After that I sew together the two outside strips, matching the beginnings at seam allowance from the side edges.  The seam is straight but the pre-tube twists around during sewing.  When done sewing, the result is a (non-twisted) primary wide tube.  (The tip angle of parallelograms/triangles is visible at the left edge of the tube.)

First steps

I first sew strips into a pre-tube at desired offset angle.  This angle determines the tip of the triangles and parallelograms.  I may pre-trim the strips at that angle, or simply offset them.  I prefer offsetting rather than pre-cutting so I do not have to deal with bias (yet).

The how of tube piecing  

  by Irena Swanson

Related links

I discovered Rita Hutchens in August 2015.  She uses the same principle as I.  Her book Totally Tubular Quilts was published in 2003 by C&T Publishing,  her web page is at, and she has several YouTube videos

There is a related method called strip tubing, developed by Daniella Stout and Georgette Dell'Orco.  Check out at the Cozy Quilt store in San Diego, which has links to their books.  Strip tubing creates small-circumference strip tubes from which one cuts individual blocks that have to be handled further.

I next trim the construction precisely at the chosen angle.  (Now there is bias along the edge.)

Back to Home.

The narrow tubes get sewn into a secondary wide tube, with specific seam shift (or edge shift), depending on desired outcome.  If I want a quilt top of parallelograms, I would probably match points to points (so seam shift zero), but for a quilt top of rows of triangles, I make seam shift non-zero, as in the photo below. 

If I wanted a top of parallelograms, I could now make a cut perpendicular to the bottom edge reaching to the top edge, sew together the right and left edges, and I'd be done.

For a top of all-over triangles, I could now cut similarly as for the first cut, but with a rotary cutter, through all the other swaths to create rows of triangles with slanted ends.  Doing that would cause me to have to do something time-consuming to the slanted ends.  So instead I first do some slitting (cutting that does not reach the outer edges), do only one more seam, then extend the slits into full cuts, to get secondary narrow tubes, which then get cut across into rows of triangles with perpendicular sides.  Or better still, the secondary narrow tubes get sewn together into a ternary wide tube, which then gets cut with one perpendicular cut to make one top of all-over triangles.

The basic outline as above of moving from (pieced) strips to pre-tubes to wide tubes to narrow tubes can get iterated several more times, for many different outcomes.

Specific further steps depend on the project; see  Projects and my developing book.

The primary wide tube gets cut parallel to the straight tube edge into primary narrow tubes of specific widths.  The widths depend determine the slants of parallelograms, or if making rows of triangles,

they determine the other angles of triangles (tip angle was chosen at the beginning)

By design of seam/edge shift, there is at least double seam allowance through the centers of all parallelograms, so that after I cut the secondary wide tube through those swaths, there is enough room for seam allowance for triangles rows.  So, I cut through one of those swaths, one layer of the tube only, and the result is a parallelogram as below.