Wednesday, March 24, 2021

Missed yarn over! Help!

Did you forget to make a yarn over (YO) in the row below? No problem, and no need to rip back your knitting. Your forgotten yarn-over can be added in the next row. 

Step 1:Work to the exact spot where the YO is missing. At this point, in the row below the one you're in, there will be a stitch tail (tail = yarn which connects two stitches). This is the tail which, if you hadn't forgotten to lift it over the needle, would have been the YO. 

With your right needle, lift this tail (blue in below illustration) in the direction of the red arrow, and put it on your left (holding) needle.  

Step 1: find the tail (blue) and put it on the holding needle, following the direction of the red arrow.

Result: you've made a mini-YO, just where the original YO should have been. Next, work into the mini-YO in the ordinary way, as shown in the below illustrations.  Make sure which arm ought to be forward by comparing to the other YO's in your work and to the illustrations below. 


Step 2: This newly-made YO is now on the holding (left) needle, with its left arm forward.  Knit into it following the path of the red arrow.


In contrast to the previous illustration, here the newly-made yarn-over lays on the holding (left) needle with its right arm forward.  Knit into this new YO by following the path of the red arrow.

The finishing step takes place once you are a few rows or rounds past the error.  Using the tip of a knitting needle, ease some spare yarn from the neighboring stitches into the mini-YO.  This enlarges the hole to better match other YO's which may appear in your work.  This "I forgot" YO may look smaller to you, but no one else will ever notice (and not even yourself, once a few days have passed). 

--TK

PS: If these illustrations look familiar, they are, being highly related to a previous post about knitted increases. Part of that post illustrates how to go back and fix a missing twist-base increase, often called an M1.  I mentioned (but didn't illustrate) that a missing YO could be added after-the-fact in almost the same way. Well, oops--my lovely readers let me know the lack of an illo was an oversight.  All fixed now, though.

Saturday, March 13, 2021

Knitted increases based on a twisted tail: Backwards loop, Yanked Increase, & Twisted Yarn-Over

Knitted increases form the foundation for a new column of knitting to appear.  Where two columns lay, side-by-side, there are now three: the two originals with a third between.  

The most simple type of foundation for the new column is a twist of yarn--a loop twisted shut, and that's (mostly) what this post is about. 


an increase in stockinette fabric
Stockinette fabric with an increase--the base of the increase column (pink) is a twisted loop of yarn. The red highlighting picks out where the actual twist of the loop occurs, and that twisted loop is considered the first stitch in the increase column.

In the above illustration, the twisted loop (red) is the foundational stitch for the new column (pink). 

Depending which way the twist is made, the top arm of the twist can "sweep up" either to the left or the right. This makes little difference structurally, but being consistent and selective about this sort of decorative detail adds to a finished appearance--handmade rather than home-made.

Top arm sweeps left
Top arm sweeps right

All the foundational twists illustrated here, regardless of which way they sweep, could have been made by one of three different methods: 

    * Backwards loop
    * Yanked increase
    * Twisted YO (yarn-over)

Although the final path of the yarn in the fabric is identical between all three methods, the result in the fabric is not: each method leaves a different amount of slack in the fabric. In my experience, it's a three-bears story: one is a bit loose (backwards loop); one, a bit tight (yanked increase); and one just right (the twisted YO--the one I usually use).

Because tastes differ and YMMV, and because in certain situations, one increase has an unexpected advantage over another, below is the illustrated how-to for all three methods.

Each of these methods share many (but not all!) operational steps, so it's mix and match on the illustrations. A blog isn't a book, where flipping back a page is easy, and with a blog, there's no printing charge for extra illustrations. Therefore, each method is fully illustrated, sometimes by recycling the same illustration, just in a different colorway. Repetitious, yes, but hopefully, easier to follow on a screen.

Backwards loop

Like all 3 twisted-loop increases, this increase can be called an M1 (for Make-1). Another name for this type of M1 increase is "e-loop." To avoid confusion, I'll call it a backwards loop increase.
 
A backwards loop is worked in a two-round (or -row) operation. First, a loop of yarn is twisted up and laid over the needle. On the following round, when you encounter the loop, you work into it while keeping it twisted shut.  This two-round operation is illustrated in four steps. 
Step 1: in the first round, work to where the increase column should start

backwards loop increase in knitting-2
Step 2: Pinch up a loop of the running yarn, twist it shut, and place it on the right needle.  As shown, you can mount the twist either left- or right-arm forward, and this determines which way the increase appears to slant in the work.  Mounting right arm in front translates to a left sweep in the fabric; whereas mounting left arm in front translates to a right sweep. As you see, the little loop looks like a lower-case "e" which gives this the alternative name of e-loop. If using the M1 designation for this increase, the sweep distinguishes whether this is called M1L (make-one-left) or M1R (make-one-right).

backwards loop increase in knitting-3
Steps 3 and 4: In the following round, when you come to the spot where the loop was parked (left illustration) then knit into the loop (right illustration) making sure that the loop remains twisted shut

The above illustrations assume you're working stockinette in the round.  If working stockinette flat (back-and-forth) the instructions for steps 1 and 2 remain the same.  Step 3 is almost the same, except you'd purl to the spot where the increase was made in the previous row.  In step 4, you'd purl into the twisted loop, again taking care that the loop does not untwist, then purl the rest of the row.  

Although this is a two-round (or -row) operation, the actual base of the column--the twist which is considered the first stitch of the increase column-- is created in the first go-round, and that is the round in which the increase appears.  The second round (or row) is actually the second stitch of the column.

Yanked Increase

As far as I know, this increase made from the tail of an existing stitch has no common name which distinguishes the fact that it is made in 1 row from a yanked tail, although this increase also can be called "make one" (M1).  To avoid confusion let's call this variation "yanked increase." 

To make the yanked increase, yank the tail of the stitch below out of the fabric with the right needle, lay that tail over the left needle, and then knit into it so as to twist it shut.  Unlike the backwards-loop method, the yanked increase is done all in one row: the tail yanked up, then parked on the left needle, then twisted shut in the knitting process.  Which way you place the tail yarn on the left needle, and which way you knit into it determines the sweep of the top arm.

yanked increase in knitting-1
Step 1: knit to where the increase column should start.  With your right needle, fish up the underlying tail (red) and place it on the left needle (blue arrow)
Step 2 (left sweep): Again, which arm you mount forward determines which way the twisted increase "sweeps."  Here, the right arm of the yanked tail has been placed forward on the left needle, making what looks like a mini-yarn-over.  When this mini YO is "knit shut" (twisted) by following the path of the blue arrow, what was a loop mounted right arm forward becomes an increase with the left arm on top, and this now-top arm of the increase appears to sweep left (per close-up). The working needle passes over the running yarn, then catches around the back leg of the loop (knit through the back loop); and then catches around the standing yarn as for an ordinary knit stitch. The running yarn is then pulled through, twisting the loop shut. If using this M1 designation for this increase, this sweep is called M1L (make-one-left).
yanked increase in knitting-3
Step 2: (right sweep): Here, the mini YO is mounted on the holding (left) needle with left arm forward.  When it is "knit shut" (path of blue arrow) what was the left arm becomes the right arm, and this arm now will appear on top, sweeping upwards and to the right (per close-up).  The needle passes over the running yarn, catches around the front arm of the stitch, and then catches around the standing yarn as for a knit stitch. If using this M1 designation for this increase, this sweep is called M1R (make-one-right).
 
Although the yanked increase is created in a one-row operation, it actually spans two rows, with the beginning of the column "exported" down a row. See, even though you're doing the work in this row, you're operating on the tail of a stitch in the row below. Specifically, the stitch you knit into that twisted tail is in this row, but the twist you made in the tail is located below the new stitch, therefore, in the row below.
 
Generally I avoid this increase as being tight. Yet, in some situations, it really shines.  
--First, if you forgot to plant the base for your increase column in the row below, this trick plants the base and knits the first stitch all at once (one operation).  Big-picture-wise, the yanked increase is a one-stitch analog of the long-tail cast on: each stitch of a LTCO is formed at the same time as its twisted-loop base.  
--In flat knitting (back-and-forth), the two-rows-in-one-operation means you can do the entire increase on the stockinette side where it's quite obvious.  No need to scan all those loopy purl-heads on the lookout for your previously parked strand. This is the only increase of the three in this post with this advantage. 
--Because this is a nice, obvious increase which requires no advance planning, it's perfect for TV-knitting. Bonus: you get a mindless row or round to catch up with the on-screen action.

Yanked increases and forgotten Yarn-Overs

A bit off-topic, but, you can knit into a yanked increase WITHOUT twisting it shut. This forms a tight sort of a yarn over--a hole in the fabric. You can knit into this hole as into any YO. Therefore, this is an excellent trick when you've forgotten to make a YO in the row below. Which arm lays forward on the left needle dictates whether the YO slants one way or another. ( Edited to add : per reader request, there is now an illustrated post on forgotten yarn-overs.)

To make an "I forgot" YO look as big as a planned YO, then, once you've gotten a few rows or rounds past the spot, pick some slack yarn from the neighboring stitches into the YO. It may look a little smaller to you, but no one else will ever see the difference.

Twisted Yarn-over

The twisted YO (also sometimes called M1) combines elements (and illustrations!) from the first two methods. Like the backwards loop method, the twisted YO increase is a two-row operation: first row = yarn parked, second row = yarn worked. The difference is that in the first row, instead of parking a twisted loop on the left needle, you park a YO.   

The twisted YO increase row 1: form a YO over the right needle.  The YO can have its LEFT arm forward (upper left close-up) or its RIGHT arm forward (upper right close-up). Depending which way the YO was formed, for the rest of the first row, it sits in the fabric with that arm forward (lower close-ups).
When you encounter the YO on the second round, it now looks an awful lot like the yanked-up tail from the yanked increase, and you treat it the same way. Like the yanked-up tail, the YO is put on the needles open, then twisted shut in the knitting process. For this reason, other than the color-coding, the illustrations below are identical between both. In each illustration, the little round close-up shows what the increase should look like on your right needle, once you've knit shut the YO on the left needle, following the direction of the blue arrow. Note again that, because you twist the stitch shut, what was the back arm becomes the top arm, dictating the direction of sweep.  If using the M1 designation, these two variants are sometimes called M1L (make-one-left) and M1R (make-one-right).



These illustrations of the twisted YO increase assume you're knitting stockinette circularly, so that you're operating from the front fabric face on both rounds.  If you are working flat (back and forth) the second step (working into the yarn over) is done via a purl.  The close-up illos are your friends for comparison purposes--take a look from the front and make sure, before you move on. 

"After-knitting" and twisted-loop increases

 The concept of "after-knitting" (intro'ed in the last post on the structure of ribbing) is the idea of acting on a parked strand by knitting it after it is off the needles, hence, the name "after-knitting." Of the three increases in this post, two--the yanked increase and the twisted YO--are actually forms of after-knitting.  This is because the twisted loop which forms the base of all these increases is not fully formed at the time of creation.  Instead, some, but not all, of the yarn for the increase is parked in the fabric, to be operated on in the following row.  

Specifically, for the yanked increase, the tail was parked (although all tails are parked if you think about it) and then the parked tail was operated on in the next row.  That operation occurred when the yarn was not on the needles, but lay loose in the fabric.  The tail of a knitted stitch is small.  It is capable of making up only a small percentage of the final twist. This is why this increase is rather tight: all the rest of the yarn in the twist has to be yanked out of the arms of the neighbor stitches. (This is similar to the transformation in fabric caused by slipped-strand after-knitting.)

Similarly, in twisted YO, some yarn was parked--that being the YO--but not enough to create an entire loop.  So, when the YO is twisted up in the following row, some of the yarn in the resulting loop also comes from the adjoining stitches.  However, a YO is longer than a stitch-tail. So, not as much of the neighbors' arm-yarn winds up in the finished twist as with the yanked increase, and that's why the twisted YO is looser. 

Now compare both of these "after-knit" increases to the backwards loop.  In backwards loop, the entire loop is formed over the barrel of the needle just as an ordinary stitch is formed.  When this stitch-sized loop is worked on the following row, the yarn of the adjoining stitches is not pulled into the loop, or at least, no more so than would occur with an ordinary stitch: no "after-knitting" occurs. Putting enough yarn for a whole stitch into a column-bottom (not anchored to anything below) well, that's a lot of unsecured yarn.  Looking at the big picture, this increase is the one-stitch version of the looped cast on (also called the e-cast on) and shares its "loose yarn" problems.

Short rows and "short columns:" anchoring to the neighbor stitches

Another way to think about tight vs. loose increases: if you're familiar with short rows, you can think of an increase as giving rise to a "short column."  So just as with short rows, we'd like to avoid holes. Making a twisted increase as the base of a column is like working a short-row without doing any connecting tricks (wrap and turn, lifted increase, etc.) This is because, like an unconnected short row, the column is not anchored.  Tightening the twisted increase by after-knitting doesn't actually anchor the increase column at the bottom, but the tightening does anchor the base-twist sideways into the adjoining columns more snugly, making less slack available to get up to hole-forming tricks.

Other increases. anchored at the bottom of the column

If you do want to actually physically connect the bottom of the increase column to your knitted fabric, here are two options.

First, the nearly invisible increase (also a species of after-knitting).  The only reason the nearly-invisible isn't more used is that it doesn't work well in a line of increases, like on a top-down raglan.  This is because, like the yanked increase, it tightens the fabric significantly, and even more so. However, for non-linear increases, like on a top-down yoked sweater, the nearly-invisible is a great choice. Full directions are at the linked post.

The second anchored-at-the-bottom increase is a KFB (knit front and back). KFB is also a kind of a twisted-stitch increase, but the twist is made in the head of the stitch, instead of its tail.  First, the stitch is knit into in the perfectly ordinary way. Then, that same stitch is knit into again through the back leg, with a twist separating the first- and second-knitting, as shown at the above linked post. (Like the nearly invisible, the yanked increase and the twisted YO, KFB is also a species of after-knitting. There's a trend here...) 

--TK

Monday, March 1, 2021

Ribbing in hand-knitting: its structure + links to "after-knitting" tricks for improvement

Why is ribbing three dimensional?  Why is it often distorted, with deformity in the last knit column before a purl column? How can ribbing be tightened into a smooth and even regularity? How to remove from ribbing, the extra slack which causes it to flare and s-p-r-e-a-d at garment edges? Our old friend, the structure of the knitted stitch, is the key to answering all these questions.

Here ▼ is a generic knitted stitch with its little tail highlighted red. 

Figure 1: Parts of a knit stitch,
the tail has been colored  red 
(view from the front
fabric face in a stockinette fabric)

In the previous post about gauge, this little red strand got the name of 'tweener strand because, as shown in the below ▼ photo of the back of a stockinette fabric, this little tail runs between stitches. Specifically, 'tweener strands run between stitches of like kind.  Knit-to-knit or purl-to-purl: these combinations are connected by 'tweener strands.


Figure 2: Stockinette fabric, top view, purl fabric face showing.
" 'Tweener strands" of current row colorized red

In this post, we are to meet another type of tail--the kind which connects stitches of unlike kind, such as a knit to a purl, or a purl to a knit.  Such unalike stitches connected together are very commonly found in ribbing.  However, basketweave, seed stitch, cables, indeed all texture stitches also have K-P and P-K connections.  So ribbing in this post stands in for texture stitches in general: everything said here about ribbing applies equally there. 

Direction of yarn travel

In getting from stitch to stitch, yarn travels in a different direction in stockinette vs. in ribbing.  To simplify: in stockinette, yarn travels along the row, but in ribbing, the yarn travels back and forth from one fabric face to another.

Specifically, in stockinette, all the stitches in any one row or round are always facing the same direction, so the yarn always enters the stitch by being drawn though the same way.  As you know, from the knitter's point of view, yarn enters a knit stitch from the back (you hold the yarn behind the fabric, drawing it towards you, all the way down the knit row) while it enters a purl stitch from the front (you hold the yarn in front of the fabric, drawing it away from you, all the way down a purl row).

In ribbing, however, you work different stitches next to one another as you work your way down any one row.  So, if you work a knit stitch (facing you) next to a purl stitch (facing away from you), you must switch the yarn from one fabric face to another.  This is how the yarn enters the stitch from the correct direction, as required. Switching the yarn back and forth from one fabric face to another means that the yarn has to pass back and forth through the thickness of the fabric as it travels between the stitches from one fabric face to another. The act of passing through the thickness of the fabric suggests we call this kind of tail yarn a "passing strand."

Passing strands illustrated

Below ▼ is illustrated a k2, p2 ribbing (also called 2/2 ribbing) in a little scrap sample.  This ribbing has been popped off the needles. The stitches appear in pairs, each pair forming two evident columns of stockinette-on-a-purl-background, and these columns appear alternately on opposite fabric faces. Stated otherwise, the hill-and-valley structure of 2/2 ribbing occurs because two columns of knit (the hills) always appear on a 2-column purl background (the valleys).  The opposite of a knit is purl, so the hills of one fabric face become the valleys of the opposite face when the fabric is flipped, and, of course, the valleys become hills.

Figure 3: Scrap of 2/2 ribbing popped off the needles

The arrangement of these alternating pairs is a bit surprising, perhaps, so here's a head-on shot, a close-up, below. ▼ As you see, pairs A, C, E and G appear on the lower fabric face in this diagram, while pairs B, D and F appear on the upper.  Pairs on any one fabric face are closer together than are those of sequential pairs on opposite fabric faces.  For one example, the heads of pair A are close to those of pair C. By contrast, the heads of pairs A and B are not nearly so close as A and C, despite pair B being originally created directly next to pair A.  There is something about the structure of the ribbing which is pushing neighboring pairs of stitches--originally made right next to one another--apart onto the different fabric faces.  Something pushes these pairs into hills and valleys. What is it?

Figure 4: These pairs were knit in sequence A-B-C-D-E-F-G,
yet once off the needles, they appear on opposite fabric faces
in sequenceA-C-E-G (lower) and B-D-F (upper).  So...why?

Below ▼ is a further closeup (figure 5) which gets to the heart of the matter. 

Figure 5: closeup of figure 4

In the above diagram, the pairs have now been individually labeled.  This closeup now starts with the second of the "A" stitches, A2, and ends with the first of the "G" pair of stitches, G1

--The    gray  line traces the line of the yarn as it travels along the current row.  

--Where the line becomes .... dotted ... the yarn of the current row is diving below the heads of the stitches of the previous row. 

--Where the line becomes   red,  this is the yarn between stitches of the same pair (which are also stitches of the same kind, whether knit or purl). We've already named this type of stitch tail a 'tweener strand. As you see, 'tweener strands travel through the fabric parallel to the plane of the fabric.  We could say that they travel horizontally along an east-west axis. 

--Where the line becomes  green,  this is the passing strand we've been talking about: it is the tail yarn as it passes through the thickness of the fabric, connecting a pair on one fabric face to a pair on the other fabric face.  As you see, the passing strand travels forward and back, at right angles to the plane of the overall fabric: we could say that the passing strand travels vertically along a north-south axis. 

Just let me drive this into the ground one more time: 'tweener strands (red) and passing strands (green) are both tail yarns as a generic category. Both 'tweener and passing strands are stitch tails which connect sequential stitches. Yet 'tweener strands and passing strands differ in what kind of stitches they connect, (same kind vs. different kinds) and which direction they travel with respect to the fabric: parallel, along the plane ('tweener) vs. right angles, back and forth through the plane (passing strands). 

Where a purl stitch and a knit stitch lie adjacent, it is the fact that the yarn travels in and out of these stitches in different directions which forces the stitches into three-dimensionality.  

Forces loaded into knitted fabric: a science-ish explanation

I've been sleuthing around in textile science textbooks and academic journal articles. It turns out there is a whole textile science universe, with such tools as "Shirley stiffness testers," serious research on the exact nature of the forces loaded into yarn at spinning, and such terms as "flexual rigidity, areal density and compressional energy."  I do not claim to understand the more abstruse terms, but what I have gathered from reading and my own little experiments is this...

Yarn is composed of individual fibers, and these fibers are often quite springy in and of themselves. Have a look at this diagram of wool's structure: it is essentially built out of springs at the most fundamental level. Other fibers may not be quite so springy, but generally speaking, fibers suitable for spinning into handknitting yarns require to have at least some bounce and body. 

In the spinning process which transforms fibers into yarn,  these fibers, generally springy in themselves, are overlaid on one another in a continuous strand, then twisted and bundled. Twisting and bundling imparts more energy to the yarn. 

figure 6:
energetic 
yarn


◄ A familiar example of the kind of energy which is stored in yarn during the spinning process (an example annoying to hand knitters) are the sorts of yarn called "energetic singles," or "overspun," or "overtwisted." These are the sorts of yarns which want to loop back on themselves as they come off the skein into the knitting. 

Not all yarns store this much energy (thank goodness!) Yet, despite such balancing processes as plying, such taming processes as steaming, heating; and even chemical treatments such as vinegar, fabric softener (or their harsher industrial equivalents) all yarns store some energy--it is inherent in the spinning process. Another familiar example of the energy stored in yarn, even plied yarn (an example again annoying to hand knitters) is fabric biasing.

So far in our account, we have springy fibers, spun into yarn which stores energy inside of itself--an energy which it retains even when knit.  Now we add more energy by knitting the yarn. You see, the loop of a stitch is a bend in this yarn. Bending adds energy by compressing the yarn and stiffening it: it is now more rigid. The thicker the yarn, or less fluffy (compressible) it is, or the tighter the twist (or all three) then the more the fibers are bunched together. Bunching stiffens the yarn, and the stiffer the fibers in the yarn are, the longer are their "bending lengths." 

In other words, the bend of the stitch-loop loads energy into the yarn and stiffens it, as well as the fabric of which it is part. An example of energy stored in knitted fabric is familiar to hand knitters: stockinette curl.  It is the stored energy in knit fabrics that makes stockinette curl a real force of nature: that curl can barely be tamed, no, not even with an iron, sometimes. 

So now in our tale, we have come to inherently springy fibers, spun into yarn which retains energy from the spinning process, knitted into bendy fabric which has more energy loaded into it from the bend of the knitted stitches.  There's a lot of energy looking for a way out in knitted fabric: a lot of pushing and jostling going on. (And here's an academic paper showing the modeling of the helixes of force within the fabric--"repulsive potential!" Oh boy!)

Looking at the closeup in figure 5 shows the pushing force in action two ways. First, the 'tweener yarns (red) are pushing the members of each pair in this K2, P2 ribbing apart. For example, although the "C" stitches along the lower fabric face are both stockinette (knit) stitches, side by side, and although both are on the same fabric face, and although both point south, yet C1 is pointing southeast, while C2 is pointing southwestIt is the force stored in the bend of the little red yarn between them that is doing the east-west pushing. 

Similarly, the passing strands (green) are also stiffened by the bend of the stitch-loop. In ribbing, we helpfully arrange the back of one stitch right next to the back of another stitch--this is the knit-purl alternation which fundamentally characterizes ribbing. When the newly stiffened yarn "sticks straight out" of the back of its stitch, the back-to-back nature of ribbing means this is the same thing as "shoving straight into" the backs of the stitches facing the other way. This straight path of stiffened yarn between the opposite-facing stitches, connecting the back of one stitch to another, means the passing strands are aligned north-south (at right angles to the plane of the knitted fabric). The force of the stiffened fibers being offered a straight-line path is what's pushing the back-to-back stitches of alternating pairs onto the opposite fabric faces, perpendicular to the plane of the fabric.

To bring this down to your own tabletop, the fact that bending (loading) this springy stuff makes it want to push back is a test you can do yourself. Lay a length of thicker yarn on the table and pinch it up into a loop. When you let go, the yarn pops into a straighter form. Anthropomorphizing, we can say it "wants to" go straight. Similarly, as this springy stuff is pinched up into a loop by knitting it, it is stiffened, and pushes back by "wanting" to go straight. When you line up the back of a knit with the back of a purl, the yarn can and does do what it wants to do: it goes straight--straight out the back of one stitch into the back of another, passing straight through the thickness of the fabric. This straight line of travel continues until it meets the next bend, and it is the force stored in these sequential bends of the unalike stitches which push the stitch heads onto different fabric faces.  

Differential length of the passing strand

A further problem with all this tail yarn entering and exiting the different stitches is, the yarn takes a different length path through the fabric on its north-south passes, depending what kind of stitch it's leaving, and what kind of stitch it's entering. Yes, the yarn between two stitch-backs is sticking straight out, * but * because of the way the stitches are mounted on the needle (usually, right arm forward), then in traveling from K to P, a passing strand has to "go around the arm." But, when traveling from the P to K, there is no arm to travel around. This means that the length of the passing strand is longer from knit to purl than from purl to knit. (At some point, TECHknitting will do an illustrated post on stitch mount explaining all  this "path around the arm" stuff  further, and when that post goes live, there will be a link here--this post is long enough already!) 

Here ▼ is figure 5, again, but this time, with blue asterisks added.  

Figure 7: same as figure 5, with asterisks added showing that, with "standard western mount," the K-P passing strands (asterisk'd) are longer than the P-K passing strands--the added length comes when the yarn travels around the protruding arm of the stitch below when that stitch is mounted right-arm-forward. 

This fabric  was knit with standard western mount--right arm forward.  As you see, the passing strand in the K-P direction marked with a blue asterisk (coming out of A2, C2, and E2) is longer then the P-K un-asterisk'd strands coming out from B2, D2 and F2. This systematic difference in length unbalances the fabric in favor of the shorter strand, being the tighter.  This means that the longer strand occurs between a knit and a purl column, as in the diagram, and that's the cause for the much-lamented "loose knit column" bordering a purl column.  However! Depending on factors such as stitch mount, yarn twist and whether the right ("regular") or left ("mirror image") needle is the working needle, the problem can also occur between a purl column and a knit column.

Knitting ribbing directly vs. "after-knitting" ribbing

So how do you tighten ribbing? How can you create ribbing without distorted columns, without uneven-length of passing strands?  How do you knit an even-looking ribbing: thick, pleasant and cushy? 

Giving a good yank on the running yarn is a popular remedy often suggested.  Similarly, working as close as possible to the needle tips is another popular remedy.  Because both these tricks pull slack out of the fabric, both these hints work, yes, but only in a limited way.  You see, both these tricks suffer a physical lower limit. No matter how tight the yank, no matter how close to the tips, no stitch worked directly can ever be made less than the diameter of the needle on which it was created, plus whatever tail is needed, physically, to travel from one stitch to another.  In other words, the size of the loop of yarn around the barrel of the needle creates is a lower limit to just how much slack can be pulled out of the fabric. You can make a stitch as small as the tightest path of the yarn over the needle, but not smaller. 

Another popular remedy often suggested is to vary the stitch mount, so called "combination knitting." This involves mounting knit and purl stitches different arms forward on the knitting needle--R arm vs. L arm. The path of the yarn traveling around the arm protruding on the stitch below is changed as the location of the arm itself changes.  Combination knitting creates a more balanced fabric as the length of the K-P passing strand is more equal to the P-K passing strand.  However, while combination knitting does undoubtedly make a more balanced fabric, it does not so much change the overall length of the tail yarn, as redistribute that length between the passing strands and the 'tweener strands. Stated otherwise, if looking back at figure 7 (the closeup of the path of the yarn) combination knitting would make the asterisk'd green (passing) strands shorter, yes, but tends to make the (red) 'tweener strands longer. Yet this is a minor issue in the overall scheme: changing the direction in which the stitch travels around the needles does not change the fact that the stitch is still being formed around the barrel of the needle. Once again, the lower bound of stitch-diameter continues to be dictated by the diameter of the needle. 

A third remedy is to simply knit ribbing tightly, on very small needles.  As noted in the previous post about gauge, tighter-knit fabric has less slack all around, and less slack means the fabric gets up to fewer objectionable tricks. Regardless of how well it works, however, your hands and arms may not thank you for this solution--it is perfectly exhausting after just a few rows.  Furthermore, a fabric stiff and tight all over may not get up to foolish tricks, but it might not be too popular to wear, either.  

So, if different methods for knitting the fabric directly are not such good solutions for improving ribbing, what is to be done?  I can offer two kinds of solutions, and, as hinted in the title of this post, each of these has already been the subject of an entire TECHknitting article, posted previously. 

Both solutions linked below share a common trait: they are species of "after-knitting," by which I mean manipulating the yarn AFTER it has been knit into the fabric.  This manipulation is done either with a knitting needle or a crochet hook. As is evident, messing with a stitch in the fabric after that stitch is off the needle eliminates the lower bound of tightness set by the needle diameter.  In other words, a stitch formed (or re-formed) from yarn which is off the needle--well! That stitch can be smaller and therefore tighter than any stitch formed over a needle, whether by yank, by working close to the tips, by mounting the stitch differently.  It can be as tight as a stitch knit on a much smaller needle, but--because you are only operating on a specific stitch--tightening the one stitch need not make the overall fabric stiff, as would happen with the "knit with a smaller needle "solution. (Note, however, that of the two types of after-knitting, one type does make a stiffer fabric than the other, more on this below.)

Another way of thinking about "after-knitting" is that you can force the yarn to bend more off the needle than on--or at least, this is so if you do not care to work tightly on small needles.  As we have seen, bending yarn tighter puts more energy into the fabric.  It is for this reason, in a mechanical sense, that after-knitted fabrics are cushy-er, springier, denser than you can get by knitting on any but the smallest needles: they are being pushed apart harder.

Have a look at the below ▼ photo, figure 8.  This is the overview of a K2, P2 ribbing created by after-knitting--it is a top down view of a ribbed hat, popped off the needles. (The hat itself is featured in figure 16, below.) The stitch pairs are labeled in the same manner as in figures 4, 5 and 7. Although the structure is the same as the ribbing shown in those figures, this fabric is significantly tighter, the ribbing significantly cushy-er and thicker. 

Figure 8: "After-formed" K2, P2 ribbing in a hat popped off the needles. This particular type of after-knitting is called "ribbing by transformation, and is discussed more, below.

Below ▼ is a closeup. The very tight passing strands (the short green ones) could not be knit directly: compare the length of these passing strands to those in figures 5 and 7: in those figures, the fabric was created in the ordinary manner: knitting the knits and purling the purls. 

In this after-knit fabric, the passing strands are much shorter than in the directly-knit fabric. Being short, they draw the stitches on opposite fabric closely together, making for a remarkably thick, energetic, and bouncy ribbing. There's very little slack between the stitch pairs, making it unlikely this ribbing will stretch out in the future. In short (har!) after-knitting tightens the fabric and specifically, shortens the passing strands. These shorter, more regular passing strands help improve ribbing by balancing it (passing strands essentially same length from K-P as from P-K).  These tricks also thicken and densify the fabric.

Figure 9: close up of figure 8.  Note the very short passing strands (green), compared to the length of the passing strands in figures 4, 5, and 7

So, how is this magic of after-knitting performed? Per above, TECHknitting has already shown the how-to in two different posts for improving textured fabrics such as the ribbing we've been examining here. 

Avoid adding excess yarn in the first place 

It is possible to avoid adding yarn at the site of the passing strand via the several tricks shown in this linked post about fixing the loose knit column in ribbing, textures and cables. At the heart of these tricks is the idea of after-forming the leftmost knit stitch bordering a purl.  

Of the four tricks laid out in that post, I here summarize only the last, although the other tricks in that post also work. In this particular trick, a strand made in the current round forms a sort of "reserve yarn" parked on the fabric surface. (The act of parking the strand on the surface is similar to working linen stitch.) In the next row, this reserved strand is picked up and drawn through to form a loop. At the same time, a new reserve strand is left parked on the fabric surface to be pulled up in the next row, and so on. Here ▼ are the illustrations of that method shown in a k2, p2 ribbing: again, the complete instructions for this trick and others like it are in the post at this link.

Figure 10

--A: Instead of purling the (darker blue) stitch adjoining the last (darker purple) knit stitch, slip that stitch instead: this parks a (red) strand on the fabric surface.   --B: Purl the next (lighter blue) purl stitch   --C: on the following round, when you come to the red strand, purl the first (darker blue) purl stitch using the red strand, NOT the running yarn. Then, purl the next (lighter blue) purl stitch.   --D: Because you did not use the running yarn to knit the (darker blue) purl stitch in step C, that running yarn has formed a new (red) strand parked on the fabric surface.   --Repeat steps C and D every round, wherever there is purl (blue) next to a knit column (purple).  

Geek note, science-ish.

 "After-forming" a new stitch out of a parked strand is an example of pi (Ï€) in action.  A knit stitch approximates a circle. The path around the outside of a circle is called its "circumference." The parked strand, the  red  dotted line in figure 11, below is a schematic representation of the solid red strand in figure 10, above .  As you see, this red strand is approximately equal to the stitch width: this width of our circle-analog (the knit stitch) is also called its "diameter."   The theory of pi says the path around the outside of a circle (circumference) is approximately three times as long as the path across its middle (diameter.) The exact relationship of circumference to diameter is that famous never-repeating irrational number "3.14159....."

Figire 11: The formula for the newly formed multi-colored stitch (bottom) is strand (red) + left partial arm  (green)+ right partial arm (green) = new stitch.  In other words, a new stitch transformed out of a strand tightens the fabric because only 1/3 of the new stitch is formed out of the strand (dotted red line).  Another 1/3 is transformed out of the partial right (green) arm and the last 1/3 out of the partial left (green) arm. Pulling these green thirds out of the neighboring stitches is what tightens the fabric with the "slipped-stitch after-knitting" trick.

Applying pi to our after-formed stitch, when you make a circle (knit stitch) out of the parked strand, the parked strand only forms about 1/3 of the new stitch. The other two-thirds come from the ( green ) arms of the adjoining stitches.  It is this act of pulling  the green yarn into the newly-formed stitch which tightens up the after-knitted ribbed fabric so very well.  Further, that green yarn could never have been pulled into the new stitch if the two neighbor stitches were still looped over a knitting needle.  Only because the neighbor stitches are off the needles and laying loose  in the fabric of the row below, can that this green yarn be pulled out of them.    

Even geekier note: This relationship of the circumference of the loop to the length of a strand is also why, when playing yarn-chicken, you can tell if you're going to win by measuring your remaining yarn against the width of the fabric.  If the strand is about four times as long as the row, you'll have enough to eke out one more row and a tiny tail left over.  I say 4-times and not 3-odd-times as wide because the circles of yarn (knitted loops) don't sit exactly next to one another--you also have to account for the connecting tail. (And, of course, as you know, to further improve the odds of winning at yarn-chicken, it helps to hold your breath, tense up all over, and knit as fast as possible all the way across that last chicken-row!)

Below ▼ is a real-world example of slipped-strand after-knitting. The ribbing on the bottom portion was knit directly in the ordinary manner, right arm forward (standard western mount).  The top portion was worked via slipped-stitch after-knitting.  I've stretched out this overview sample so the purl columns aren't hiding on the fabric back. You can really see how much tighter overall is the after-knit ribbing.

 

Figure 12--slipped-stitch after-knitting, overview, stretched out to show purl columns

Here ▼ is a further closeup, unstretched this time. Compare the columns of numbers on the side--these are the identical labels from the fabric, just copied-and-pasted onto a plain background. This illustrates that forming the leftmost stitch of each knit column out of the parked strand has the effect of tightening down the entire fabric.  Just that much slack is pulled out of the fabric! (Note that, of the two tricks shown in this post, this one--slipped-stitch after-knitting--tightens down the fabric the most.)

Figure 13: slipped-stitch after-knitting, closeup


Ribbing by Transformation

The second solution is "ribbing by transformation." This amounts to dropping a column into a ladder, then latching the ladder rungs up again as a knit column on a purl background.  If you prefer video, here are links to two tiny you-tubes about that process.

 Releasing a ladder: https://youtu.be/qpLCBX3MZmA 

Latching up a ladder:  https://youtu.be/LDIy4VIfi_Q

This transformation forces 'tweener strand which originally connected a K-K or P-P pair of stitches to become a passing strand connecting a K-P or P-K pair.  Because 'tweener strands are shorter than passing strands in directly knitted fabric, this trick of transforming a 'tweener strand into a passing strand tightens a transformed fabric. Note that this trick tightens the fabric less than does the slipping trick discussed above: re-orienting a 'tweener strand into a passing strand does indeed tighten and thicken the ribbed fabric but not as much as forming a loop out of a strand. There's less pulling yarn out of the arms of the neighboring stitches with transformation, and, as shown in figures 8 and 9, above, the overall row gauge does not change much with this transformation trick (although it does with slipped-stitch after-knitting).

Transforming stockinette into ribbing is a trick which has also been previously shown in a TECHknitting post.  Although the original post appeared in the context of fixing a curling scarf, the idea of transforming a stockinette fabric into a ribbed one is identical, so have a look at that post for the full details. Note that the very tight fabric in figures 8 and 9 above (showing the super-short passing strands) was created by transforming knit fabric from stockinette to 2/2 ribbing.

Here's ▼ another example of ribbing by transformation.  The first photo shows a half-and-half fabric clipped onto a magnet board, right half made in 1/1 (K1, P1) ribbing worked directly, left in stockinette.

 

Figure 14
(Side note: Although you can't tell from this photo, the two fabrics are more or less the same length, the apparent difference is due to stockinette curl.  Further, this fabric has not been blocked--more about blocking below.  Further note: the V's of the stitches in this yarn appear unbalanced.  This is because this yarn is a single yarn--Brown Sheep Lamb's Pride worsted-weight.  This single yarn is not overspun, and so does not kink back on itself, nor does it bias in the fabric other than to form uneven V's in each stitch.)

The second photo ▼ shows that same fabric, now with the left-hand stockinette portion transformed into 1/1 ribbing to match the right side.  Because the fabric is unstretched, you can't see the purl columns, but they are between each of the stockinette columns. On each fabric, six knit columns (11 total stitches, if you include the 5 separating purls on the back) have been  marked out with pins--red for the transformed fabric, white for the directly-knit. Even over this short of a run, there is a quarter-of-an-inch difference between the two ribbings. This photo does show the tighter nature of the fabric, what it cannot show is the greater bounce or cushy-ness of the transformed fabric, but that does exist. 

Figure 15: Even over so short a run as this, there is a marked difference between the directly knit (white pins, right) and after-knit (by transformation--red pins, left). Because not as much slack is pulled out of the fabric by transformation (this trick) , the row gauge does not change as much as it does with slipped-stitch after-knitting, figures 12 and 13, above (previous trick).


Which trick where?

For texture knitted fabrics like basketweave, where the pattern changes every few rows,  then in my experience, after-knitting as-you-go is quicker, meaning slipped-stitch after-knitting with its parked strands, figure 10, above. In fact, if you practice at it, this method of knitting becomes second nature. The downside is, this works best in circular knitting (↻) because when worked back-and-forth (↔) the strand parked on the fabric surface on the way there (← ) turns out to be parked on the reverse fabric face on the way back (→ ) making it harder to get at.

For 1/1 (K1, P1) ribbing, then in my experience ribbing by transformation is somewhat easier, and the sort of work you can do mindlessly while watching Netflix. However, transformation takes longer. 1/1 ribbing is shown in figures 14 and 15.

NOTE that if planning to transform stockinette, the transformation will go easier if you do NOT block the stockinette before transforming.  The reason is, blocking sets yarn into distinct and directional waves--it "kinks" or "sets" the yarn.  Transforming a wave that has been set to curl one way (a knit stitch) into a wave curling the other way (purl stitch) is like trying to brush your hair to curl under when you absentmindedly used your curling wand to curl an outward flip.

If dealing with only a few loose column of knits on a purl background, such as with cables or broad ribs, then, in my experience, there's not a lot of difference, efficiency-wise. Mess around and see which trick you prefer.

Real world uses

One practical use in every-day knitting is to densify the outer edges of your ribbing.  Even if you don't need ALL the ribbing in a sweater-bottom or cuff to be stiff and straight and very thick, you may wish to consider either knitting the edge stitches on a smaller needle, or doing one of these after-knitting improving tricks for the rows of ribbing closest to the outside (first few rows if knitting bottom up, last few rows if working top down.) As few as five rows or rounds is plenty enough to make a real difference.  It's the edge which suffers snags, rough surfaces, and teeth used to pull on a mitten, so making this denser is worth the extra work. A further advantage: after-knitting (or knitting on a smaller needle) not only toughens the outermost edge, but has the further advantage of preventing flare and splay-out over time.  

 For cables or wide ribs, working the last of the K-columns-before-a-purl via slipped-stitch after-knitting is a reasonable procedure, especially because that trick becomes faster and faster as you get in rhythm. It's also not a bad trick to try if you have the problem of a large hole developing just past the twist of your cable. As mentioned above, this trick also makes a truly noticeable difference in basket-weaves.

As for ribbing by transformation, well, transforming small areas like a cuff-edge goes quickly, but transforming large areas of stockinette into ribbing takes a long time.  Large-scale transformation is so time-consuming that it is a trick to be trotted out mostly for special occasions--like rescuing the curled up scarf many beginners mistakenly knit in stockinette

Figure 16

◄ Another special occasion for ribbing by transformation is if you want a  *serious * thick ribbed hat for winter weather. In a few more posts, TECHknitting will offer a free pattern for a seriously thick and cushy winter hat with ribs made in this manner.  Figure 16 is a little preview of this "out-standing" hat: so thick and heavy, it easily "stands out" there, upright on its own.



Til next time,

--TK


PS: Are you perhaps wondering why there's a whole post here which ends up just sending you to two previous posts?  See, if TECHknitting blog were a book, and if that book had a competent editor, the structure of the knitted stitch, its two types of tails--'tweener and passing would have been edited together with the idea of "after-knitting" from the previously-published posts. However, TECHknitting is not a book, but a serial blog, written higgledy-piggledy over many years. Futher, there is no editor, competent or otherwise. Net result = after-the-fact summations like this one, tying up all those loose ends.

Going forward, I think the passing and 'tweener strand vocabulary is going to make it easier to coherently analyze stitch mount (direction of wrapping over the needle) which is a post coming up at some point. So, this post and the previous form a bit of background prep, as well.
 
PPS: Questions? Comments? I've turned off the comments for the blog--oh so much spam for sex toys, car parts, who-knows-what. For some reason, Facebook has prevented me from posting there for years--something about being anonymous as well as too commercial--so that's not a good way to contact me, either. Instead, DM me at Ravelry where I am TECHknitter, or use the e-mail link on the lower left sidebar of the "more info" profile page.

Down the rabbit hole
(related links on TECHknitting blog)