#34 - A Bumbling Breakthrough

Getting Natural Glue to Stick to Natural Fibers

An Engineering Self-Study is supported by readers like you. Thank you for reading, sharing, and subscribing.

---

It’s mind-blowing that you folks read my ramblings. It’s even more mind-blowing when some of you spend time and effort to share your knowledge with me.

I suppose I shouldn’t be shocked. One of the inspirations for this newsletter was this post about learning in public, which claims, “People notice genuine learners. They'll want to help you.” Put differently, generous readers take pity on me when they see me floundering. Whatever the reasoning, it’s really freaking cool.

To recap my flounderings: previously on An Engineering Self-Study, I wrote about my first attempt to combine a natural fiber fabric (linen) with a natural glue (casein). The result was frustratingly floppy. I identified three main problems:

1. The fabric lifting up off the mold.

2. The fabric not being fully wet out.

3. An unknown issue with the fabric or with the glue/fabric interface.

Those first two problems were solved in a subsequent post by using a vacuum bag, some wax paper, and a squeegee. 

It was time to tackle the third issue, and I didn’t have a clue where to start. Just then, in the depths of my confusion, I heard celestial harps -- the storm clouds parted, a beam of light shone down from the heavens, and my phone was illuminated by a notification. A miracle! I had an email from a reader.

The reader was Miriam Dym and the email would be crucial in my progress, but it wasn’t as simple as: “Try this. It will solve your problems.” “That worked, thanks!”. 

Initially, Miriam and I considered the particulars of the fabric, like the yarn and the weave. But the more I thought about it, the less likely it seemed that the fabric itself was to blame. The linen was about the same weight as the fiberglass I was comparing it to, and the linen’s flax fibers were about the same stiffness as the glass fibers as well. If the fabric was at fault, I should’ve seen a smaller difference between the linen and the fiberglass composites (theoretically). So it seemed more likely that the glue/fabric interface was at fault.

Miriam also suggested pre-washing/scouring the fabric to get rid of any existing sizing (coating on the fibers). This is why reader suggestions are so cool. I would’ve never thought of scouring in a million years, but it’s an important step in Miriam’s world to prepare fabrics for dyes. I liked the idea a lot. Of course, instead of scouring with washing soda as was suggested, I decided there must be a more complicated technical solution. 

So I subjected myself to a painfully boring slog through scientific papers. I found some relevant papers, like this one that suggested an ultrasonic treatment of the fabric with a lye solution. I didn’t have ultrasonic equipment, so I decided just to soak the fabric in lye overnight. The resulting composite was slightly stiffer, but nowhere near the fiberglass. 

(To test stiffness, I make a swatch of composite roughly 3”x4” [arbitrary], using the same glue mixture each time. I put the composite in the oven on a “Warm” setting until dry and then flex it in my fingers to gauge stiffness. Not exactly quantitative, but I’m only looking for big changes at this point.)

So again, I found myself with a floppy composite and without any ideas on how to un-flopify it. Maybe I could just buy my way out of my problems. There are a ton of linens being marketed specifically for composites like this one or this one or this one or this one. I figured there must be something special about them.

But the majority of companies don’t cater to small consumers like me or, if they do, the fabric is priced outrageously ($40-$60/yard!). I guess the thought is that since people are looking for carbon fiber, they’ll be willing to pay carbon fiber prices. Not me! I was looking for something priced more like linen and, finally, I found it here from a company based in France. I don’t like to rely on a single supplier for anything -- I’d rather have a more universal solution -- but I needed to see if there really was a difference in these more niche linens.

While my environmentally friendly linen was being flown halfway across the world, another package showed up. This is what is known in industry as a “parallel path”, a much-loved way to get results half as quickly with double the resources. This package was from Miriam, who not only shared knowledge, but also sent fabrics for me to try along with a note that said, “Hope they are useful or the failures are interesting!”, which is about the nicest thing anyone’s said to me. 

I tested each fabric -- which included hemps and linens of different kinds. I accidentally created a blind study by assigning codes to each fabric and forgetting what they meant. When the samples had dried, I found that the two stiffest fabrics were scoured linens. And boy were they stiff -- they felt as good as a fiberglass/epoxy composite. 

A massive step forward, then, with scoured linens. Scoured. Yes, the thing Miriam had suggested right at the beginning. Why didn’t I try it earlier? You must remember my mind is a cesspool of different ideas -- these posts, as much as I try to show the deadends and backtracks, are a simplified version of events with the benefit of hindsight. Still, I was extremely lucky that Miriam sent me those fabrics or I would’ve missed a great solution that was sitting right in my inbox.

A few days later, another email in that inbox triggered another promising development. Every Monday, I get an email from The Prepared, a fantastic manufacturing-related newsletter. (Coincidentally, I’ve written for The Prepared. Here’s my issue if you’re interested.) In the Nov. 16th issue, Spencer included an interesting link about sizing in woodworking, where a thinned layer of glue is applied before the final glue-up in order to allow the glue to soak into the wood. 

For my fabrics, I had only considered sizing as an enemy -- something that was getting between my glue and the fibers. But what if I could add my own sizing? Wet out the fabric in a thinned version of the glue and dry it before applying the main coat of glue as before. It was easy enough to try, so I did. 

And lo an behold, it seemed to work!

The sized fabrics were stiffer and, interestingly, slightly darker than their unsized counterparts. Visual proof that the fabrics were being wet out more. The sized material was also a little duller, as if the glue was being drawn in rather than just sitting on top. So I now had two promising processes -- scouring and sizing.

By this time, the technical linen from France had arrived. It had a slightly more open weave than the other linens I’ve used, was slightly darker than them, and had a twill weave compared to their plain weaves. How much of a difference would this fabric make? 

Turns out, it was difficult to tell. The composite piece that I made from the French linen felt very similar to Miriam’s scoured linen and responded just as well to sizing. But of course, this wasn’t a true apples-to-apples comparison. I was comparing this technical fabric, which had only been sized, to a fabric which had been scoured and sized. I needed to scour my technical fabric to isolate the effect of the fabric. 

In other words, was the technical fabric performing well because of its weave and yarn? Or was it doing well because it had been scoured from the factory? If the scoured fabric improved significantly, I would know that there was some magic in the fabric itself. So I set about scouring.

Apparently, you can scour fabric by either throwing it in a washing machine with washing soda or leaving it in a pot of hot water with washing soda for half an hour. I chose the second method, since I didn’t have much fabric.

The green-tea color is all the sizing that was stripped off the fabric.

I tested two fabrics: the technical linen from France and the original linen I had (and struggled with in that post I mentioned previously). For both fabrics, after scouring, I applied sizing. Here’s what I found: the technical linen showed a minor improvement from its un-scoured counterpart. The original linen showed a massive improvement. 

That tells me that scouring and sizing can turn any natural linen into a respectable base for a composite. The differences between the fabrics -- the original linen, Miriam’s linen, and the technical linen -- are minor when they’re all processed the same way. That’s a really exciting development, which makes me cautiously optimistic that I can successfully make an all-natural composite part.

Let’s hope I don’t get added to those “celebrating too early” compilations.

Thanks again to Miriam and The Prepared.

---

Corrections? Questions? Comments? I’d love to have your input. Leave a comment, email me at surjan@substack.com, or find me on LinkedIn.

---

Drawing exercise #23. If you missed it, here’s why I’m learning to draw.