Friday, November 11, 2011


There is a long standing convention in the textile industry on wraps per inch (wpi). Pack to refusal!  They wind the thread into a slot or gap and they pack to refusal.  (It is a gentle and careful packing that does not  deform the yarn.)  This results in an accurate and precise number that can be related to yards per pound (ypp).

Knitters on the other hand, when making wpi measurements wind their "wpi" loosely, (well not too loosely),  and they get a number that is "fuzzy".  See for example  or It is a more modern method, but it will not give a precise and accurate number. And, it does not differentiate between worsted, woolen, and cotton/silk  It is a different culture. Then, they call the textile guys, "wrong" for using the old, precise method.  The only thing the fuzzy method is good for is putting the yarn in a Craft Yarn Council category.  And, that does not provide a lot of useful information. Particularly for somebody like me who is likely to be spinning fine yarns.  Trying to wrap 20,000 ypp singles (35s) loosely, but not too loosely, is an exercise in futility.   However, with the right gauge one can pack to refusal, and get a precise number.

So here is a solution. Use a gap gauge, wrap the yarn around it, and pack to refusal.  Then, for worsted add 10% and take the square of the number.  That will be a good estimate of the yards per pound.  For Woolens, add 16% and take the square of the number and you will have a good estimate for the ypp.  For silk and cotton add 7%, take the square of the number and you will have a good estimate for the ypp.  That is useful information.

Divide the number of wraps per inch packed to refusal by 2, and you will have a good estimate of the number that is defined as "wpi" by the Craft Yarn Council  and used by American knitters as "wpi.

If you have the yards per pound, take the square root and divide by 2, and you will have the knitters' wpi. Or, you can take the square root of  yards per pound, subtract 10 percent (16% for woolen, 7% for silk and cotton) and you will have the wraps per inch, which is the reciprocal of the thickness of the yarn. This gives the diameter of the yarn. This tells you something useful about the your yarn or thread.

In Judith MacKenzie 's the intentional spinner, she does not tell how to do wpi, but when we look at her projects we see that she says a  semi woolen yarn at 2475 ypp has 27 to 30 wpi; 1460 ypp has 18-20 wpi ; 300 ypp has 5 - 6  wpi.  She does not even try to get a number more precise than a range of 10%.  Shannon Okey's Spin to Knit,  on page 57 says that if  want to know more about grist, see Alden Amos's Big Book of Hand Spinning.  Well, Alden says, "Pack to refusal!"  Then Okey contradicts Alden on page on her page 126 by saying one can just wrap the yarn around a ruler.  No, that is not "pack to refusal".

This having two different measures in related fields, with the same name causes some confusion.  For example, yarn, that I say is 100 wpi, most knitters would call 50 wpi, and then they would say I do not know what I am taking about.  Well that is yarn spun at 22 hanks per pound or 12,000+ ypp.  I work with those kinds of threads every day.  Do they?  Measuring the yarn as 100 wpi tells me something useful about the yarn.  Measuring it as 50 wpi tell me that it is "lace weight", and I knew that without picking up a ruler.

Alden Amos gets it correct in his Big Book of Handspinning. I just wish that Alden had put in corrections for yarn construction and yarn fiber.  Then the wrap per inch  number in the table on his page 383 would be 7% to 16% less.    Peter Teal in Hand Woolcombing and Spinning gets it correct, and his Appendix III is the best conversion table. (However, he only addresses worsted yarns, and grist is stated in spin count.)  Better is scanned at


=Tamar said...

You say "pack to refusal" doesn't deform the yarn. But on the websites you linked to, their descriptions of the wrapping process are identical to your description. If ptr gets twice as many wraps per inch as their method, then it must deform the yarn. There is no other explanation for the difference.

Aaron said...

It is like a Japanese subway. How many people can fit in a subway car? With packers, they can get more commuters on the train, and yet nobody is so deformed that they end up in the hospital.

That is what ptr means. Full body contact from head to toe. Without ptr, you will not see that kind of full body contact. Unless there is full body contact, the packers can likely pack more people into the car, ie the car is not full. As long as you let the passengers decide how much space there is between people, nobody knows how many people are on board the train. Once everybody is packed on in full contact with each other, then we have a good estimate of how many passengers are on board.

Until you have packed your gap to refusal, you do not know how many more warps of yarn will fit into the gap.

At ptr we have a uniform fiber density across the gap at a thickness of 1/wpi. That tells us the cross-section of the yarn, in a simple and repeatable way. From the cross section of the yarn, we can calculate its grist.