Experiments with Shadow Weave

 Reno Fiber Guild and friends has been involved in a study group about shadow weave.  I decided to keep my experiment for the group to an 8 shaft draft that is available on the 1000+ Pattern book by Marian Powell.  All of the towels were woven using the 8 shaft profiles from  8-28 on page 235 of the book.  The profile shown is 8-28-1 and is tromp as writ.  In the book there are 12 different treadlings for this design and I used quite a few and incorporated another treadling from another pattern as well.

Since I have enough shafts, I added a basketweave selvedge for simplicity in weaving with two shuttles. After converting the profile to a draft, a portion of the threading is as follows.  The basketweave selvedge threading isn't shown in this draft.

The warp was alternating dark green and pale gray.  Each towel had two color wefts that were different from the warp. 

It was difficult for me to get good photos of the towels that follow, but I hope you can see that many possibilities were available with one threading.  I didn't even come near to exhausting the different patterns.    Also note that since these are actual pieces of cloth, there are some distortions in the photos of the pattern because the cloth was not absolutely flat.    Click on each photo to get an enlarged view.

It would have been fun to see what these patterns would look like in black and white, but it was also fun to play with all sorts of colors to see what I could get by with.  The more value contrast, the better but I didn't find any weft colors that I couldn't make work as long as the value was much different.  Bright yellows and chartreuse were both favorites, along with some very bright orange.  Quite a fun project.

Two block twills with multiple tie up opportunities

I'm a big fan of using a different tie up to alter the look of a design on my current warp.  I'm locked into my threading because I don't ever rethread for another design, but what I can do is change the tie up and the treadling to create a whole new look.

 

I started my experiments with two block twills after watching an episode of JST's School of weaving.  I suspect this episode was all about changing a profile draft into a two block twill and then weaving it on 8 shafts.  I used some of Jane's tie up suggestions and then looked for more.  I came up with many possibilities and will someday soon, weave a whole bunch of variations for towels.

 

In the meantime, since I haven't posted in a very long time, I'm giving readers of this blog some of my ideas.

 

Click on each photo for a larger view of the tie ups. 

 

 

 

  7/25/2024 I'm updating this post because something else occurred to me while talking about block twills at a weaver's meeting.  You don't have to abide by the four quadrant tie up approach I suggest in the drafts above.  Why not use an 8 shaft twill tie up?  Weave tromp as writ or change the treadling to a point.  For that matter, try other treadling sequences to see if any work!

This approach means you can put on a very long warp and weave many different designs - or if you don't mind changing tie ups during a piece, you could create a spectacular textile using several tie ups! 

 

 

Designing with Blocks Using Mathematical Sequences

Using mathematics to design weaving profiles or sequences is a theme that I return to over and over.  I have relied on writings about this subject because my mathematical knowledge is relatively rudimentary.  I was very lucky to have found this document in the Ralph Griswold collection of weaving documents stored at the University of Arizona website.  "Algebraic Expression in Handwoven Textiles" by Ada K. Dietz  

This is my journey into designing a profile and weaving drafts based on the mathematical expression of the cube of a binomial  (a + b) ³  My interpretation of this expression is aaa + aabaabaab + abbabbabb + bbb. I regrouped this into aaaaa b aa b aa b a bb a bb a bbbbb  Below is the two block profile created from my interpretation.

 

The next steps are to convert this into weaving draft using block substitution. With four thread blocks, the draft uses 8 shafts. With 5 thread blocks, the draft uses 10 shafts - and so forth. The most common substitutions are with 3/1 and 1/3 twill blocks, but I’ve found that more interesting designs can be had
when you think outside the box.  The following draft interprets the profile using 3/1 and 1/3 twills. But  this tie up could be altered in one or two quadrants to a broken, 4 thread twill. What fun that might be.
 

 The block conversion below uses a 3/2 twill and an altered 5 thread twill, (which doesn’t have a regular twill line). See the little squares in the design where the altered twill is weaving in block B

This conversion uses a regular 5 thread satin and an altered satin tie up

 

The following draft uses two altered 5 thread twill blocks. I discovered this option in Handweaving.net. To me, the lines look wavey. The surprise is that my weaving software (WeavePoint) analyzed the threading and found it could be woven on 8 shafts.  In checking further, I also found that Handweaving.net will compact this draft to 8 shafts as well. How great is that!

Below Block A uses an altered satin tie up and Block B uses a 5 thread altered twill tie up. Busy, but the profile is still visible.

The next draft uses 5 thread satin blocks. The dark block is flipped horizontally and reversed so that dark now becomes light. The draft doesn’t have clean cut edges where the two blocks meet.

There are, of course, many more options to interpret this profile.  Feel free to experiment and enjoy the results!

Shadow Weave and Weaving Software

It's been an interesting few days spent exploring weaving software and how to convert a profile draft to a shadow weave draft.  I thought I would share what I learned in a post.

 The conventional ways to convert profiles into shadow weave involve having a profile draft with a tie up that lifts 1/2 of the shafts such as a 2/2 twill or a 4/4 twill tie up.  Most weaving software programs don't give you a block substitution for conversion, but I found three that do.  WeavePoint, Proweave & Windows Weaveit Pro.  

It all started with  a profile draft on Handweaving.net  (#32432) that I liked and wanted to see how it would look in shadow weave. 

Because my weaving software of choice is WeavePoint, I tried translating it to shadow weave using the Translate Profile option.  WeavePoint gave me what I was looking for.  A four shaft, shadow weave design.  Great!

I had my draft and could have stopped there.  But, I didn't.  I wanted to see what other software programs would do and this is where it got interesting.  I found that Fiberworks didn't give me any options to convert profiles to shadow weave.  I did find that there were tutorials to help you do this - but with one caveat, that the tie up be in the format of 2/2, 4/4, 6/6 etc.   I didn't have this in my original profile and if I put a tie up like that into my profile the whole design would be different.  

Then, I tried Proweave.  I don't use this software very much, but it does have a lot of translation options to different weave structures.   I found a bunch of different options, so I took the one for four blocks and no incidentals.  

 

 

 

 

 

 

Great - but this is an 8 shaft draft and the block lengths have been expanded.  I could work on the draft to downsize the length of the blocks, but it didn't seem like a better choice than WeavePoint gave me.

Off to Windows Weaveit Pro.  This program gave me three options for converting my profile.  Atwater method, Powell method and Lang and Voolich method.  This was interesting because I hadn't heard of Lang and Voolich before - something new to me.  My first tries were with Atwater and Powell.  Both of the translations gave me a draft - but not for the original profile.  Instead, the software had evidently changed the tie up to a 2/2 twill because when I changed my profile tie up and looked at the pattern, the translated pattern looked identical to the thread by thread draft.  Next I tried the Lang and Voolich option.  Yeah - it gave me a correct draft, but again, one for 8 shafts.

I could have stopped there, but I really wondered why WeavePoint would give me a 4 shaft draft and the other two programs gave me an 8 shaft draft.

As I looked at the WeavePoint version, I realized that actually the profile was three blocks and Weave Point had analyzed this before it translated my profile.  And, where there was a transition from Block A to Block B, there was a double thread inserted so that the color sequence of dark and light wouldn't be interrupted (examine the first draft in this post).

So, I went back to the three programs to see how they would deal with my new, three block profile.  WeavePoint was happy.  It translated  exactly as it had the first time.  Proweave, translated it to a 6 shaft draft and Windows Weaveit Pro wouldn't give me an option to translate it at all.

 I think if you examine the profile and the solution that WeavePoint gives you, you could probably figure out other profiles.  I also tried to find Lang and Voolich's monograph, Parallel Shadow Weave  but couldn't find any current source for it.  Even Complex Weavers didn't have a copy in their library.  It would be interesting to see how their solution worked - even though it wasn't as elegant a solution as WeavePoint gave me.    

 Yeah for the WeavePoint programmer, Bjorn Myhre.

Anatomy of this year's Christmas towels

 Quite a few years ago, I read a paper called "Designing with Farey Fractions" by Ralph Griswold.  Ralph was a computer scientist with a penchant for weaving drafts and weaving design.   I did some designing using a Farey sequence as the threading for an 8 shaft draft and wrote about the process here in this blog. https://bannermountaintextiles.blogspot.com/2013/07/farey-fraction-towels.html    The results were interesting (to me) but recently I began to think about other things that might be done with this Farey sequence used as a profile draft. 

I digress a bit here to tell you that I don't understand the mathematics involved.  Yes, I've looked at the papers and most of it is just out of my league.  However, using mathematical sequences in weaving design is very much something that I understand and can use.

I started thinking about how I might weave green and red towels for Christmas sales without the problems of green and red crossing one another and blending into brown.  The solution appeared to be to use blocks of weave structure so that the colors of red and green would appear more or less solid against one another.  Which comes back to using a profile draft and translating it to satin and block twills to weave my towels.    I selected the Farey sequence above for my threading profile and converted it into a series of drafts using 5 thread satins and 5 thread twill blocks. 8 blocks times 5 thread structures translates to 40 shafts.  How lucky for me that I have a loom that will weave these designs!

In preparing the ideas for my warp color, I used stripes of red and green which used portions of the profile that would give a different design for the two color stripes.  Red and green yarns were another problem.  I didn't have any of the specific value in my stash that I wanted for the warp, so I wound off 2800 yards for the green warp stripe and 2800 yards for the red warp stripe. I dyed each of the 2800 yards in the same pot so that I would get exactly the same colors - although, a variation in the warp color stripes could have been interesting too.

 1/5 and 5/1 twill blocks

 This towel was woven with 5 thread satin blocks making it a damask cloth.  And all without a Jacquard Loom!

Accidental Christmas trees appeared in this variation.  Again, the structure is 5 thread satin.

I was able to weave 11 different towels on this 12 yard warp and still had designs left that I could have woven.  This last towel was woven with a dark green weft which almost changed colors once it was planted on the lighter green and the red warp stripes.  I'm always amazed at the differences you can get from variations in weave design and weft colors and must say that I didn't tire of what kept emerging as I wove.

I guess the final question is whether or not I'm done with Farey Fraction sequence weaving.  Probably not.  I expect that the next warp will expand the sequence into a larger design - but who knows, it may take me years to come around to the next iteration in my fascination.

Postscript!

For more about Farey and his fractions, here is a Wikapedia article  and a photo of the Farey Fraction sequence up to 8.  Use the denominators for your own sequence work.

If you are interested in satin weaves and  their various counters, here are a couple of resources that help to explain how they are formed. The documents contain almost identical information but in somewhat different format.  If you eyes aren't glazed over when you finish with these documents, I don't know what it will take☺

http://www2.cs.arizona.edu/patterns/weaving/webdocs/mo/D/SatinCounters.pdf

https://www2.cs.arizona.edu/patterns/weaving/webdocs/gre_satn.pdf

Using WeavePoint to design an amalgamation draft

 I've been fascinated with amalgamation drafts since I first read Alice Schlein's Network Drafting: An Introduction.  Recently, Alice published a new monograph  called Amalgamation: Double your Dobby and I eagerly purchased a copy directly from her.

As I read the monograph, it appeared that only some weaving software programs were capable of copying the drawdown portion of the draft and pasting it into the threading.  I use WeavePoint for most of my design work, so I went to AVL to see if the programmer of WeavePoint, Bjorn Myhre could help.   At the same time, Bob Kruger from AVL sent the question to Jannie Taylor.  Between the two responses from Bjorn and Jannie, I was able to amalgamate to my heart's content!  So - thanks to both of them for making my software just a bit more valuable to me.

The following is my process.  I've also tried it with Fiberworks for PC and it works well there too.  Choose a threading that you want to amalgamate.  You will notice that I have selected a threading that can be duplicated, in other words, I haven't ended on shaft one so that it repeats without double threads.  Alice's method is just a bit different and she does put in threadings that begin and end on the same shaft.  (More about that later).

 

The next step will be to add a satin tie up and select "tromp as writ" to fill in the treadling.  In the example below the counter is 11.  (Satin counters are explained in Alice's monograph which I recommend buying if you are interested in doing this kind of design work.)  A plus to WeavePoint software is that you select Fill from your options bar - then select satin.  This feature will give you all the counters that will work with the number of shafts in your design.  Otherwise you will need to research satins for the number of shafts you are working with.

Then, change from tie up mode to liftplan mode.  This will give you a liftplan that you can paste into your threading.  So - copy the liftplan and paste it into the threading to the left of the original 

 

 

 

 

 

 

Change back to a tie up and treadling and do another "tromp as writ". 

 

Now you are ready to get rid of the satin tie up and experiment with twill tie ups.  Check the float length when you have one that pleases you.  Sometimes it can be hard to get something that has  floats that are 5 or less in length.  WeavePoint makes searching for a tie up easy too.  In the Fill option, select twill tie ups.  You will be able to set the float length for the tie up.  I usually start with 3, sometimes need to go down to 2 in order to find a tie up that won't result in too long floats in the drawdown.

 

 

You can use the draft as is now, or do another little fiddle bit.  I often copy the entire threading and then use "paste special" which is found in the Edit menu.  Click on "flip vertical" and then paste that threading to the left of the original threading.  This option gives you more designs.

 

 

The final bit of information that I gleaned while working with Alice's monograph and my WeavePoint design solution is that when you copy the original liftplan with the amalgamation in it, you can paste it into the threading by doing  a vertical flip (again use paste special in the Edit menu)  This allows you to start and end on the same shaft with your original threading.

 

3/18/23  This is a postscript to this post.  I recently found that Handweaving.net has a feature that automatically will make a series of amalgamation drafts online.  Watch the tutorial https://www.youtube.com/watch?v=P4cysOpJSKk   and see what you think.  You can always upload one of your own drafts to the site and manipulate it online as well.  Amazing tools that make our weaving design life so much better.

 

 

Warping with a Stationary Paddle

I have been wanting to learn how to warp with a paddle using multiple yarn sources for a very long time.  Recently, I found a Victorian Video from 1997 with Sallie Guy called "Warping and Loom Preparation with Sallie Guy". She has a very good visual instruction of the process using a stationary paddle. Unfortunately, this video is very hard to find.  If you find a VHS tape, it may be worth your while to convert it to a DVD.  I did find one copy on Amazon.

 
 I followed Sallie's instructions to a "T" or so I thought.  My first warp had terribly twisted warp threads and it was a real struggle to weave off.  So, I tried another and had the same result.  I went back to the video and listened to Sallie say several times that the hand position is very important when transferring the cross from your hand to the warping board.  I didn't understand what I was doing wrong but guessed that there was something in the video that wasn't easy to see (at least I couldn't see it) so I started more experiments.

I finally solved the answer to the hand position conundrum and am here to share my experiences with you. 

There are several ways to set up your yarns so that you can warp with a stationary paddle.  A very common method would be to have the yarn wound off onto bobbins and placed on a spool rack.  Frequently, I choose that method.  But for the demonstration I recently did for my local weaving guild, my husband and I put together a Rube Goldberg warping station. 

 

The pvc cone holder helped to keep the lighter weight cones from falling over.  The yarn from each cone was threaded through a 6 dent reed and I deliberately kept the reds and the oranges alternating with the other colors so I could see the colors in the cross more easily and make sure what I was doing wasn't twisting the warp threads.

 

The next step was to thread these same yarns through the stationary paddle, which is nothing more than a piece of plastic rigid heddle with a handle, mounted on a pole.  Again, the pole is piece of pipe flattened on the top and the paddle is clamped to it.

 

 

You thread  the rigid heddle piece - slot, hole, slot, hole - etc.  For my demo warp I used 6 threads.  You could easily warp 20 warp ends at once if you had bobbins and a spool rack. To get a color sequence that I could follow on the warping board and would show up my errors., I put all the red/orange threads in the slots and the others in the holes. When you are practicing and learning the technique, I would recommend using three similar colors and then three that can be anything else.  Once you look at the cross photos you will see why. 

 

After you have threaded your rigid heddle paddle, you tie them all to the first peg of your warping board.  Sallie's video goes into planning a route for the warp, winding a raddle cross, etc.  Since those details are easily picked up from other sources, I'm not going to discuss them here..

 

Before we get down the to nitty gritty, I need to make another comment.  I have plastic hooks mounted on the top of each end of my warping board.  These are plastic picture hanging hooks and when I'm picking up my cross, I keep the yarns from slipping off the warping board by anchoring them in the hooks while I'm manipulating the cross. The second from the last photo shows one on the left top corner of the loom.  Enlarge the photo to see it better.

 

Next step - pick up the cross at the stationary paddle. 1. Holding the threads with your right hand, raise them (see photo above) and put your left index finger in the space created.  2. lower the threads (see photo below) and put your thumb in the space created.  You have created your threading cross.  In these photos you will see that the orange colored threads are all in one half of the cross and the other colors are in other half.

 

Then, keeping tension on the warp with your right hand, slide your left hand towards the warping board.  Once you are close, wrap all the threads clockwise around the end peg.

 

The photo above shows the hands holding the cross - but it hasn't yet been seated on the warping board pegs.  This is the critical hand position with the thumb on the right.  Then, you tilt back your hand so that the threads that are closest to the warping board go over the top of the two cross pegs. 

This is the most awkward part of the process.  If you have done it right, the lighter threads will be on the top left and the red/oranges on the right top.  A color sequence similar to this is good to do while you are learning how to manipulate the cross at the warping board.  You can easily see if you are doing it correctly.  A warning.  The first times I did this, I didn't tilt my hand backwards and in the case of the photo above, I put the orange threads on the top of the left peg and the light threads on the top of the right peg.  In the video, Sallie calls this the downward pass and this is where the video failed me.  I couldn't see this hand position.

The photo above shows me coming back in the upward pass towards the cross.  I've completed making the cross at the stationary paddle in the same exact sequence as above.  I've unhooked my threads from those plastic hooks at the right side of the warping board and am approaching the cross pegs.

This cross transfer is easier than the downward pass.  The palm faces you, the thumb is on the left and you don't need to tilt your hand.  From here - rejoin your all your threads and go clockwise around the end peg. (as above).  Anchor the threads in the plastic hooks and start the whole process over again.

And, if you have been doing everything correctly, this is the way the cross will look.  All of the red/orange warps on one side of the top, all of the other colors on the other side and when you look at the bottom of the cross the colors are reversed.  Another thing you will notice at the beginning peg is a false cross.  Ignore it.  Once you have cut that end loop, the threads will be in a normal configuration again.

I recommend the Sallie Guy video highly.  There is nothing like being able to watch the process over several times before you attempt it.


Now, let's go on to the question - why in the world would you want to do this?  Well, if you have delved into the book by Marian Stubenitsky,  Weaving with Echo and Iris, you will want to warp four or eight colors at once.  This technique is a great solution because you will pick off your color sequence one by one at the cross making everything very straight forward.  You can warp four threads at once - yes, but I like this method better.

You may have lots of odds and ends that you want to use up.  Instead of constantly cutting and tying on new ends you just warp along with multiple colors.  Want a draft for this kind of warp?  See Alice Schlein's garbage warp draft. (I use the provocative word garbage - you may wish to call it a mixed warp)!  Maybe you have an interesting stripe sequence that requires frequent color changes, this can easily be achieved by warping the stripes all in one pass. 

Remember to calculate your sett and take that into consideration when you are making your raddle sections if you warp back to front.  If I am warping at 24 epi, I might use 6 threads per pass.  Then, I know that in one inch I will be making 2 downward and 2 upward passes.   If you warp from the front, and don't have a raddle cross, just keep track of the number of threads you have warped.

This post must naturally be a work in progress.  I suspect that there are details I have left out and may need to add later.  If so, I'll just update it.