Welcome to Rod Repair!

The following is reprinted from the January/February 2000 issue of RodMaker Magazine. Author Ralph O'Quinn is a staff writer for RodMaker as well as the formulator for the Trondak Company's line of U-40 rod finishing products, and Ferrule Lube. Our sincere thanks to RodMaker for sharing this information!


Putting It Back Together
Part Three

Ralph O'Quinn
Story and Photos by Ralph O'Quinn


Problem #2

Number 2. The break shown last time was in an area that will not see much flexure or bending. Under heavy load, the rod at this point will be mostly straight, except when casting the line. The act of casting is probably the peak load that this repair will ever see. As we go further up the rod the bending becomes more severe. The break in No. 2 is about 14 inches from the tip. The bending moment is more pronounced in this area. Your sleeve needs to be slightly longer in order to accommodate the longer moment arm and slightly stiffer sections being joined. So increase the length of this sleeve to about 1 & 1/2 inch to 1 & 3/4 inch over all, which will mean that about 5/8 to 3/4 of an inch on each end will be structure. Don't forget the tapered ends. The bend moment on this area will be quite severe. Maybe enough for the graphite sections to rupture the low modulus sleeve even though it is wrapped and coated. To counter this shearing force, we will install a very small piece of either fiberglass or graphite on the INSIDE. This piece must fit snugly, and be about 1/4 inch long. If it is any longer it will be involved in the bending force and we don't want that. Be sure that this piece is in place before you position the outer sleeve. When the adhesive is cured, the sleeve is cleaned up and wrapped and finished like any guide or ferrule. Grasp the rod about a foot on each side of the repair sleeve and bend it into an arc with the repair in the center of the arc. The arc should be constant - no noticeable flat spot at the sleeve location. If there is a noticeable flat spot, then your sleeve is either too long or too heavy, i.e., too thick a wall section.

TAPER TROUBLE

At this area on most any fly rod is where you will have to start allowing for the taper of the blank when fitting a sleeve. The further back we go, the worse the problem becomes. No problem fitting the rear section - the largest OD - but depending upon the degree of taper - the front section can be quite challenging . This particular area is still so far toward the tip that the problem is very minimal. Your sleeve will fit very nicely over the rear section, but to fit it on the forward section you will find that the ID of the sleeve is smaller than the OD of the section that it must cover. Some repairs on fast taper rods require the removal of all guides and the tip in order to slide the sleeve into place. But what we have here is merely a snug fit. The sleeve must be pushed a tad but that's all.

Problem #3

Number 3. Now that we are into real structure, this one has to be good in every aspect or the rod is doomed. This is in the area of greatest flex which means greatest bend load and fatigue factor. But mostly, it is an area of sensitivity. You FEEL the rod in this general area and you want to maintain that feel. You don't want a 6-weight fly rod to feel like a 4-weight or an 8-weight. It will feel like a 4-weight if we use the same pattern of structure that we used in #2, because that design does not have sufficient stiffness to work with the bend load of the graphite in this area. To increase the stiffness using the same materiel, we merely have to increase the 'I' moment, and this is easily accomplished by adding an internal doubler. Calculating the amount of overlap for this internal doubler or plug can be rather tricky. Too long and it will impede the flex of the graphite and change the feel of the rod. Too short and it will not support the external sleeve, making the rod feel wimpy and sluggish and it will probably break after very little usage. I wish I knew of some simple formula to apply here, but it dwindles down to a gut feeling combined with experience plus a little logic thrown in for good measure. Any aerospace stress engineer experienced in calculating design loads for wing structures could whip out his Slide Rule, (whoops, that one dated me) er, I mean calculator, and give you firm answers. But in the absence of calculable direction, let's find out where logic leads us and if there are any aerospace stress engineers out there - sound off. Take another piece of graphite tip section of this same approximate weight and flex the same area where we are making our repair. Flex it over a linear scale - 12 inch ruler will do fine. As you flex the rod, note on the scale and ask yourself the question, "If a piece were inside, how long (length) could it be before it resisted the flexing?" I judged this one to be about 2 inches, but past experience has taught me that about 1&1/2 inches is best in this location for this weight of rod. Nevertheless I inadvertently went ahead and made the internal plug 2 inches long. Another way to determine this length is to figure that the overlap for the doubler will be between 3 and 4 times the diameter. Remember now, this plug is made of fiberglass and it is a smaller diameter than the rod, which means that by itself it will effect the graphite about as much as a wet noodle. However, combined with the external sleeve which we are about to construct, the graphite will have met its match. See photo above. This plug is acting exactly like the plug 'ferrules' on those rod blanks which design this type of ferrule such as Fisher, Scott, etc. After all, a ferrule is merely a joint and that is what we are doing here - making a joint. If it were going to be a ferrule, we would have to beef up that area of the blank where the ferrule is inserted, both ends. Then leave the end with the decreasing taper free to insert and disjoint at will. But since this joint is not intended to be a ferrule, it must be supported with an external sleeve similar to those we used in #1 and #2.

EXTERNAL SLEEVE

The external sleeve must ALWAYS overlap the internal plug. A general rule of thumb for the overlap is twice the diameter at the overlap point. I like to round things out to nice even numbers, so I used a 1/2 inch overlap in this instance. Then add almost another 1/4 inch on each end for the taper and you have a sleeve about 3 & 1/2 inches long. This overlap is critical to the integrity of the joint, much more so when the plug is graphite, but even as fiberglass it will exert stresses that could rupture the rod blank if the load is not properly distributed. Remember now, there is maximum bend at this point.

Your internal plug will be sized by inserting the selected piece of fiberglass through the nearby ferrule and out the broken end. The broken ends will have been smoothed and squared off prior to this point. Carefully mark and cut the piece so that you wind up with a plug with 1 inch inside the ferrule end and 1 inch extending, that will fit snugly inside the tip end. As you do more and more of this type of repair on different sizes and types of rods it will become apparent to you that the forward section should have a little less plug length than the rear section and this ratio will vary with the wall thickness, degree of taper, and location on the rod. In our example here, the optimum is about 13/16 inch, but let's not quibble and just make it 1 inch for now. With your plug cut and sized, be sure and radius the ends. Round them off with a file at about a 45 degree angle. This will prevent a sharp edge from gouging into the rod wall.

CONSTRUCTING THE SLEEVE

Select a piece of fiberglass rod section from what appears to be of the same or similar taper, and preferably non-painted. The painted fiberglass rods are the cheapies with thick walls and very coarse glass cloth construction. These make very poor external sleeves. Match your selected piece with the butt end of your joint - the larger end. Cut your fiberglass so that it will fit over the end and extend past your internal plug about an inch, which makes it about 2 inches total. Now trim your piece so that you have an equal amount for the other end. You should have a piece for a potential sleeve which is about 4 inches long at this point. Now you've got a problem. How do you get that tapered sleeve over the tapered tip section when the ID of the sleeve end is smaller than the OD of the rod end? You can take off all the guides and slide it on from the tip. There are 6 guides and the tip top and this is almost always the quality way to do it. So now it becomes a judgment call because there is a quicker (and easier) method which does the job quite nicely. (Just don't let that stress engineer from the wing group know what we're doing, cause he might get out his slide rule and squash the whole deal.) First, size the sleeve to a net fit on the butt end, which means that you will trim it to about 1&5/8 inch overlap. Next, trim it to the same dimensions for the tip end. Touch the end of the sleeve to the tip section end and see if the sleeve will slide over and onto the tip section. It had better not! It easily slides over the butt section, so put it there for now. The taper in the rod makes a mismatch in diameters between the section of the tip end and the reinforcing sleeve. The end of the sleeve will match the rod some 1&1/2 inches inboard - which is where we want it to wind up. But how to get it there? Try this - bevel the end of the sleeve that attaches to the tip section to about a 15 degree angle, or as shallow as you can handle. A power disk sander or belt sander is ideal for this operation. Rotate the sleeve in your fingers as you bevel and grind the end to zero. This beveled end will be quite flimsy, so match it to the tip section again and this time force the tip section into the beveled sleeve. Enter at a slight angle then push the rod section further into the beveled area which will split the sleeve at this point. This splitting will facilitate further movement until the sleeve is fully installed on the tip section. The other end of the sleeve which attaches to the butt section can now be beveled to the same degree. The splitting of the sleeve has done nothing to impede its structural integrity. The split is usually 1/4 to 1/2 of the length and is neutralized when the area is bonded, wrapped and finish applied. This is considerably faster than removing and replacing 6 guides and a tip top!

BONDING

The external surfaces of all pieces are prepared for bonding in the usual manner, i.e., light abrasion of the surfaces with Scotchbrite or 400 grit sandpaper. The inside of the rod pieces must be cleaned of loose particles by swabbing them out with a small bottle brush, a doubled up pipe cleaner or some similar instrument. Sometimes I use a small round file and on the larger blanks - a small notched dowel with a piece of sandpaper inserted into the notch and the other end chucked in my winding lathe jaws and rotated inside the blank. You will need a piece of rod or a stick of some kind to poke the internal plug all the way through the butt piece. Cover it with the paste epoxy and insert it in the ferrule end and push it through with your stick. Be sure and clean out the ferrule immediately. With the plug snugly in place, add more epoxy to the exposed end and bond in place the tip section The sleeve is loosely attached to this section, so slide the sleeve toward the tip and away from the repair area - then liberally coat the area toward the butt with epoxy, slide the sleeve over this epoxy and rotate it several times in order to thoroughly coat the inside of the sleeve. Now apply another thin coat of epoxy in the area of the repair and slide the sleeve into its position. Wipe off the epoxy that is forward of the sleeve, using IPA. With the sleeve in position, it must be wrapped. The wrap will be a temporary one, as its only purpose is to hold all parts snugly together while the epoxy cures. Since it must be wrapped very tight, I normally use D thread. The split end will close and epoxy will be squeezed out. When the epoxy has cured, there will be cleanup to do in the area of the split and at each of the beveled ends. When the cleanup is completed, apply your final wrap with a nice A thread to match the color scheme of the rod and finish off with your favorite epoxy wrap finish. ~ R. O'Quinn

Next Time

Next Time!


Previous Rod Repair Articles
Part One
Part Two
Part Three
Part Four
Part Five

[ HOME ]

[ Search ] [ Contact FAOL ] [ Media Kit ]

FlyAnglersOnline.com © Notice