Can you prove it?

Jan. 1, 2007
WHEN adhesive foam tapes were introduced in light trailer manufacturing, they provided a number of product and production advantages, but sidewalls tended

WHEN adhesive foam tapes were introduced in light trailer manufacturing, they provided a number of product and production advantages, but sidewalls tended to bow out and delaminate.

Engineers had ideas on how to reduce or eliminate the delamination and allow the trailers to retain the smooth sidewalls that customers want. But how could they best prove that their proposed solutions would do the job?

The idea behind the trailer design was simple: Eliminate any mechanical fasteners from penetrating the sidewall panel and allow space for the panel to move and grow as the aluminum heats and cools creating an unconstrained panel attachment. Furthermore, prove that the adhesive tape is more than durable enough to hold sidewall panels to posts and accommodate any expansion that occurs.

But the devil is in the details. What variables are involved? How much freedom to expand is required? How should that freedom best be provided? What options are there?

To get some answers, 3M conducted tests at the Bosch Automotive Proving Grounds in New Carlisle, Indiana. The goal of the study was to compare the performance of various enclosed cargo trailer side wall attachment methods — evaluating their durability, thermal, and leakage properties.

The durability testing simulated up to 100,000 miles of use over a range of road conditions. The thermal testing simulated repeated high heat and sun exposure. The leakage testing utilized high pressure water directed at the overlap seams.

What they tested

To address the issue, two nine-foot tall, 28-ft long enclosed cargo trailers were built. Each side of the trailer had a different assembly method.

Using Bosch's system of accelerated road testing, 36,000 miles of durability testing was performed. Each trailer was inspected on a daily basis, with a thorough inspection conducted at the end of the test.

The trailers were then subjected to a thermal test, which involved 30 heating/cooling cycles. The 50-minute cycles started at room temperature, ramped up to 180-200°F and returned to room temperature. Focusing on overlap seam locations and panel to post attachments, trailers were inspected for delaminations after each cycle.

A leakage test was performed using a variable water pressure sprayer. Each overlap seam was subjected to a range of water pressures sprayed into the overlap seam. The seam was monitored from the inside of the trailer for leakage into the trailer. The mechanically-bonded trailer leaked when water was sprayed at the seam at less than 75 psi. By contrast, the taped seams withstood 3200 psi.

The VHB tape-bonded trailers with an unconstrained design outperformed the mechanically fastened trailer in durability, managing thermal expansion, and leakage, according to the results of the testing.

Different options

The two 28-ft trailers were fabricated with nine-ft sidewalls:

  • Trailer 1 had sidewalls that were mechanically fastened using 10-20 self tapping screws placed on six-inch centers along each post and overlap seam location.

  • Trailer 2 used multiple fastening methods. One side was a partially constrained design. Panels were bonded with 3M VHB acrylic foam tape and constrained at the bottom with fasteners. Panels were unconstrained at the top.

Side 2 of the trailer was a fully unconstrained design. Panels were bonded to the side posts with 3M VHB acrylic foam tape, but they were left unconstrained at both the top and bottom.

Other than the way the side sheets were secured, the sidewall construction was identical. Hat section side posts were installed on 16" centers. The trailers used the same 8" bottom rail and 6" top rail extrusions. Both had 108"-high sidewalls that were lined inside with 0.375" luan plywood fastened to the hat channels with self tapping screws.

Two 5000-lb axles were used with Goodyear Marathon ST 225 75R15 radial tires inflated to 65 psi.

Attaching the panels

To constrain the sidewall panels, fasteners were inserted through the upper and lower rails. To “unconstrain” the sidewall, a couple of changes were required:

  • Researchers determined that the sidewall panel could be shortened by a couple of inches and still have the ends covered by the upper and lower extrusion.

  • They found the existing extrusion had enough space adjacent to the sidewall to allow the mechanical fasteners to be moved down or up on the extrusion and allow for free movement of the sidewall.

  • The final adjustment was to place shims behind the extrusion to prevent any pinching of the panel. (shims were 0.188" thick to allow for the thickness of two sheets of aluminum and an overlap tape joint) This type of modification may work within an existing design or perhaps minor changes to an existing extrusion is required. The key was to leave approximately 0.25" of free space at the top or bottom edge of the panels to allow for expansion.

Exterior sidewall panels were 0.030" thick aluminum, pre-coated with a black polyester paint on the outside and a clear epoxy wash-coat on the inner surface.

To make sure bonded joints received a fair evaluation, researchers examined the peel strength of the tape before beginning formal testing. They found that the best adhesion is produced when both metal surfaces are abraded. The wash-coated surface was scuffed with Scotch-Brite 7447 Clean and Finish Hand Pad at each post location and the overlap area of the exterior black paint.

Adhesion promoter AP111 was applied to these same locations. 3M's VHB tape (CV62F, 0.062" thick, one inch wide) was placed on the posts, and then the side sheet panels were moved into position.

Panels were typically positioned on the trailer by aligning the leading/forward edge of a panel with the appropriate post. The vertical position of the panel was then addressed.

Once properly aligned, the panel was tacked into position, either by placing screws at the top, in the case of constrained panels, or peeling back 6-8 inches of the VHB tape liner and attaching the panel to the tape. Next, the overlap seam was prepared by sanding, priming, and applying the tape (0.045" thick, 1 inch wide). The remaining tape liner was removed and seams rolled to firmly anchor the panels at the seam and post locations.

Testing for durability

Testing of the enclosed trailers was conducted at the Bosch Automotive Proving Grounds. Bosch is an independent organization that provides facilities and experience in vehicle performance testing. Durability testing consisted of several steps:

  • Baseline profiling of trailer response to road conditions:

    The first part of establishing a baseline road profile was to instrument the trailer with triaxial accelerometers, which measure G-forces in the longitudinal, lateral, and vertical directions. Three accelerometers were used and placed at the rear axle, frame rail adjacent to the axles, and on the hitch A-frame. The trailer was then loaded with more than 12,000 lb. Once instrumented and loaded, the trailer was taken over a variety of real world road conditions.

    The second portion of the road test involved driving the loaded trailer at various speeds over various features located on the Bosch durability track. These features provide a range of simulated road conditions which is then compared to the real world data to formulate a test plan.

    In addition to accelerometers, string potentiometers were mounted at each post/panel interface. String potentiometers are spring loaded spools of wire that are instrumented to detect the travel, in either direction, as the string moves. The base of the device was mounted on posts and string mounted on panels to determine the relative movement of panel to post. The potentiometers were meant to provide insight to the relative movements of posts and panels and how this influences the tape performance.

  • Developing a road schedule. Bosch Engineers evaluated the accelerometer G-force data for both the real world conditions and the test track conditions. They then evaluated the test track events to determine which feature and at what speed produced similar G-forces as the road testing. Once the speed and event was correlated, they could then look at how many times they needed to operate the vehicle over these events to provide accelerated testing.

    The acceleration ratio determined by Bosch was 60:1, meaning every 1 mile on the test track was equivalent to 60 miles of normal driving. The method of translating accelerometer data to simulate road conditions is a very common and proven method not only used by Bosch, but also by other proving grounds.

  • Durability testing results. Once the course was established, the test was initiated with the mechanically fastened/constrained trailer. The trailer was run on the course for 600 miles. At each shift and at each break in the shift, the trailer was inspected. Thorough inspections were performed at the beginning of each day. These inspections focused on identifying any issues with the sidewall attachment.

After completion of the testing, researchers inspected all significant trailer components, such as axles, wheels, hitches, and welds on frame. After the mechanically fastened trailer had completed 36,000 miles, the taped trailer was tested.

No delaminations or gaps were observed on the taped trailer panels. By contrast, the mechanically fastened trailer panels had 0.045" gaps along the overlap seams, and 31% of the screws worked loose.

Once the 36,000 mile test was completed, testing of the taped trailer continued to a total of 100,000 miles of durability testing. No side panel delaminations or gaps were reported on the taped trailer sides, even though other areas of the trailer failed — including a broken weld on the front corners of the trailer, the loss of front upper trim, and two instances in which the hitch clasp bolt broke.

In addition to the visual inspections, considerable data was captured regarding the relative displacements of the sidewall attachments as measured by the string potentiometers. Typical deflections for both trailers were 0.030".

After observing the string potentiometers in action, researchers turned their attention to the level of noise or influence on the string pot that originated from “off axis” flexing of the sidewall. In order to sort this out, three triaxial accelerometers were mounted on two panels and a post that were adjacent to each other and adjacent to two of the current string potentiometers.

The accelerometer data was integrated twice to transform the data from acceleration to displacement. The differences between post and sidewall movements in each axis were studied and compared to the data provided by the string potentiometer.

The largest component of movement was lateral, which corresponds with the in and out movement of the sidewall between posts. This movement influences the attachment point of the string — essentially a small post mounted on the sidewall — and does not reflect relative movement at the post/panel interface, the researchers reported. True movement at the tape interface is much less than the 0.030" typically seen on the string potentiometers.

The other significant conclusion from the string potentiometer data is that there were no measurable differences between the two trailers. This indicates that the sidewall attachment, for this particular trailer design, had little influence on the structural rigidity of the trailer.

Thermal testing

The intent of thermal testing was to simulate a black trailer sitting in the sun in a hot region, such as Arizona. A thermal chamber provided the heat the testing required.

Within the chamber, six 35,000-Btu infrared heaters were spaced at five foot intervals along the length of the trailer. Within approximately 15 minutes, the trailer side was heated to 180°F. The heat uniformity was fairly consistent, with a maximum temperature of 205°F at the top of the panels and a minimum temperature of 165°F at the bottom of the panels. This correlated to field observations in which a 30 degree difference, top to bottom, was measured on a six-foot tall trailer.

Once 180°F was reached, temperature was maintained for 15-20 minutes. Then the heaters were turned off, the heat chamber opened, and the side wall allowed to cool back to room temperature, approximately 72 degrees. This cycle was repeated 30 times for each side of the trailer.

Researchers inspected overlap seams for delamination after each cycle and examined sections of the post to determine how well the bond to the panel performed.

After the cyclic heating test, digital topographical imaging and video imaging of the trailer side provided valuable information about the deformation created by thermal expansion stress.

Overall the two unconstrained tape conditions provided the best results. For the taped trailers only one minor post delamination was observed on one post.

For the mechanically fastened trailer, eight screws continued to loosen beyond what occurred during the durability test. The average gap at the overlap seam did not change from the initial measurements of 0.045".

Topographical mapping

In order to better understand the buckling characteristics of thermally expanded panels, experiments were conducted using a commercially available white light digital image correlation system. This system uses two digital cameras focused on the panel surface from two different angles. A speckle pattern is painted on the panels with a water soluble ink or paint. As the panels expand and buckle, the digital cameras monitor the relative motion of the painted spots, and the system's software converts this data into a topographical map.

Using an enclosure similar to the one used for thermal cycling, a section of the trailer side was heated to approximately 190°F. At this point, the doors to the enclosure were opened and the digital images were captured. The camera's field of view was approximately a 4 foot wide by 9 foot high section of the trailer side. The horizontal center of the pictures was always an overlap seam.

Topographical maps were generated for each of the side attachment methods. The maps are excellent tools for comparing the level of surface distortion created by thermal expansion, according to the researchers. The technology also provides insight into the stresses placed on the overlap seam and panel to post bonds.

Each map uses the same scale, with the X-axis ranging from -800 mm to 800 mm (+/- 31" to either side of an overlap seam) and the Y-axis ranging from -1300 mm to 1300 mm (+/- 51" above and below the vertical center point). The Z-axis represents the magnitude of surface distortion or buckling. It is color coded over a range from -10 mm to 24 mm (-0.40" to 0.950"), with the negative values referring to movement inward to the sidewall plane. Each change in color grade represents a change of a little over 1 mm (0.040") in magnitude.

Checking for leaks

The trailers also were tested for leak resistance. A variable-pressure washer wand was used to direct a stream of water towards the overlap seam at a 10° angle. The tip of the pressure washer wand was held 12" from the seam. The seams were monitored for leakage from the inside of the trailer.

On the mechanically fastened trailer, all of the overlap seams leaked at garden hose pressure (less than 75 psi).

The unconstrained tape-bonded trailer was subjected to increasing pressures starting at 1500 psi up to 3200 psi. No leakage was observed on any of the six overlap seams, even when the 3200 psi water stream was directed at one location for the duration of 30 seconds. This testing was performed after the 100,000 mile durability testing and thermal testing.

The integrity of the taped joints received additional testing. With two 12.5" × 11" taped panel assemblies simulating an overlap seam as targets, the pressure washer wand was set to the maximum of 3200 psi and directed at the overlap seam at a 10° angle from a distance of 12". No damage to the seam was observed at this setting. The tip of the wand was then moved in 1" increments towards the overlap seam. One of the panels did not display any degradation even when the wand was held for 30 seconds at a distance of 1". On the other panel, some trimming of adhesive that extended beyond the seam was observed at a distance of 4".

At a distance of 1", approximately 0.5" - 1" of each edge separated and partial cohesive failure was observed on the remainder of the seam. This partial cohesive failure extended approximately 0.125" into the 1" tape width. However, the bond stayed intact even under these conditions.

Running the course

What the test trailers encounteredEvent Dimension Spacing Intent Inverted Chatter Bumps 1.0" deep
24" long
full road width 5 sets of 10 bumps, each set having a variable 5-9 feet spacing between bumps. 350 feet of recovery area between sets. Moderately severe vertical accelerations. Bumps located on inclines, creating large pitching movements and fore/aft accelerations. Chatter Bumps .75" height
15" long
12" wide Sinusoidal shape bump with peak to peak distance of 3.5 feet Small vertical displacements at high frequency Impact Bumps 1.25" height
18" long
½ road width Staggered side to side with variable spacing between 10 and 20 feet Harsh vertical displacement with side to side bending due to staggered placement Undulating surface Variable wave forms in both down and cross road directions. 1500 feet of variable frequency wave forms with a maximum height of 12 inches. Produces undamped resonant frequency vibration. Typically exhibits increasing fore/aft pitching and very high suspension deflections. Rumble Strips 0.5" deep
4.0" long
12 feet wide 4" spacing Small vertical displacement,high frequency event Resonance Road 0.875" deep
18" long
½ road width Variable distance of 3.5 to 9.5 feet between pairs. 10 Oriented, 10 and 15 degree angle steps with a total of 247 steps in each wheel path. Vertical displacement with varying frequency. Staggered Blocks 6" high
23 feet long
½ road width Six staggered bumps alternated over a 126 foot path, inducing frame twist. Strategically placed to torsionally twist trailer body in both clockwise and counterclockwise directions Gravel Roads Range of low level inputs Varied, graded periodically