Yarn Strength Distribution, Weak Place & Safety Factor
Put this calculator on your own site
Paste this where you want the calculator to appear. It works on any site — WordPress, Squarespace, Webflow, Ghost or plain HTML — and needs no JavaScript of yours. It carries a link back here, which is the only thing we ask for it.
The mean has a safety factor of 5.5. The weakest place in a kilometre has 3.3.
Safety Factor at the Weakest Place
—x
Expected weakest segment against running tension
Percentiles, Weak Place & Margin
Expected Weakest in this Length
—cN
Safety Factor on the Mean
—x
5th Percentile Strength
—cN
1st Percentile Strength
—cN
Standard Deviation
—cN
Independent Segments
—
Tension as a z-Score
—
Normal-Model Break Probability
—%
Segments are treated as statistically independent, which they are not entirely - strength varies with count, and count has periodic and drift components, so neighbouring lengths are correlated. Correlation reduces the effective number of independent segments and makes the true expected minimum less extreme than the figure here. The normal distribution understates the lower tail for yarn: Weibull is the better model and would give a weaker expected minimum, so the safety factors here are optimistic rather than conservative in the deep tail. Strength is at the stated gauge length and standard atmosphere; a different gauge length gives a different mean and CV for the same yarn, and the two cannot be compared. The break probability output is included for completeness and should not be used as a breakage-rate prediction, for the reasons given in the questions above.
Using this calculator
About the Yarn Strength Distribution, Weak Place & Safety Factor
The formula
This is the expression the tool evaluates. Every term is named underneath, with the unit it must be supplied in.
CV to an absolute spreadstandardDeviation = meanStrength x cvStrength / 100
Everything downstream needs the standard deviation in the same units as the strength, and CV is how the test report states it. A 12% CV on 320 cN is 38.4 cN of spread, which is the quantity the weak places are drawn from.
Strength at a stated percentilepercentile = meanStrength + standardDeviation x z(p)
The normal quantile is evaluated by the Abramowitz and Stegun 26.2.23 rational approximation, accurate to better than 0.0005 in z. The 5th and 1st percentiles are the practical description of the weak tail that a mean and a CV together conceal.
The weakest of manysegments = yarnLength / testLength expectedWeakest = mean + sd x z( 1 / ( segments + 1 ) )
A kilometre of yarn is two thousand independent gauge lengths, and the expected minimum of n samples sits near the 1/(n+1) quantile. This is why long lengths fail at loads that a tensile test says are perfectly safe - the test measured one segment, and the process loads all of them.
The margin that actually matterssafetyFactorWeakest = expectedWeakest / spinningTension
Comparing running tension with the mean flatters the yarn. Comparing it with the expected weakest place in the length under load is the honest figure, and it is typically a third to a half of the apparent margin.
Symbols used above
Symbol
Stands for
Unit
CV
Coefficient of variation, standard deviation over mean
%
z(p)
Standard normal quantile at probability p
—
n
Independent gauge lengths in the length considered
—
RKM
Breaking length, an alternative strength unit for yarn
km
How the result is derived
Step by step, from the values you type to the figure on screen.
The 5 inputs are read from the form on every keystroke: Mean Breaking Force, CV of Breaking Force, Gauge Length, Running Tension and Length Considered.
Each value is checked against the accepted range in the input table below. A value outside its range stops the calculation rather than producing a misleading figure — the results blank out and a message appears.
The validated values are substituted into the expression above, which resolves Safety Factor at the Weakest Place together with every supporting figure in one pass — no value is carried over from a previous entry.
The supporting outputs — Expected Weakest in this Length, Safety Factor on the Mean, 5th Percentile Strength, 1st Percentile Strength, Standard Deviation, Independent Segments, Tension as a z-Score and Normal-Model Break Probability — come from the same pass, so they always describe the same case as the headline figure.
Results are rounded for display only. The full-precision value is used throughout the chain, so reading a rounded intermediate figure back into the tool by hand can shift the last digit.
What each input means
Where to read each value on the floor, the unit it must be in, and the range the tool accepts.
Input
Unit
Accepted range
Default
What it means
Mean Breaking Force
cN
20 to 5000 cN
320
CV of Breaking Force
%
2 to 40 %
12
Gauge Length
m
0.01 to 5 m
0.5
ISO 2062 uses 500 mm
Running Tension
cN
1 to 2000 cN
58
Length Considered
m
1 to 100000 m
1000
What the tool returns
The headline figure and every supporting value it is built from.
Output
Unit
What it tells you
Safety Factor at the Weakest Place (headline result)
x
Expected weakest segment against running tension
Expected Weakest in this Length
cN
Safety Factor on the Mean
x
5th Percentile Strength
cN
1st Percentile Strength
cN
Standard Deviation
cN
Independent Segments
—
Tension as a z-Score
—
Normal-Model Break Probability
%
Worked example
Given
0
Mean breaking force 320 cN at 12% CV
1
500 mm gauge length per ISO 2062
2
Running tension 58 cN
3
1,000 m of yarn considered
Substituting
sd = 320 x 0.12 = 38.4 cNsegments = 1000 / 0.5 = 2,000z at 1/2001 = -3.291, so weakest = 320 - 38.4 x 3.291 = 193.63 cNsafetyFactorWeakest = 193.63 / 58 = 3.34 against 320 / 58 = 5.52 on the mean
Answer
0
Safety factor at the weakest place 3.34, against 5.52 on the mean
Running tension is 6.82 standard deviations below the mean
The mean says the yarn is running at a fifth of its strength. The weakest place in a single kilometre says a third. Over a full package of a hundred kilometres the expected minimum falls further again - which is the whole reason breakage is a rate rather than an event.
How to use it
Work through the input groups in order — Strength Data and Load & Length at Risk. The defaults are a realistic case, so you can change one value at a time and watch what moves.
There is no calculate button. Every figure recalculates as you type or drag, which is what makes this usable for a what-if sweep rather than a single answer.
Read Safety Factor at the Weakest Place in the dark results panel — that is the headline figure, expressed in x.
Check the supporting rows underneath (Expected Weakest in this Length, Safety Factor on the Mean, 5th Percentile Strength, 1st Percentile Strength, Standard Deviation, Independent Segments, Tension as a z-Score and Normal-Model Break Probability) before acting on the headline — they are where an implausible input usually shows itself first.
Reset to defaults returns every field to the reference case, which is the quickest way to check whether a surprising result came from the tool or from an input you had changed earlier.
Where this is used
Process planning — establishing Safety Factor at the Weakest Place before a trial is booked, so machine time and material in Spinning, Winding & Yarn Package Engineering are committed against a calculated figure rather than an estimate.
Costing and quotation — Safety Factor at the Weakest Place is an input to the cost sheet, and quoting from a worked number rather than a remembered one is what keeps a margin intact.
Troubleshooting — when the floor result drifts from plan, entering the measured values (starting with Mean Breaking Force) shows how much of the gap in Safety Factor at the Weakest Place each variable explains.
Teaching and study — the accepted ranges bracket normal Spinning, Winding & Yarn Package Engineering practice, so moving one variable at a time shows the shape of the relationship rather than a single answer.
Reading the result
Typical bands and what each one is telling you.
Value
What it indicates
CV 8 - 12%
Good ring yarn single-end strength variation.
CV 12 - 18%
Rotor and coarser carded yarn.
Safety factor above 3 at the weak place
Comfortable running margin.
Below 2
Expect a breakage rate that no machine setting will fix.
Assumptions and limits
Segments are treated as statistically independent, which they are not entirely - strength varies with count, and count has periodic and drift components, so neighbouring lengths are correlated. Correlation reduces the effective number of independent segments and makes the true expected minimum less extreme than the figure here. The normal distribution understates the lower tail for yarn: Weibull is the better model and would give a weaker expected minimum, so the safety factors here are optimistic rather than conservative in the deep tail. Strength is at the stated gauge length and standard atmosphere; a different gauge length gives a different mean and CV for the same yarn, and the two cannot be compared. The break probability output is included for completeness and should not be used as a breakage-rate prediction, for the reasons given in the questions above.
Every input is bounded to the range normal practice occupies (Mean Breaking Force 20 to 5000 cN, CV of Breaking Force 2 to 40 % and Gauge Length 0.01 to 5 m, and so on for the rest). Those bounds are guard rails against typing errors, not a claim that the formula fails one unit outside them.
The calculation is deterministic: the same inputs always give the same result. It carries no allowance for machine condition, operator skill, ambient conditions or lot-to-lot material variation unless an input above explicitly represents one.
Nothing is sent anywhere. The maths runs in your browser, so the numbers you type never leave the page.
Standards and further reading
ISO 2062 - Textiles, Yarns from packages, Determination of single-end breaking force and elongation, which specifies the 500 mm gauge length.
ASTM D2256 - Tensile Properties of Yarns by the Single-Strand Method.
ASTM D2904 - Interlaboratory Testing of a Textile Test Method, for the variance structure of repeated tests.
Abramowitz & Stegun, Handbook of Mathematical Functions, 26.2.23 and 26.2.17, for the quantile and CDF approximations used.
Questions people ask
The break probability shows as zero. Does that mean the yarn never breaks?
No, and this is the most important limitation of the model. Running tension here is nearly seven standard deviations below the mean, and a normal distribution puts essentially no probability that far out - so the number rounds to zero. Real yarn breaks anyway, because real breaks come from defects the normal distribution does not describe: thin places, foreign matter, a bad piecing, a slub. Use the safety factors and the percentiles, which describe the bulk of the distribution well; do not use the break probability as a prediction of breakage rate.
Why does the expected weakest place get weaker as I consider more length?
Because it is a minimum of more draws from the same distribution. Two thousand segments give a minimum near the 1/2001 quantile; twenty thousand push it to 1/20001, which is further into the tail. Nothing about the yarn changed - only how much of it is exposed. This is the weakest-link principle and it is why strength must always be quoted with a gauge length, and why a warp of thousands of ends over hundreds of metres fails at loads that would look safe on a single-end test.
Should I use the mean safety factor or the weak-place one?
The weak-place one, set against the length actually under load in the process concerned. For ring spinning that is the short span between the front roller and the traveller; for warping it is the whole beam length of every end simultaneously, which is why warping exposes weak places that spinning never found. The mean safety factor is useful only for comparing yarns with each other, and it systematically flatters a yarn with a high CV.
Is the normal distribution the right model for yarn strength?
It is a reasonable description of the bulk and a poor one of the lower tail, which is exactly where failure happens. Yarn strength is better modelled by a Weibull distribution, which has the weakest-link behaviour built into it and a heavier lower tail, so it predicts more low-strength segments than the normal does. The normal is used here because CV is what test reports give and it makes the percentiles directly computable; treat the deep-tail figures as optimistic and prefer measured low-percentile data where it exists.