Priscilla: the Burning Bolero

Deborah and her two friends came to KOBA as they were hunting for garments to wear for their first excursion to Burning Man festival in South Africa, in the middle of a desert. At the Burning Man festival, visitors are known to dress up extraordinary to celebrate the “burning man”.
They liked the idea of wearing garments with lights as it will show well at night, especially desert night will be very dark otherwise. Few weeks after their first visit to KOBA, Deborah came back again with some ideas of her costume. She has gathered some images on Pinterest to share her inspirations/ preferences with us. This was very helpful for me to understand what kind of style she likes. As I looked through the images, it reminded me of the work of Rachel Freire. I showed her the website of Rachel, and she liked the style very much. So we decided to base the design on Rachel’s boleros, and place optic fiber on the frill to add light to it.

The challenge of this garment is that at the festival it will be very hot during the day (over 40 degrees) and very cold at night (0 degrees). So the design should be adaptable to wear directly on body and on thicker jacket.
Also, the burning man is held in a middle of desert and the sand will get in everywhere and may collude metal parts. The electronics needs be shielded made robust to stand the sand dust.

The above sketch is the first sketch of the bolero. Let’s see if we can get close to it.

Development

Toile test

test wearing

Black fabric test

Fitting

some alteration on base harness. this will be made with thicker material.

Fabric sampling
These are samples from Maybachufer. I am not sure about the color…

And here are things I found at Kumasch Stoffe. I took only samples today.

I am thinking of the second one here. It has nice suede like texture. Once side orange-ish shiny gold and the other side is matt suede. The other option will be to go with the first one, both side shiny orange-ish gold taffeta.

They also had nice translucent fabrics, but I am not sure if it goes well together…

On Monday, we finally decided on the fabric and got 2m of the Golden Suede like fabric. started to make 150cm frill for test.

It drapes very nicely and shows the details of gathering very well. The fabric is quite soft, and we are not so happy that it “sags” a bit.

As the sagging was an issue, we also tested to make the frill with golden taffeta fabric. We added thin reflective material inside the edge tubing to give more character.

The first idea was to mix both types of frills, but it did not look good together.

Left side frill is taffeta and the right side is suede. The taffeta holds the shape well, but it looks a bit plain.

The next test was to make frill with 2 layers, sew 1 layer of suede and 1 layer of taffeta together. The result is that it has the look of golden suede material, but is much more rigid and holds the shape better. I was afraid that it is becoming too heavy, but this was not an issue.

For harness(base) part, we purchased 2nd hand old leather coat as material. It was on sale and the zipper was broken.. so I think it is fair that we disassemble it and use it as materials.

Leather glue test with fabric core to achieve non-stretch and stiffness.

In the mean time, Ben has been working on the harness design and pattern. The harness has to hold a battery and electronics, which needs to be shielded from sand. Here is a sketch by Ben.

and the current harness pattern by Ben

The textile pushbutton:

The taxidermy looking circuit:

Light tube and circuitry embroidery:
SMD mini neopixel LEDs (WS2812B MINI) are mounted on a 4 line strip protoboard facing 3mm brass tube. (1.5cm long) LEDs are mounted vertically on the protoboard and connections to the top side of the LED is made with copper tape from the side. We also tried with bumblebee connectors, but it takes very long to make them with brass tubes and as it worked well with protoboard and it is relatively fast to make them, we decided to use this design for the bolero.

The circuit is made with sewing machine regular running stitch using copper conductive thread as bobbin thread. When using the conductive thread in this way, the actual conductor side will be on the back side when you stitch. So, I traced the circuit mark to the back side first and stitched from back side so the copper thread is placed on the facing side. At the end, we noticed it was actually better if the copper thread was on the back side than the facing side as it will be isolated with lining fabric and leather facing it. This will be a revision that needs to be made for next version. After the copper thread is stitched with sewing machine, we still needed to thread the copper thread and make the connection to the LED mounted protoboard by hand. This took few hours to finish all of them.

Almost finished:

Light up:

We were very happy when turned the light on when we finally finished it. Additionally this happened at 4am in the morning after hours of hard work and debugging. (Hannah, Mika and Ben worked around the clock to finish it in time!) We still needed a name for this project, and both Mika and Ben mentioned the movie “The Adventures of Priscilla, Queen of the Desert”. So this is the name for this bolero: Priscilla, the queen of the desert!.

Hand-written instruction and stamping the KOBA stamp:

Repair

After Deborah came back from the festival, we asked her to bring it back to check how “wear and tear” happened to the bolero. Also, we finished the bolero in a rush to catch her flight to the festival and some parts were done in a “quick and dirty” way. We wanted to check and repair the bolero.

One issue was that as the neopixels draw a lot of power and creates some peak in power line, it messes up the readings of the microphone module. We checked this with oscilloscope and to prevent this, the best way is to separate the power source using DC DC converter. We used SW1- 505 S DC DC converter that is specialized in isolating power sources.

The pinout of the SW1- 505 is 1) – V input 2) + V input 3) -V output 4) +V output

First I cut out the + and – connection between neopixel and microphone module + teensy. Then connect connection from the battery to 1) -Vin 2) +Vin, connect neopixel + and – to 1) and 2). I have also added extra 1000uF capacitor between + and – of neopixel.
Then connect microphone module + teensy + and – to 3) and 4). Now the power source is disconnected.

But now, it did not work, as the neopixel data line from the Teensy is floating. So, I connected the GND of teensy to neopixel GND (-). I am not super sure if this is correct thing to do, but now it works.

The second issue was that some piexels were showing strange color when you move the fabric. I opened it and looked the connection inside. Some power connections that were made with the copper conductive thread was almost broken, and resulting to disconnect once in a while. This was due to rub against protoboard on the edge. I added extra threads and covered with fused fabric.

Also, Deborah said sometimes optic fiber came off form the tube. so I added heat shrink tube to fasten the connection between the brass tube and the optic fiber.

The fabric button was not working good as it is not easy to find where you have to push. We decided to switch to button button
original fabric button

The connections are made with elitex silver thread. after the connection is made, I used normal cotton thread to wrap around the silver thread for isolation and protection.

Then the plastic film (cut out kapton sheet) is added to extend the push surface, and covered with power mesh net. The final layer of the button is made with the same fabric as the bolero so it matches the color. When you push the button surface, it pushes the mechanical button inside and makes a click sound. and we can get a reliable “push” each time you press the button.

All these repair was made by opening the seam on the neck line. after the repair is finished, I hand sewn the seam back.

Because the microphone module was not working well due to power problem, we had to work around in the code when we first finished in April. Also code was quite messy. Now that the microphone works as it is supposed to be, I have also updated the code, cleaned up and make sure it respond well to the music.
Here is the code on github >> https://github.com/KOBAKANT/KOBA-11-Priscilla_Bollero