Maleficent Tutu

Every year a few of my friends from my ballet school pick a day close to Halloween and dress up in dance appropriate costumes for class; we call it Balloween.

I was made redundant last year and decided to take some off. The problem is that when I have a lot of time on my hands is that when I have an insane idea like making a Maleficent inspired ballet bodice and romantic tutu, instead of writing it off completely I say “well, I’ve got the time.” This is never actually the case.

Work finishing the embellishments ran on into February 2016, but it took me six weeks (two for prep, four for construction) working full time to get the tutu ready for it’s intended outing.

I spent the first few weeks trying to pick apart the key elements of Maleficents costume to see what could be adapted. The most obvious and recognizable outfit is the christening dress with its high structured collar, however, I wasn’t entirely sure if I could pull off the detail and still have an outfit that I could dance in successfully. Certainly it wasn’t going to be a first time project.

Eventually I settled on the ombre velvet dress as the basis for the outfit and accepted that I was going to have to make some fairly significant changes.


Photo from Style Fritz.

Without a shop like Tutus and Textiles where I could buy a pattern and appropriate fabric and notions I doubt I would have had the guts to pull it off, so thanks to Suzanne for enabling me!


The hardest part of the design was working out what to do with the embellishments. I spent two weeks of prep and a good chunk of the toile build time trying to translate the obvious aspects of the film costume over to a ballet bodice; including looking for villain appropriate lace and trying to work out if I could do something with feathered epaulets.

There isn’t much (or any) literature covering the design and construction of ballet costumes; I resorted to scouring the internet for high quality images of tutu bodices to try and get a better idea of what was usual, and possible, so that I could emulate it.

Eventually I decided that I’d have to take a less literal approach; my big break came when I came across a photo of Margaery Tyrell’s wedding dress from Game of Thrones and found  Michele Carragher’s description of how the briars are made using Czech glass spikes for thorns. I thought that worked quite nicely with the wall of thorns in and how several of Maleficent’s head wraps are edged with teeth and claws.

Let the games begin

As I’d never made a boned bodice before, or worked with velvet, I decided to make a full mockup (which will be referred to from here on as the toile for ease of typing) so that I could work through any construction related snags. There was one party at the two week to D-Day mark that I could wear it to which would allow me to test it for functionality. That would give me just enough time to make any adjustments before I started on the final version.

I soon learned that cutting velvet is fun. Super fun. I was very glad that it was the top fabric so it didn’t matter as much if it was a bit out because if you so much as look at velvet it moves 6 inches.

One of the main problems that I found (and this was true for the final version as well) was that I’m not used to working with fabric this thick so I spent a lot of time getting stabbed by pins and needles, particularly as I speeded up. There’s a fair amount of blood on the inside of the bodice but thankfully I avoided bleeding on anything that would be visible.


Attaching boning casing to velvet seams and getting them straight is also fun. Even though I hand basted them all I still managed to ruin one by ripping it out to re-sew so many times that I trashed the edge.

I placed the bust seam casing too low, it didn’t give enough support to the top inch of the bodice and it deformed the line at the bottom which gave me all sorts of trouble when I got around to the piping.

The pattern for the romantic tutu called for 25 meters of net and tulle to be gathered in four layers. The gathering itself took an entire day and if I hadn’t had a dress form to work with this is where the project would have died.

By this point I was getting very close to the first deadline and did I mention that I’d never made or installed piping before? I didn’t have time to get hold of the correct foot for my sewing machine but I worked out that I could use the invisible zipper foot with much the same effect.

This was also the first time I’d sewn hooks and eyes as the main closure and I’d had a very hard time getting time to sit straight. I decided to try a heavy duty hook and eye tape instead, I tested it on the toile and it mostly worked so I went ahead with the plan for final version.


Ballet bodices fit with almost no ease, they need to cling to you as you move in all sorts of directions, and as I’d been finishing the bodice in a rush I hadn’t tried to put it on my dress form fully closed. When I did, I discovered that while the bust, waist and hip dimensions matched I still couldn’t close the bodice, it was out by a good 2-3”. I suspect this means a new dress form in my future.


Once more with feeling, at least I knew what I was getting in to cutting velvet this time around. I handled the boning placement in a way I’m much happier with. However, I did get a piping foot and I’m still not sure if the problem is me or it but I was much less happy with the results and with no time to fix them.

I adjusted the length of the tutu layers so rather than coming to a straight line at the hem they are graduated to mimic the ombre effect of the film costume. I like how it turned out, on even numbered days at least, but I think that if I was to do this again with more time I’d try for a dagged edge.

Moving on

The next day I unpicked 2/3rds of the piping and made/installed new using the invisible zipper foot again. There are some minor distortions in the top fabric as a result (I should have basted above the seam line before I did anything, you live and learn) but I’m much happier with the revised piping.

This also necessitated ripping out the hook and eye tape so it’s a good thing I wanted to replace that as well. I wish I could say that with a bit more practice it was easier to get the hooks and eyes straight but they were a bloody nightmare from start to finish. For the hooks I prick stitched through both layers of fabric to hold them on without being too visible. I’m quite pleased with the final effect but would like to never ever do it again on velvet.

Sparkle harder

I’d also had to unpick most of the crystals I’d put at the neckline, they were just a placeholder as even though I knew I needed something sparkly I hadn’t quite had the time to pull it off properly. True to form I spent several days looking at options before realising that I was never going to be able to get the variety of sew-on stones (both size and colour) that I needed and switching to beads. I ended up planning a 94 bead monstrosity in four different shades with varying types and sizes which went through several iterations on paper as well as some last minute changes while I was sewing.


One of the things I was able to see from the Great Ballet Photo Hunt of 2015 (which was then backed up by an acquaintance who apprenticed making ballet costumes in the usual way) is that most stone embellishments are sewn onto netting and then transferred to the costume. This is fine if you’re sewing for theatre but I’m planning to wear this so I ended up sewing the beads directly to the bodice.

I also wanted to add more of the briar and figured that it’d look nice if I put loops on each of the side seams as a continuation of the briar running down the front of the bodice. Most tutus don’t have decoration in that position as it causes problems with pas de deux work, but as this is just for me I was able to be a little more flexible.

I dyed the straps to match my skin tone with tea, I didn’t have enough elastic to do a proper patch test but I came pretty close anyway.

And there you have it! Innumerable hours (and bleeding fingers) later I have my very first tutu. I’m quite proud of what I pulled off for my first attempt and I’m planning a second.


Averaging one outfit a year

2014 was a terrible year in which I was unable to do any sewing due to stress and drama and projects and funerals, but I have returned.

I have another soft circuit project in the works, however, it was put on hold due to the Maleficent inspired ballet bodice and romantic tutu that I made for Balloween1. I ended up having to do some fairly significant work on the bodice afterwards as there were a number of aspects that I wasn’t happy with but I’m nearly finished and a proper write-up is in the works.

1) Some of my ballet friends pick a day close to Halloween and dress up in dance appropriate costumes for class, Halloween, ballet, you see where we’re going. This year our teacher played Halloween appropriate music and it was wonderful.

The Vault of Heaven

If ballet has taught me one thing, and sometimes it feels like it has only taught me one thing if you don’t count “how to spend all your money on leotards”, it’s that failure is an integral part of the learning curve.  Just over a year ago and freshly armed with this knowledge I decided to learn how to sew, partly because I have some very awkward proportions and thus finding clothes which fit me can require patience in the face of overwhelming odds, and partly because it’s a skill which has slipped out of general use and I find that interesting. On the understanding that I was unlikely to make anything I’d be seen dead in straight away I’ve been sewing basic pieces as toiles for about a year now before I finally made one up and thought that if I did it in real fabric I might admit to having made it.

Given my recent discovery of Adafruit’s wearable electronics platform it seemed logical at the time to combine the two. In retrospect I may have been a little ambitious, not that realizing this would have stopped me. In the end I think that it took me the same amount of time to create the skirt as it did to sew the electronics and neither were particularly quick.

The pattern that I’d decided on for the skirt was Hollyburn, it’s a deceptively simple piece and having worked the toile in a medium weight calico I knew I’d be able to use a fabric that could handle the weight of all the conductive thread. This point will become very important later on. In order to make the pattern work with the sparkle skirt project I was going to have to add an overlay, I spent a reasonable amount of time going through the My Messings Hollyburn sew-a-long trying to work out if I could (or should) use a similar technique to the one used to add a lining. Eventually after doing some sense checking with Kathryn I decided that I was going to have to attach the overlay at the pocket seams and zip as well as at the waistband in order to get the look I was after at the same time as disguising the conductive thread and softening the LEDs.  Given my penchant for science fiction it was always likely that I’d end up with a sports mesh, after working my way through Berwick street in search of the perfect fabrics I with with airtex along side a medium weight cotton poplin for the base. As this was the first time working with a mesh, or any sort of fabric with that much flexibility, my first chunk of project time was taken up by sewing a series of test swatches. Safety first.


Sewing the mesh into the seams along the pockets worked out rather well; I managed to keep the crisp line that I was hoping for without creating a bulge at the beginning of the airtex seam. I put the zipper in with my sewing machine but as I had to sew through the poplin only to avoid an ugly line on the airtex it was hard to sew as close to the line as I would have liked, so I ended up hand picking it as well to try to keep things neat. The waistband was also a fairly challenging due to the sheer bulk of the fabric in some areas but I managed to wrestle it into submission.


Having finished the poplin hem I moved onto my circuit diagram. I wanted to sew FLORA and the accelerometer onto the inside of the skirt so they wouldn’t either show through the mesh or snag on it. As you can see from my original diagram below (spot the sadly not deliberate mistakes) I’ve put FLORA underneath the pocket to protect it and me while the skirt is being worn.


I ended up moving the accelerometer closer to FLORA but as I still wanted to keep the positioning that I had I ended up having to jump the connecting line over the VBATT – thankfully with sewing this is much easier than it would be otherwise. I gave myself a 5 inch section of the skirt (below the line where the overlay was attached to the zipper) for the LEDs and created a random number generator in excel to work out the placement within that space at regular intervals.

But here is where I reach the most irritating part of my story. I’d originally planned to use 3-ply thread, and in fact I sewed the first 10 LEDs on by hand. Unfortunately at that point it became obvious that the thread was impacting badly on the drape of the dress, so I ripped out all of them and started again with 2-ply. It was not my finest moment. Thankfully 2-ply thread can be used in the bobbin of a regular sewing machine so this enabled me to sew the power and ground lines on different sides of the material, cutting down the likelihood of them touching and shorting out. It’s also easier to sew, and crucially, to knot than 3-ply although that still doesn’t make it easy to work with. Using single runs of 2-ply thread works for the sparkle skirt because the LEDs are only on in tiny bursts, but when I run them using the regular strand test they just don’t hold up because they’re not getting enough power. I’m going to have to do some serious thinking about how to manage my next project.

dsc_0119 2

Soft circuit, front

dsc_0124 2

Soft circuit, back




Twinkle twinkle

Even after moving to 2-ply, running meters of stainless steel thread through the fabric has noticeably sharpened the silhouette.

And this is the skirt in action.

All in all not bad for my first real project.