Introduction: Suitability Testing of Two Kinds of Loop Pile (Velcro®) Material for an IPad Case Refurbishing Project

About: Math/Science Educator and writer with more than 30 years of experience in science and industry.

Two types of loop pile material (Perfect® Knit Loop and Perfect® Tricot) were evaluated for use in an iPad case refurbishing program. Perfect Knit Loop was shown to have superior adhesion and wear properties making it a better choice for long-lasting repair of worn iPad cases.

Mobile device use in K-12 classrooms has increased dramatically recently in the United States and many districts are implementing on-to-one mobile device programs [1]. As more schools implement these programs for students, costs are incurred for maintaining the devices.

At one middle school, approximately 750 iPads will be in use at the end of a three year phase-in. The devices have protective cases and these cases have been shown to have a life span of 1-3 years, eventually failing to hold the iPads when the hook and loop material (Velcro®) that attaches the iPads in the inner case to the outer case becomes worn. This leads to user difficulties as the devices cannot be stably oriented upright in either the landscape or portrait positions. Worn hook and loop pile also increases the chance of damage, such as broken screens, since the devices are more likely to fall out of the cases.

The iPad cases in question are shown in Figure 1. The inner case that encloses the iPad has a full covering of loop pile on the back. The loop pile attaches to an approximately rectangular section of hook material on the inside of the outer case. As shown in Figure 2, the location on the loop pile that is repeatedly attached and detached from the hook side eventually wears and the inner case no longer adheres to the outer case.

After considering various ways that the iPad cases could be refurbished, replacement of the worn loop pile seemed like a viable method. Through an Internet search a company was found that provided potentially suitable material in bulk. Two samples of loop pile material were provided by the company.

The experiment described here was designed to answer the question: Which of the two types of loop pile material would be most effective for refurbishing worn inner iPad cases?

The following hypothesis was developed: If the pull force required to detach both thick-pile and thin-pile loop fastener material from the hook material is measured under equivalent conditions, then the thin pile loop will lose its ability to adhere to the hook side faster, as detected by a reduction in required pull force.

Step 1: Materials and Methods

Material Tested. Two types of loop pile material were evaluated. The “thick” material was Perfect® Knit Loop and the “thin” material was Perfect® Tricot (both from (PERFECTEX, LLC, Huntington Beach, CA). The thin sample looked and felt similar to the original loop pile used on the iPad cases.

Test Setup. The experimental apparatus is shown in Figure 3. For each type of loop material, a 1cm x 10cm (±0.1cm) piece was placed on the hook portion of a used iPad case. The iPad case was fastened to a work bench with three clamps.

A flat piece of wood was placed over the sample and a 1.13 kg (2.5 lb) weight was rolled across the wood twice to set the loop pile on the hook material (Figure 4). This kept the attaching force consistent for each run of the experiment. Pull force measurements were made with an Ohaus Delta spring scale, part number 8075-30 (Flinn Scientific, Batavia, IL). The accuracy of the spring scale was checked with a set of calibration masses and was within 3-5% percent of the indicated readings.

Procedure. Loop pile samples were affixed at one end with a piece of 24-gauge brass wire formed into a loop. Each sample was placed on the test setup and pulled at an angle of approximately 70° from the horizontal as shown in Figure 3. Whenever the material under test detached from the underlying hook layer with a distinct “tearing” sound, the force in Newtons indicated by the spring scale was recorded. Data were collected for the entire length of the 10-cm long samples while leaving the end of the material still attached for the last reading. Each sample was tested in this way three times consecutively and the data from all three runs was combined for analysis.

After the first triplicate run of each material was completed (“sample 0”), each sample was pressed against the hook layer and torn free 50 times and the hook layer was then cleaned with a nylon bristle brush. After the second triplicate run (“sample 50”), the samples were again pressed and torn loose an additional 50 times, the hook layer was cleaned, and the final series of triplicate runs was done (“sample 100”).

An additional triplicate run was done with a new piece of thin material (“new thin”) to compare its adhesion with the original sample.

Macrophotography. Images of the loop material were taken with an Exolabs digital camera system and iPad app (Exolabs, Seattle, WA).

Data Analysis. Graphical and statistical analysis using one-way ANOVA with Dunnett’s ad hoc post-test, and Student’s t-test were performed with Kaleidagraph software, version 4.1 for Macintosh (Synergy Software, Reading, PA). Results were considered significantly different for p-values ≤ 0.05.

Step 2: Results

Figure 5 shows an unworn region of the original loop pile. The loops are well formed and intact with loop widths of about 0.03-0.06 cm. Figure 6 shows an example of a worn region of the original loop pile with significant fraying and few intact loops. Figure 7 shows the new thin material with a loop width similar to that of the original pile (0.04-0.06 cm). The new thick material with larger loop widths of about 0.10-0.11 cm is shown in Figure 8. The loop widths are considered typical but some are obviously much larger than the ones measured.

Table 1 shows the statistical summary for each data set resulting from the experiment. The original data can be accessed at

SampleThick 0Thick 50Thick 100Thin 0Thin 50Thin 100New Thin
Minimum 1.5 1.6 1 0.1 0.1 0 0.1
Maximum 5.1 5.8 5.7 1.4 1 1.2 1.8
Sum 135.3 124.8 132.63 14.1 14 12 29.9
Points 41 40 42 31 43 46 52
Mean 3.3 3.12 3.16 0.45 0.33 0.26 0.58
Median 3.4 3 3.15 0.40 0.20 0.2 0.5
RMS 3.42 3.25 3.33 0.54 0.41 0.35 0.70
Std Deviation 0.90 0.94 1.06 0.30 0.24 0.24 0.41
Variance 0.82 0.88 1.11 0.09 0.06 0.06 0.17
Std Error 0.14 0.15 0.16 0.05 0.04 0.04 0.06
Skewness -0.05 0.88 0.13 1.15 1.30 2.27 1.13
Kurtosis -0.58 0.64 -0.05 1.28 0.80 5.26 0.73

Table 1. Statistical summary of the results. Data rounded to the nearest hundredth if needed.

Figure 9 shows the results of the repeated pull force experiments for both the thick and the thin samples. The thick sample had an initial mean detachment pull force of 3.3 N which reduced slightly after 50 and 100 attachment-detachment cycles.

The one-way analysis of variance (ANOVA) and post-test results are summarized in Tables 2 and 3. ANOVA with Dunnett’s post test showed there was no significant difference in mean detachment force for the thick sample after 50 cycles (p = 0.61) and after 100 cycles (p = 0.73) (see Table 2). Dunnett’s was chosen as the best ad hoc post test because the variances for the data sets were not equal [2]. Results obtained using Tukey’s Honestly Significant Differences test (Tukey’s HSD) were essentially the same (data not shown).

Figure 9 also shows the mean force required to separate the thin sample from the loop was much lower than that required for the thick sample, beginning at 0.45 N and reducing to 0.33 N after 50 cycles and to 0.26 N after 100 cycles.

The ANOVA with Dunnett’s post-test showed that the mean detachment force for the thin sample after 50 cycles did not quite meet the significantly different criterium of p ≤ 0.05 (p = 0.068) but was significantly less after 100 attachment-detachment cycles (p = 0.003) (see Table 3).

Figure 10 shows that the change in separation force after repeated attachment-detachment cycles was not due to wear on the loop side since a fresh piece of thin material had a mean pull force similar to that of the original piece. The mean detachment forces for both of these samples were not significantly different based on Student’s t-test (p = 0.13).

Analysis of Variance Results
Total 122 113.61275 0.93125201
A 2 0.73443934 0.36721967 0.39038821 0.67765
Error 120 112.87831 0.94065255
Dunnett's Multiple Comparison
ComparisonMean Difference|q|P95% Confidence Limits
Thick 0 vs Thick 50 0.18 0.8351 0.6146 -0.30248 to 0.66248
Thick 0 vs Thick 100 0.142143 0.6676 0.7287 -0.33448 to 0.61877

Table 2. ANOVA with Dunnett’s post-test results for thick sample.

Analysis of Variance Results
Total 119 8.6099168 0.072352242
A 2 0.70171679 0.35085849 5.1908692 0.00692
Error 117 7.9082 0.067591453
Dunnett's Multiple Comparison
ComparisonMean Difference|q|P95% Confidence Limits
Thin 0 vs Thin 50 0.129257 2.1101 0.0677 -0.0079002 to 0.26641
Thin 0 vs Thin 100 0.193969 3.2107 0.0033 0.058698 to 0.6.32924

Table 3. ANOVA with Dunnett’s post-test results for thin sample.

Step 3: Discussion

Clear differences are evident in the images of the new and old loop pile materials. Although the widths of the old loops and the Perfectex Tricot thin loops were similar, the Tricot loops appear to made of thicker strands (Figures 5 and 7). The thick Perfectex Knit Loop had what appeared to be an even thicker strand diameter and clearly wider loops (Figure 8). The focus of Figure 8 is on the upper surface of the material and additional levels of loops exist beneath this upper level.

Several actionable conclusions can be drawn from the pull force results.

Since both the thin and thick material have larger strand diameters that the original material, either one would likely be an improvement. However, as shown in Figure 9, the thick sample offers significantly more adhesive strength than the thin material. This provides an advantage in that the iPad will be more securely attached, reducing the chance of damage.

Another advantage of the thick material over the thin, is that is it more wear resistant. This is illustrated by the significant reduction in the force required to separate the thin material from the hook side after 100 cycles of attachment and detachment (Figure 9). As shown in Figure 10 , this reduction in adhesive power was not the result of wear of the hook material, and thus resulted from wear of the loop material under test (see also frayed worn loop in Figure 6). The thick sample experienced no significant loss of adhesiveness after 100 attachment-detachment cycles.

It was a bit surprising that the fresh sample of thin material required a greater average pull force than the original sample, but this was most likely due to the width of the sample being slightly greater than that of the original sample. It is difficult to accurately cut loop pile material to size.

The data could be improved by using a load cell force sensing system with automatic data recording. But, the results are more than adequate for making a material choice for the given application.

Although the thick material is slightly more expensive than the thin material, there is no need to completely cover the back of the iPad inner case. Because of its greater adhesive strength, two small strips of loop material, attached with contact cement, allow an iPad case to be returned to serviceable condition a cost of less than $1 each (see Figure 11). This is a significant savings for a school, which otherwise would have to pay about $15 each for replacement cases.

At the time of this report, about 30 cases had been removed from service at my school because of loop pile wear. Refurbishment is underway using the thick material to result in a savings of approximately $400 over purchasing replacement cases. Over the life of the iPad program, which is expected to last for at least five more years, the savings will likely surpass several thousand dollars. Furthermore, this experiment has allowed us to choose the better material, such that once repaired, a case will likely never have to be refurbished again, at least not for worn loop pile on the back of the inner iPad holder.

The hypothesis posed regarding the inferior wear characteristics of the thin loop pile material is supported by the results. The results suggest that, in general, any kind of mobile device case using hook and loop attachment can economically be repaired with significant cost savings for schools and businesses.

Step 4: References

[1] 2013. Nagel, D. Report: Students Use Smart Phones and Tablets for School, Want More. The Journal. <accessed April 25, 2014>

[2] 2011. IBM Corporation. One-way ANOVA Post Hoc Tests. <accessed April 25, 2014>

Scientific Method Contest

Participated in the
Scientific Method Contest