“Automated” shell shapes

March 2020: in attempting to fill a request for a punchcard repeat, I am finding added ways of looking at the topic, will share in a later “revisit” post.

This technique combines holding with the slip stitch setting. When KC is set to slip it is the punched holes/black squares that result in needle selection and stitches knitting. Blank areas in cards or mylars are slipped/ skipped.

my mylar repeats for each segment

the swatch before pressing knit side

the purl side

the dimensional texture is flattened out  when pressed

the purl side, flipping the shell shape horizontally

using a yarn with memory and tighter tension would help retain the 3-dimensional quality if that was the  original intent

A bit on method: the repeat used for the hand technique in the previous post was changed to an even number of rows, with other adjustments.

The execution was in a brick configuration

all knitting begins on and moves right to left; needle bed may be marked to help track repeats

. for straight side edges program second mylar repeat first; I knit my sample on needles 22L – 34R

. some needles will need to remain OOW, cancel end needle selection = KC II

. COL: first selection row is done L to R with the yarn color used for the next shell sequence in A feeder

. COR: the machine is set to hold non-working repeat groups, and the KC is set to slip <->; in the half-repeat working the first half-shell takes place on the first 7 needles on right

. COL: when the top of the repeat is reached the orange row will be selected left to right

. COR: after needle selection of previously held stitches happens on that row, bring the total number of stitches for the next repeat on the left into work manually, knit one row across the 21 needles

. COL: stitches in the yellow area will be in B position; bring all stitches to their immediate right to hold, then the “yellow group” to work by hand, continue to knit in the same process across until the horizontal row of shells is completed

. COL: program machine for full, alternate pattern repeat (bottom of mylar) for a row of all whole shells

. COL: depending on personal preference, holding may be canceled for the first selection row, or stitches may be pushed back to D position and carriage kept set to slip <-> before knitting back to right and resuming working on each pattern unit

. COR: repeat process, working on groups of 14 stitches at a time from left to right

. COL: on completion of the row of full shells return to the first program, continue the process until the desired length is reached

To match casting on and binding off  I often start with waste yarn, make the decision as to how to end the piece in a way that I like based on my test swatches, then rehang the stitches from the first row and treat them as I did those in the last row of knitting.

An entrelac pretender

“Automating ” normally labor-intensive hand techniques cannot truly duplicate them. Below is one effort to produce an “entrelac-like” fabric using the slip stitch setting. The biggest advantage of this is the knitting speed as opposed to creating the individual cells using holding and picking up stitches. Some drawbacks: slip stitch floats on the reverse make for a dense fabric, and may be an issue in a garment. There are elongated stitches along the edges of the shapes at the color changes, these cannot be avoided. I am not a fan of the Brother single bed color changer so I opted for changing the color by hand. “Air knitting” helps one sort out where to set up the pattern on the machine. The fabric is knit using the slip setting, so I chose placement on the KM where I had needles selected on each side of the knit on my first row.
The pattern is begun on row 8, card locked, COL.
The first-row preselection is to the right.
COR, card to advance normally, KC set to slip in both directions, change color after 8 rows of knitting, then every 14 rows.
As in any slip-stitch design, punched holes will knit stitches, unpunched holes will skip them, creating floats on the purl side between knit stitches.  The green dot on the right of the card indicates the starting row for the first repeat, row 8, subsequent lines are reminders to change color before knitting across that particular row, taking into account the fact that the card reader is reading below the line of sight on the machine’s exterior.   Note that before the color change the yarn in use will create a float in the area that will be skipped by, and knit in the same area as the subsequent color will; at first, this may seem counter-intuitive, but it is correct, and the effect may be seen on both sides of the swatch.
If changing color by hand, remember to “close the gate” after each color change to avoid dropping the knitting off the machine if the yarn in use moves too far forward to be “caught and knit” by the appropriate needles.
With necessary adjustments in starting row and settings, technically this fabric may be knit on any KM brand.

The yarn in this test is the same alpaca/silk blend used in my shawls, at tension 6. If a color changer is used, the KC row must be from right to left with knitting rows moving left to right and back to the color changer.

2023
The same repeat may be used double-bed as well, using single or multiple colors, with the potential for variations in striping.
The minimum electronic size repeat is 12X28 pixels.
The red cells in the chart indicate the greatest number of rows, 12, that any one stitch will be held until a black pixel knits that same now very elongated stitch on the next pass.  The main bed is set to slip in both directions, and the ribber is set to knit in both with the first and last needle in use on the ribber.
The skipped stitches get quite long, thin yarn may be used but watch for any breaks or issues with gate pegs, adjusting the tension as needed.

Symbols to punchcard 1

In the chart below pattern repeats take into account the punchcard limitation of 24 stitch maximum repeat. In the sections separated by the color stripe, the bottom shows a purl side symbol chart for slip stitch, the center the purl side symbol chart for brioche/ tuck stitch, the top the repeat punched out for use on the KM with black dots representing punched holes. Keeping in mind there is a 36 row minimum for the card to roll adequately through reader, this repeat would need to be punched 9 times. If one wants to use the color changes in other than totally random manner, then the pattern repeat must be an even number of rows in height. One option is to use double length, but unless the yarn used in the repeat below is very thin, for tuck stitch that may be beyond the limit of the KM. I also prefer when knitting lengths of fabric not to use elongation; for me that makes it easier to correct mistakes. A reminder: the punchcard selection mirrors the design horizontally (particularly noticeable in letters), so the hand knit repeat need not be reversed for a match.

Below is a more manageable tuck repeat reconsidered for color changes (shown in change of ground behind punched holes). The first row selection needs to be from right to left toward the color changer in Japanese machines (Passap is on right, but console takes that into account). This is not the only color change sequence possible, only a place to start.

With very rare exceptions, tuck stitches generally must have a knit stitch/punched hole on either side of the unpunched square. This is because side by side loops jump off on the next pass, rather than knitting off in a group, making a long float in in some cases an interesting mess. Because slip stitch skips needles creating floats rather than depositing loops in needle hooks, the tolerance for side by side slipped stitches if far greater, and the number of rows that the individual stitches are not knit is limited by the strength of the yarn, and the tolerance in the machine. Both tuck and slip stitch fabrics benefit from being evenly weighted, with weights being moved up regularly during knitting. Canceling end needle selection and having the pattern repeats line up with tuck/slip on each edge may produce interesting side edges. If texture is the goal yarns that can be “killed” by pressing/steaming should be avoided.

Pleats: automating “pleating”, single bed

This post was originally written in 2013. I found myself reviewing it in 2018, and editing it with the intent to attempt a pleat sample using the Ayab interface. In those days I was using Intwined for some of my charts. The program has since been unsupported for Mac, without upgrades and a series of problems that led to my switching over to excel completely for my charts, and now to Numbers in the latest Mac OS.

I work primarily on multiple model Brother machines, in this instance the 910. Some of the information provided below may need tweaking for use in other brands KMs. “Automation” of some holding functions may be achieved using slip setting to knit required stitches. Below is the mylar sheet repeat used for my sample swatches. The setup is on working needles 21L, 19R, program for color reverse and twice the height; KCII (cancel end needle selection); first-row needle selection from left to right, and with carriage ending on the right the first set of needles selected will be those that knit, the remaining bed will slip. This is opposite to the configuration familiar in holding, where needles out to E are held, those in B or D will knit (Brother needle position jumped the letter C, they are  A, B, D, E). In this type of knitting, all needles in use on the bed should be cleared with each pass of the carriage.

I have gotten used to keeping programming numbers for  locations on mylar around a square to correspond to the lights surrounding the house icon on the 910 and worked with the following 2 options for my test swatches

In the samples below the first set of every other stitch/ black square in the increasing/decreasing angles were slipped on the non-selected needles to create/mark the inner purl fold, the second set in the area that does straight knitting tucked on non-selected needles to create/mark the knit outer fold. For the latter to occur, the cam buttons need to be switched to tuck <-> for 2 rows, and then back to slip <-> for the remaining knit repeat. Small holes are created at edges of slipped areas as miters are created, as would appear were the fabric created through holding.

A: the alternate fold areas as they appear on the machine during knitting

the fabric after some casual steaming shown on the knit side, the pleating just about doubling on itself

B: with the added knit rows, the swatch before steaming

here is the purl side after some steaming showing the change in overlap

its knit side

both fabrics allowed to “hang”

The look may be varied considerably by changing the sequencing of the number of knit stitches, and the number of plain knit rows between the EON slipped or tucked ones. Electronics facilitate that, and with machines capable of accepting programming of the whole needle bed, there is even greater freedom. The sample above was knit loosely in acrylic, the holes would be less apparent in a tighter knit. If bothersome, they may in theory be “avoided” by factoring in “wrapping” if every row of the repeat is drawn rather than every other. With only 60 squares available on the mylar it is possible to go twice as wide and produce fabric width that may suffice for a skirt’s length. With software that allows for programming the width of the needle bed, possibilities expand even more. When going double wide with a repeat such as the one above, some problems result not so much in the inner fold slipped rows, but in the outer fold of the tuck ones. As in nearly any knitting when needles have 2 side by side loops resting on them, these loops will create a float/ ladder, so the tuck rows will essentially behave like the slipped ones as the double loops are dropped rather than anchored on the subsequent pass of the carriage. In the configuration here if the double-wide button is used, one way to solve that issue would be to bring consistently the same of either of the 2 non selected needles out to D or E position (still faster than hand-selecting repeats by hand for holding), resulting in knit stitches and single tuck loops.

5/18/18  Working out the kinks: the same repeat is redrawn to factor in the reduction of the eyelets at edges of slipped areas: a partial graph showing “automatic wrap”. However, the “wrap” will fail with color reverse, and that first row would have only one stitch knitting. If double wide is used, then the wrap will happen around 2 needles, creating yet another “float”. 

its color reverse 

Plain knit rows (blanks on mylar) may be added at the bottom or top of the repeat to change pleat depth. Here is an amended repeat, discarding the “automatic wrap” idea, first prior to color reverse, then after

 
Those all blank rows would result in no stitches being knit, essentially a 2 row “free” carriage pass, and are removed. The repeat is  now also capable of being knit twice as wide if desired, adding the hand technique on tuck folds

Below the edited repeat begins design using one pixel per stitch and per row. “Automatic” wrap happens as one row is knit, while the next preselects an extra needle. The yarn travels under the extra needle prior to its being knit on the next pass from the opposite side. There will still be an eyelet, a bit larger than hand wrapping, but smaller than without any wrap at all. The EON pixels in the center of the shape are tucked. For that to happen the knit carriage needs to have the cam buttons switched from slip to tuck, and back to slip after the 2-row sequence. The fabric folds out to the knit side. The EON (every other needle) set at the top of the shape is knit in the slip stitch setting, resulting in the fabric folding into the purl side. The process could be reversed to have each fold to the opposite side of the knit.

The carriage needs to be on the opposite side of increases in the number of stitches worked. Here the first preselection row should be from left to right, with carriage traveling to and from the right for the subsequent pairs of rows. 

More variations: the slip stitch configuration below is changed to the same EON selection for 2 rows. The same could be done for the tuck fold. For this repeat, the first preselection row should be from right to the left, making that the side the knit carriage returns to after each pair of rows knit. The sample was knit in a too thin acrylic. Both eyelets and their “wraps are visible, as well as the direction of folds.

These are pleats in a skirt made eons ago, that have retained the fold over the course of many a year 

An added alternative for fold lines: create outside crease transferring EON to make a row of eyelets, create inside crease by knitting the desired row double row with 2 strands of garment yarn. “Automatic” repeat must be adjusted accordingly, whether by redrawing or punching, or using cam settings for the correct stitch formation.

Seaming should be planned on the inner fold of the fabric, depending on whether the purl or knit side is used as the “public” side, the least visible join being one that is grafted ie with Kitchener stitch.

The same principles may be applied to punchcard knitting but because the repeat size is 24 stitches the resulting fabric is one suitable for portions of garments/ accessories, in details such as ruffles and edgings. Punched holes would match the black squares in the color reversed repeats.  “Air” knitting helps determine the exact needle location required, and to decide which side the first row must be selected prior to actual knitting. The image below shows the start of only one possible partial repeat based on the last one above

I usually sample my repeats by using hand selection and holding before committing to drawing on mylar or punching holes. An even number of rows is required in this technique.

Slip/ tuck stitch experiments

These scarves were designed using the same method as described for mazes and mosaics, they are knit in rayon chenille, fringes are composed of  i_cords applied to cast on and bound off edges. The smaller shape/repeat allows for more control over fabric width while retaining full repeats

blue ovals: 11X58 inches excluding 3 inch fringe

BW: 9.5 X 58 inches excluding 4.5 inch fringe

time to stop playing and get back to winter inventory production!

hot off the presses 11/10: tuck stitch 11 X 60 rayon chenille

an unplanned “mutation”

scarves measure and average 9(+) inches in width, 60(+) in length after blocking

11/17/2012: some colorways

.

Mosaics and mazes: machine knits_ from design to pattern

Maze patterns have long vertical and horizontal lines broken by regular gaps and the pattern lines change course from the vertical to horizontal, and vice versa. Maze cards can be identified by completely punched sections, some alternating with every other square marked for two rows, usually geometrically shaped. Areas of stocking stitch produce horizontal colored stripes, and alternating pattern stitches that slip or tuck cause the vertical stripes, which are sometimes pulled nearly diagonal by the influence of tuck or slip. The fabric will be unbalanced because the number of needles slipping or tucking will not be the same on every row. Odd rows form 2 color horizontal stripes, even rows vertical stripes, with color changes occurring every 2 rows.

Mosaics have a brick arrangement (tessellae), with clear perimeters and cores, and stepped diagonals (frets) that are partially formed bricks, their positive and negative spaces are created by the use of contrasting colors. The stripe sequence is not as obvious. The punchcard looks even less like the original design.

In single bed work, the reverse of the fabric will show the original design in the texture of its slip or tuck stitches. There usually will be no floats longer than one or two stitches.

The knit side may look like a fair isle but the back lacks the usual long floats, hence the name “float-less fair isle”

The row gauge is compressed. Tuck fabrics are short and fat, slip ones tend to be short and thin. Some patterns elongate in washing. The tension used is usually one number higher or more than that used for stocking stitch for slip patterns to reduce their narrowing, tuck patterns may also have to be adjusted to suit. Smooth yarns in contrasting colors are the easiest to establish and test the pattern, then the choices can be far more personal.

Designing your own: traditional “rules”

  1. if scale matters consider that the height of 2 rows may equal the width of one stitch
  2. start small, let each square on your graph whether on graph paper, in a design program or spreadsheet/vector program cell equal one stitch, each line on the graph represents 2 rows of knitting, when knitting the pattern double-length specific to KM may be used. The unfilled squares represent the lighter color/color1, the colored squares represent the dark/color2
  3. no more than one stitch to tuck, two to slip at a time
  4. row 1 and all odd-numbered rows (most stitches knit) can have any number of squares marked, the slipped (tuck, or slip/part tuck in alternating directions) are represented by blank grids (no more than 2 side by side for slip, single for tuck), they are generally knit in the lighter color/color1
  5. even-numbered rows must have single squares marked, they are generally knit in the darker color/color 2, there should be no more than 2 “light squares”/ unpunched holes side by side, the slipped (tuck or part/slip tuck in alternating directions) are represented by marked grids
  6. vertical lines must begin and end on odd-numbered rows
  7. vertical lines must always consist of an odd number of rows in total
  8. the finished design must be an even number of rows to allow for traveling back and forth to the color changer for picking up and carrying the subsequent color
  9. if the design is not to be elongated check to see that every light square to be worked in the dark color is present in the row below, that every dark square in the row to be worked in the light color is also present in the row below

Susanna’s chapter on mosaics has information on fabrics where “rules” get broken. Changing the order of the colors or introducing a third color may yield pleasant surprises. Knitting is started on a non-patterning row with first-row selection toward the color changer in Japanese machines. If you have a machine that preselects needles: color must always change when the needle selection changes. Four movements of the carriage are required to produce two rows of knitting.

One approach with a design that breaks rules:

masking alternate rows and “separating them”: odd rows knitting in color 1

dark squares get punched out/ drawn, light ones tuck or slip depending on cam settings

color 2 knitting even rows:

light squares are punched out/drawn and will knit, dark squares ones tuck or slip depending on cam settings

colored areas below are those to be punched overall

I used Excel to eliminate yellow fill on odd rows, darker fill on even. Many articles on this subject date back to graph paper, pencil, and eraser days. Quick color fills including empty make the process quicker with software. Still finding the image above confusing, it may be easier to decide what to draw on the card/mylar if all areas to be punched are dark, blank squares can then be more easily identified and marked, punching everything else or coloring them in and using color reverse if your machine has that ability. In the image below the lighter color is replaced by a darker one

the resulting card, which needs to be elongated X2

The swatches were knit using both the slip and tuck settings (also breaking the usual rule). Some of the tuck rows have a bit of color scrambling likely due to the amount of side by side tuck loops in the repeat not knitting off properly in those spots

slip stitch front

the back

tuck front shows the repeating trouble spots

tuck back

point grids for developing designs are of 2 types

in turn, the pattern may be drawn over them

staggered units may require some cleanup and “erasing”, as represented by pink squares

when the shape is what one desires, color separation follows as for the design at beginning of the post

Susanna Lewis at one time did publish a technique that could be entered in the E6000 that essentially did the separation; wincrea does not presently download techniques, there are other programs that can, and/or a combination of card reader sheet and computer download may be used, but that is for another day.

Mosaics and mazes from “FI” “universal” patterns

Many punchcards that obey the usual restrictions for tuck, in particular, may be used to create “random” mazes and mosaics, with color changes happening every two rows. Test swatches will show differences in surface texture, patterning, width, and height of the knit. It is useful to use clearly contrasting colors to study how the structure of the fabric is affected by different techniques. This test series explores the quality of the stitches created, along with using different knit carriage cam settings, although this repeat does not produce designs typical of either mosaics or mazes.
The swatches were knit during a class demo, for easy visibility, not as studies for finished garments or accessories. They were produced on Brother punchcard KM, using a single bed color changer. Electronic KMs advance a row for each pass of the carriage regardless of its beginning position/side. As noted in later blog post shares, such fabrics are produced more quickly and easily if an electronic is available with  2 compatible carriages for use.
Yet another single color variation, missing here, would be to use opposing buttons for tuck/slip.
The first preselection row is toward the color changer.  The FI pattern front with a bit of bleed-through where floats were hung up on the purl side

tuck 1 color slip 1 color  tuck 2 colors slip 2 colors  last but not least, slip stitch adding a third color front, still changing colors every 2 rows

This is a factory punchcard design, a smaller repeat. The repeat enlarged and punched for use in the pictured samples

Mosaics and Mazes: knitting on the machines

Two great books on the subjects by Kathleen Kinder:

another article/printable PDF resource by Susan Guagliumi (this link no longer works, subscription and login are presently required on her site for access to her free pubs)

Susanna Lewis:  “A machine knitter’s guide to creating fabrics” pp. 71-78, 1986. “Designing your mosaics” and “Figurative designs in machine knitting” were published in To and Fro Magazine, and presented at Passap University app. 1992

Denise Musk: “The technique of Slipstitch” pp. 36-46 1989

Barbara Walker Mosaic knitting companion CD by Morgan Hicks 381 motifs “charted and converted for your electronic knitting machine or crochet, .pat or .pcx format”.

A DVD stitch compendium salute to Barbara and an interview with her. Most reviews online describe as it being most suitable for beginning knitters. For a treasure of her treasury patterns including some cable stitches, one may visit The Walker Treasury Project. The group’s photostream may be found on Flickr.