I have decided to work a blackwork sampler to become more familiar with building stitches and to see how some stitches work at a fairly small scale. I’m using the same 22 count hardanger I used for the thread sampler, with one square (two threads) of the hardanger per stitch.
I did a very brief thread trial (these threads will eventually be removed but I’m leaving them there for the time being in case I want to change threads for emphasis and/or outlines). From left: Gütermann S303 (30 tex / 35 wt); Signature 40 (25 tex / 40 wt); Wondefil Konfetti (20 tex / 50 wt); Gütermann natural cotton (30 tex / 35 wt); Wonderfil Efina (17 tex / 60 wt). As previously noted, tex and weight correlate to density, not diameter.
As you can see, the Wonderfil Efina is distinctly finer than the other threads, but isn’t as slippery as the Silco (which was horrible to work with). I actually bought it at the same time as the others but promptly mislaid it so it didn’t appear in the thread sampler. I’m glad I found it again, as I think the finer thread will allow more pattern details to be visible at this small scale.
The design is my own: a fairly generic fan (or shell) in art deco style. I drew the design on paper, pinned it to the fabric and tacked through the lines. Unfortunately I didn’t block the fabric beforehand. I made sure to align the design with the vertical grain, but the horizontal grain is slightly off-true — I tugged it straight when I pinned it to the design but there was some slippage, and you can see that the horizontal lines are not on grain. I’ll correct this on the fly for the base and the centre blade of the fan, but the rest of it can remain as is. I’ll block it once it’s all done (memo to self: organise some sort of base for blocking embroidery).
I’ve made a small start on the central blade, and I’m noticing the same tendency to slide under the hardanger threads exhibited by the 100-wt Kimono and 60-wt Silco. I guess I should have expected it when I opted for the finer thread, but it’s still annoying. I shall have to make the effort and seek out even-weave linen in cream or white for any future work at this scale. I don’t expect the same issue when I start working at doll scale because that will be on plain-weave batiste or voile and I’ll be working over two threads at least.
I compared the four non-perle cotton threads that I used in the blackwork thread sampler. As you can see, weight (or tex) does not correlate well with diameter. The most slender cotton thread in the photo is the Gütermann natural cotton, which is technically the second heaviest of the four. The thickest is the Signature 40 which is technically the second lightest. Although we can see that the Konfetti (20 tex) is very slightly finer than the Silco (40 tex), I don’t think anyone would claim that the Konfetti is half the diameter even if it is half the weight. Edit: having just taken the time to do the maths, two threads of the same density would have weights in proportion to their cross-sectional areas, not their diameters. If the area were halved, the diameter of the thinner thread would be √2 of the thicker, or roughly 70%, so maybe they aren’t so far off after all. The Gütermann and the Signature are still anomalous, though.
Unless we can get a direct measure of diameter there will be no way to assess embroidery threads without buying a sample reel and trying it out on fabric.
One of the historical outfits that I would like to make for a doll is a sixteenth century gown with blackwork on the sleeves. I know this will be difficult to carry out at doll scale, but I’m going to do my best.
The last month or so has been devoted to trying out various threads and stitches in preparation for a blackwork sampler. I bought a half-metre of 22-count hardanger (I wanted evenweave linen but the shop only had it in very strong colours, not white or cream) and cut two rectangles approximately 25 x 20 cm so that I can put them into plastic pockets for ease of reference. The first one, shown here, is more of a thread sampler than a stitch sampler. For each thread I started out with a very basic diagonal / cross / asterisk stitch, and then two designs, one using only right angles (in the early sixteenth century style) and the second using diagonals as well (late sixteenth century style). For most of them I also included a degree of shading, with mixed success.
Unlike in other forms of embroidery, there are no set stitches for blackwork. Designs are built up using a combination of running stitch, backstitch and cross-stitch and can be adapted to simulate shading. I referred to several different books for patterns, some of which were tweaked for this sampler.
I’m using a Q-snap frame to hold everything in place, and I used a 5x magnifying lamp to help with the smaller stitches.
I tried out four perle cottons, four cotton sewing threads and two silk threads. They are discussed in order of weight (with one exception because I made a mistake). For ascertaining weight and tex where those units weren’t given on the reel I used the Wikipedia conversion table. Note that Nm is equivalent to weight according to the page on yarns (that is, weight = 1000/tex and Nm = 1000/tex so Nm = weight).
#3 perle cotton (Sue Spargo Eleganza) = 333 tex / 3333 dtex
This size is way too thick for the fabric, even using four pairs of thread per stitch. It has a knobbly feel to it, like rope, and a very high friction while pulling it through the holes. It would be interesting to use that rope-like effect in a design but it would need to be on something very coarse, like hessian or a low-count Aida.
#5 perle cotton (Sue Spargo Eleganza) = 200 tex / 2000 dtex
Better than #3, but still too thick for the fabric, unless an almost-solid effect was desired. The thread handled well with little tangling.
#8 perle cotton (DMC) = 165 tex / 1650 dtex
Still a little thick, but not too bad when using four pairs of thread per stitch. Handled fairly well.
#12 perle cotton (DMC) = 83 tex / 830 dtex
This was definitely the right weight of perle cotton for this fabric if using four pairs per stitch, though it’s still a little on the thick side — best for outlines rather than fillers. Relatively smooth, handled well.
#35 Gütermann natural cotton sewing thread = 30 tex / 300 dtex
Trying to ascertain the weight of this thread was difficult. The Gütermann website lists it as 300 dtex which converts to 33 weight, but the reels are marked C Ne 50, which would equate to 85 wt or 11 tex using the conversion table in wikipedia. Frankly, neither of them looks right to me. There is almost no difference in the thread diameter between this and the #50 Wonderfil. It was a little tangly when sewing, and I know the thread doesn’t have a lot of strength in it, so I wouldn’t use it in any area that might get stretched.
#40 Signature 40 = 25 tex / 250 dtex
Oh, this is a lovely thread. Good size, an excellent level of friction, not much tangling. It’s the best cotton thread for this fabric and I’ll almost certainly use this when I’m sewing a blackwork border on a human-size shirt or shift. It’s a bit on the heavy side for doll clothes but I’ll use it if I can’t get the fine silk thread to work on batiste.
#50 Wonderfil Konfetti = 20 tex / 200 tex
Nice thread, good friction (though not as good as Signature), not much tangling. There is nowhere to anchor the loose thread end, though, so you either cut into the spool or use a net cover.
#25 Wonderfil Silco = 40 tex / 400 dtex (out of order because I thought it was #60 wt)
This was not a fun thread to use. It’s a sateen finish lint-free cotton, very slippery and prone to tangling. It was difficult to apply any sort of tension as the threads kept slipping under the hardanger threads. I finished the design but I will never use this for hand sewing again. (To be fair to the thread, it wasn’t designed for hand sewing). The store listed this as a 35 weight (28 tex) but the reel says 40 tex (25 wt).
I also tried out two silk threads, as silk was a commonly used embroidery thread in the Elizabethan era.
#33 wt S303 Gütermann = 30 tex / 300 dtex
This is a beautiful thread to use, just the right amount of friction, and a very good size for this fabric. I’ll definitely use this again for costume embroidery — not on batiste, as it would be too heavy, but on a medium-fine cotton or fine linen.
#100 wt Superior Threads Kimono silk = 10 tex / 100 dtex
I was very disappointed with the way this thread behaved on hardanger — it was very slippery and kept sliding under the threads of the fabric, especially when turning at the ends of the rows. I ended up having to anchor the threads at each end of each row to keep the stitches straight, and because I had to keep the tension very loose I ended up with multiple loops and tangles on the back that required repeated bouts of unpicking. However, this thread is not designed for embroidery, particularly not cross-stich or blackwork, and when used for appliqué or fine piecing it’s absolutely fantastic. I might try it on batiste and see how it fares with a finer and tighter weave.
To summarise: best threads for blackwork on 22-count hardanger are Signature 40 cotton and Gütermann S303 silk, with DMC #12 perle for outlines.
1. Weight does not correlate to diameter.
2. Slippery threads are not suitable for hardanger fabric.
1. An actual stitch sampler on the other cut piece of 22-count hardanger
2. A tryout using Kimono silk on batiste.