Ring Roller Dies for forming a curved V shape

For the Path Light project I needed to make curved corner pieces for the roof. The roof panels are roughly triangular and the corners have a roughly 1/4×1/4 angle of “12 oz” copper to cover the corner. The twist is that the roof panels are not straight but instead somewhat concave giving the roof a flare at the bottom. The corner pieces cannot be bent to shape by hand, hammering to flare them takes forever, and using a chisel as a punch to stretch the edges is both slow and prone to cutting through. So a variation on a ring roller is needed. I have a Harbor Freight Ring Roller . It is not exactly a precision instrument but will suffice for this work and save a lot of time vs. building the mechanicals to hold and turn the dies from scratch.

Ring Roller mounted to some wood blocking so it can be secured in a bench vise.

I had run across a youtube video by “Stuff Made Here” https://www.youtube.com/watch?v=WuY2-OrT9ig where he used 3D printed dies to bend some steel parts. Forming the copper strips would require far less stress, so I decided to give it a go. The first goal was to prove that the pieces could be formed as I desired, second would be to improve durability if needed . That could be solved by changing plastics or turning the wheels out of metal on the lathe. The first step was to pull the C-Clips that hold on the factory rollers and get some measurements .

Factory roller pulled and v-rollers fitted

Next the v-roller were designed in Sketchup and 3D printed. They were done in PLA at 50% infill with 3D honeycomb for the infill pattern. No support material was used even with the 45degree overhang (yes there are imperfections at the seam between layers because of this). It did take a few tries to get rollers sized such that they gave the relatively shallow curve that is needed and not split the V in the copper strips when inserted. If I had “properly” 3D modeled everything I could have probable avoided printing a few rejects.

V rollers – Note these had already been used to bend over 30 strips. Minimal wear, mostly from the edges of the strips which had serrations form the snips.

The V-rollers as printed did need a bit of clean up on the inner diameter and the keyway. The adjustable reamer did well for the bore. Yes, I know that you machinist purists will say this is will throw the bore off but I do not have spiral chucking reamers. If my printer had been better dialed in I might not have had as much clean up, but this is close enough. If you don’t have an adjustable reamer sandpaper and a dowel works too. You don’t want too tight a fit as the rollers do interfere with each other and all 3 have to be put on or removed all at once. You can see some faint faceting of the circumference in the wheels and the copper, due to the way Sketchup does circles, but once the part is finished they are completely invisible.

Roller in action bending the v-pieces for the corners of the roof of the lights.

For short run parts such as these the 3D printed parts provide a relatively quick and low cost / effort option to forming the parts. A nice thing was that I could design, kick off the printer and then go back to designing and building the rest of the light components while the part was printing. Also without the right tooling (broaches, which I do not have ) making keyways in aluminum or steel on the lathe is a bit of a pain.

I hope this gives you some food for thought on making 3d printed plastic metal forming tools.

Path Lights Concept

We are on our 4th (or more) generation of path lights for along the front walk and driveway. We live in the country and there is no street lighting. So we need path / landscape lights to provide illumination for safety (of course nothing is level here) and it just looks nice. However the commercial lighting products are pretty cheap and flimsy unless you are willing to spend hundreds of dollars per light. We have the wiring in place and a large 12 v transformer from the initial incandescent path light installation. We had converted to LEDs a while back but want things a bit brighter. Besides the current lights are again falling apart and the glass is breaking either on its own or due to “incidents”.  Solar / battery lights are nowhere near bright enough and the rechargeable battery replacement is an ongoing expense. 

So now it is time for new lights (again). Being recently retired, I have the time to pursue this, after getting more pressing projects out of the way. I started researching looking for designs that i could use as a starting point and to show Teal as I needed her approval. The lights will be mounted on top of 18-24″ posts with some sort of spike anchor base. We started closing in on some promising ideas and then it was time to start making prototypes out of construction paper. This would allow us to see the final size in the garden setting and help inform what the construction process would entail. I had decided to use copper and brass for the shell of the light and a cear “hammered” finish stained glass for the panes. LED light would be used.   The metal would then be powder coated.  The initial  batch for the front walk will be 8  lights.   If this works  out well then 10 or 12 more will be needed for the driveway.

I wanted something of an Craftsman or “English cottage” sort of look with a peaked roof and 4 or 6 sides. So the first things to tackle were over all size and the roof design.

The first prototype looked like this.

First paper and copper foil prototype

It was a start and could be made in copper but the look was not quite right. I wanted a curved / flared roofline, reminiscent of a thatched roof. The copper foil / flashing I had on hand was also flimsier than I wanted. So purchased a roll of 12oz copper flashing to see if that would be thick enough. Then it was time to play with the roof design to see if I could get the look that I wanted and that Teal would agree with. More time was spent with paper, straight edge, compass and french curves. Now I had 3 paper roof prototypes.

Paper Roof Prototypes

We ended up choosing the middle shape. It has the flare I wanted but I was still debating which one. . Teal did not like the slightly upturned corners of the one on the left. So she had the deciding vote. As it turns out fabricating the curved roof in the middle was hard enough.

More materials were ordered. My old stained glass supplier was out of business and I had to try a new one. THis is where I got a few varieties of brass came and the class itself . More metal was ordered from SpeedyMetals for the bottom panel. I ended up using sheet brass for the bottom.

Now was time to start the fabrication and build the tooling. The Press Brake was already done but more was needed.

The prototype light looks like this.

Path light prototype

I’ll cover the construction details in future posts.

Low Cost Press Brake

  • For some upcoming projects I need to neatly bend sheet metals. In the past I had done this for small “one off” items by blaming in a vise or between some pieces of angle iron or using a “tinners pliers” / metal seamer. However now I needed make dozens of bend that are crisp and repeatable. Options ranged from a full size press brake, sheet metal brake (the type you lever up to fold) or kits to build one to go in my 20 ton arbor hydraulic press. All seemed to be a bit overkill and the tool budget had been going wild lately so I needed to make something with “parts on hand” rather than purchase another tool. Besides I am pretty much out of space for another large tool that would normally get infrequent use.

After reading and watching several youtube videos on brake construction, I had a basic design in mind. The goals being ot keep it simple and use materials on hand:

  • ~12″ width / capacity
  • “Air bending” with movable die jaws. WIth Air bending there is not a lower die but the metal is pressed down into the gap between some bars / jaws . It can provide nice crisp bends
  • Hand operated or able to fit in my bench top arbor press for additional force. This avoided having to build an outer frame and the hydraulic press was too slow
  • Fixed blade / upper jaw rather than removable fingers

Digging around in my metal stock / scraps I found some:

  • 1.5×1.5×1/8″ steel tubing which would become the top rail.
  • 1/4 x 1.5″ steel flat stock for the jaws and blade
  • 1/2×5″ steel flat stock for the base
  • 1/2″x 6″ bolts that would be the columns
  • Springs and miscellaneous hardware

The first step was to shape the edge of the blade on the mill. I used a 3/4″ “corncob” or roughing end mill. The head of the Bridgeport mill was inclined to ~37 degrees. You want the blade edge to be less than 90 degrees to allow for some degree of overbending to allow for spring back . This could be done with a hand held grinder and file instead

Next the top bar was cut and the blade was welded to it. I did my best to center the blade and keep it vertical but there was undoubtedly some error. So to drill the holes for the columns the blade was clamped in the vise. I wanted a close sliding fit to minimize error in the bends and did not have stock for bushings on hand. The holes were then reamed for a decent fit on the bolts with an expandable hand reamer.

Reaming the post holes

Next the holes for the column were drilled in the base plate. The digital readout of the mill helps in setting the exact spacing

Each hole was hand tapped with a tap follower held in the drill chuck to ensure that the threads were perfectly aligned and the posts would be vertical .

Hand tapping 1/2×13 with follower.

There were also the jaws to make and the holes for them to be drilled and tapped. However for these I have so called “gun taps” that can be power tapped. This is much easier and faster.

Power tapping for the jaws
Press brake

The brake was now assembled and the initial tests show it can make nice crisp bends. Depth stops were added after this photo to allow for better repeatability of the bends.

Bending the copper corner pieces for the path light roofs

Exterior Light Refinish

Our exterior lights were looking pretty shabby. The original antique / brushed brass was tarnished and dull. 26 years of UV damage had destroyed the original clear coat. There were house paint splatters on them as well.

This is one of the smaller lights, by my shop patio.

image of light before
Before
refurbished light
After

This was also the first major powder coating project after the test pieces. I figured that getting a home improvement project done before tackling the workshop powder coating projects would help win some brownie points and help justify the equipment investment.

The lights are made of thin brass stock. The knurled balls on the corners and for the removable top (for light bulb change) were about 50% frozen on due to rust. They are screwed onto about 1.5″ pieces of threaded rod which were crimped into the corners of the lights. So in the cases where they were stuck the threaded rod came out as well. This allowed fo disassembly of the class panels and removing of the top. From then on it was a matter of starting at the finials and working my way in. Each rounded section had its own nuts and / or coupling.

Some parts, notably the threaded rod pieces needed an electrolysis treatment to remove the rust or corrosion. The parts were then bead blasted to remove all of the old finish and corrosion. Bead blasting for each lamp took about an hour. The small parts were mounted to a piece of masonite so they could be held while blasting. A wire was zig-zagged on the back side to connect all of the parts for grounding while powder coating.

small hardware on board after powder coating
Back side of the small hardware board showing the grounding wire and retaining screws or nuts

This board was simply placed on the bottom of the oven for baking or on the lowest rack. The other parts were hung via stainless wire from oven racks for powder coating and backing. Please remember to wear a dust mask / respirator while powder coating and blasting. The dust is a really nasty lung irritant. I neglected this at first (like you see in many of the youtube videos), and my lungs hurt for 3 days. Now I use my Miller Eclipse P100 respirator religiously when doing the blasting and powder coating.

The many other flat and domed pieces were hung prior to coating and there are a few important tips:

  • Only load 1 or 2 parts at a time prior to spraying the powder. Having a rack full of parts limits your ability to maneuver the gun and get even coverage. Load 1 or 2, spray, load a couple more, etc. Besides the overspray will help finish build on the previous parts.
  • Add a ground clip to each part rather than relying on the hanging wire and rack for grounding. With the lamp parts the center hole is an ideal point to add a clip lead ground. It wont show later, it is easy to coat after clip removal and you can get some leverage from the insides of the domes when removing the clip. I bought a new batch of alligator clip leads for just this purpose.
  • Hang the bigger parts with 2 wires. They WILL swing as you load them into the oven and this will lead to bare spots where the powder is knocked off.
  • Don’t be greedy and overload the rack with lots of parts. They will swing and bang together leading to bare spots and other defects. 4-6 at a time seems to be the limit. 8-10 is just asking for trouble. The bake time on these thin parts with the powder I was using – Eastwood Architectural Bronze is only 23-35 min (20 min after the part hits 400F). So you can still do over 2 batches per hour and have a nice break in between.
  • It is better to transfer full racks of parts in and out of the oven rather than singles and trying to hook them on and off individually (or maybe I am just a superb klutz with limited motor skills).
  • If possible hang the parts with >3″ and preferably 4+” gap from the rack (another variant of “don’t get greedy”). You need the space and don’t want to be spraying downwards through the rack if at all possible.
  • The corollary to this is don’t push your air compressor too hard with long bead blasting sessions. My DeVilbiss “6 HP” 60 gal unit gets really hot after 30 min of continuous run time (which happens while bead blasting even at 50 PSI reduced pressure to not cause the beads to disintegrate). At that point, the air is hot going into and out of the tank. Then the moisture (and oil) make it past the moisture separators and start showing up as discoloration on the parts while blasting. That is the signal to stop and let things cool down. The blast cabinet requires 9-12 CFM @ 50PSI for glass beads continuously while blasting (far more than a spray gun). I don’t have a chiller / condenser for the air lines (another $700+ investment) which would eliminate the condensation issue but not the overheating compressor .
Oven before I switched to a special rack for the panel dividers which are laying on the non-stick foil

Lamp parts hanging after initial fusion of the powder coat
First 2 domes attached Note the spacer bracket between the pipe sections
Base attached to wall bracket
Cushioned pliers to attach nut and threaded rod. Blue tape or electrical tape work well to avoid scratching the parts.
Glass in, ready for the top

I used a variety of textured clear “stained glass” to replaced the old beveled glass panels. I like the look of the textured glass better than the clear. Plus I did damage a few of the beveled glass pieces slightly during disassembly. The glass cuts are simple straight line snaps. However the seedy glass often deviates a bit from straight if the cut is near the bubbles. So a glass grinder or coarse grit diamond stone is a big help. I had not done any stained glass work for a few years. The internet supplier I had used previously used, was out of business and Milwaukee Art Glass is currently only open by appointment (as I found out on arrival). Some of the glass came from Hobby Lobby (which I detest, but is the only other local alternative I could find). Aside from the philosophical and general quality aversions, the small sheets they sell, would only yield 3 panels (and I need 6) per light (and there were 5 lights) with a lot of waste. However, some may be usable as I do the upcoming path light project.

“Large” light completed

Overall, this was a very satisfying project while building a variety of new skills. Replacing these lights with new ones would have cost much more than the powder coating equipment investment. There were 5 lights in all. Next up will be new scratch built path lights for the front sidewalk.

Aside from new tools, material cost for the 5 lights was:

  • Glass bead blast media ~$20 for about 10 lbs consumed
  • 1 lb of Architectural bronze powder coat powder $12.95
  • Glass ~$50 – mixed sources
  • New LED lamps from Menards (FEIT 100w equivalent) $25

Powder Coating Startup

I have been interested in doing powder coating for a number of years but this does entailaddtioinal equipment and space (and spousal buy-in). I have been frustrated with conventional finishes (paint, black oxide, etc.) for my metal working projects. Paint takes a long time to dry, requires multiple coats, decent temps to spray and is not friendly for spraying indoors (and we live in Wisconsin – so this is important for half of the year).

Basic setup L-R, rack for powder coating, oven, powder coat gun in drawer, blast cabinet, dust deputy cyclone

With my recent retirement (yeah!), I was nearing hte completion of redoing my daughter’s kitchen. At this point I was looking to the next few projects (redoing house exterior lights, redoing bridgeport and SBL controls, new exterior path lighting, David’s motorcycle, etc). These would mean either lots of spray painting and future redos or we could get this done with powder coating. For example, I cleaned up and then had professionally powder coated the crash bars/ engine guards for my bike (Honda VTX-1800) which cost $120. Talking with the shop guys at HyTech Powder Coating in Waukesha they said the typical minimum is $80-120 or small projects. While they did a great job, I restarted thinking about a DIY approach and being able to do powder coating as needed.

Powder coating involves:

  • Mechanically cleaning the part to enable good finish adhesion. This typically means abrasive blasting and chemical wipe.
  • Electrostatically spraying a paint powder on the part
  • Baking the coated part to fuse and cure the powder into the final finish

This is basically similar to conventional spray painting. However there are some similarities and differences. :

  • Prep in either case is key. However Powder Coating does seem to place greater emphasis on mechanical bonding and hence the need for abrasive blasting.
  • Spraying the finish requires a special gun but this is in the same price range as a decent HVLP spray gun. Masking of areas not to be painted requires either silicone plugs or special tapes due to the high temperatures used in the curing. I picked up the Eastwood dual voltage powder coat gun.
  • Oven for curing . This can be as small as a toaster oven or a room sized monster used for powder coating car chassis. For the work I will be doing, a home oven is big enough. DO NOT think you can use your regular home food / baking oven for both purposes. Plus you will need additional ventilation for the fumes. Fortunately, used stoves / ovens (esp. wall ovens ) are dirt cheap. You do want a convection oven. Check your local Habitat for Humanity Restore or Craigs List. I got mine at Habitat Restore in Waukesha, WI . A 30″ GE Profile convection oven for $25 (~1% of the original price) . This wall oven did require a basic cabinet in which to reside (which would also get drawers and drawers eventually).
  • Rack to hold the pieces while spraying. I picked up a stand from Eastwood but it was designed for elves or dwarves and had to raise the top by 3 feet to bring it to a workable height.
  • Blast cabinet. This was the most costly item both in initial purchase price and parts for modifications. I bought the Harbor Freight 40 lb blast cabinet (with the 20% off coupon). However doing the necessary mods basically doubled the price in added parts. Check youtube for the many videos on souping it up. Must haves: caulk ALL the joints form the inside with Vulkem – it leaks powder like a sieve, metering return line off of the dump chute to feed the blast gun, lowering the grate inside, adding a separate pressure regulator for the gun, additional baffling for air intake and extraction, dust deputy cyclone ahead of the shop vac, new lighting, wheeled caster base, additional outlets for shop vac and powder coat gun, EZ open wing nuts for the window for replacement / maintenance. Yes this sounds like a lot but think of it as a “partial kit” and go from there. Besides my 4yo grandson – Sawyer had a lot of fun helping. Good thing it was all metric as I taught him to grab the 8, 10mm wrenches vs trying with imperial fractional sizes.

Entryway bench

Teal has wanted a bench by the front door to use when putting on shoes. This would be better than sitting on the stairs.

Last time we were at Kettle Moraine Hardwoods, the local lumber mill, we came across an interesting Elm slab that was about the right size. So it went in the truck with the rest of the lumber.  Our floors are Red Elm, so this complements them nicely.

The slab had cracks in it due to the knots. The largest cracks was filled with some  matching fine sawdust. All of the cracks were then flooded a few times with thin Cyanoacrylate glue.   The slab was then planned and belt sanded.

Teal wanted the front edge left as a live edge, but it still needed some clean up.   This was done with spokeshaves and sandpaper. The edges were rounded over with a block plane.

The top was finished with: two coats of Bona DTS sealer, lightly scraped between coats to get rid of the raised grain fibers, and then sanded with 220 grit. Then, two coats of General Finishes Enduovar Polyurethane Matte lustre finish were applied.  These were all brushed on as this small surface area did not justify breaking out the sprayer.

The legs were a bit of a problem as there is a heating duct coming up right were one of the legs should go.  This lead to an unconventional design idea. The legs would be curved, and support it in the front and a ledger bar screwed to the wall in the back.

The legs are made from two 1/8″ sheets of steel. The template was made fr0m 1/8″ hardboard. The curves were made with the aid of a thin batten (wood strip)  that was bent to shape and traced.  The template was cut out on the bandsaw and the edges sanded smooth.  The steel sheets were stacked and the template clamped to them and the pieces were then cut with the plasma cutter (and it was snowing again that day).

The edges of the steel were ground clean. It is very hard to weld through the plasma cut edge as is as the steel forms a nitride / oxide coating during the plasma cutting.

The bottom end of the “Y” was also spaced apart. This gives the legs a bit of a flair. 1 small scrap of 1/8″ steel was placed in for the first 3 inches or so.

The edges were then TIG welded shut and ground to a nice curve.

The top of the legs is a piece of 1/8×1″ steel  and the feet are 1/4x1x3″ steel. These had the corners rounded, holes drilled in the top pieces and then were also TIG welded onto the legs.

Given that I am new to TIG welding, through the process I stuck the tungsten tip more than once. The TIG welding process has a bit of learning curve, especially for someone that does not chew gum, as that would preclude walking for me.  Managing the torch and then adding the filler rod without getting the electrode contaminated is a bit of a trick and I also roasted my fingers a few times in the process. Once the electrode is contaminated, definitely stop and change it out, trying to make do, just makes a mess and things get way too hot.  I can safely say my tungsten grinding skills are now quite good. Welding will take more practice, but I did get a few really nice beads along the way. I think getting a really nice bead now and then helps suck  you in and makes you forget the frustrations of the learning process.

Afterwards the legs were given one more pass with an 80 grit flap disk and were ready for painting.  One coat of primer and 2 coats of satin black. I like the Rustoleum Pro spray paints for this.  They dry fast and hard and have a decent re-coat window. I have used these for most of my tool  builds & rebuilds.

Crib – Planing and Plasma

Completing the rails

Today’s tasks included completing the front and back top rails and making the bottom plates that fasten the ends to the front and back .

The top rails needed to be transformed into the sleek shapes in the drawing. This meant making the under-hanging lip, adding it to the rail and the generating the sweeping curve of the rail top.

The underhanging lip is a 1/2″ thick semicircular segment which is then glued to the rest of the top rail. To safely and accurately make a shape like this you must start out with a larger piece, shape the edge and then rip it off.  Below you can see the stock being run through on the router table with a 1/2″ radius round over bit. The fence is set flush with the front edge of the router bearing. The fence is needed as the bearing of the router bit will not be landing on un-cut stock  and it provides the needed support for the cutting depth. The feather boards  help to both guide the stock and keep my fingers clear of the spinning router bit.

 

After the profile is cut with the router, it is then ripped off on the table saw.

This profile was the glued on to the top rail. You can see it as the bulge in the lower left of the rail as seen below. The drawing of the rail end was the printed out square with the end and with no perspective in Sketchup. This was then printed life size (which took several tries). The print out was the cut out and placed over the ends of the rails and the outline traced with a  sharpie.

The corners and excess were then saw off on the bandsaw and table saw. At this point the goal is to have a rough approximation of the curve which is ready for hand shaping.   The bandsaw with the table tilted offers a safer alternative to the table saw when there is a small land / support area under the base of the stock and the cut has no support directly underneath it.   Be careful here, greater overhang under the cut can fling the stock or break the blade if you lose control.  In retrospect a feather board behind the blade would have been a good idea here.

Now comes the exercise part.  There was a LOT of hand planing required to get to the final profile. Remember this is Oak.   It took just over an hour to plane the rails and another 1/2 hour of sanding and touch up planing.  I started with a #6 plane set for a fairly aggressive cut.   The shavings piled up quickly and my heart rate rose as well.  I think I was excused from skipping my usual workout on the elliptical (the shirt did not stay on long after this photo).

When planing a curve like this, you start out with the facets cut on the saws approximating the curve. With the plane, you basically bisect each facet, adding new ones and incrementally going from a rough set of angular faces to an ever better approximation of the curve. The sound of the plane and touch of your fingers guides where to make each cut, angling each one differently than the prior one.  After planing, then the sanding starts with 80 grit cloth backed paper on a long stick.

Side brackets

The next step was to start cutting the brackets which hold the end pieces to the legs. I wanted to minimize the visible hardware on the final bed, sacrificing a bit on having more hardware showing on the crib. The ends are held on with 12 gauge steel plate brackets (about 0.1″ thick) . The brackets are cut out with the plasma cutter (much more fun than a saw).

The brackets are then drilled to 1/4″ for the screws and then the locations are marked with a transfer punch. The holes are drilled and brass threaded inserts are screwed into the wood.  Below you can see the frame with the bottom bakets in place and ready to start making the top end brackets.

 

Recumbent trike – idlers and longer ride

The trike is now mostly done, but the drive train was giving me grief. The idlers were not working out and I replace the rear cassette due to skipping.  The previous owner of the components had beat the bike more than I initially thought with damage to the cassette and outer crank chain ring , both of which had to be replaced.  Maybe there was a good reason he cracked his Trek carbon fiber frame?

After having problems with:

  • Metal pulleys as idlers . These were garden tractor style A size v-belt pulleys form Northern Tools. There were metal shavings everywhere and the chain would pop off.
  • Hybrid pulleys. I used sprockets from the old gear cluster and 3D printed nylon sides and spacer to center them over the ball bearings. They held together, but after 5 miles were wobbling badly and the low gears were skipping. The chain would rock off the sprockets, ride on the sides or pop off all together. The nylon withstood a huge amount of force and merely deformed without breaking. However with a 9/32″ wide bearing, there was too much lateral force.
  • Hybrid pulleys versions 2&3. I used PLA for the test models. However, it blew apart under stress in less than 3 miles.  There were problems with geometry and the bearing to sprocket interface.   I was not thrilled with the thought of making new parts out of aluminum to withstand the stress and support 2 bearings per pulley. Nylon flexed too much and I would have needed to bake the spool again due to the humidity. to drive off the absorbed moisture prior to printing.

I then again turned to Google. This lead to Terra Cycle idler pulleys. http://t-cycle.com/idlers-chain-management-c-41/idlers-c-41_9/sport-power-idler-p-134.html They were highly recommended on the AZ website and elsewhere.   I ordered a couple of these and one of thier 28″ chain tubes as an upgrade from the garden hose.  With the research, I also decided to redo the chain tube supports and let the sprockets and chain tube slide freely on the shaft rather than being held side to side in a fixed location.   The resulting setup is seen below.  I used 1/16″ x 3/4″ aluminum stock and bent it to match the chain tube. This was easier to make and the integral chain retention is a huge benefit.

You can also see above, that when the old front idler shattered, the chain was then rubbing on the underside of the handlebars.

With these installed, I dressed for cycling (cycle shorts, not my regular cargo shorts) so as not to appear to be flashing the passersby. The padding of the cycling shorts was not necessary but the snug fit was.  I then did an intermediate length “shake down” ride. Just shy of 20 miles and nothing fell off or bound up. I still need to do some derailleur adjustment (some skipping under hard stress) and front brake fiddling (some squeaking/ rubbing of he disks) and front wheel toe-in adjustment as well as tying down the right front brake cable which is rubbing my calf.

However, it rode very well overall. It was comfortable and fun. There were plenty of interested looks on the bike trail, as you might expect.  Max speed was 26 MPH and average was 12.  Still below my road bike, but I hope with a bit more tuning and remembering to top off the tires, I should break even.  My back, wrists and hands felt much better than on the road bike. Conversely, my shoulders and biceps were a bit stretched as my hands are below and behind my back due to the reach for the handlebars. Legs were pretty good, but my shins are a bit sore tonight.  I expect that there will be some “human break in” for the new riding position.

 

Recumbent trike – Paint

So now I have striped the trike back down to the frame. There was more touch up welding and grinding going on. The Atomic Zombie website motto seems to be “weld, cuss, grind, repeat” and it fit my welding skill level. So after things got close, I fixed any questionable spots that appeared and finished at 80 grit with a flap disk. Then I got out the Bondo to fill the last gaps, divots and make the fillets pretty. This took a few coats not counting the one where I mixed in spot putty rather than hardener  (same size tubes and color – darn). So I also had to clean off the non-curing mix with lots of acetone and paper towels. Then after recoating with a proper mix, it was more sanding filing and then priming. The fillets all look nice now and errant grinder marks have disappeared.

For the paint,  I used Rustoleum self-etching primer and the color coats were Rustoleum Professional High Visibility Yellow.  While a 2 part auto paint or powder coating would be more durable, this paint combination has served me very well on multiple machine tool builds and rebuilds (14″ Radial arm saw, Southbend 13″ lathe, Bridgeport Mill and the CNC router).  Undercarriage parts were painted Gloss black for contrast. Unfortunately a week later it is still rather soft and scratching during assembly.

This paint has a less <1 & >48 hour recoat window. I can generally push this to < 3 hours but at 12-18 you will definitely get blistering much of the time. So the painting was done on a weekend where I had a full day available for prep, prime and paint.

I had some business travel scheduled so this forced a week to allow the paint to harden fully. Plus I was getting started on the new idler pulley design. The original pulleys from Northern tool are a bit narrow and wearing badly.   So I will make some similar to what you can buy from TerraCycle for a fraction of the cost using spare gears and 3D printed parts.

Recumbent Trike – Steering pull solved

The steering pulling to the right bothered me.  The 1.5 to 2 degree difference in the caster angles was there and it had to be the clue I needed. However, the angle readings were varying on each try but overall the right wheel had less caster than the left. I really did not want to cut off the arm and realign the whole thing again as that was not easy the first time. Searching at lunch for caster steering problems, I came across a Quora post that said if the caster is uneven the car will pull towards the side with less caster. https://www.quora.com/How-caster-angle-affects-to-the-vehicle-dynamics. This was the answer I needed!

So, I decided to cut the front tube perpendicular to the long axis. This would give me the ability to adjust the caster without messing up the camber.  WIth a 2 degree angle error and 1.5″ tubing I would need to have a slot 0.078″ or just over 1/16″ . One slice with a cut-off blade in the angle grinder would be about right.  I placed a straight edge against the back of the spindle tubes and marked across the arm.

Measuring the angle exactly is very hard. So I enlisted my wife’s help and used a pair of 2 foot long winding sticks placed against the front edges of the steering bearing cups. Sighting along these from the side I could see the angles were plainly different.  Winding sticks are used in woodworking when flattening benches, aligning jointer beds and other areas where you are looking remove twist in a surface.  Below, I am holding a ruler in the position where each of the winding sticks was placed. 

After slicing through the top and then the sides, a bit of downward pressure closed the gap.  A quick check with the winding sticks again showed that they were now almost exactly parallel – the difference was gone! Now it was time for a couple of hot tack welds and a test ride. You can see the gap at the bottom, where the cut is folded and that it is tight at the top.

This did the trick. I can now go 100-200 feet hands free with very little drift. I will still tighten up the steering a bit to add some friction as bumps will cause a deviation (yes more caster would probably help but I am going to leave this one alone now.  So it was now time to weld over the length of the joint and start to grind flush. It gets dark early now so I was not able to completely finish and it will be easier when the bike is disassembled for painting.

Now I need to figure out the rear chain skipping…