How to start working with block designs

Planning a summer lace scarf: Part 2

I settled on a block design for my summer lace scarf. What does that mean?

The word block is often used by weavers in several slightly different but related ways, and this can be a little confusing at first. I’m going to approach it from two directions, first visually and then technically.

Lace scarf with a pattern of plain weave blocks

A visual representation of blocks

The visual representation of blocks is summed up in the profile draft. The drawdown in a profile draft is like a view of your cloth from a distance. Imagine your weaving pinned to a wall on the other side of the room – you can’t see the detail of the threads interlacing, but you can see the overall ‘look’ of the pattern.

I’ve chosen lace as my example to work with here, because it is typically woven in a single colour – or in closely related shades of a colour – so we can more easily focus on the pattern as it is expressed in weave structure. In this case it is areas of plain weave contrasted with areas of lace (click to enlarge the photo to see this more clearly).

When we want to design a block pattern, we often start with a sketch of that overall look. Squared paper isn’t essential, but it is extremely helpful for the analysis we’ll do later.

Design ideas sketched on squared paper
Playing with design ideas on squared paper

The technical aspect of working with block designs

There are no limits on our imaginations when it comes to sketching out an idea. However, there are technical limits to block designs.

The essential property of a design block in your woven fabric is that you can manipulate it independently of other parts of your design.

You can do something different with each block that your design contains. What that ‘something’ is will depend on the structure you are working with. In the case of lace, we have a choice between ‘weave it as lace’ and ‘weave it as plain weave’.

You might want to look at my earlier posts on twill blocks or double weave blocks for more examples of what this means in practice.

For a weaver to be able to manipulate one part of the design separately from other parts comes down, as you might expect, to a question of shafts. The number of shafts we have available imposes the technical limit on our use of blocks.

Matching the visual and the technical

Ultimately, we will turn our profile draft into something we can weave by assigning shafts to the individual design blocks. How many shafts we need to assign will depend on the structure we’re using. This gives us some flexibility. A design which is ‘too expensive’ (i.e. uses too many shafts) in one structure may be ‘affordable’ in another. Essentially we need to act as negotiator for our own design process, going back and forth between the ‘ideal’ and the ‘possible’ until we are happy that we have the best result for our design.

So far we have a sketch of what we would like our cloth to look like. The next step is to create a profile draft out of our sketch. This will record how many different blocks our pattern needs.

A simple example of a block design

Let’s take the ‘window panes’ draft I have sketched above and approach it methodically.

Step 1: Isolate one vertical slice of the design and give it a label. Using letters of the alphabet is good as numbers might be confused with shafts. So we’ll call it A and colour it in for good measure.

Step 2: Now look at the other vertical slices. If you’ve used squared paper, it will come into its own right about now. Which of them are identical to the one labelled A? i.e. which have the same pattern of filled and empty squares?

Every vertical slice which has the same pattern gets the same label, in this case A. Label them, colour them in and move on.

Step 3: Repeat steps (1) and (2) for any vertical slice that isn’t already labelled. Give it a new label, B, and maybe give it another colour too.

Our draft is now completely coloured in, so we can stop. But if there were vertical slices still without labels then we’d keep repeating steps (1) and (2) until they were all identified.

What has this exercise achieved?

We have determined that there are two distinct blocks in our design. Areas of the sketch labelled A need to be manipulated independently of those labelled B in order to achieve this pattern in a woven fabric. Our profile draft records this by adding a ‘threading’ row which shows the pattern of A and B blocks.

Designs with more blocks

The windows design is a nice economical one. We only need two blocks, which opens up lots of structure options.

What about some of the other designs? The ‘alternating windows’ pattern, for instance, needs four independent blocks.

And my temperature charts for the Highland Summer scarves? The threading profile has seven rows, indicating that there are seven different blocks. We can see this if we analyse the drawdown for the East of Scotland profile. It is worth noting here that block A, the ‘background block’, is not shaded at all in the drawdown, but it is still a block!

Note also that my sketch here doesn’t show the widths of the blocks as they are realised in the final design. That isn’t essential for this analysis. This is the stage of the fruit salad where we identify that we need Apples, Bananas and Cantaloupe: the exact quantities of each are less important than naming what’s involved.

Negotiating with your block design

I used 16 shafts to weave my Highland Summer scarves in huck lace. In terms of shafts, huck lace might be called a ‘mid-price’ structure (don’t tell it I said that!)

One blog post is insufficient to give an exhaustive discussion of all the options, but we can quickly summarise a few examples, as long as you hold firm to my favourite motto: “it depends”.

Expensive structures: Twill, double weave both need at least four shafts per block. On 8 shafts, therefore, you can have two independent blocks.

Mid-price structures: Huck lace needs two shafts as a base, and then two shafts per block (in general). On 8 shafts, therefore, you can have three independent blocks.

Affordable structures: Bronson lace, summer and winter need two shafts as a base, and then one shaft per block. On 8 shafts, therefore, you can have six independent blocks.

Hmmm, six independent blocks on an 8-shaft loom, and my design calls for seven. This tells me that I will need to adjust my ideas.

Reducing the number of threading blocks

I’ve sketched out some options below.

  • I could limit my design to five months’ data instead of using six (left)
  • I could drop the ‘background’ block so that the squares in my pattern are connected rather than separated. Note that (at this stage) I haven’t ruled separating them horizontally (middle)
  • I could adjust the design to make two similar blocks the same (right)

Note that the profile draft for the North of Scotland has this feature already. There are only five distinct blocks here (can you verify that claim?) so I can simply choose to focus on this design and drop the others. I kept it as a seven block threading so that I could weave it on the same warp as the other two designs.

It is worth noting here that it may be possible to keep all seven blocks if I am careful in the way I implement my chosen structure, but that’s something I’ll look at next time.

I could, of course, take things further. The design process isn’t a linear journey to a single ‘correct’ destination, it is experimental and playful. I could take four blocks (let’s say three pattern blocks plus a background block) and rearrange them. I can repeat them, mix them up, and I will still have a four block design.

Liftplans and weaving blocks

We’ve done a lot of work already, but there is another essential aspect of the design to consider. How are we actually going to make the pattern of blocks appear?

The threading you construct determines which parts of your warp will work together – making pattern at the same time – and which parts will work independently. However, it is the sequence of lifts that will actually deliver that pattern.

On my ‘window panes’ sketch I now need to look at the rows rather than the columns and ask myself the same question as before: how many distinct and different blocks do I have? The process is fundamentally the same as outlined above, but we are looking at horizontal rather than vertical slices.

Once you get used to this process you will be able to leap right over this step to the tie-up below, but for now I am going to give these blocks their own labels. To avoid confusion with our threading blocks, I have started at the other end of the alphabet and labelled these Z and Y.

Identifying the lifting blocks needed to accomplish the ‘window panes’ design

The profile tie-up

To add the lifting blocks to our profile draft, we need to determine which of the threading blocks are active in each lifting block. ‘Active’ will mean different things depending on the structure you are using, but we will come back to that next time. For now, we’ll say that the active blocks are the ones coloured in, while the background is the area left blank.

  • In lifting block Z, both A and B blocks are coloured in.
  • In lifting block Y, only block A is coloured in, while block B is blank.

This information becomes my profile tie-up, as shown below, and my sequence of lifting blocks is indicated in the treadling.

Complete profile draft for the ‘window panes’ design

My sketch can now be entered into my weaving software, which is a good place to start to playing with the scale of the blocks. How much A do I want and how much B?

Constraints on weaving blocks

The good news for table loom weavers is that there are no constraints for you: go forth and weave whatever you wish!

The same applies if you are weaving on a computer dobby loom. A mechanical dobby loom, such as my 16-shaft Megado, is somewhat constrained; I’ll look at this a little bit more in my next post.

However, if you are weaving on a loom with treadles then you will need to examine your lifting blocks as carefully as your threading blocks and take the ‘price’ of your structure into consideration here as well.

It’s slightly harder to generalise about the number of treadles needed for a given structure, but the price table is similar to that for shafts.

Expensive structures: Twill needs four treadles per lifting block. Double weave is best accomplished by splitting the tie-up for the two layers (as shown here). It therefore needs four treadles for the plain weave base plus two treadles for each lifting block.

Mid-price structures: Huck lace needs two treadles for the plain weave base and then two treadles per lifting block (in general). Summer and winter* can be accomplished with four treadles for the plain weave base and tie-downs and then one treadle per lifting block.

Affordable structures: Bronson lace needs two treadles for the plain weave base, and then one treadle per lifting block.

You have to love Bronson lace, don’t you?

* There are many different ways to treadle summer and winter: this is what I have tended to do to maximise weaving potential on my countermarche loom. Have a look at this example to see what I mean.

Revisiting the negotiation with your block design

Guess what? We can negotiate with our lifting blocks in the same way as with our threading blocks and apply the same tricks:

  • drop one or more blocks completely
  • adjust blocks to be the same as others
  • take a section of our design with a small number of blocks and play around with it

If working with block designs now seems awfully long-winded…

I have broken this process into two distinct halves, considering the threading and the lifting blocks separately. This makes it easier for me to describe what is going on, and I hope it also makes it easier for you to follow. However, we don’t typically work in such clearly defined instalments, and I don’t intend to suggest that you should.

As you get used to working with blocks, you will develop a feel for the number of blocks you are creating and your own preferences for managing their number. But until that becomes second nature to you, it is worth having a process you can turn to for checking and clarifying. On the one hand, it is good to let your imagination run wild. On the other hand, it can be frustrating to spend a long time developing an idea that you later realise you don’t have the tools to weave.

Oh, and I have focused on structure in this discussion, but don’t forget you have colour too! That’s a whole other topic, which is beyond the scope of this series, but if you are feeling constrained then you might want to check out this example of a two-block twill that looks like much more.

Treadle reduction:
is this the answer?

Now it may be possible to reduce the number of treadles that you need for your design, if multi-pedal treadling is an option for you on your loom. In fact, multi-pedal treadling is what I am suggesting anyway for double weave and summer and winter.

However, not all multi-pedal treadlings are equal! In the examples above, we are working with the structure to split it into logical units for weaving. This is good practice and the underlying logic makes it reasonably easy to treadle accurately.

If you have done this and you still don’t have enough treadles for your design, it may well be possible to make further reductions algorithmically, but at the cost of an artificially complex treadling sequence.

This is a perfectly valid option, but I’d suggest that it should be the option of last resort. Before you decide it is inevitable, take another look at your design and see whether you can be more economical there.

What’s next?

We’ve looked in detail at how to create profile draft, and negotiated our way to the sweet spot between the constraints of our loom and the ambition of our design.

The next step is to turn it into a threading draft which we can take to the loom. That will be the topic of my next post.

Happy weaving!

First posted on © Cally Booker