Buckle up, this one’s pretty interesting.

I was about to throw in the towel tonight. And by that, I mean the project of using the CLO software. But, just as I was about to give up, I made an interesting connection that turned my perspective around completely…

Let’s rewind a bit. For context (I think all the readers of the blog are aware of this), a couple of years ago, I took this jacket I own from Acronym, a J28-K:

(I love this jacket. I bought it from Saint Alfred in Chicago, I think in 2016. I remember chatting with the sales associate, who felt I was a big enough fan of the Acronym brand–or maybe he just wanted me to stop talking his ear off and leave–to give me a discount on the jacket. At the time, this was sitting on the shelf… but a few years later the model, a collaboration with Nemen, became very sought after)

Anyway, back in 2021/2022 I used masking tape to make an attempt at reverse engineering the pattern so I could make another one out of a different fabric. Here’s what the pattern looked like once I was done:

Overall it worked well. I used this pattern to make the jacket I introduced in the first post:

However, there was always one issue that plagued my copy of the jacket. It was the underarm gusset.

No matter how much I tried, I couldn’t understand how to get the triangular underarm gusset to work. I had done my best to faithfully reverse-engineer the 3D shape of my real-life jacket into flat, masking tape patterns. I had figured out the body and the (very tricky) sleeve. But the gusset just wouldn’t work, and I couldn’t figure out why.

For literal years I’ve been wanting to figure out what was wrong with the pattern I reverse-engineered. CLO has felt like an opportunity to tackle that problem.

First, I needed to digitize the pattern.

I duplicated and flipped the photo of the masking tape pattern shown above, then printed both copies out on letter-size paper:

Then I traced the pattern in Affinity Designer (essentially Adobe Illustrator):

Probably not the most efficient way to do this. But it was fun. Photocopies of photocopies of photocopies…

Then, I took the digitized version of the pattern and traced it in the CLO software as a pattern:

The pattern looked something like this on the 3D avatar, before simulation:

flight attendants, please take your seats and prepare for takeoff

(In this version, I omitted the triangular armpit gusset.)

When I simulated the jacket in CLO, it just wouldn’t work. I encountered the same exact problem I did in real life. And I didn’t feel like I had an understanding of how to edit the pattern to fix it. Which… left me exactly where I’ve been for the past 2 years. Argh.

All of that time spent learning new software, just to arrive at the same place I started? 😩

As I complained to my wife and considered scrapping the whole project, an idea popped into my head. I remembered something I had read from an Acronym product description:

🥷
Nothing extra. Nothing standard. J32-GT is a cross-section of current Acronym concepts. From the fit and mechanics of the KG fit block, to the architecture and operational rhythm of the detached entry pockets and integrated fast hood, every aspect of the jacket has been addressed from our unique perspective. Built with next generation Gore-Tex Pro fabric technology for improved ruggedness, increased breathability, and greater comfort in more conditions. A combination of old world and new, the KG fit block is a hybrid of vintage Japanese kimono and modern technical sportswear pattern making. Generous (zero resistance DFMA fit) yet economical (minimal seams, tape, and weight), it feels light and effortless

– 3L GORE-TEX PRO [Next generation technology]
– MICROGRID backer
– adjustable stowaway Fasthood
– ForceLock
– removable JacketSling
– EscapeZip
– SpeedLock shockcord hem
– left and right Gravity Pockets
– detachable logo tape
– 5 external pockets
– adjustable velcro cuffs
– front zip closure

source

Specifically, the thing I remembered was this:

A combination of old world and new, the KG fit block is a hybrid of vintage Japanese kimono and modern technical sportswear pattern making.

The pattern was based on a kimono block.

Note that the pasted description above is from a different jacket, the J32, but I think it uses the same, or a similar, pattern block to the J28. The way there’s no shoulder seam, but rather crinkling around the shoulder mirrors what I see in my original J28.

I began to realize that maybe the comparison to a kimono could explain how my pattern shape was wrong!

Here’s a photo of a vintage kimono:

Notice that, when laid flat, it’s in the shape of a capital “T.”

My pattern, in contrast, doesn’t follow that shape.

I looked at my pattern again and realized that maybe it should be T-shaped.

To test my theory, I made some changes in CLO to change the slant highlighted above, into a flat line:

It looks good. After creating the sewing relationships among the pattern pieces and running simulation, I examined the underarm area. Shockingly, it seems to look very similar to what I see on my original jacket. I don’t want to get too excited before I can truly validate this, but I feel like it’s on the right track. At the very least, this is something new to experiment with.

Something about the pattern just feels more simple and “right” now. So I think there’s a good chance it’ll work out.

Thanks for reading, and see you next time!