Ortlieb Organizer

Having a decent idea what the pattern should be, it was now feasible to laser cut it. I cut these out:

• Nylon: 2x base, 2x tall, .5" seam margin
• Canvas: 2x base, 2x tall, 1" seam margin

I chose to start with an all-nylon bag. I realized later I might have used canvas for the outside and nylon for the inside. All-nylon was a good choice to start with though.

I sewed the nylon pieces together, knowing that it would be easier to sew the other bits on first, but not knowing exactly where to put them. The fabric was thin enough that I could manipulate it enough to sew things on later.

So here's what I did:

• Sew base A to tall A to make pocket A
• Sew base B to tall B to make pocket B
• Sew pocket A to pocket B
• Insert plastic pieces into pockets
• Pin velcro to both sides of each pocket (to ensure tight fit)
• Remove inserts and sew on velcro
• Find locations to attach lock strip, pin webbing in place
• Sew on lock strip

Whereas ideally I would have done step three last.

So everything generally works, but the lock is so heavy, it drags the fabric down too much, and it also puts a lot of stress on the webbing-fabric attachment (A single box x stitch). The strip needs more support, and fortunately the plastic insert can do exactly that.

After the cardboard prototype, I decided I was ready to take on the sewing project. I bought some off-the-shelf fabric that seemed appropriate, and then proceeded to think about how to make the thing for a few months. I spoke with some very helpful people at the Sew and Tell meetup, and finally accepted that I just need to try making it a few times. One of the suggestions was to make that first fabric prototype with cheap scrap fabric, so I tried that.

The design was based on the cardboard model, with two broad sides and a thin strip connecting them to give it some volume. I got one big design suggestion at the meetup that allowed me to make progress on this: each side should have a sealable pocket to contain the plastic insert. With that in mind, this first version allowed me to test cutting the pattern, and the sewing process. It turned out that this fabric was awful to work with (very stretchy and very ragged edges), but I definitely learned enough to gain the confidence to work on the next version, with the good fabric.

I was proud of myself for documenting my learning process with the first bag, but it was not that helpful afterward. I changed the design, and I've already internalized the lessons I learned about sewing techniques. Probably most significantly, I abandoned the side strip idea. A laptop fits inside the bag without that added width, and I had a hard time keeping the three pieces of fabric properly aligned. I might want to use a similar approach for a different project later.

Lessons learned:

• Began to understand how to think about designing the inside-out object.
• When my mental model of the inside-out object fails, I can pin fabric together and turn it inside out to test.
• When sewing around a corner, and you need to raise the foot to reposition the work piece, you can rotate it in either direction - it doesn't have to follow the direction of the seam.
• If I try the thin strip idea again in the future, I will cut it maybe four inches longer than needed, THEN sew it, then trim excess. That will fix half of the alignment problem, but still need to be sure that the two side pieces are lined up properly. It might be easier to use several smaller thin strips, but that would also be uglier.
• When cutting the pattern, following the curved corners isn't important: cut a polygonal hull, sew whatever seam you want, then trim. The seam doesn't even need to be curved, because the bag will take on an organic shape anyway.
• The four main parts are attached with three seams, and I put the third one in the wrong place, so the first two were unnecessarily exposed. It occurred to me that I could have just used a single seam, but having multiple seemed like a good idea anyway.

My first attempt was a cardboard mockup, held together with zip ties. This worked well enough to prove the concept. It disintegrated after a month or so.