Ferrule Tool Alignment

I have made more than 200 of the ferrule shrinking tools and have had only positive feedback on their performance, until recently. Paul had borrowed a tool from me and was able to tighten the ferrule fit on several rods. He returned the tool and then later bought one. But the tool he bought did not work as well as the borrowed tool. His description is that the ferrules were “unevenly swedged”. I can accept this description since he has successfully used the borrowed tool.

When the tool was eventually returned, it was obvious that one bearing race had been dragging against an end plate.

Future tool production will be with spacer shims on the two fixed bearing pins.

While waiting for the tool to return, I tried to imagine what the problem could be. I obtained a Fujifilm product called “Prescale” that, when squeezed between surfaces, produces an image of the pressure pattern. I made up a tool alignment procedure that would slightly improve the uniformity of the pressure pattern.

My procedure for realigning your tool if you feel that it is not correct or if you have dropped it:
1. Loosen the two radial screws. Loosen the 3 screws that clamp the two sides together. This only needs to be a quarter turn or so. Tap the edge of one plate with a mallet to be sure that it is not adhering to the spacers.
2. Tighten the two radial screws, bringing the moveable roller into contact with the two fixed rollers. Bring both screws into contact with the bearing pin before tightening either.
3. Re-tighten the 3 clamp screws.
In the photo are before (left) and after (right) alignment impressions made on film strip. There is a slight improvement.

In use, the tool is working on a hollow ferrule that is much more compliant than bearing-on-bearing contact, so pressure is much more uniform than what I have shown in the photo.

Posted in Alignment, Ferrules | Leave a comment

Alaska 2019

Last summer, we had a good partial day of fishing on a creek near Juneau. We decided to make a longer trip this year, and so went to a fly-in lodge northwest of Anchorage. The trip was by float plane.

There is no way to get there by road, but you might make it by dogsled in the winter.

The lodge is on Lake Creek, a tributary of the Yentna River. Lake Creek is not fed by glacier melt, and so runs clear.

The lodge building:

Guest cabin:

Rainbows would take sculpins, mice, and several types of streamer.

There were also grayling.

King salmon were present. This one may have had an encounter with a seal before going upstream.

One day, we went up nearby Fish Creek to a lake full of northern pike.

The highlight of that day was getting close to a moose.

Posted in Fishing | 1 Comment

Power Hacksaw

Before milling or turning, stock typically needs to be cut to length. My first tool for this task was a handheld power bandsaw lashed to a plywood frame.

This worked quite well, and is still functional. The reason that I have tried other means for cutoff is that it did not do well with large (say 3 inch diameter) round stock; the blade did not travel straight in the vertical plane. Part of the reason for this is that it is still a handheld tool and it is difficult to maintain constant pressure during a lengthy cut. Also, the wood frame is too compliant.

Next I purchased a bandsaw from Little Machine Shop.

This was better because the frame was stiffer. But after a year of occasional use, the rear guide assembly failed. Again, this is still a handheld tool and I was probably forcing it too much. LMS sent me a new guide assembly and it is again working.

Large, free-standing metal bandsaws have hands-free operation and should perform better than either of these tools. They also have limit switches to shut down the blade at the end of the cut. But I do not have room in my shop for this.

If you search for “power hacksaw” on Youtube, you can find videos of many homemade saws. Invariably, they involve a crank to achieve reciprocating motion. This kind of gadget has quirky appeal; I decided to try my hand.

Here is the result.

What is different in my design is the use of a large (NEMA 34) stepper motor. This avoids the cost of a gearmotor and the complexity of two stages of belt drive.

Two more photos:

The limit switch:

The motor control: a 48 volt supply, a step motor driver, and an Arduino Nano to generate pulses at a ramping rate.

This is an auxiliary fixture to hold large round stock.

Here is an extra weight for the front end.

Design and Construction Notes:
1. The frame for the saw blade is from a Stanley hand saw (STHT20138). The additional bow under blade tension is appreciable and must be considered when drilling holes in the mounting brackets, i.e., have the blade under tension when you spot the hole locations.
2. The fit of the blade end holders in the frame is loose. I was able to improve the squareness of the saw cut by inserting shim stock into the gaps.
3. The base is made from 1.5 x 1.5 inch “T-slot” material. Guide rails are 5/8 diameter steel shaft.
4. I tried Oilite for the linear bushings, but could not align well enough to prevent binding. Final bushings are teflon filled acetal.
5. The motor is rated 640 in-oz (at 5.5 amps/phase) but I think this means holding torque. The relevant rating is 475 in-oz driving at 100 rpm and 48 volts. 48 volts is much more than needed to push 5.5 amps through the 0.43 ohm winding resistance, but is needed to obtain a sufficient rate-of-change of current (4 mH/phase).
6. The motor has 200 full steps/rev but its driver is set to “microstep = 2”, so 400 pulses are needed per revolution. I run at 1.0 rev/sec, so the pulse source is a maximum of 400/sec. On acceleration, I ramp from 40 to 400 pulse/sec. I think that a 400/sec constant source would be OK (maybe a 555 timer), since there is no problem recovering from a stall when the source remains steady at 400/sec.
7. In operation, 0.7 amp is drawn from the 48 volt source, or about 34 watts. Most of this is accounted for as ohmic loss: 2 phases * 0.43 ohm/phase * (5.6 amp)^2 = 27 watts. The supply for the Arduino is just a 5:1 resistive divider from 48 volts. Wiring for the limit switch should be shielded.
8. The crank length is 2.0 inch, so the push available to the blade is 475 in-oz / (16 oz/lbf * 2 in) = 14.8 lbf. The crank grips the motor shaft with a steel shaft collar. Link pivots are shoulder screws running in bronze bushings, oil lube.
9. The two guide shafts together weigh 3.5 lbm and are nearly centered over the cutting point. Including the saw frame, the total down force at the cut is about 5 lbf. The machine runs without stall at this cut pressure. But if I add the auxiliary 1.8 lbm weight at the front end (increasing the down force at the cut to about 8.5 lbf), stall is a problem. I am using a 14 tooth/inch blade, but did not see much difference with 24/inch. NEMA 34 step motors of twice this rating are available and should allow use of the extra weight. I assume that this would increase the cut rate.
10. Cutting is slow but square. I can do other tasks while the saw runs.

Here is a parts list. I have not been careful about fastener quantities, and I did not even try on small items like standard flat washers, wires, switches, etc.

Update 15 Oct 2019: I have removed several previous updates concerning stall problems with the hacksaw. Problem turned out to be misalignment of the linear bearings causing jam. Every time I loosen/retighten the blade, I have to realign. Four bearing blocks are held by 2 screws each; loosen and retighten these to realign.

Here is the vertical saw that I made from my first bandsaw. Will keep 24t/inch blade on it for fine work.

Update 7 Nov 2019: To get satisfactory operation, I had to make two modifications to my original design. Mod 1 was to relocate the center of crank rotation. This I did by drilling new mounting holes in the motor plate. Now the original dogleg crank is replaced with a straight crank.

With the original crank location, the tangent of the crank angle (relative to the direction of blade frame motion) was too large and reciprocating motion would lock up even without a cutting load.

Mod 2: I still needed additional torque from the motor. Since the motor could supply plenty of power (speed times torque), a gearbox was needed. I bought two involute cutters of China source and made a mandrel to hold them.

Setup for cutting the 48 tooth output gear:

Setup for cutting the 17 tooth pinion:

Both gears grip shafts through shaft collars.

This is the gearbox frame. It is designed for NEMA 34 mounting at both input and output and so can be simply bolted in between motor and crank.

Finally, a different view of the modifications:

I reprogrammed the Arduino chip to give 1100 pulses/sec (was 400) so blade speed is unchanged. With the larger crank torque, I can use the extra weight at the front end of the saw.

Note on saw frame mounting: The Stanley saw frame is bent from an oval steel tube. Bending distorts the oval, making it thicker on the inside of the bend. When the frame is screwed to the two hangar brackets, they will twist as the screws are tightened. This will lock up the linear bearings. I used shim stock at the interface to prevent the twist and preserve free sliding motion.

The poor design choice here was to make the reciprocating frame from the Stanley saw and two hangar brackets (aluminum 3/8 x 1). It takes a lot of attention to keep everything aligned so that the 4 linear bearings do not bind. A better design would be to make this one piece, which would require a 3/8 aluminum plate about 8 x 15 inches. I may pursue this change if I figure out how to profile the aluminum plate.

Update 13 Dec 2019: Problems with binding have continued, so I have replaced the Stanley saw frame with a homemade frame. This keeps the bearings aligned and seems to have solved all problems.

This is working so smoothly that the gear reduction may not be necessary.

Posted in Cutting | Leave a comment

Inefficient Use of Material

I am starting a reel with a winding plate, similar to a “Perfect”. Here are the aluminum parts.

Left: spool, 25 grams from a 233 gram blank (10.7%).
Right: frame, 42 grams from a 391 gram blank (also 10.7%).

Posted in My Reels | Leave a comment

Faceplate for Minilathe

I was working on a new reel design, and realized that several operations would best be done with the parts clamped to a faceplate.

The problem, however, is that on my Minilathe, the carriage cannot travel far enough left to get the toolpost in position. The travel is OK for a part held in a 3 jaw chuck, but not for a part clamped to a thin faceplate. The large handwheel for carriage travel interferes with the box of variable speed drive electronics.
In this picture, you can see that I long ago added an acrylic plate as a chip shield. It also restricts carriage travel, but is no worse than the handwheel.

The first thing that occurred to me was that the electronics box needed to be relocated. It was really stupid of the designer to put it in this position.

The rack and the leadscrew are both capable of moving the carriage far enough left to allow faceplate use, it is only the electronics box restricting travel.

But then I saw that it would be much easier to just move the handwheel.

So I made this extender for the handwheel shaft.

Here the extender is installed.

I made a 4 inch diameter faceplate on my mill.

Here is the faceplate installed and finish turned.

Posted in Fixtures, Turning, Work Holding | 2 Comments

San Juan River

Was at the San Juan yesterday for the first time, did a float trip from Texas Hole to Crusher Hole. That must be the standard route, as there is even a shuttle service to spot the guides’ vehicles and trailers downstream.
It is a tailwater fishery. Water was clear; I have never seen so many trout. They get fat on the abundant midge life coming out of the bottom draw of the dam.

It is bobber fishing over tiny nymphs. I had a “San Juan Slam” of rainbow, cutbow, and brown. Many hookups, many brought in, most of nice size. Guide Scott Warren of Durango CO.

Posted in Fishing | 1 Comment


My family took an Inner Passage cruise. Son-in-Law and I used the Juneau day to go fishing. I did not research what to do ahead of time, we just signed up with the Cruise Line standard. Bear Creek Outfitters and guide Sam did a fine job: supplied waders and rods, and tied all knots, etc. Also carted the special items needed for Alaska fishing: satellite phone, bear spray, firearm.

We flew out on a Beaver to Slocum inlet and then slogged our way up the creek.

Sam gave us Clouser Minnows, but of colors that I never saw in Michigan. A “school” of Cutthroats and Dolly Varden readily took these streamers. A little father upstream were Pink Salmon, and they wanted the Clousers also.

Granddaughter clearly does not believe my lie about a big fish.

Posted in Fishing | Leave a comment

Ball Turning

My “standard” bronze frame reel has two surfaces that are cut to a spherical shape, one is the spool retaining screw head (convex) and the other is the waist on the knob (concave; at least, concave motion of the cutter). When I had a Sherline mill I could use it as a lathe by turning the headstock 90 degrees and mounting the cutter to a rotary table. But now I have the more robust Mini Mill, and axis of the headstock can only be vertical. It is still possible to turn, but it seems quite awkward.

So I sought a way to do the turning on my Mini Lathe. First I bought the standard ball turning tool:

But it failed on both counts; it was not big enough to make an 0.8 inch convex radius or a 1.8 inch concave radius.

On Pinterest, I have been seeing many home made ball turning tools. So I have made my own, of a generally similar design.

The cutter is a 3/8 inch round carbide insert.

Cutting a convex surface – screw head.

Cutting a concave surface – knob.

Posted in Turning | Leave a comment

Tool Post Grinding

I have been selling “Ferrule Shrinking Tools” again; they are used by bamboo rod makers to reduce the diameter of a female nickel silver ferrule and improve the fit. Since I have pretty well satisfied the total rodmaker demand for this tool, it is hard to justify ordering another large batch of parts from a machine shop. So I decided to make a few myself.

The main structural parts are two aluminum disks, 2-5/8 diameter and 3/8 inch thick. The slabs are sawn from a 2-5/8 rod and have to be faced off with a lathe.

I did the facing with my Minilathe because the Sherline lathe does not have enough low speed torque, making the operation very tedious. But the faces of the disks were coming out about .004 inch out of parallel. This would be OK for the purpose, but I felt that I should be able to do better.

It is easy to see why the disks vary in thickness; the height of the three chuck jaws vary by a total of .004 inch.

So the chuck jaws need a little adjustment. But these jaws are hardened and cannot be trimmed with lathe bits that I have. This a a job that calls for a tool post grinder. Not wanting to lay out money for the real thing, I decided to try my Dremel tool.

Instruction on how to do this are on Varmit Al’s Mini Lathe Page. First step is to square up a Dremel-sized stone. I did this with a diamond point tool.

Here is the grinding setup. The chuck jaws are clamping a scrap disk so that the jaw surfaces that I want to trim are at the right radius. The Dremel tool is held in a bracket that I made a long time ago for use with a Sherline mill. All 3 jaw surfaces have been colored with a Magic Marker. I turned the chuck by hand.

Since I have removed the compound assembly in order to mount the Dremel tool, I have to advance the stone using the lathe’s rack drive. A dial indicator shows the carriage axial location. A clamp for the carriage is needed, also described by Varmit Al. I had to take very light cuts, about .0003 inch, or the Dremel tool would stall. But the final result was quite satisfactory.

Posted in Abrading | 2 Comments

New Band Saw

This has been my cutoff saw ever since I started metal working. It is just a handheld bandsaw lashed to a homemade plywood frame.

It has been quite satisfactory on flat bar stock and rod stock less than 1.5 inch diameter.

It fell short, however, when I started making bronze frame reels. The material that I wanted to use (c544) comes only as rod stock, and I needed 3 inch diameter. It was difficult to get a straight cut because the plywood frame was not sufficiently rigid.

When I needed to make a lot of cuts on 2-5/8 inch diameter aluminum rod stock, I decided that it was time to upgrade the saw. This is my new Model 4829 saw from Little Machine Shop.

It appears that the same saw is offered by Grizzly. Grizzly stocks repair parts but LMS does not. I know, because the first time I changed blades I ruined a rubber tire.

LMS sells a 10 tooth/inch blade that is much better for large stock than the 14 tooth/inch blade that comes with the saw.

So what is the deal with the 10 pound weight?

The saw hinge has a built-in spring that lifts the saw. The extra weight partly overcomes the spring torque and makes cutting easier. The manual for the saw says to not push down on the handle, but instead let the saw’s weight provide the cutting force. That is a ridiculous instruction, since the spring entirely overcomes the saw’s weight.

Update 20 July 2018: For dealing with larger diameter round stock, it helps to have a custom vise:

Update 26 Oct 2018: After 11 months of service, the new saw has failed me. It has turned into a blade eating machine. Can no longer make cuts, blades are quickly destroyed. I think that it has to do with the 45 degree twists that the blade has to make coming off the pulley and into the guide rollers. The blades are quickly formed into an arc (in the 0.5 x 0.020 cross section) then wander off to make an arcing cut to the left. They will wedge in the material being cut (2.63 diameter aluminum 6061). I cannot see that anything is misaligned inside.

Update 14 Nov 2018: So here is the problem

The blade is slipping to the side of the front backup bearing and running against the side of the bearing outer race, rather than on its O.D. This quickly deforms the blade.

Here you can see the groove being worn into the bearing block by the back of the blade.

Posted in Cutting | 3 Comments