDA bending test. The tailored parts have narrow transition zones, and are spot weldable, both in hard and soft zones. The technology is used in the B-pillar reinforcements of several BMW models.^{M-2, K-53, R-27}  

Figure 26: Tailored heating of galvanized PHS1500 in the indirect process: (a) Blank temperature evolution in hard and soft zones, in a roller hearth furnace using absorption mass in the soft zone; (b) hardness distribution and approximate tensile strength in hard and soft zones (re-created after Citation K-53).

Tailored heating technologies are beneficial for their energy efficiency, as the soft zones are heated to lower temperatures. The technology may be applied to uncoated and Zn-coated blanks; however, AS-coated blanks are at risk for incomplete coating diffusion in the soft zones. For these reasons, similar technologies (excluding conduction heating) also are used in a secondary heating device after the furnace.^O-13  These techniques are listed in the Intermediate Pre-Cooling section below.

2)      Tailored Quenching

In tailored quenching methods, the whole blank is austenitized in the furnace and the cooling rate is controlled such that the soft zones cannot develop high percentages of martensite. This can be achieved by two main process routes:

• Intermediate pre-cooling, where a secondary furnace is employed where the temperature of hard zones is maintained, but soft zones are allowed to cool.
• In-die cooling, where a fully austenitized blank with uniform temperature distribution is placed on the tool, but the part is cooled at different cooling rates through several process routes.

Intermediate Pre-Cooling

Complete coating diffusion does not occur in AS-coated blanks subjected to tailored heating profiles. To ensure the full coating diffusion and uniformity of the coating all around the blank, the blanks must be fully austenitized. One of the earliest approaches kept the hard zones in the roller hearth furnace, while extending the soft zones out of the furnace. This technology produced a part with two zones only, with a linear transition zone (Figure 27). AS-coating is fully developed for weldability and e-coat adhesion. Tailored properties are reproducible. For this furnace-extending method, no extra investment is necessary other than automation programming.^L-56

Figure 27: Simplest pre-cooling technology: extending the soft zones out of the furnace. (a) Schematic of extending out of furnace (not to scale, from Citation B-55), (b) B-pillars made by this technology.^A-74

Intermediate pre-cooling can also be done in a divided furnace. In this case, contrary to Figure 23a, the uniform heating temperature is set over 885 °C. The soft zone area is then set to a lower temperature and thus pre-cooled. The rear rails of the 2013 Ford Escape shown in Figure 20b are produced with this technique.^M-59

Most of the tailored heating strategies discussed so far are suitable only for larger soft zone areas, but not for small areas. Intermediate pre-cooling by extending out of furnace and pre-cooling using a divided furnace strategy can only produce a two-zone tailored part, such as in Figure 28b. Since 2011, there has been an interest in producing three-zone tailored parts. By 2015, the Audi Q7 employed a three zone B-pillar with soft flanges for joining purposes. Soft spot weld areas are also under development.^{H-62, A-74, B-68}

Figure 28: Tailored B-pillar evolution: (a) monolithic, (b) two-zones tailored, (c) three-zones tailored, (d) soft flanges, (e) soft spots (re-created after Citations A-74, B-68, P-3).

To address these challenges, several intermediate cooling systems have been developed. AP&T uses multi-layer furnaces, with an addition of a TemperBox®. The blanks are austenitized in the multi-layer furnace. Before being fed into the press, the blanks are first moved into another layer (the TemperBox®) where re-heating is done with masking. Masked areas cool below 700 °C, whereas the unmasked areas are re-heated to 930 °C. The cycle time varies between 30 and 70 seconds, depending on the thickness of the blanks (Figure 29). For continuous production, one TemperBox® supports five-chamber furnaces.^K-41

Figure 29: Time-temperature evolution in the TemperBox®.^K-41

Similar technologies have been developed by other furnace makers: Schwartz has developed a thermal printer which can be a stand-alone unit or installed at the end of a roller hearth furnace.^L-56  EBNER has developed their PACC module, which can be integrated to a roller hearth furnace and cools the areas of interest by contact cooling.^O-13

In-Die Tailored Cooling

In this process, the blanks are fully austenitized in the furnace, but the cooling rate is locally adjusted. Areas with a local cooling rate over 27 °C/s are expected to transform to nearly 100% martensite. In soft zones, cooling rates should be lower than this critical number. The cooling rate is a function of the thermal contact conductance (see Figure 31a) and the temperature gradient (ΔT) between the tool surface and the blank. Thus, lower cooling rates can be achieved by^{B-14, M-61}:

• Heated die inserts,
• Die relief method, or
• Local die inserts with low thermal conductivity.

If a segment of the die is heated, sections of the blank in contact with this area have a smaller temperature gradient (ΔT), leading to reduced heat flow and lower cooling rates. In addition, sometimes this phenomenon occurs unintentionally if the dies are not cooled efficiently and hot spots are observed.^B-14

In the automotive industry, heated die inserts are used typically between 300 °C and 550 °C. Typically electric cartridge heaters are used, Figure 30a. If the inserts are heated over 420 °C (the martensite start temperature for 22MnB5), no martensite formation occurs while the blank is in contact with the dies. For productivity purposes, sheets should stay in the dies as short as possible. After industrial quenching times (10-15 seconds), soft zones may still have phase transformation during air cooling in the exit conveyor. This may cause distortion in the final part. One simulation study found that 80 seconds of air cooling was needed to transform all the austenite into other phases.^{B-14, M-61, B-69, B-70}

Figure 30: Tailored parts with heated die inserts: (a) Simulation model with cooling channels in hard zones and heating in soft zones, (b) phase transformation may continue in soft zones.^B-70

This process has been applied as early as 2009 (if not earlier) in  the Audi A5 Sportback.^B-20 The car had a three-zone B-pillar, similar to the sketch in Figure 28c. Since then, several complicated geometries have been realized in an industrial scale with “heated die insert” technology. In 2015, the 10^th generation Honda Civic was equipped with complicated rear rails, shown in Figure 19. These components are also made with heated inserts.^C-22  Also debuting in 2015, the Audi Q7 was equipped with B-pillar reinforcements with soft band and soft flanges (similar to Figure 28d).^H-62  As of 2021, heated dies appear to be one of the most common process routes to create tailored parts.

Another method to get lower cooling rates is to reduce the contact pressure or introduce an air gap between the blank and the die. As seen in Figure 31a, as the contact is lost, thermal contact conductance (h_c, the amount of heat passing through the unit area of blank to the tool) is reduced significantly. For example, at 5 MPa contact pressure, h_c is equal to 1.5 kW/m²°K. As soon as the contact is lost, the value is less than 0.3 kW/m²°K.^B-70  A Schematic showing an “air gap” design for soft flanges is presented in Figure 31c and compared with a conventional die in Figure 31b.^C-4

Figure 31: (a) Thermal contact conductance is less than 0.3 kW/m2°K, once there is an air gap (own work, raw data from Citations O-14 and M-62); (b) schematic of a conventional press hardening die, (c) introducing air gap to obtain soft flanges.^C-4

Use of insulated die inserts is another method to obtain tailored cooling.  These reduce the heat flow from the blank to the die. Typical hot forming tool steels have a heat conductivity of 27-32 W/m²°K. When ceramic insulators with less than 6 W/m²°K conductivity are used (Figure 32a), the inserts will heat over 200 °C after a few strokes. In the meantime, tool steel temperature is around 60 °C, as they can dissipate more heat energy. The strength in the soft zones may be as low as 650 MPa, corresponding to approximately 200 HV hardness.^K-55  The method may not be feasible for mass production, as the first few parts will not have the same strength/elongation level until a “steady-state” is achieved, shown in Figure 32b and 32c. In real production conditions, the production may be halted for maintenance, safety, or work hours reasons.

Figure 32: Insulating inserts: (a) experimental die set at TU Graz, (b) hardness distribution of the first part, (c) hardness distribution after a few cycles (re-created after Citations B-14 and K-55).

3)      Tailored Tempering (Post-Process Annealing)

The last method for obtaining tailored properties is to produce soft zones by annealing a fully hardened part. This can be done by induction or laser, as seen in Figure 32. Post-process annealing is relatively simple to implement, as the blank is heated and quenched uniformly in press hardening line. Annealing is added as a follow-up operation, which adds cost, but gives flexibility. The number of soft zones, their geometries and mechanical properties can be varied during the project timeline. Soft zones could be adjusted for different cars/variants that share the same component but require different soft zones.^{B-14, G-48, J-22}

With this approach, however, final properties of soft zones may vary significantly depending on the temperature-time curves. Several studies have shown yield strength may spread from 450 MPa to 1300 MPa, and tensile strength between 550 and 1350 MPa.  In addition, geometric distortion may also occur, since the heating and cooling is done in a local area.  Finally, surface and coating conditions may change, affecting weldability, corrosion resistance and/or e-coat adhesion.^{L-40, M-61, B-72}

BMW has been using induction annealed B-pillars in their 3-Series Sedan/Touring (2011-2019) and X5 SUV (2013-2018)^R-27, and possibly other vehicles. Volvo has studied the technology with induction annealing, Figure 33a.^H-61  Benteler has been using induction annealing for secondary bending of bumper beams.^L-40  Gestamp evaluated laser tempering on prototype parts, Figure 33b.^B-71

Figure 33: Post-process tailored annealing can be done by: (a) induction^H-61, and/or (b) laser.^B-71

Post Hardening Processing: Trimming 

Options include setting the trim line with a developed blank, laser cutting, soft zones, hard trimming, and hot cutting.

A developed blank might be appropriate for areas which can accommodate larger tolerances.

Soft zone development to aid in easier trimming and joining is discussed in the prior section.

Laser trimming leads to improved fatigue strength and raises the failure strains, but suffers from relatively long cycle times, high capital investment and maintenance cost.  More powerful high-end lasers measurably reduce cycle times.

Hard trimming is not usually the best long-term option for high volume applications since the hardness of the PHS part is about the same as the hardness of trim steels. A harder, more wear resistant tool steel would now fail by chipping.  

Hard trimming creates burrs, large shear zone, and microcracks in the fracture zone.  Each of these lowers the fracture strain required for failure, which lowers component crashworthiness.

Trimming of fully hardened PHS needs extra consideration since the scale formed on uncoated PHS leads to abrasive wear and Al-Si coatings may stick to the tool creating galling conditions.  A PVD coating improves wear resistance. World-class operations which use hard trimming specify advanced powder metallurgy tool steels with advanced tool coatings.

In-die hot cutting may occur after heating but before forming, and is usually limited to approximately 90° cuts.  Microstructural changes occurring during the heating and cooling cycle influences flange position tolerances.

Hot cutting may also occur after forming but before quenching.  This approach requires less force and causes reduced damage to the tooling and dies since the steel is softer to cut. One challenge is that the formed part is cooling during the cutting operation, with the part changing dimensions as it cools. Hot cutting improves cycle time and may reduce capital investment.  Grains of fine ferrite rather than martensite form at hot sheared edge, reducing the delayed fracture risk associated with hydrogen embrittlement.

Honda developed this process to both trim the part as well as create holes during hot stamping to replace time- and energy-intensive laser cutting.  (Figure 33, Figure 34).  Their approach uses a high-speed hydraulic system, and further reduces cooling time by spraying the part with water. The first production application was the 2012 Honda N-Box Center Pillar Reinforcement. ^{H-49, S-112}

Figure 33:  Honda N-Box Center Pillar Reinforcement created with In-Die Trimming during Hot Stamping. ^H-49

Figure 34: Internal Die Trimming Process at Honda.^H-49

Thanks are given to Eren Billur, Ph.D., Billur MetalForm, who contributed to this article.

Back To Top

• Introduction
• Tailor Welded Blanks (Laser Welded Tailored Blanks)
• Tailor Rolled Blanks
• Patchwork Blanks
• Tailored Properties