Odds 'n' Sods of Information and Opinion

Advanced Fly Design

The current steelhead season has been a frustrating one for me as the flies that had been so effective in past seasons, have failed to hook fish this fall. I have had numerous tugs from fish, sometimes three plucks in a row from the same fish or a tug followed by a smashing take, but few hookups. Most of those I did hook got off. However there is a solution.

We've had low flows for most of the season which has had two effects, one definite and one probable. In lower flow rates, most fly designs, especially long ones like my tube flies, tend to swim with the hook sagging badly. So a fish that hits the wing misses the hook. Secondly, with the low flows the fish are probably less secure so therefore less aggressive, which results in more plucks and tugs than full takes. Since all of my flies have wings that are up to twice as long as the body, there's ample opportunity for a plucking fish to tug the wing and miss the hook.

I realize that a stinger hook style fly would solve all of my fish missing problems, but I don't like them for a number of reasons and I won't use them. For one, if a fish does inhale a stinger equipped fly, there's a good chance the hook will end up in the gills. That said, I am willing to extend the hook out to the end of the wing, but no farther.

Having become truly fed up with all of the misses, I decided to redesign my tubes so that the hook was at the back of the wing and the fly would ride level in slower flows. The very tip of the wing would still extend a teensy-weensy bit beyond the gape of the hook, so technically it wasn't a stinger hook design. The big thing though, the fly body and hook would have to ride level so that the hook would almost be touching the wing.

Yesterday on its very first usage, the redesigned fly hooked the only fish that touched it, a nice 5 lb. chrome hen which stayed on and was landed. One of my fishing buddies saw me use it and asked for the specs which I supplied. Today he caught three on it after batting an o'fer yesterday with multiple tugs and no hookups on his usual flies. Even though it’s a bit of a fiddle to make the tube, all of my tube flies will be done this way from now on. If I had started the season with this style, I probably would have 30 or more fish to my credit instead of the handful I have to date.

First the theory:

There are two points on every wet fly or streamer that controls how it swims: the centre of gravity (CoG) and the centre of pressure (CoP). The CoG is the point through which the forces of gravity and the pull of the leader act. The CoP is the point through which the hydrodynamic forces act.

The pull of the leader, or tow angle, significantly affects how the fly rides. In this frame capture from a video I shot, two identically made flies are shown swimming. Both hooks are the same size and weigh the same, the only difference, one is tied on a down eye streamer hook while the other is on an up eye salmon hook. The image clearly shows the effects of the tow angle.


Throughout the video, these flies consistently rode in this fashion, with the streamer hook version for the most part riding above the salmon hook version thanks to its higher drag and lift produced by its greater angle to the current.

This video dramatically shows how the tow angle produced by the tie in position of the leader, changes the ride angle of the fly. Since tube flies have the leader running down the centre of the fly it makes sense that the typical tube fly will run at an angle somewhere between these two examples.

The tow of the fly pulling through the CoG plus gravity pulling down through the CoG, in a sparse fly like the one shown here, usually overwhelms hydrodynamic forces acting on the CoP. However, with a more heavily dressed fly, the hydrodynamic forces acting through the CoP can influence how the fly rides to a much greater extent. A perfect example of this is an unevenly tied or poorly designed fly that will ride upright in a slow current, but kick over onto its side in a faster current where the hydrodynamic forces are much greater.

In this image, the two centres are illustrated:

CoG and CoP

The CoG is fairly easily determined through the balance of the fly, but the CoP has to be approximated and it will move around a bit as the current influences the shape of the fly. A generalization of its position is sufficient for our purposes. It can best be estimated to be at the place on the fly with the greatest cross sectional area and drag. That telltale hump in the wing of the fly of the salmon hook fly shown in the underwater shot, is usually the best indicator of its position. In a fly that produces a teardrop shape in the current, the CoP tends to be in the neighbourhood of the widest point.

Assuming a neutral tow angle, like most tube flies possess, the relative position of these centres will affect whether the fly runs nose up, nose down, or level:
  • Placing the CoG in front of the CoP causes the tube to plane downward. We can do this by moving the CoG forward via adding lead clouser eyes, tungsten cones or beads, or using a copper tube as I have done. We can also tie the fly with much of the material farther back on the shank to move the CoP rearward
  • Placing the CoG behind the CoP will cause a fly to ride nose up and rise in the current. This is the common condition of most typical tube flies that are tied without some sort of weight added to the front or without moving the material rearward to compensate
  • Placing the CoG and the CoP at the roughly the same spot will produce a level riding fly
  • The farther apart these two points are, the more pronounced the effect

We can control the position of the CoG by:
  • adding or removing weight along the length of the hook or tube
  • using shorter or longer hooks or tubes
  • using heavier or lighter wire hooks

We can control the position of the CoP by:
  • Varying between sparsely and heavily dressing the fly
  • Moving the bulk of the materials forward or aft along the hook or tube
  • The buoyancy of the materials and their relative positions
  • The shape of the fly both in cross section and length

Two fly examples of how important these issues are: my
Brown Trout Weamer and the Intruder. The Intruder is tied with clouser eyes at the front and a hackle at the rear of the shank. Both of these features act to place the CoG in front of the CoP producing a fly that will tend to plane down and stay down when swung. The stinger hook remains high and close to the wing so that when a fish strikes the wing, it also gets a mouthful of hook.

Brown Trout Weamer uses a different approach to achieve the same thing. The fly has two aspects that do the same job as the Intruder in moving the CoG and CoP, but instead of weight and a rear hackle it uses a palmered hackle over the body and a very light wire hook. Both act to bring the CoG close to the CoP.

I never understood why the original recipe was so effective and variations less so, until one day I dropped one in the river without it being attached to the leader. I had dulled the point so badly on a rock, that resharpening wasn't an option so I cut it off the leader. In the process of tying on a new one, the damaged one got dropped in the river. I wished I had had a camcorder with me for what happened next I have not seen in any other streamer or wet fly design. On hitting the water, the fly immediately oriented itself nose first into the current and dead level. It then swam slowly backward and down in the current, perfectly stable and swimming level. It looked just like a natural. I've since tried this with other
Brown Trout Weamers and they all swim the same. No wonder it is so popular with the fish.

So onto the tube design:

I wanted my tube design to have some of the properties of my Brown Trout Weamer and the Intruder so it had to have these priorities:
  • The hook had to be at the end of the wing
  • The hook had to ride very close to the wing
  • The tube had to be relatively narrow so that it would accommodate my usual sparse designs
  • I would not be adding weight via tungsten cones or lead eyes as these would cause anchor extraction and turnover problems when fished with my usual long leaders

To get the weight distribution the way I wanted it, I had to build my tube with a weight forward bias. This tube design moves the centre of gravity (CoG) close to the centre of pressure (CoP) so that the fly rides level. It does not require any rear hackling to ride level, though there is no reason why one couldn't use this tube style to tie Intruders.

The design uses four components:
  • a ⅛" O.D. copper tube
  • a plastic liner that will fit inside the copper tube
  • a silicone tube for the rear body
  • a silicone tube for the junction tubing

The copper tube is roughly 40% of the length of the tube and is glued to the forward part of the liner tube. The front end of the liner is melted to flair it as added insurance to keep the copper tube on. The silicone tubing is then pushed onto the liner, leaving about ⅜" bare to accept the junction tubing. We have the option of tying the fly body over both the copper and the silicone tubing or just the copper tube. In the latter case, we can use different coloured silicone tubing to compliment the fly colours. The silicone tubing is available in many colours from
The Canadian Tube Fly Company.

So if you're finding that steelhead are hitting your fly but continually missing the hook, remember this blog post and get your flies riding right!