Most of the time, I have a pretty good idea what I am going to write before I sit down and start clacking away at the keyboard. I’ve gathered information and, if necessary, run tests. The two are then given a few days of mental marination. I figured a column on knots would follow this general theme. Boy, was I wrong.
The testing process started innocently enough — a couple of hours in my garage measuring the strength of various knots tied with a selection of tippet materials. There’s nothing new or unique about that. The problem was that my knot-strength data was inconsistent. I didn’t seem to be able to replicate the results others had claimed. Intrigued, but somewhat stumped, I stepped away from the garage and spent a few hours poring over some technical papers.
At first, it seemed as if the technical stuff wasn’t going to be of much help. Most of the papers involved a branch of mathematics called knot theory, which has become a powerful tool in biology, chemistry, and medicine. Unfortunately, the mathematics of orientation-preserving homeomorphism in three-dimensional Euclidean spaces doesn’t really help fly fishers tie better knots. At least not as far as my math skills can tell. I was about to give up when I came across a paper in a physics journal. With careful reading, it was apparent that the researchers had uncovered something that might have real practical value to fly fishers. Unfortunately, the only way to determine if this could indeed help us tie better knots would require a fast digital video camera and a digital microscope. Thankfully, like most fly fishers, I had both.
By adding a macro lens to a GoPro camera, I could shoot close-up videos of knots at 120 to 240 frames per second. Getting the camera sorted out was the easy part. The real challenge was building a platform that kept the camera focused on a tiny knot while yanking on the tippet hard enough to make it break. That took a couple of hours, an eight-inch C-clamp, and a two-by-four.
The macro lens on the GoPro provided enough magnification to see how the overall knot behaved at high speed, but wasn’t enough to pick out subtle details. Since most tippet material is measured in hundredths of a millimeter, something in the order of 100 times magnification would be needed. My digital microscope has the necessary power, and it also shoots decent videos.
I contacted the folks at Scientific Anglers and RIO, let them know what I was up to, and asked if they’d send me some factory-fresh tippet material. I wanted to make sure that any findings were based on materials that hadn’t spent months sitting on a shelf. Ultraviolet light and heat can be quite unkind to tippets, and I thought this might have been part of the reason why my garage testing produced inconsistent data. Both companies were happy to be involved, and a few days later, I had several spools of brand new tippet.
It took almost two days to shoot a few dozen videos using common fishing knots and several tippets, including hours chewed up painstakingly reviewing the videos. Given the time and equipment needs, it was obvious why others had not gone down this particular path. It’s not exactly cheap, and it’s a lot of work. To keep things simple, most of the tests were run on 5X tippet, which according to SA and RIO is the most popular tippet strength. The results exceeded my modest expectations and revealed a number of things I’d never suspected or even heard of.
Initial Seating Force
The first revelation occurred before I even pushed the shutter button. Even with the relatively modest magnification of the macro-adapted GoPro, it was obvious that the first knot (Improved Clinch) had not fully seated, despite a firm tug. It had looked OK to the naked eye. I switched off the camera, grabbed a spring balance, and ran a few tests. To seat Clinch, Improved Clinch, and Uni Knots on 5X tippet took just over two pounds of force. The Trilene Knot, which is basically a regular Clinch Knot with two passes through the eye, required two and a half pounds. Apparently that double pass increased the knot’s overall resistance to tightening.
No one I know seats knots with a spring balance, so it’s probably not a bad idea for folks to have a tactile sense of how two-plus pounds of tug on a fly actually feels. This is easy enough. Tie a fly onto some tippet and attach the other end to a spring scale or something that weighs about two pounds, such as a quart of milk. Grab the fly between your thumb and forefinger, as you would when tightening a knot, and pull on the scale or lift the weight.
Expect to pull harder if you are using heavier tippets. My tests with 16-pound tippet (a common strength for salt water) revealed that it took eight pounds to seat an Improved Clinch Knot fully. This is why it often makes sense to wear gloves when tightening knots in heavier tippets. That much force can produce painful garroting cuts in water-softened digits.
Unseating and Reseating
As the Clinch, Improved Clinch, Trilene, and Uni Knots came under tension, they pulled away from the hook eye by a fraction of a millimeter and then quickly reseated themselves. The whole process took less than 10 milliseconds. It’s hard to know for sure if this stretch/ snapback movement increases the likelihood of knot failure, but it’s possible that it might create small, localized areas of heating that could weaken the knot. In contrast, the Palomar and Orvis Knots exhibited a less dramatic circular movement of the tag. Interestingly, these two knots had performed better than the others in my earlier garage tests.
Twisting Clinches
As the Clinch, Improved Clinch, and Trilene knots came under tension, their twists rotated as the knot tightened. The internal twist (the one you make with your fingers) was overlaid by an external twist created by the section that forms the tag. This dual twist is a bit like a miniature version of the Bimini Twist.
For some of the knots, the tag-twisting process halted before the knot reseated, creating an uneven knot. Looking at the videos, it seemed as if the problem was caused by the internal twists. Adding extra twists to a knot clearly exacerbated the problem. Was twist-induced friction to blame?
To test this, I lubricated several knots with glycerin or Armor All prior to seating. Glycerin seemed to help with the Clinch Knots, but the Trilene stubbornly maintained an uneven profile. Presumably the Trilene’s double pass through the eye has a significant effect on the knot’s overall resistance to tightening. Armor All clearly reduced friction; the twists in the Trilene Knot rotated smoothly and quickly seated. In fact, Armor All reduced the friction so much that the Clinch, Improved Clinch, and Uni Knots all slipped undone.
Gluing Knots
The Armor All test suggested that a product that goes from liquid to solid might allow a knot to seat down more easily and, when it hardens, make it less likely to slip. Such a product would be applied to the knot before it was seated, instead of the usual lick of saliva. This would ensure that the product coated both
the internal and external surfaces of the knot. Two products seemed to fit the bill. Superglue was certainly easy to apply, but went from liquid to solid a bit too quickly. On two occasions, the fly became securely attached to my thumb and forefinger. This would seem to suggest that supergluing hook knots isn’t a great idea with patterns featuring soft materials such as hackles, hair, or CDC. It might be workable with large saltwater flies, but you may still end up adding some human skin to the pattern. It’s probably best left on the fly-tying bench.
Loon’s UV Knot sense, which remains liquid until exposed to UV light (a UV flashlight or the sun) did a decent job. The knots cinched down smoothly and did not unseat and reseat once they had been zapped with ultraviolet. The glue also seemed to eliminate postseating rotation of the turns that had been so evident with untreated knots. It didn’t change the strength of well-tied knots, which is probably due to the surface chemistry of modern tippet materials. Josh Jenkins at Scientific Anglers noted, “Both nylon and fluorocarbon have low surface energies, so they don’t take to adhesives that well.”
I suspect UV adhesives could help tie stronger knots when the weather is really cold. Maybe it’s just me, but I find it very difficult to tie a good knot when my fingers are Novocain numb. Midwinter steelheaders may want to consider adding a tube of UV Knot Sense to their kits.
Tight Turns
The Clinch, Improved clinch, and Trilene Knots all changed shape as the tension increased, going from a fairly uniform width to distinctly tapered, with the fat end at the hook eye. The terminal twist (farthest from the eye) became particularly tight. The paper in the physics journal had shown that strings break in the region of highest curvature, which they found typically occurs at the ends of a knot. According to the authors, the tighter the turn (the higher the curvature), the greater the elongation on the outside edge of the turn. When this elongation reaches a critical amount, fissures form, reducing the string’s strength and ultimately resulting in failure. You can see this sort of thing when you take a live branch or twig and bend it over a knee. The outside edge starts to splinter before the thing finally snaps in two. So the physicists seem to have disproved the popular idea that monofilament cuts into itself. Instead, the tight turn is actually tearing the tippet apart at the molecular level.
This suggests knots that produce particularly tight twists are more likely to fail. Judging from the videos, the Clinch, Improved Clinch, and Trilene Knots all produce very tight terminal twists. The Uni Knot has less curvature at this point, most likely due to the fact the tippet and tag end both exit the knot at this location. The Palomar and Orvis Knots seemed to have the least curvature of the knots I tested. I was beginning to see why these two knots had performed better in the garage strength tests.
The tight-turn hypothesis goes a long way toward explaining why Wind Knots are such bad news. A Wind Knot is a fancy fishing name for an Overhand Knot, which the researchers found produced the tightest turns of any knot. My garage testing showed that an Overhand Knot reduces tippet strength by about 50 percent. Bear this in mind the next time you decide that the couple of minutes it takes to fix one is too much hassle. Murphy’s Law pretty much guarantees this is when the biggest fish you’ve ever seen takes your fly.
Flow and Rotation
Even after the knots had reseated, the tippet material continued to f low through the knot as tension was added.
The effect was quite mesmerizing. The central (fly-to-leader) section exited the knots like toothpaste out of a tube, while the twists of the Clinch and Improved Clinch showed signs of further rotation and tightening. The Uni Knot seemed the most prone to flow, probably due to lower twist friction. This likely explained my garage strength tests with the Uni Knot. I had found that a 6-turn Uni Knot tied in 5X tippet was consistently weaker than a 6-turn Clinch or 6-turn Improved Clinch. It took 10 turns for the Uni to match the strength of the Clinch Knots. Of course, with that many turns, the knot was harder to seat.

But what really caught my attention was the tippet material inside the hook eye. Prior to the videos, I had assumed this part of the knot was rigid under tension. It most certainly was not! Instead, the monofilament rotated around the eye like a rope on a pulley. All knots exhibited this behavior, though it was most pronounced with the Clinch and Improved Clinch Knots. The Palomar and Orvis Knots exhibited the least rotation.
Dodgy Eyes
This rotation means the hook eye acts as a bearing surface for the tippet. Any defects with the eye could potentially cut into the tippet. Rusty hooks would clearly be a problem, but what about brand-new hooks. Do they all have smooth eyes? It was time to put some hooks under the microscope.
It took a few minutes to get the lighting and focus right, but every regular (ring-eye) hook had at least one defect on the inside surface, toward the back of the eye. There was a sharp-edged flat spot where there should have been a smooth, curved surface. From what I can tell, these defects are caused by the eye-forming process, which involves bending the wire around a small, yet very hard metal post. Cheaper hooks tended to have more pronounced defects, presumably due to softer hook wire or less stringent quality control during the eye-forming process.
That said, in most hooks the defect(s) reached only to the midline of the eye, leaving the location where most knots are tied nice and smooth. But that wasn’t always the case. In some old Eagle Claw 413 jig hooks, the defect went completely around the inner circumference.
Another thing that stood out was the end of the wire that forms the eye. In several cases, it had a sharp edge, and there was a noticeable gap between it and the hook shank. Only loop-eye salmon hooks seemed to have defect-free eyes, presumably because the eye-forming process is somewhat different.
If the tippet material moves over these internal flat spots or the sharp end of the wire, it could conceivably get damaged. Conventional fixed knots tend to stay toward the front of the eye, well away from the sharp wire end and flat spots. But what about Loop Knots? The loop, unconstrained by knot friction, could easily move over the defects and do so multiple times as the fly is cast and retrieved or when a hooked fish turns or jumps.
This Loop Knot–damage hypothesis seems to jibe with my own experiences, especially with weighted flies. Clousers tied on jig-style hooks can be incredibly effective. They sink fast and have an up-and-down motion that activates the kill switch with a lot of fish. But I’ve noticed that sometimes, after 45 minutes or so of casting, the fly will pop off midcast. Examination reveals that the knotted part of the Loop Knot is still intact, but the loop itself has been converted into a V. Presumably, this is possible with other weighted flies tied on Loop Knots. There’s a simple solution, however. When using heavy flies, forget the Loop Knot and use metal snaps such as Scientific Anglers Stay-Loc snaps or RIO’s Fly Clips.
Getting Personal
You may think you tie strong knots. I sure did, and I was wrong. You may also have believed that a specific knot yields XX percent of tippet strength, because that’s what knowledgeable folks have said or even demonstrated using sophisticated machines. That’s also wrong. The fact of the matter is that knots are somewhat personal. The way you tie and tighten a knot isn’t exactly the same as the way someone else does it. This was something that Josh Jenkins at Scientific Anglers and Simon Gawesworth at RIO had both noticed while conducting knot testing on Instron materials-testing machines. They told me their tests were as much about the person tying the knot as about the knot itself.

In addition to the personal factor, you also have to take into account tippet chemistry. Tippets are a made from a mix of chemicals, which varies by manufacturer and according to the desired characteristics. As such, two seemingly similar tippets can have different knot strengths. And then there’s the issue of the environment. You can’t expect to tie great knots when the weather is cold or the light is poor.
This means it’s a really, really, really good idea to figure out which knots work best for you. Thankfully, you don’t need an Instron machine or a high-speed camera or microscope. For hook knots, just tie an appropriately sized hook to each end of a length of tippet, one with your usual knot and the other with some other type of knot or the same knot with more or fewer twists. Secure one of the hooks to something solid, such as a screw or cup hook in a fence post, and do the same thing with the other hook using the finger guard of some forceps. Put on some glasses to protect your eyes and pull on the forceps until one of the knots breaks. Repeat the test a few times to eliminate any tying mistakes. It helps to tie pieces of bright yarn onto the hooks. They’re really tough to find in grass or a carpet.
For tippet-to-leader knots, get two identical leaders (preferably new) and connect them with a 12-inch section of tippet. Tie one leader with your usual knot and the other with something different. Wrap the leaders around gloved hands, leaving the knots and tippet between your hands. Pull the leaders apart until one of the knots breaks. As with the hook knot tests, repeat a few times. In as little as 20 minutes, these two tests will tell you more about your knots than any expert. While knowing the actual strength of your knots is obviously useful in all flyfishing situations, it seems to be especially important for folks fishing 3X or lighter tippets in waters with trout over three pounds. The momentary force of the strike or a violent headshake from the fish can exceed the knot strength. I’ve personally experienced this on several very frustrating occasions, and I’ve seen it happen with friends, too. When there’s a real chance of hooking trout over three pounds, I’ve taken Denny Rickard’s advice and use nothing lighter than 10-pound test. The fish don’t seem to care.
Based on these tests, I have switched over to the Palomar and Orvis Knots. I’ve also ditched Loop Knots when fishing weighted flies. These changes are based on the way I tie knots and the kind of fishing I do. They may or may not apply to you. If all you ever do is fish for small trout in mountain streams, stick with an easy to tie knot such as the Clinch. There’s no point in making things complicated. I have done my best to explain the videos, but realize that words can go only so far. Six videos have been uploaded to YouTube so that you can see for yourself what is going on. Simply type the following into YouTube’s browser and it should take you to each video:
Clinch Knot Microscope Video
Orvis Knot Microscope Video
Palomar Knot Microscope Video
Four Turn Trilene Knot Microscope Video
Six Turn Uni Knot Microscope Video
UV Knot Sense Microscope Video
Perhaps you will notice something that escaped my attention. I wouldn’t be at all surprised. If nothing else, I hope this gives you a better understanding of knots, which, let’s face it, are one of the most important, yet underrated aspects of fishing. Tight knots and tight lines.