News - Hydroforming.net Search - Hydroforming.net Industry - Hydroforming.net Forum - Hydroforming.net Subscription - Hydroforming.net Careers - Hydroforming.net Schedule - Hydroforming.net Home - Hydroforming.net
Hydroforming.net    
back             All rights reserved. © Hydroforming.net by Boerries Burkhardt .

 

 

Tool and Part Design for Tube Hydroforming

Dipl.-Ing. Arndt Birkert, Dipl.-Ing. Jörg Neubert
Krupp Drauz GmbH

Abstract

In designing the hydroforming process for the production of automotive structural components, a number of factors must be taken into account to ensure technically and economically viable production. The component design and the required mechanical component properties determine both the process flow and to a large extent the design of the necessary forming dies and required machines.

The following paper discusses key design and production aspects and illustrates them by reference to selected examples.

1. Introduction

A wide range of components can be hydroformed from tubular blanks for use in body, chassis and powertrain. Tube hydroforming provides many of the properties required above all for automotive structural components.

The high level of dimensional and design accuracy required for automated body-in-white manufacture is facilitated in particular by the naturally closed component cross sections and the capacity for a "minimum plasticization" of all cross-section areas and consequently low springback.

Completely closed cross sections over the full length of the component result in a high degree of torsional and bending stiffness. Similar stiffness can only be produced on shell designs by continuous welds (e.g. MIG or laser welding), which can lead to a considerable cost penalty versus largely conventional spot welds.

The parts count for complex-shaped components in particular can be reduced since the same components produced by conventional drawing techniques frequently have to be put together from several parts to avoid material waste. Moreover hydroforming permits the integration of certain additional shapes which would otherwise take the form of separate add-ons.

The process chain is shortened mainly due to a highly integrated hydroforming process which, in addition to the actual forming of the components with possible preforming and bending operations, can also include cutting functions such as piercing, partial flange trimming on extruded sections as well as cutting component ends. In addition, the naturally closed sections mean that the joining process otherwise required for half shells, with all its disadvantages as requires tooling expense and tolerance-related problems, can be dispensed with.

Requirements for minimized weight and packaging are met by the aforementioned improvements to the component properties and by the elimination of flanges.

Despite these and other advantages, the natural limits of the process must also be pointed out, since these can in part severely restrict design latitude, also in direct comparison with deep-drawn parts. Essentially this includes the variation of the component diameter over its length as well as completely different force and tension conditions for plasticizing and forming the components. The latter impact in particular on the equipment requirements and thus the cost-effectiveness of the entire process.

In the following, key aspects with regard to the design of components, processes and tools are described. This paper aims to improve understanding of the process and thus encourage engineers in particular to design components with the hydroforming process in mind.

2. Designing process-compatible component cross-section geometries

The required component cross sections are essentially produced by preforming, expanding and calibrating, with preforming being performed either on a separate tool or in the hydroforming die itself. In the following, all three process steps are discussed separately.

2.1. Expanding without axial force

When talking about expanding it is important to differentiate between expanding with and without the application of axial force. In both cases the blank must meet the conditions to permit error-free insertion in the hydroforming die. A major aspect in designing the diameter of tubular blanks is that the smallest component diameter should largely match the tube diameter of the blank. If the diameter of the tubular blank greatly exceeds the smallest component diameter, the cross section will be collapsed when the die closes, or blank material will be squeezed into the parting line of the die halves. If the diameter of the tubular blank is too small, it is possible that the component cannot be fully formed due to excessive circumferential elongation. This will cause premature component failure from necking or bursting.

When expanding under axial force, the expansion process is supported by axial compression. This permits considerably greater circumferential elongation than when expanding under internal pressure only. While expanding without axial force is shown almost as a plane-strain curve in the forming limit diagram (j2=0) and failure thus occurs at low levels of j1, expansion under axial force involves a distinct compression-tension forming process (j1>0, j2<0).

However, on comparatively long automotive structural components which in addition can also be bent, the possibilities for introducing axial compressive stresses to the required component areas are very limited, Fig. 1.

Fig. 1: Characteristic stress distribution when expanding under axial force

As the pressure increases and with it the contact between die and workpiece, bringing with it a simultaneous increase in normal contact stresses, the compressive stresses introduced at the ends of the workpiece are increasingly weakened by frictional stresses between workpiece and die the greater the distance from the blank ends. Depending on component geometry, it is possible that the forming area will stop being supported by axial force at only a short distance from the ends of the workpiece. With regard to the maximum achievable circumferential elongation, these areas are governed by the parameters for virtually plane-strain.

The maximum circumferential elongation in percentage terms which can be achieved in the plane-strain condition for various materials with a virtually square cross section is shown in Fig. 2. The fact that the manufacture of the required cross-section geometry for the strain conditions observed is based - unlike expansion under axial stress or deep drawing - exclusively on reducing the wall thickness, makes it necessary to meet close material-specific design guidelines.

Fig. 2: Maximum circumferential elongation when expanding without axial force

When hydroforming tubes or sections of aluminum alloys in particular, the maximum expansion limits are very low. It is essential that this be taken into consideration when designing components. The problem is exacerbated if expansion is to take place in pre-bent blank zones. Due to non-linear strain curves, the inability to predict the beginning of necking by means of the forming limit diagram in conjunction with the otherwise extremely useful finite-element simulation is also particularly problematical.

Extensive practical experience and above all targeted practical model tests used to quantify the influence of pre-bending on different materials with different characteristic basic cross sections is particularly helpful here. If the test design is suitable, the results can be transferred to practical components with a sufficient degree of approximation. This problem is the subject of research work currently being performed by Thyssen Krupp Automotive in conjunction with Thyssen Krupp Stahl within the ThyssenKrupp organization and also in industrial cooperative research projects in which Krupp Drauz is involved.

The question as to whether lubricants need to be used when expanding without axial force can be answered by observing the required circumferential elongation and above all the wall thickness distribution after forming, Fig. 3.

Fig. 3: Wall thinning through expansion with / without lubrication

Even with compact cross-sectional geometries, comparatively homogeneous wall thickness distributions are only achieved under suitable tribological conditions. The higher the friction between workpiece and die, and the higher the percentage circumferential elongation in a given cross section, the more inhomogeneous the wall thickness distribution over the circumference after forming.

To allow cost-effective volume production, compact geometries with large corner radii and low circumferential elongation are desirable. This allows a reduction in lubricant requirements. Ideally it should be possible to dispense with any further treatment subsequent to initial greasing of the tubes in the rolling mill.

2.2. Preforming without internal pressure

Based on the above, preforming is illustrated in the following using box-type cross sections as an example.

As already described above, when expanding round tubes of common aluminum alloys in a die of e.g. square cross section, depending on the tribological conditions and material it is possible to achieve circumferential elongation of 8-12% in straight-sided component areas with a degree of reliability suitable for volume production. This is nothing like enough to fill a surrounding square with a corner radius of e.g. 3xs0.

Similar problems are encountered in pre-bent zones, where the remaining maximum circumferential elongation even of high-quality, extremely formable steels can be well below 10% depending on bending geometry.

Producing cross sections of this type requires the use of a larger tube blank which is preformed to an intermediate geometry close to the final cross section, Fig. 4.

Fig. 4: Preforming of cylindrical tubes without internal pressure

There are two main variants for preforming with or without internal pressure. In the simpler variant, the preforming area of the die is wider than the insertion diameter of the tube. In this case preforming can generally be effected simply by closing the hydroforming die, Fig. 4, top. In the second variant the die width is lower than the diameter of the tube being inserted. In this case preforming - if a compression method is used - is made possible by providing angled die faces, Fig. 4, bottom. A decision must be made as to whether preforming is to be effected in a separate die or integrated in the hydroforming die. If preforming takes place in the hydroforming die, the component radii in the upper half of the component cross section cannot be integrated in the upper die as negative contours due to the problem of how the die is split. The radii are formed freely during the hydroforming process, which has the disadvantage of larger radius tolerances than is the case when forming is controlled fully by the die shape.

The limits of this process - in particular without internal pressure - are reached when the wall of the workpiece starts to wrinkle. Depending on requirements for the surface quality of the component in specified areas, very slight wrinkling may be permissible. Tolerance limits are generally formed by a wrinkle height which would lead to irreversible wrinkling during subsequent expansion or calibration, Fig. 5.

Ideally cross sections should be dimensioned in such a way that the wall of the workpiece does not come away from the die during preforming, and expansion of approx. 5% can take place after forming so as to plasticize the overall cross section to a large extent.

Fig. 5: Irreversible wrinkling when preforming without internal pressure

In terms of the cost-effectiveness of the two variants w>d0 and w<d0, it should be noted that an additional preforming stage - in particular for small components - can frequently be performed using the same press slide as for the hydroforming die. This involves only a slight increase in investment for tooling; cycle times (assuming suitable process design) and thus component price remain the same, disregarding the very slight addition to machine costs for the preforming die. Large-surface components such as subframes and similar are an exception.

To ensure the stability of the production process, preforming in a separate die is generally to be preferred in the case of w<d0.

2.3. Preforming with internal pressure

In particular when processing blanks which are by necessity larger than the die width, and when forming irregular component geometries, preforming with internal pressure may be required in order to allow the blank circumference required for the component to be accommodated in critical die zones without the risk of irreversible wrinkling. By way of an example Fig. 6 shows two such critical component zones on a chassis component.

Fig. 6: Influence of internal pressure on the preforming of a selected example

Preforming with internal pressure tends to diminish the cost-effectiveness of the overall process. For example, preforming with internal pressure requiring a longer cycle time has an adverse effect on unit costs for the components. Less serious are the additional costs for the preforming or hydroforming die due to the necessity of sealing the workpiece ends during die closing. This is generally effected by using an upper die of at least three parts, featuring spring-mounted sealing inserts which move in advance of the remaining die cavity space during closing, Fig. 7.

Fig. 7: Die design for preforming with internal pressure

2.4. Calibrating at the end of the expansion process

Calibration with internal pressure as the final stage of the hydroforming process has already been described in sufficient detail elsewhere /1-4/. Accordingly it is possible to determine the internal pressure required to form a required blank in a given die geometry with sufficient accuracy by analytical methods at the theoretical process design stage. The correlation between the outside component radius ra/s0 and the internal pressure required to achieve it is presented in Fig. 8 for three different materials under plain-strain conditions.

Fig. 8: Calibrating pressure curves for different materials

The maximum internal pressure required to calibrate the workpiece is generally based on the minimum outside radius required for the component. This internal pressure has a decisive influence on the design of the hydroforming die in terms of permissible stresses in the die as well as the elastic deformation of the die under maximum pressure.

There is also a linear relationship between the maximum calibration pressure and the required die clamping force, and thus the design of the clamping mechanism, e.g. in the form of a hydraulic press.

Within the framework of the technical process requirements determined by the required component properties, cost-effective process design therefore requires that the internal pressure be kept as low as possible. Minimum outside component radii for steels should be no lower than approx. 4xs0, especially as smaller component radii created exclusively by internal pressure place greater demands on the tribological system, Fig. 3.

3. Integrating cutting processes in the hydroforming process

There are two basic strategies for integrating hydropiercing into the hydroforming process, Fig. 9.

Fig. 9: Hydropiercing

For piercing from the outside inwards, there is an opening in the tool cavity which is closed by the piercing punch. During the piercing process the punch - normally hydraulic - penetrates the component wall. In this variant the internal pressure assumes the function of the female die in conventional shearing. The level of internal pressure in this case only affects the quality of the component's surface geometry directly around the hole.

Fig. 10: Hydropiercing from the outside to the inside at different internal pressures

Piercing from the inside outwards is different. In this variant the opening in the tool wall is also closed by a punch, which however in this case serves only as a pressure pad or ejector. Piercing is effected by the retraction of the counter-pressure punch when piercing pressure is attained. The forming tool now serves simultaneously as a female die, with the pressure medium assuming the function of the piercing punch. In this procedure the pressure must be high enough to cause the material to shear off along the cutting edge of the tool. The quantitative correlations are shown in Fig. 11.

Fig. 11: Correlation between hole size and necessary internal pressure when hydropiercing from the inside to the outside

In the case of a round hole, the square of the hole diameter is used in the force to be applied by the pressure medium, while there is a linear relationship between the length of the cut edge and the required cutting force. The required piercing pressure thus increases clearly as the size of the holes decreases. This means that the hole diameter is a determining factor in designing the die and the machine.

With a view to achieving a cost-effective process design for piercing from the inside to the outside, wherever possible hole sizes should be selected which can be made using pressures below the calibrating pressures required for component forming.

4. Designing dies for hydroforming and hydropiercing

4.1. Basic die design

On the one hand hydroforming tools must be suitable to produce the required component properties, on the other hand a cost-effective design must be aimed for, with size a determining factor. When it comes to die design, an optimal balance must be found between these contradictory parameters.

With regard to die size, the horizontal forces arising from the internal pressure are of major significance. Whereas the forces exerted in clamping/holding direction are compensated by the machine force, horizontal forces acting on the die result in stresses and corresponding die deformation. Both stresses and deformations should be kept to manageable levels through suitable die design, as the stresses can cause damage to the die over the operating period and deformations can lead to undesirable variances in component dimension. Fig. 12 shows a schematic die design.

Fig. 12: Cross section of a hydroforming die with inserts

It is vital that critical notching points (R) be rounded to avoid die fracture. Furthermore an O-shaped die cross section should be aimed for in which the bending arms support each other mutually and thus minimize die bending. Active die inserts should be parted at critical notching points to avoid die fracture possibly involving plastic deformation.

Forces between the die blocks are transferred via matching plates. These plates are fitted free of play with zero gap and should be designed in terms of material and surface quality to prevent wear. The fitting of inserts is of particular importance for irregular die geometries, where it is not possible to match all the parting lines by surface grinding. Even the smallest of gaps between die inserts and other active parts, such as slides used to prevent undercutting etc., will cause active parts to be displaced or tilted under high internal pressure and have a lasting effect on the dimensional accuracy of the components. Instead of exerting the surface pressure for which the die elements are designed, the displacement or tilting of individual inserts causes high localized stresses. Stresses of this nature can in turn cause plastic deformation which lead to an increasing deterioration of conditions in the parting lines with each further stroke.

To allow dies to be put into operation quickly and achieve lengthy lifetimes with high component precision, careful manual finishing of contact surfaces is essential. This finishing should initially be effected in no-load condition outside the machine in the die shop, followed by a second manual finishing phase in the press.

The choice of materials for dies is based on the function of the individual die elements, the quantities to be manufactured over the total production period, the maximum internal pressures involved and the other wear-related conditions. The workpiece material is also of decisive importance.

Risk of wear exists primarily in the sealing zone at the component ends, in exposed active areas such as concave contour zones and above all on cutting edges. In selecting die materials, criteria to be considered in addition to basic functionality (strength and toughness) include aspects such as machining, polishing and hardening capacity and suitability for required coatings. Further requirements relate to the weldability of die materials, preferably through electrode arc welding, to allow a certain scope of modification and repair work to be performed without the need to replace major parts of the die.

For the active inserts standard materials are used such as those employed for deep-drawing and cutting tools. The basic die bodies are a different matter; increasingly, heat-treatable steels from the field of die-making are used, such as grades 1.2714, 1.2738, 1.2344. These steels already combine high strength and good machinability prior to heat treatment. Even in the heat-treated state, where strengths of between 1,200 and 1,400N/mm² are possible, cost-effective machining (milling, boring etc.) is still possible with corresponding metal-cutting tools. Furthermore these materials can also be used for active zones where loads are not too high and relative movement between workpiece and die is low, as surface layer hardening by means of laser or inductive hardening processes can be effected without any problem. In this way hardness levels of ³55 HRC can be achieved.

The high strength levels indicated are required on the one hand due to the possibility of notch stresses; on the other hand it is important that insert settling is avoided over the lifetime of volume-production dies.

At present, insufficient reliable findings are available on the suitability of cast materials for the basic body of volume-production dies. Cast basic bodies have already been used successfully in prototype dies at Krupp Drauz. The use of cast materials is particularly expedient when deep pockets and cavities have to be milled out of the basic body, as is the case above all when a large number of inserts and slides as well as hydraulic piercing cylinders have to be accommodated. High metal-cutting volumes are also generally required for the production of components with multi-level three-dimensional curvature.

Structural steel grades St52-3, C45 or 42CrMoV4 are suitable for the manufacture of base plates, axial cylinder retainer plates and similar lower-stressed parts.

4.2. Designing dies with the help of finite element simulation

The complex nature of die geometries makes analysis of the stress and elastic strain conditions in the die virtually essential. Finite element simulation is a useful aid in the design of dies.

Programs such as ProMechanica, Ansys and ABAQUS are used.

Fig. 13 shows a problem zone with regard to maximum stresses. The sharp edges of the corner area shown result in very high notch stresses which would probably cause die fracture in volume production. Inserts must be provided at points such as these to alleviate the stress condition

Fig. 13: Stress distribution at maximum pressure

Finite element simulation can be used to compute the optimum dimensions of basic die blocks, inserts and slides. The integration of openings to accommodate hydraulic piercing and forming cylinders is of particular importance in the FE analysis. In addition to the sharp contour edges of the inserts, zones of this kind are subject to high stress concentrations. In extreme cases it is necessary to dispense with the piercing of particularly critical points during the hydroforming process.

5. Summary and outlook

Hydroforming has currently advanced to a stage where the various processes can be used to produce a wide variety of automotive structural components. In terms of technological and economic process design it is not always possible to optimize the integration of several process steps in the hydroforming process.

Designing and engineering hydroformed components with the production process in mind is a basic prerequisite for process optimization and should be effected at a very early stage, targeting an acceptable level of technical and equipment outlay. The decision as to which production steps are to be integrated in the hydroforming process depends on a wide variety of factors. Some of these were discussed in the above, such as the piercing of critical points in the hydroformed component.

In many cases, multi-stage production processes - e.g. the manufacture of hydroformed components in progressive dies - will enjoy dominance over single-stage component production in economic terms since such methods frequently reduce wear and scrap levels. Despite this, developments are currently under way which will allow the integration of complex end-cutting operations in the hydroforming process. To this end Krupp Drauz is performing investigations into two fundamentally different process variants.

Current research and development activities in industry and at institutes all over the world point to a further rapid growth in the applications spectrum of the hydroforming processes.

References:

/1/ Birkert, A.
Abschätzung der Kalibrierdrücke beim Innenhochdruck-Umformen
In: Blech, Rohre, Profile 9/1997
Bamberg: Meisenbach Verlag, 1997

/2/ Birkert, A.
Berechnung erforderlicher Kalibrierdrücke beim Innen­hochdruck-Umformen
In: Blech, Rohre, Profile 3/1999
Bamberg: Meisenbach Verlag, 1999

/3/ Birkert, A., Neubert, J.
Perspektiven des Hydroformens in Karosserie und Fahrwerk
In: conference publication "Prozeßkette Karosserie"
Stuttgart: Fraunhofer Institut Produktionstechnik und Automatisierung, 11./12.02.1999

/4/ Birkert, A.
Werkzeugauslegung für das Innenhochdruck-Umformen von Strangpreßprofilen aus Aluminium
In: conference publication "Das Umformen von Al im Automobilbau"
Bad Nauheim: Technik+Kommunikation Verlags GmbH, 1999

This report is published with the permission from Dipl.-Ing. Arndt Birkert and Dipl.-Ing. Jörg Neubert of Krupp Drauz GmbH

 

last update: 2000-10-13 10:45
Contact Hydroforming.net

Copyright © 1998-2000 Hydroforming.net by Boerries Burkhardt.
Hydroforming.net is not responsible for the correctness of this information.
All rights reserved.