Related posts on creating loops:
long-loops-a-bit-on-method.
long-stitches-on-km/
some-long-stitch-swatches/
For double-bed work with long loops in various designs in single or multiple colors, see drop stitch lace
I became curious about creating long loop shapes on a knit ground using continuous strands of yarn, forming loops in the same direction, and allowing for knit rows between them, resulting in returning the yarn to the starting loop formation side. My own preference is to make the loops with the knit carriage on the right, anchoring them with the following carriage pass to the left. To continue doing that, the yarn would either need to be cut, leaving a yarn end, or it needs to travel back to the left, where it can form the next row of loops from left to right.
In these tests, the movement of the yarn back to the left was achieved by having it weave on every other needle from right to left.
Triangular shapes are easily recognized, and I began with a moderately large one, at first planning for loops on every other needle. Black cells on the right represent loop locations, and green the weaving pattern formed as the carriage knits right to left. 
The length of the loops determines end-use; long loops can become fringe, and short ones the macro version of pile knitting.
Tools to aid loop formation that are fixed in height across the needle bed are needed if the pattern is not used as a single motif, but rather as a recurring one. Separate strands of yarn on bobbins or balls, or cones would need to be used for each shape. The loops in this swatch are large, so there are extra knit rows between each set.
I use colors and yarns that are randomly on hand for swatches unless I am planning a specific finished item.
The first swatch is worked with loops on every other needle, the second with loops on every needle, as shown in the photos
Worked as a single motif:
KC II: preselection is made from left to right
COL: make loops moving from left to right using ribber gate pegs, I like to bring needles to be wrapped out to hold as I move across each row, wrapping frequency may be varied apart from needle preselection,
knit a row to left, anchoring them
COL: knit to right, the needles for the weaving pattern to the left will preselect
COR: check that weaving brushes are down, place yarn over preselected needles,
knit a row to the left, securing the weaving pattern, the yarn end will be on the left again
COL: knit a row to the right, needles for the next row of loops will be preselected
COR: use the yarn now on the left again to form loops from left to right, knit a row to the left to secure loops. Repeat the process. The resulting shape brings holiday knitting to mind.
The loops worked on every other needle
and on every needle. The original preselection repeat may be used. The only yarn ends that need to be dealt with are at the beginning and end of the shape, the reverse of the knit has the appearance typically seen in knit weaving, where the thicker yarn forces the stitches in the background one apart.
What of recurring shapes with reversing directions? It occurred to me that plastic straws may prove to be handy tools for creating the loops and easier to manage than rulers. The maximum length of the straw is a limitation. Exploring the technique in a single motif gives one the opportunity to explore issues and limitations.
My straw is 10.25 inches wide, made of plastic, is not rigid, and can be squeezed flat easily. When using it as a loop guide, it was held low enough below the knitting on the top bed on the carriage side to allow some ease in the loops so the straw can have an easier time dropping below the sinker plate when the knit carriage passes over it. It can even partially collapse if needed.
I found managing the straw soon became rhythmic and easy.
The lengths of the straw on both sides of the loops serve as a handle on the left, and on the carriage, right side, the extra length can be guided down a bit to keep it from angling up and getting caught in the sinker plate.
Once the carriage began to move across, I encountered no problems. The straw may get pushed slightly to the left with the carriage pass.
Planning possible actions and a starting repeat with single rows knit between loops. In this instance, loop formation happens when the carriage is on the right, and weaving happens when the carriage is on the left. The latter means the weaving yarn is laid over selected needles in the wrong direction, with the long end of the yarn away from the knit carriage. That said, because weaving happens only on every other needle X3 and on a max of 7 on every other needle, the first selections were no problem, and the second was manageable with some caution. The arrows indicate the direction of the carriage movement after the loops are formed, and after the yarn is laid over the preselected needles for weaving.
The image on the right is the repeat tiled in Gimp to evaluate its vertical alignment. It is 14 pixels wide and 20 high, and could be used in a punchcard and placed at the center of the card, making it centered in each 24 stitch location on the needle bed.
If an attempt is to be made to knit several repeats, vertical columns of plain knit between the pile knit columns are desirable.
The needle preselection will begin on 14 stitches, so only for the first 2 rows, the needles preselected for the first 7 stitches of the 14, shown as grey cells, push the needles involved back to B.
Depending on the machine model and software used for the download, the repeat may need to be mirrored to achieve the result planned in the drawing, which is true in my 930. The starting PNG
may be placed on a wider canvas in Gimp; here is a 30 stitch mirrored version for use on my 930.
The resulting design is quite dense but successful.
Planning moves on for weaving to happen with the yarn in the proper direction, particularly if more than one design is to be attempted horizontally, and to make the loops less dense. Extra rows of plain knitting are added between loop formations.
Here, the design is placed once again on a 30-stitch ground but is now 40 rows high, drawn in the desired direction, not yet mirrored. It is expected that the aspect ratio for the shapes will change, becoming elongated. Again, preselection for the first seven stitches on the right for the first two times selection occurs on those needles is canceled with needles pushed back to B.
Clothespins, as seen in the lower right, can come in handy to put slight weight on yarn ends to help hold them in place at the start of the piece or as it progresses.
The proof of concept swatch shows a very different appearance in the distribution of the loops from the first sample.
It is possible to knit more than multiple series of shapes. Enough spacing between the forms is required to allow for the use of more than one straw.
I had the easiest result by leaving the straws uncut.
Each straw end on the carriage side again needs to be kept down until the sinker plate begins to pass above it.
It is a good idea to try the method first and practice some hand selections.
The rows where all 14 stitches in each group were worked required a bit of extra care, but knitting was quite manageable.
Long loops may also be created with i-cords, strips of knitting or fabric, strung beads, and imagination is the limit, depending on preference, available materials, and the specific design. Even tiny beads may be used to create loops, as seen in this swatch. Dental floss may be used as the very strong “thread”, secured with e-wraps on the knit background where needed.
For threading the beads onto lengths of the chosen thread, a tool I have used (which also comes in handy for threading a serger) is an “EEZ-THRU” floss threader, recently available in this brand 
Here strips of knitting are used along the edge and knotted, strips are also applied to the body of the swatch during knitting for contrasting color interest
I cords may be applied to the edges of a knit ie scarves, but there are a lot of ends to weave in. There are trade-offs if techniques are explored to use continuous strands with limited or no cutting.
There is a hand knitting pattern called foxpaw, for which some illustrations may be found in the photos related to the advertisement for this class https://stitches.events/shop/classes/west-2022/intro-to-stacked-stitches-6/.
A while ago, I was interested in trying out a simplified version using the knitting machine; the project has remained a UFO (unfinished object) since then.
The graduated size cordings here were created by knitting lengths of i-cord on 4 stitches, knitting X number of rows, folding the cord upon itself, and joining the second half to the first using seam as you knit, continuing with the next shape immediately after completing the join. When applied to a cast-on edge, if the goal is to use the technique for a substitute fringe, the double cords are a bit dense compared to the body of this particular knit and too close together. Perhaps creating a trim by applying the double cord to a side edge of a vertical strip and then in turn to where needed would produce a better result, or limited numbers of graduated cord lengths might provide interest in the body of a knit. 





























































A large variety of pleats may be knit on the single bed as well, one being shadow pleats. The resulting knits also need to be gathered on one of the 2 edges in items like skirts at waistlines, and the number of knit rows needed for the volume required can be daunting and a large commitment of time if not effort.
Both yarns used are wool, a yarn that has memory and spring-back. In the potential fold rows, the red color was replaced with a thin ply with the same fiber content. The first folds were knit in an *8 with thin, 14 with thick, 8 with thin** color sequence. Watching the knit as it progressed showed the thicker fair isle areas folding inward, with the thinner areas folding outward. The remainder of the swatch used a 4 thin, 8 thick sequence. The knit was steamed and pressed, the folds are soft but permanently present with the lower edge of the piece lying quite flat after a considerable amount of time. 

The blocking saga: I do own blocking pins but honestly have only used them in demos, and on rare occasions such as this, or to cut them down when I needed a fast replacement for a ribber cast on comb lost wire.


after steaming and pressing 

Trying for a bit more detail, there is a see/peek-through quality.
At first, I thought it might be a version of the stitch using 3 colors, but later, looking again, I came to believe it to be a 2 color 
I wanted to add solid color drop-stitch stripes, the solution being to add all black rows. Every needle will select every row in those areas, colors are changed every two rows, creating the desired effect when the corresponding stitches are dropped.
A visual summary with an added color change column and row numbers with rows on which dropping all loops can occur marked with different color numbers. In this case, the ground is in the dark color, the shape in the lighter one.
Determining whether the shape or the ground stripes knit the dark color is simply done by beginning the first all knit rows with the dark color or the light.
but I would like to have stripes in the solid lighter colors as opposed to 2, 


Here some of the peek-through quality of the fabric is shown again. When any fabric is gathered sideways, it is likely the repeat will appear narrower and longer.
Using a 16/2 cotton at nearly the same tension produces a fabric that is even sheerer. 
And now those solid stripes in 2 rows knit are really bothering me, thinking they should be switched to only one row of loops, making the original repeat without those added black rows the better repeat?
A potential use is for the thinner knit is for it to be used as a layer over a different pattern or the same design as seen here, in the thicker yarn 
seeking a rounder, smaller donut shape:
The red is a thicker yarn than the pale yellow, the fabric had an interesting slight curl evocative of shadow pleats. Other possible changes include the elimination of the top and bottom row in the original design.




This placement is tested and kept in mind in other explorations.
Using a spreadsheet: the same sort of chart may easily be created quickly in Numbers. In this instance, my table is still planned for 24 stitches in width, but 54 rows in height. An extra column is added and used on the far left to mark rows to be hidden. Beginning at the bottom left two rows were filled in a different color, the third row is left blank, all 3 cells are selected. When multiple cells are selected, depending on which side of the selection box one hovers over with the mouse, a yellow dot/ handle will appear.
The tool acts on the selection. Clicking on it and dragging it with the mouse will, in this instance, repeat the selection until the mouse is released. This may be done in any direction and quickly fills in whole tables. It is not necessary to perform this extra step before hiding rows, but I find having that extra color makes it easier visually, especially when working on long repeats. It also makes for easy return to selection if hiding rows is done in shifts.
continue to the bottom of the chart, and under the Table menu, choose to hide 36 rows. With rows hidden the mesh repeat shrinks from 6 rows to 2.
A new 4 cell table is created. The 4 interior cells were chosen, copied, and pasted in the lower-left corner of the reduced height table. Once pasted, selecting the repeat again will allow one to use those yellow handles to fill the contents first toward the top, then to the right. The image on the left shows the results, with only the numbers for the unhidden rows shown on the left. In turn, the heart was the pencil tool to draw it using a third color in the chosen location.
The unhiding rows function produces the expanded repeat with all knit stitches in red on the mesh grid ground. The column with the yellow cells is deleted before converting the Numbers repeat to an electronic or punchcard, the how-to discussed in other posts. Comparing the hand-drawn heart in Gimp alone on the left, to the spreadsheet results on the right, there appears to be a difference in the starting rows, and in only one other row, two black cells appear that can easily be altered in either repeat. Brother machine knitters would need to shift those 2 blank rows at the bottom of the repeat on the right to its top or to to start lace patterning on row 3. 

More choices exist, continuing to place the heart 3 rows up from the bottom, and not using multiple layers. 
The layer-to-alpha image may be saved as a png with transparent background for any future use.
Even easier, working with the full-color, white ground heart repeat placed 3 rows up from the bottom, A, set red as the foreground color, and then use the bucket fill tool B to fill its ground with the mesh pattern, seen in C
Comparing the all in Gimp Drawing to either of the last 2 patterns, two differences appear, an extra row of transfers before removing transfer stitches to start the heart shape, and those 2 extra black pixels/eyelets 


Developing brick repeats or half drop is possible with offset and brushes in Gimp, but, to my mind, easier in Arah. Using the Arah drawing in repeat, the design is now 24 stitches wide by 120 rows high in a brick arrangement offset by 12 stitches. 
The same heart, in half drop repeat, offset by 30 rows, now double wide, 48X60, suitable only for an electronic machine
Changing the background grid for other stitch types: the heart is rescaled for use by 2 or 4 times in height. Again, the differences between Gimp’s incorrect scaling, A, and the Arah drawing in repeat, B 
Possible applications 
This begs the question of working on larger images. For use on a lace mesh, simpler designs apart from overall size are best, but if a mesh base is not your favorite fabric or goal, tuck, slip stitch, thread lace or even fair isle patterns may be created with more overall flexibility, using the same principles on backgrounds.
This is the first attempt at a partial repeat test proof of concept. I have been telling people serger monofilament withstands ironing and light pressing, and periodically I test advice I have given formerly. To start with, the darn monofilament, which I even used in double bed garments, but nearly 2 decades ago, refused to feed properly or at all. Because it is nearly invisible, my knitting started with it pulling too tight, and I wiped out 12 needles out of 72 in different places on the needle bed in a single carriage pass. Determined, I sorted how to hand feed it, got the rhythm, complete a swatch. A marks an error I made in loading the second track in img2track, resulting in an added, wrong pattern row. The holes, B, C, and D were nonexistent until I tried flattening the fabric a bit with an iron, and the monofilament simply melted in various places. 
I had interesting issues with having needle selection visually appear correct throughout, and the pattern itself appearing correct when checked at 1800 magnification. While the smaller sample was accurate until the filament began to melt, here I had 2 needles not knitting the yarns alternately but together, and an odd change in the center that looks as though yarn selection in those areas was reversed. The thicker yarn here is cotton, the thinner rayon. Troubleshooting is required. Unplanned tucking is a sign of damaged needles, but because the second yarn used herd is so thin, the effect was not obvious until the vertical pattern in the same area on the bed became apparent.
The cam buttons and undercarriage were oiled. The first selection tests involved programming this 
The finished test swatch: because of the disparity between the number of all knit stitches vs patterned ones where one of the colors is slipped on every pattern row for every other stitch, the fabric shrinks dramatically when off the machine. The knit areas pop out a bit, and a lengthwise tug makes enhances the effect
If the goal is a flat fabric panel, then blocking is a necessity. Here the piece is casually pressed, no pins, letters point out issues:
The mesh effect was noticeable after resting, the piece measures 25 inches in width by 25 inches in height.
a new file was opened, bucket-filled with the ground 2X2 pattern, the above was copied and pasted in the chosen position, resulting in this
The 100X92 png
which in turn needs to be color reversed for use as a thread lace pattern 
Nearly all my previous dbj pieces have been knit on a Passap E6000. The 930 experience for such repeats is new to me. With some help from Tanya Cunnigham in reviewing the steps required when using img2track, I returned to cellular automata repeat saved years ago.
That provided an opportunity to decide I preferred the reverse color placement as well as wanting a thicker ply for the white, resulting in twice the fun with 2 colors, and another scrapped sample 
I was able to achieve a reasonable repair on the knit side, but the birdseye pattern on the reverse is a bit scrambled. 
There is also potential for exchanging colors to get a sense of how the pattern might appear in different colorways 







This is from a Japanese language manual for the 891(1987-89) punchcard machine, which appears to introduce the idea of replacing the use of the knit carriage and plaiting feeder with one specifically designed for intarsia 


Since the knit carriage needs to move to and from the left-hand side of the machine with each color change, an even number of rows in each repeating segment is recommended, but not necessarily required. The first preselection row generally moves from right to left. End needle selection helps the edge stitches knit. At times end needles will need to be pushed forward to the knitting position by hand. Depending on how the repeat is placed on the needle bed, with some experience with a tuck or slip stitch, one can decide whether keeping the end stitches in the pattern creates a better effect at vertical edges. Analyzing 2 random repeats
and their respective PNGs.
Four tucked rows are probably the limit on Brother machines unless one is working in fine yarns. In the first interpretation, the ground knits for 2 rows on all preselected needles. In the second, the surrounds of the interior striping are knit for single rows only at the top and bottom of the repeat. Electronic repeats A and B on the right may be as small as a single 4X12 unit.





The “corrected” repeat without additional rows, some possible color changes can follow the colored chart suggestions.
Moving on to electronics, playing with symmetry, the repeat now becomes 30 stitches wide, the tiled image check for the unaltered version on the left. 


Tuck knit in a solid color 
Slip stitch in all variations has purl side edges which curl to a greater degree, the short skipped areas were probably due to too tight tension of too quick a carriage pass to the opposite side 




slip stitch with the addition of a third color in the rotation 
The associated punchcard: 

Brother yarn changers are numbered, from right to left, and their published card designs color suggestions reflect that. The lace extension rail must be used as the knit carriage needs to clear the color changer on that side in order for the colors to change properly. 
Some of the Studio punchcard pattern books showed both sides of the expected fabric associated with each card. Here is a repeat that breaks the tuck rule of no more than one blank square side by side in any row.
This swatch pattern from a Japanese magazine illustrates the difference in the formation of the tucked loops when two blank squares exist side by side. The repeat is 10X22, and the colors are assigned letters rather than numbers here as well.
Another rule breaker: odd numbers of tucked rows with no added all-knit rows. Experimenting with such repeats results in less organized all-over patterns, here colors are changed every 2 rows, 

These repeats take shapes in another direction which becomes more textural and interesting when blank stitch areas are expanded for use in slip stitch setting 
This pattern, with color changes every 2 rows and two-row tuck sequences has an offered interesting pattern shift.
The tiled X2 horizontal repeats lined up side by side show that extra knit stitches have been added, shifting tuck stitch rows by one stitch in alternating directions, but just because it is published, it does not necessarily make the repeat correct. Those striped areas can only occur if there are solid all-punched areas.
Keeping the constraints of a 24 stitch repeat, reducing the width of segments to 12 stitches rather than 24, the original repeat as amended
The tiled results for each
An approach to designing such patterns can begin with a template for color changing every 2 rows and taking colored squares away to indicate stitches that will be slipped or tucked.
Japanese magazine publications, which often recommend many color changes, each for varying numbers of rows. Sometimes more is less. The number of colors may be reduced, and changing the number of rows used for each color as well can expand the number of fabrics produced from a single repeat. Good note-taking is a necessity if the intent is to easily reproduce the fabric at a later time.
Adapting punchcard designs for use in electronics becomes easier once one is familiar with the stitch structure. This is a cousin of 328, 13 stitches X 52 rows. Tiling, as in all designs, helps sort out errors or missing pixels.
and slip stitch changing colors alternately every 2 rows, then every 4:
My blog posts on working with and designing mosaics (suitable for tuck and slip) and mazes (slip stitch only, multiple side by side unpunched holes or white squares in any row), in reverse historical order
while in slip stitch the strand of yarn bypasses the non-selected needles as the row is knit, forming floats between stitches
Symbols commonly used for both
Some of the options:
and not often used, but worth exploring, the use of opposing tuck and slip buttons at the same time.
As with any knitting, for needle selection to occur the knit carriage (also known as KC) needs to engage the belt using the change knob set to KC. End needle selection or not depends on the goal fabric. If KC is in use but no cam buttons are pushed in, there will be needle selection, but the fabric produced will remain stocking stitch.
Converting the cards to black and white pixels: the small single repeats for each card are highlighted with a red border. Depending on the method for programming the electronic machine, however, the single repeat may have to be repeated horizontally to match the number of stitches to be used in the piece. The third repeat is a hybrid of the previous 2, the start of making what is published more personal 

The blank vertical areas may be arranged moving across the repeat’s canvas in a variety of ways. In this chart the tuck symbol is evident, some of the knit stitches around each tuck series are highlighted at the bottom of the chart in green, the single electronic possible repeat is 4 sts by 12 rows 

Few tuck stitches amidst lots of plain knitting are likely to not distort the fabric very much or produce a noticeable texture. The fabric will lie fairly flat, and approach a width proportionately closer to that of stocking stitch using the same yarn. The outlines can serve as markers for the introduction of additional hand techniques ie tying objects or beads in the center of the shapes after knitting and prior to felting in order to obtain surface bubbles of non felted stitches, or marking areas for duplicate-stitch or other embellishments.
With the basic structure recognized, weaving punchcards may be suitable, not all need be color reversed. 

251 is a bit more problematic. Half the repeat is wider than 24 stitches. Here it is readjusted to 24 stitches, the height is 32 rows which may just barely squeak by punched only once
the repeat is tiled to check proper alignments
This repeat is from a Studio mylar sheet. It also may be used in punchcard machines after removing 2 columns, since only half the repeat is necessary, and it is 26 stitches in width. The color reverse option is necessary. In electronic machines, it is easily accomplished with a command or the flip of a switch/ push of a button. The white squares as given would produce loops on all the corresponding needles, with no stitch formation in those areas.
I chose to eliminate 2 columns from the blocks on the left. Tiling shows the amended repeat’s appearance, with the color reversed image for actual knitting to its the right.
Repeats with a balanced number of black and white squares provide all-over textures in fairly balanced fabrics. As the number of black squares on a field of speckled tuck stitches grows, the knit shapes may actually poke out from the surface of the knit, since those areas are not gathered in the same way as their surroundings. Yarn properties and tension also have an effect.

Designing your own can begin with the choice of a template, such as this one, 24 stitches by 36 rows. 
Working with a different shape, using copy and paste to place it, adding a brick variation on the right, for punchcard full repeats of 24X40
Testing tuck stitch limits, breaking the side by side white square rule in all over patterning with moving blocks of 2 by 2 blank squares 