BRIEF DESCRIPTION OF DRAWINGS
FIG. 1 is a prior art illustration taken from Sturrus U.S. Pat. No. 5,395,036, showing a B beam.
FIG. 2 is a perspective view of a first embodiment of the present B-shaped beam.
FIG. 3 is a cross-sectional view taken along line III-III of the B-shaped beam in FIG. 2.
FIG. 4 is a three-point bending test fixture.
FIGS. 5-6 are top and cross-sectional views of a second embodiment B beam with power ribs.
FIG. 7 is a cross section of a prior art B beam similar to the inventive B beam of FIG. 5-6, but having a cross section with a vertically-linear front wall.
FIG. 8 is a graph showing the results of a three-point bending test conducted on the B beams of FIGS. 5-6 (B beam with power ribs) and FIG. 7 (B beam without power rib).
FIG. 9 is a photograph of the top of the straight B-shaped beams after the test shown in FIG. 8, the damage showing a different stress distribution and impact deformation, the B-shaped beam with power rib (shown at a top of the picture) having a wider stress distribution and wider region (less localized region) of impact deformation than the B-shaped beam without power rib (shown at a bottom of the picture). FIG. 9A is a line drawing of FIG. 9.
FIGS. 10-11 are computer-generated front views of the B-shaped beams in FIG. 9, the FIG. 10 showing an FEA analysis of stress distribution during bending of the B-shaped beam with power rib (FIG. 9, top of picture), and the FIG. 11 showing an FEA analysis of stress distribution during bending of the B-shaped beam without power rib (FIG. 9, bottom of picture). FIGS. 10A-11A are line drawings of FIGS. 10-11.
FIG. 12 is a graph of displacement versus bending load comparing test results on a three-point bend test (see FIG. 4) of the B-shaped beam with power rib (see FIGS. 5-6) as compared to a B-shaped beam without power rib (see FIG. 7), the comparison being made using FEA correlation techniques to show weight-equivalent B-shaped beams.
FIG. 13 is a top view photograph of two B-shaped beams (see B beam with power rib in FIGS. 5-6 and B beam without power rib in FIG. 7) after a 5 mph flat barrier physical impact test, the top beam in the photograph being of a B-shaped beam with power rib, and the bottom beam being of a B-shaped beam without power rib.
FIG. 14 is a graph of intrusion distance (movement of a center of the beam toward a vehicle's radiator) versus load, comparing test results of a 5 mph flat barrier physical impact test of the B-shaped beam with power rib and of the B-shaped beam without power rib.
FIG. 15 is a graph of intrusion distance versus load, comparing test results of a 5 mph flat barrier physical impact test of the B-shaped beam with power rib and a standard B-shaped beam without power rib (cross section with vertically-linear front wall), but the data for the B beam with power rib is adjusted (using FEA correlation techniques) to account for a reduced wall thickness in the B beam with power rib so that the B beam with power rib has an equal mass to the illustrated B-shaped beam without power rib.
FIG. 16 is a graph of intrusion distance (rearward movement of the beam during impact) versus load, comparing test results of a 10 km/h IIHS bumper barrier physical impact test of the B-shaped beam with power rib and the B-shaped beam without power rib (i.e., flat face wall).
PRIOR ART
FIG. 1, taken in part from Sturrus U.S. Pat. No. 5,395,036, is exemplary of B-shaped bumper reinforcement beams having a transverse cross section with a vertically-linear front wall. The illustrated B beam 200 in FIG. 1 includes a “vertically-linear” front wall 201 formed by co-planar edge portions (“wings”) 202, 203 welded to a center web 215. It is noted that many B beams include a single continuous portion of sheet forming their entire front wall. In such B beams, the weld(s) is located in another location on the B beam. The B beam in the Sturrus '036 patent includes a cross section with two tubes 205 and 206, one spaced above the other by web 215 when in a vehicle-mounted position, such that four walls (213, 214, 216, 217) extend horizontally from the front wall, with the coplanar walls 212A and 212B closing a rear of the tubes. The B beam in Sturrus is swept (i.e., longitudinally curved), however it is noted that many B beams are straight (i.e., longitudinally linear).
Detailed Description of Preferred Embodiments
As will be understood by persons skilled in this art, in a pure bending condition, the theoretical beam maximum bending stress is predicted by the following equation: σ=M/Z, where M is the bending moment and Z is the plastic section modulus. When σmax≦σyield, the beam will theoretically not buckle under bending moment M. Therefore just before beam buckle, Mmax=σyield×Z. Mmax is often referred to as section flexure rigidity. This theoretical value M must be correlated to actual test results (actual Mmax), since actual values vary. For example, as illustrated and discussed hereafter, a ratio of the actual Mmax value to the theoretical Mmax value can be as low as 50% to 60% in a B beam with a cross section having a vertically-linear front wall, such as the prior art B beam shown in Sturrus U.S. Pat. No. 5,395,036 (see FIG. 1 herein and discussion above).
We have discovered that a ratio of the actual Mmax to the theoretical Mmax value can be raised to about 70% to 80% or higher in a B beam 20 incorporating an integral channel-shaped reinforcement rib 33 (referred to herein as a “power rib”) into the unsupported portions of an otherwise generally vertically-linear front wall in B beams. Our testing shows that this rib is preferably at least about 8 mm deep, and at least about ⅓ of a height of the unsupported portion of the front wall extending over individual tubular sections. This is considered to be an extra-ordinarily surprising and unexpected result, given that the (vertically-linear) front wall of a B beam is already supported near its center by the middle horizontal walls of a typical B beam. This is especially surprising when the unsupported span in the vertically-linear front wall (i.e., that portion of the front wall that extends across a tube section) in bumper reinforcement beams is typically only about 40 mm to 65 mm, and yet a dramatic improvement in actual bending strength is still achieved. As result of the present inventive concepts, new design choices exist. For example, existing B-shaped bumper reinforcement beams can be reduced in wall thickness (i.e., to save weight while still providing a same impact strength). Alternatively, the impact strength of existing B-shaped bumper reinforcement beam designs can be increased without added weight or cost (i.e., simply by adding the power rib to a flat front wall without changing sheet thickness or part design). Alternatively, new B-shaped bumper reinforcement beams can be designed with thinner front-to-rear dimensions yet with equal strength to other “deeper” designs (thus saving package space at a front of the vehicle and also reducing intrusion distance during an impact).
The illustrated B-shaped bumper reinforcement beam 20 (FIGS. 2-3) is rollformed from a sheet to define a pair of vertically spaced tubes 21 and 22 (when in a vehicle mounted position). The B beam 20 includes a front wall 23 that extends from top to bottom of the beam and that defines a front of each tube. The unsupported front wall portions over each tube are generally vertically-linear and aligned, however the front wall 23 includes a channel-shaped rib 33 located on the front wall centrally over each of the tubes 21 and 22. The ribs 33 stabilize the unsupported front wall portions over each tube in a way that provides improved impact strength, as discussed below. The illustrated rib 33 is formed inwardly so that it does not protrude in front of the front wall of the beam 20. By this arrangement, the rib 33 is not initially impacted by an object (such as a pole or tree). Thus, the ribs 33 are not bent during initial impact, allowing them to stabilize the front wall of the beam for a longer period of time during initial impact. However, in a broadest sense, a scope of the present invention is not believed to be necessarily limited to inwardly-formed ribs (33). Also, the illustrated ribs 33 are centrally formed over each tube 21 and 22, and the illustrated tubes 21 and 22 are similar in size and shape, as are the ribs 33. However, in a broadest sense, a scope of the present invention is also believed to include a B beam where the two tubes are not of equal size and/or shape, and where additional tubes may be present, and where the ribs are not necessarily centrally located over each tube, nor where the ribs are of equal size and shape.
The illustrated B beam 20 of FIGS. 2-3 is preferably formed from a sheet of material, such as 1.0 mm to 2.2 mm steel (or more preferably 1.1 mm to 1.6 mm thick, or most preferably 1.2 mm to 1.4 mm thick, depending on functional requirements of the bumper system). The sheet has a tensile strength of 40 KSI, or preferably 80 KSI, or more preferably 120 KSI (or in some circumstances 190 KSI). The upper and lower tubular sections 21 and 22 are spaced apart and connected by a pair of juxtaposed intermediate vertical walls 23 and 24. The upper tubular section 21 includes horizontal walls 25 and 26 interconnected by front and rear vertical walls 27 and 28. The lower tubular section 22 includes horizontal walls 29 and 30 interconnected by front and rear vertical walls 31 and 32. The illustrated vertical wall 23 is made by coplanar edge portions of the rollformed sheet that are welded at a center location to web 24 to form a “vertically-linear” front wall. However, it is contemplated that the vertical wall 23 could be formed from a continuous single portion of sheet material (in which case edges of the rollformed sheet would be joined in a different area along a perimeter of the B beam). A pair of mounting brackets 22′ are attached to the rear walls 28, 32 near each end. The illustrated mounting brackets each include flanges welded to the swept beam 20 and each bracket further includes coplanar aligned portions with apertures adapted for bolted attachment to a vehicle's frame rails.
In the illustrated arrangement of FIG. 3, the tubular sections 21 and 22 have a vertical dimension D1 of about 1.5 times a depth dimension D2 of the tubular sections. The illustrated beam 20 itself has a vertical total height D3 of about 3-4 times a depth dimension D2 of the tubular sections, and the power ribs have a vertical dimension D4 that is about 33% to 50% of a height of the respective tubular sections and a depth dimension D5 is at least about 10% to 35% (and more preferably about 25%) of the depth dimension D2. The illustrated B beam in FIG. 3 has the following actual dimensions: individual tube height dimension D1 of each tube is about 65 mm, total beam depth dimension D2 is about 40 mm, total beam height dimension D3 is about 150 mm, rib height dimension D4 is about 20 mm to 30 mm, and rib depth dimension D5 is at least about 8 mm (or more preferably 10-15 mm).
It is noted that the present invention of ribs 33 in the unsupported portions of the front wall of B beams is particularly important when B beams are made from thinner material, and/or when made from high strength material, and/or when the B beams cross section has a high height-to-depth ratio. The reason is because B-shaped bumper reinforcement beams are often made “stronger” by using ultra high strength steel, because the material's high yield point enables higher section flexure rigidity. This allows lower thickness materials to be used, saving weight. B beams with high height-to-depth ratios provide a wider impact face while still providing good bending strength. However, it has been observed that in B beams with vertically-linear front walls have increasingly poor actual bending strengths, especially at lower material thicknesses, (such as 2.2 mm or less, and especially at 1.4 mm-1.2 mm or lower thicknesses) and/or at higher material tensile strengths (such as 80 KSI to 190 KSI or higher) and/or with cross sections having high height-to-depth ratios (such as where the beam is 150 mm high, 40 mm deep, each tube height being about 65 mm high and the tubes being spaced about 20 mm apart). In such B beams, our testing shows that the B beam's actual bending strength is substantially below the theoretical bending strength, often only 50%-60% of the theoretical bending strength. This is apparently due in significant part to the local instability of the front wall in unsupported regions of the front wall over each tube in the B beam. This local instability reduces the actual Mmax significantly below the expected theoretical value . . . such that the actual strength of these B beam falls to only about 50%-60% of the expected theoretical value.
In the testing described below, the actual Mmax value of B beams were raised significantly from about 50%-60% of their theoretical bending strength to about 70%-80% in a B beam having power ribs. In at least one test, the actual bending strength was raised almost to the theoretical bending strength. We believe that this can be explained in part by the different type of failure mode exhibited between the B beam 20 and the prior art beam of Sturrus '036 patent. In B beams having cross sections with vertically-linear front walls (and no “power rib”), the front walls appear to kink and collapse prematurely during an impact due to compressive longitudinal forces developed in the unsupported portions of the front walls, which results in localized instability of adjacent walls and then premature total failure of the beam. Contrastingly, in B beams having cross sections with front walls having power ribs (i.e., channel ribs formed in unsupported front wall portions extending over the tubes), the front walls appear to better resist premature kinking and collapse. This results in a stronger beam (i.e., a B beam having an actual bending strength closer to its theoretical bending strength). Notably, we believe that this premature collapse due to kinking from compressive longitudinal forces is due to a somewhat different failure mode than a theoretical bending failure. Specifically, the theoretical bending strength increases when a beam's bending moment M value increases. However, when material from the front wall is used to form a channel-shaped rib into the face of a beam, it actually decreases the beam's theoretical bending moment since material is moved from the extreme front of the beam (where it contributes a greatest amount toward beam bending strength and bending moment “M”) and is moved toward a center of mass (where it contributes a lesser amount toward the beam's bending strength).
To test the present theory, a three point bending test fixture 300 was used, as shown in FIG. 4. The test fixture 300 included lower supports 301 spaced apart 880 mm and having a curved upper surface 302 for engaging the beam. The test fixture 300 further included an upper head 303 having a lower surface 304 defining a radius for pressing against a center of the beam under test. The beam (illustrated by beam 305) was positioned on the supports 301 for engagement at its mid-point by the upper head 303.
Early experimentation was conducted using two similar beams, one having power ribs (see the B beam 20 with power ribs 33 as shown in FIGS. 2-3) and one not having power ribs. The beams were identical in every aspect except for the power rib (33). Specifically, they were made from exactly the same material coil (i.e. same material properties and thickness), had a same longitudinal curvature, and a same total vertical height and depth. The beam 20 with power rib 33 had a dramatically improved bending strength by about 20% at bending displacements near failure. This was extremely surprising to us.
To further test the present concept, a second beam 20A was constructed with power ribs 33A in its front wall 201A over its tubes (FIGS. 5-6) and a second beam 320 was constructed with vertically-linear front wall 321 without power ribs (FIG. 7). The beams 20A and 320 each had a total height of 115 mm, and a total depth of 70 mm, and mounts 22A′ welded to their rear surface. The beams were both made from sheet material having a tensile strength of 190 KSI and a 1.16 mm thickness. The beams 20A and 320 each had top and bottom tubes with a height of 45.5 mm and depth of 70 mm, and that were spaced apart about 24 mm. The top and bottom tubes 205A and 206A define four horizontal walls (213A, 214A, 216A, 217A) (when in a vehicle-mounted position), with each horizontal wall having a slight bend at its mid-point, with the forward half portion of the horizontal walls being relatively parallel and horizontal, and with the rearward half portion of the horizontal walls being tapered inward toward a rear of each tube. In the beam 20A, the front wall had power ribs 33A formed centrally over each tube in the unsupported areas of the front wall, the power ribs each being about 15.49 mm deep and about an equal width of about 15.49 mm (at their mid-depth level). The front wall included a radius R7 of about 7 mm that occurred in several locations, including on the top tube at the upper corner from the top wall onto the front wall, at the upper corner as the front wall transitions into the top power rib 33, at a bottom of the power rib 33, and at a corner from the power rib 33 onto the front wall near the center web. The front wall portion over the bottom tube includes radii R 7 at similar locations as the top tube. As noted above, the beam 320 (FIG. 7) had a cross section with a vertically-linear front wall (i.e., no power ribs). The beam 320 was otherwise similar to beam 20A.
A three-point bend test (see fixture in FIG. 4) was conducted on the swept B section beam 20A with ribs 33A (FIGS. 5-6) and on the swept standard B beam 320 with flat face (without rib) (FIG. 7). In the three-point bend test (FIG. 8), the B beam 20A with power rib 33A gave an improved actual max load=60.2 kN. Contrastingly, the standard shape B-section 320 (without power rib) only gave an actual max load=43.9 9 kN. Also, the B beam 20A with power rib 33A provided a larger deformation area (see upper B beam in the photographs in FIG. 9), while the standard B beam 320 showed evidence of kinking and provided a more localized buckled area (see lower B beam in FIG. 9). This is well-shown by the FEA analysis (see FIGS. 10-11) which gives a visual image of stress representing a three-point bend failure mode. Specifically, stress was distributed over a much large area A1 in the B beam 20A with power rib 33 (FIG. 10), resulting in higher load carrying capacity. Contrastingly, stress was more concentrated in a much more localized area A2 resulting in premature buckling, a sharper buckled point, and a lower load carrying capacity in the B beam 320 with vertically-linear front wall (FIG. 11).
The maximum bending moment was determined on the beams 20A and 320 to better understand the present test results. As noted above, the theoretical maximum bending moment equals the plastic section modulus times the yield strength. (i.e. Mmax=Z×YS.) For the B beam 20A, the theoretical Mmax=13938 mm3×1224 MPa=17060 Nm. For beam 20A, the actual Mmax=PL/4, where P=load, and L=span of test fixture. The actual Mmax therefore was (60.2 kN×880 mm/4)=13244 Nm. Therefore, the ratio of the actual/theoretical Mmax=(13244/17060)×100%=77.6%. For the B beam 320, the theoretical Mmax=13494 mm3×1224 MPa=16517 Nm. For beam 320, the actual Mmax=PL/4, where P=load, and L=span of test fixture. The actual Mmax therefore was (43.9 kN×880 mm/4)=9658 Nm. Therefore, the ratio of the actual/theoretical Mmax=(9658/16517)×100%=58.5%. We conclude that, by decreasing the amount of premature thin wall buckle in the front wall, the B beam 20A with power rib 33A is able to get much closer to the theoretical Mmax value than the B beam 320 vertically-linear front wall (i.e., without a power rib). We believe that on thicker beams (i.e. beams with a deeper horizontal section depth), this ratio will go even higher, such as to 85% to 95% or above, due to the type of failure and stresses when bending such beams.
To further illustrate the present inventive concepts, we wanted to compare two B beams of equal weight, one B beam being like B beam 20A with power ribs 33A in its face, and one B beam like B beam 320 having a cross section with a vertically-linear front wall (and no power ribs). Notably, the B beam 20A must be made from a slightly wider sheet since it must include additional material in order to form the channel-shaped power rib 33A. Thus, an “equal weight” B beam 20A requires a thinner wall thickness in order to be equal weight to a B beam 320 with no power rib. We used finite element analysis to generate data for a hypothetical B beam with power rib (identified as a B beam section with power rib, called the “WESWPR B beam”) but having a reduced wall thickness so that it had a same weight as a B beam without power rib (identified as a B beam section with no power rib (called the “WENOPR B beam”). The result was an WESWPR B beam (with power ribs) with a wall thickness of 1.15 mm was a same weight as an WENOPR B beam (no power rib) having a wall thickness of 1.23 mm. We refer to the WESWPR B beam and the WENOPR B beam as “weight equivalent B sections.”
The data in FIG. 12 compares the strength of this hypothetical WESWPR B beam with power rib (i.e., wall thickness 1.15 mm, sheet material of 190 KSI tensile strength) against the WENOPR B beam with linearly-vertical front wall (no power rib, wall thickness of 1.23 mm, 190 KSI tensile strength material). Specifically, the WESWPR B beam had a weight/length of 0.0045 kg, an actual max load of 56.1 kN, and an actual Mmax of 12342 Nm. The WENOPR B beam has a weight/length of 0.0045 kg, an actual max load of 43.9 kN, and an actual Mmax of 9658 Nm. This shows a surprising 25% or more increase in actual Mmax for a WESWPR B beam (with power rib) over an equal-weight WENOPR B beam (no power rib) at significant displacements of over 25 mm.
We also dynamically tested the present inventive B beam. One commonly used dynamic test is known as the “5 mph flat barrier physical impact test.” Such tests are commonly known and do not require a detailed explanation for those skilled in the art of automotive bumper design. Basically a vehicle-simulating wheeled sled supports a bumper system including a B beam attached to its face, and a polymeric energy absorber 345 attached to a front of the B beam. The sled is impacted against a flat barrier while moving at 5 mph. (Alternatively, the sled is stationary, and a pendulum impacts the sled/bumper arrangement at 5 mph.) In the present test, the sled weight (“vehicle mass”) was 1800 kg (60% at the front and 40% at the rear). Another commonly used dynamic test is called the “10 km/h IIHS Bumper Barrier Physical Impact (100% beam to Barrier Overlap).” In this test, bumper B beams are impacted against an obstacle with an impacting structure simulating another bumper. Again, this test is understood by those skilled in the art of bumper design, such that a detailed explanation is not required for an understanding of the test. In our test, a same 1800 kg sled weight was used.
FIG. 13 is a photograph of a B beam 20A with power rib 33A and a B beam 320 without power rib after a 5 mph flat barrier physical impact test, as described above. Both beams 20A and 320 included an identical polymeric energy absorber 345 attached to and abutting their front wall. As can be seen, the 20A B beam with power rib exhibited a distributed impact zone Z1 without any well-defined buckles (see center region). Contrastingly, the B beam 320 with vertically-linear front wall (i.e., no power rib) includes a well-defined buckle near its center at location Z2. This result occurred despite the presence of the polymeric energy absorber on the face of the B beams. Notably, the polymeric energy absorber tens to help soften an impact and spread stress. Yet, the premature buckling problem still occurred in the B beam without rib, and did not occur in the B beam with ribs 33.
FIG. 14 shows the data from the 5 mph flat barrier physical impact test on the beams 20A and 320 shown in FIG. 13. The data shows that the B beam 20A provided a significantly higher impact strength (i.e., about 129 kN total load) than the B beam 320 (which provided a 110.5 kN total load). Also the B beam 20A with power rib had a front face intrusion of 53.8 mm and a back face intrusion of 31.5 mm, while the B beam 320 without power rib had a front face intrusion of 62.2 mm and a back face intrusion of 54.2 mm. It is noted that both beams 20A and 320 were impacted with the same energy. Therefore, as shown by the data, the B beam 20A recovered from its maximum back face intrusion of 53.8 mm to a recovered final position of about 23 mm permanent set . . . while the B beam 320 recovered from its maximum back face intrusion of 62.2 mm to only about 37 mm permanent set.
FIG. 15 uses the data from FIG. 14, but is modified using FEA analysis to generate data for comparing weight-equivalent B beams under the 5 mph flat barrier test. In FIG. 15, the B beam (20A) with power rib (using data from the correlated FEA model) had a 1.15 mm thick material, and generated a maximum load of 131.6 kN, a front face intrusion of 51.4 mm, and a back face intrusion of 26.5 mm. Contrastingly, the weight-equivalent B beam (320) without rib had a 1.23 mm thick material, but generated only a maximum load of 110.5 kN, a front face intrusion of 62.2 mm, and a back face intrusion of 54.2 mm. Notably, the B beam 20A had a 49% decrease in back face intrusion using an equal mass beam to the B beam 320.
Also, FIG. 16 shows the results of a test conducted on beams 20A and 320 having an equal wall thickness under the 10 km/h IIHS (Insurance Institute of Highway Safety) Bumper Barrier Physical Impact test with 100% beam to barrier overlap. The B beam 20A with power rib 33A provided a maximum front face intrusion of 111.7 mm, a maximum back face intrusion of 40.4 mm, and a maximum load 131.8 kN. Contrastingly, the standard shape B beam 320 with flat face with equal thickness material provided only a maximum front face intrusion of 121.6 mm, a maximum back face intrusion of 83.2 mm, and a maximum load of 97.6 kN. Thus, the B beam 20A with power rib again significantly outperformed the B beam 320 without power rib (i.e., with vertically-linear front wall).
To summarize, we have discovered that a B-shaped bumper reinforcement beam with power rib in its front wall centered over each of its two tubes has a dramatically and significantly improved actual impact strength as compared to a similar B-shaped bumper reinforcement beam with cross section showing a vertically-linear front wall. The improvement in the B beam with power rib is shown by significantly improved: increased actual bending strength, increased actual dynamic impact strength, photographs showing more distributed deformation at a point of failure and showing greater spread of stress in the beam with power rib, reduced actual back face intrusion, and reduced actual front face intrusion. We conclude that the addition of power ribs in unsupported portions of the front wall over tubes of a B beam is significant. As a result, the actual impact strength of B beams are much closer to theoretical values when power ribs are added. Surprisingly, this is true for B beams having tubes where an unsupported portion of the front wall spans only 40 mm, and is especially true where the material thickness is 2.2 or lower (and especially at 1.4 mm or lower), and when the material strength is above 40 KSI tensile strength (and especially at 80 KSI-190 KSI tensile strengths or greater), and when the rib is at least about 8 mm or more preferably about 10-15 mm.
It is to be understood that variations and modifications can be made on the aforementioned structure without departing from the concepts of the present invention, and further it is to be understood that such concepts are intended to be covered by the following claims unless these claims by their language expressly state otherwise.
Patent Application
20080093867
