How To Buy A Quality Sofa—Part 4: Fabric

This week we’ll begin to talk about the fabric used in your sofa – which we (of course) think is a very complicated and important topic! One thing to remember is that there is much more fabric used in constructing an upholstered piece of furniture than just the decorative fabric that you see covering the piece – a typical “quality” sofa also uses about 20 yards of decorative fabric, plus 20 yds of lining fabric, 15 yds of burlap and 10 yds of muslin, for a total of 65 yards of fabric!

So what do people look for in an upholstery fabric?

After color, fabric durability is probably top of everybody’s list.  Durability translates into most people’s minds as “heft” – in other words, a lightweight cotton doesn’t usually come to mind. But more important in evaluating durability than the weight of the fabric is the length of the fibers.  Cotton as a fiber is much softer and of shorter lengths than either hemp or linen, averaging 0.79 -1.30 inches in length.  Hemp’s average length is 8 inches, but can range up to 180 inches in length. In a study done by Tallant et. al. of the Southern Regional Research Laboratory,  “results indicate that increases in shortfibers are detrimental to virtually all yarn and fabric properties and require increased roving twist for efficient drafting during spinning. A 1% increase in fibers shorter than 3/8 in. causes a strength loss in yarns of somewhat more than 1%.”[1]    In fact, the US textile industry has  advocated obtaining the Short Fiber Content (SFC) for cotton classification.  SFC is defined as the percentage of fibers shorter than ½ inch.  So a lower cost sofa upholstered in cotton fabric, even one identified as an upholstery fabric, could have been woven of short fiber cotton, a cheaper alternative to longer fiber cotton and one which is inherently less durable.

Patagonia, the California manufacturer of outdoor apparel, has conducted  tests on both hemp and other natural fibers, with the results showing that hemp has eight times the tensile strength and four times the durability of other natural fibers.   Ecolution had a hemp twill fabric tested for tensile and tear strength, and compared the results with a 12-oz cotton denim.  Hemp beat cotton every time:   Overall, the 100% hemp fabric had 62% greater tear strength and 102% greater tensile strength. [2]   And polyester trumps them both – but that’s a whole different ballgame, and we’ll get to that eventually.

There is a high correlation between fiber strength and yarn strength.  People have used silk as an upholstery fabric for hundreds of years, and often the silk fabric is quite lightweight;  but silk is a very strong fiber.

In addition to the fiber used, yarns are given a twist to add strength. This is called Twist Per Inch or Meter (TPI or TPM) – a tighter twist (or more turns per inch) generally gives more strength.  These yarns are generally smooth and dense.

So that brings us to weave structure.  Weave structures get very complicated, and we can refer you to lots of references for those so inclined  to do more research (see references listed at the end of the post).

But knowing the fibers, yarn and weave construction still doesn’t answer people’s questions – they want some kind of objective measurement.  So in order to objectively compare fabrics,  tests to determine wear were developed (called abrasion tests), and many people today refer to these test results as a way to measure fabric durability.

Abrasion test results are supposed to forecast how well a fabric will stand up to wear and tear in upholstery applications.  There are two tests generally used:  Martindale  and Wyzenbeek (WZ).  Martindale is the preferred test in Europe; Wyzenbeek is preferred in the United States.  There is no correlation between the two tests, so it’s not possible to estimate the number of cycles that would be achieved on one test if the other were known:

• Wyzenbeek (ASTM D4157-02):  a piece of cotton duck  fabric or wire mesh is rubbed in a straight back and forth motion on a      piece of fabric until “noticeable wear” or thread break is evident.  One back and forth motion is called a “double rub” (sometimes written as “dbl rub”).
• Martindale (ASTM D4966-98):  the abradant in this test is worsted wool or wire screen, the fabric specimen is a circle or round      shape, and the rubbing is done in a figure 8, and not in a straight line as in Wyzenbeek.  One circle 8 is a cycle.

The Association for Contract Textiles performance guidelines lists the following test results as being suitable for commercial fabrics:

According to the Association for Contract Textiles, end use examples of “heavy duty contract” where 30,000 WZ results should be appropriate are single shift corporate offices, hotel rooms, conference rooms and dining areas.  Areas which would require higher than 30,000 WZ are: 24 hour facilities (like transportation terminals, healthcare emergency rooms, casino gambling areas,  and telemarketing offices) and theatres, stadiums, lecture halls and fast food restaurants.

Sina Pearson, the textile designer, has been quoted in the Philadelphia Inquirer as saying that 6,000 rubs (Wyzenbeek) may be “just fine” for residential use”[3]   The web site for Vivavi furniture gives these ratings for residential use:

Theoretically, the higher the rating (from either test) the more durable the fabric is purported to be.  It’s not unusual for designers today to ask for 100,000 WZ results.  Is this because we think more is always better?  Does a test of 1,000,000 WZ guarantee that your fabric will survive years longer than one rated only 100,000 WZ?  Maripaul Yates, in her guidebook for interior designers, says that “test results are so unreliable and the margin of error is so great that its competency as a predictor of actual wear is questionable.”[4]  The Association for Contract Textiles website states that “double rubs exceeding 100,000 are not meaningful in providing additional value in use.  Higher abrasion resistance does not necessarily indicate a significant extension of the service life of the fabric.”

And of course, any company can skew results in their favor.  This is an image I found on Google images, with abrasion test results from a company selling leather motorcycle clothing.  They did say that “leather will sometimes score up to 100,000 cycles or so on the Wyzenbeek test, but testing to destruction (over 50k cycles) doesn’t always prove much.”  No comment on these results !

There are, apparently, many ways to tweak test results. We’ve been told if we don’t like the test results from one lab, we can try Lab X, where the results tend to be better.  The reasons that these tests produce inconsistent results are:

1. Variation in test methods:       Measuring the resistance to abrasion is very complex.  Test results are affected by many factors that include the properties and dimensions of  the fibers; the structure of the yarns; the construction of the fabrics;  the type, kind and amount of treatments added to the fibers, yarns, or      fabric; the time elapsed since the abradant was changed;  the type of  abradant used; the tension of the specimen being tested,the pressure between the abradant and the specimen…and other variables.
2. Subjectivity:    The  measurement of the relative amount of abrasion can be affected by the method of evaluation and is often influenced by the judgment of the operator.  Cycles to rupture, color change, appearance change and so forth are highly variable parameters and subjective.
3. Games Playing:     Then there is, frankly, dishonest collusion between the tester and the testee.  There are lots of games that are played. For instance, in Wyzenbeek, the abradant, either cotton duck or a metal screen, must be replaced every million double rubs. If your fabric is tested at the beginning of that abradant’s life versus the end of its life, well.. you can see the games. Also, how much tension the subject fabric is under –  the “pull” of the stationary anchor of the subject fabric, affects the  rating.

In the final analysis, if you have doubts about the durability of a fabric,  will any number of test results convince you otherwise?  Also, if your heart is set on a silk  jacquard, for example, I bet it would take a lot of data to sway you from your heart’s desire.  Some variables just trump the raw data.

REFERENCES FOR WEAVE STRUCTURE:

1.  Peirce, F.T., The Geometry of Cloth Structure, “The Journal of the Textile Institute”, 1937: pp. 45 – 196

2. Brierley, S. Cloth Settings Reconsidered The Textile Manufacturer 79 1952: pp. 349 – 351.

3. Milašius, V. An Integrated Structure Factor for Woven Fabrics, Part I: Estimation of the Weave The Journal of the Textile Institute 91 Part 1 No. 2 2000: pp. 268 – 276.

4. Kumpikaitė, E., Sviderskytė, A. The Influence of Woven Fabric Structure on the Woven Fabric Strength Materials Science (Medžiagotyra) 12 (2) 2006: pp. 162 – 166.

5. Frydrych, I., Dziworska, G., Matusiak, M. Influence of Yarn Properties on the Strength Properties of Plain Fabric Fibres and Textile in Eastern Europe 4 2000: pp. 42 – 45.

6. ISO 13934-1, Textiles – Tensile properties of fabrics – Part 1: Determination of Maximum Force and Elongation at Maximum Force using the Strip Method, 1999, pp. 16.