Page dedicated to ongoing work and minor projects
The vertical supports for the launchers look a bit out of proportion, but with anything smaller diameter the LEDs would not have fit inside. Even with these, it was a tight fit with all the wires, especially the ones attached to the brass tubes for touch sensing.
This is a part of the project I was not looking forward to. It turned out to be a half-day project, but it turned out to be relatively stress-free. The key was to make a jig for each part in the sequence. The first jig was to hold the cut-off tool so I could rotate the brass tubing for a clean cut. I cleaned up each piece, both longer and shorter, on the lathe using sandpaper in 2 grades. The second key was another jig to hold each short piece for cross-drilling. That required a fitted dowel inside the brass tube to prevent the tube from collapsing, and allowing a clean hole. I then used that same dowel to hold each assembled piece for soldering. I heated each assembly with a blow torch before applying solder to each quadrant. The last part was to sand and polish each assembly after it had cooled.
Each assembly will be mounted in the “launcher” pattern with a programmable LED at the junction. The light will reflect inside the horizontal tube, producing a sequential light show.
I’m learning Arduino programming. The circuit and software in the image above is the interim result of a week’s work. The microcontroller is an older Arduino Nano. The connector is a bit flaky, so I ordered several more. In that process, I found that the Raspberry Pi foundation now also has a microcontroller, the Pico. That one runs on C or Python, the Arduino on a proprietary language modelled on C. If I didn’t already have the Arduino and was starting from scratch, I would probably have used the Pico. However, the ecosystem, particularly add-in libraries, for Arduino is much more mature.
I now have 5 Arduino Nanos and an Uno, and 4 Raspberry Pi Picos. Given that they run between $6.60 and $16, it’s not a big deal if I damage one. So far they’ve proven to be robust.
In the photo above I’m doing a battery run-down test. The small flashlight/power cell is rated at 2600 mAh. It’s been running for almost 9 hours. At this rate a fairly standard cell phone battery bank could run this circuit for several days. If the test runs successfully for 24 hours, I’ll get a good battery bank and power this circuit with that, not worrying about a wall power supply.
Labour Day was a very productive day. The most immediately important task was to laser-engrave my favourite urn box for a friend of Jayne. The second-most immediate was making the trinket tray in the photo above. The design file is from Beki at Vectric. It was last month’s Monthly Project file. I ran it unmodified apart from the dedication on the base.
I also successfully uploaded an Arduino sample file to the breadboard circuit for Harsh Mistress. I had to revert to my old Surface Pro 3, since my Surface Book wouldn’t recognize the Arduino Nano. The sample file works great! Now I can start modifying it to give the patterns I want for Harsh Mistress.
Here is the result of several iterations and days’ worth of work on this concept. Mare Crisium on Luna’s north-east quadrant is the principal setting of Robert A. Heinlein’s classic novel “The Moon is a Harsh Mistress”. This sculpture will be based on a Lunar Reconnaissance Orbiter image with LEDs highlighting places mentioned in the novel. The LED locations are a best guess based on repeated careful readings of the source material. The sculpture will be about 970 x 570 mm, designed for wall mounting.
The next step in the process is diving into Arduino programming for the LEDs.
By the way, I have a Patreon profile, which is a simple way to support my efforts.
For several years I’ve been using a cardboard box for wood scraps. Meanwhile, I had some leftover flooring taking up space. Both situations are now resolved. The bin has skids on the bottom, and is quite heavy. It’s not something one would try to pick up when full.
My challenges with Solaris continue. I’ve spent some time almost every day working on it, and am now getting close to calling it “officially done” whether it works or not.
I did achieve another goal though. It occurred to me that the “Oh My Darling!” and “Cold Trap” sculptures would work just fine as ceiling lighting fixtures. I modified their bases, made a support frame, and suspended each in turn. “Cold Trap” is now re-posted on Etsy as a Chandelier. “Oh My Darling!” is the new light in my office.
Solaris is “done”. After quite a few disassembly/reassembly cycles the parts work fairly well together. I had to sand the gear wheels again, but even so there is still too much friction and stiction. It free-runs fairly well, but adding the escapement and hour/minute wheels causes it to stop at certain points. I know that part of the problem is that the hour/minute wheels are too tight, and the main drive gear appears to be overloaded. I’m also pretty sure that the weights should be a little weightier.
My conclusion after uncounted hours of fiddling and fettling is that designers and makers of wooden clocks have more ingenuity and patience than I do!
A week or so ago I was asked by my neighbour if I could make a rustic bench. The wood was reclaimed from an old barn, and must be well over 100 years old. There were mortises cut into the beams, which I made the undersides. Smaller wood timbers had tenons cut at the ends, which fit into the mortises very well. Being able to repurpose the work of craftsmen working in the late 1800s in Lanark County was rather inspiring. I also reused the dowels, which only required cutting to length and bevelling with a hand plane. The original barn builders would have used pull shaves. In order to protect the wood but keep as much of the original character as possible, I only did a light sanding, then applied rattle-can polyurethane.
Solaris is partly assembled. There are some alignment issues, but the main problem is the idler wheels for the weight lines. I turned them from dowel stock. The instructions suggest patio door rollers. I will have to purchase a set and retrofit them. The friction is too great with just the dowels, even though I waxed them. Once that is done and the main bolts are tightened the mechanism should run more freely – hopefully well enough for this to be a clock, rather than a semi-mobile sculpture.
After 3 months hiatus, I’m working on the Solaris clock again. I had cast the epoxy into the acrylic tube for the weights months ago. Clearly using the dregs of my bulk epoxy supply wasn’t the best idea, since the bottom of the tube had not set fully. Cutting the sections made a royal mess of my mitre saw, which took quite a while to clean up. I also had to make a new pair of bottom supports because I considered the originals too fragile for the weight.
The left weight in the photo has the bottom at the top, showing the Walnut plug covering the knot in the fishing line.
This presentation box is for an old friend. He will mount his father’s and grandfather’s military service medals in it.
This was a fun little project. The client wanted to permanently mount a gift hammer. His wife saw the gift box I made for the live-edge furniture and liked the knot effect. At KJP I found a perfect piece of Walnut, and purchased the two feet with the knot. The greenish epoxy holds the hammer securely.
This project took longer than I thought it should. Now that it’s done, it satisfies all my requirements. The hinges are repurposed from Xplornet satellite dishes, the aluminium tubes were once parts of a vertical-axis wind turbine (my University degree project), and the pulley was intended to control a storage unit in the shop. Even the wood was left over from other projects. The only purchased component is the sun screen sail itself.
The build log for the live edge furniture project has been moved to a dedicated page.
The baby gate is done and delivered, so back to Solaris. I had a clear 50 mm diameter Acrylic tube that’s been moving with me for decades. The ChemTec epoxy Part A had crystallized again, but a half hour in a warm water bath liquefied it again. At first I mixed a small batch to test the hot melt glued end block I had put on the tube. There were no leaks, so I mixed a larger batch and poured it in.
From prior experience with shallow pours, I delayed adding the tints, because they tend to settle over a few hours. So I added the remaining Part B to the Part A jug, mixed thoroughly, and poured it in to the top. There were of course a lot of entrained air bubbles. Then I added the first tint colour. It started to settle, but not very quickly. A few hours later I added a second colour, then eventually three more. The epoxy is of course slowly setting, even though the mixture was about 4:1 instead of the specified 2:1. (Clearly I haven’t been measuring the ratios properly, since I did it by weight instead of volume.)
This morning the original colour had only settled about half way down, and very tenuously at that. Only the top has a strong tint. Lesson learned – do staged pours if filling a column. However, the air bubbles never cleared from the lower part of the column, so the end result will be a clear bubbled cylinder for one weight, and various shades in the other.
The upper image is more or less how it would appear to an adult from the top of the stairs when it’s closed. The wall will be at the left. The lower image is the opposite view, from the stairs when it’s closed. The wall will be at the right and the stair banister at the left. When not in use, a magnet at the end of the latch handle would engage with a steel plate on the wall.
Aaaron got renamed Walter Miles by the time he entered the world yesterday. The gate will be ready for him. After working most afternoons this week on it, all that remains is the hinge post and the latch post. The Maple for those is at the top of the picture.
The Walnut trim wasn’t planned. It hides various “oopses”. I clearly need practice with half lap joints. Mind you, these are the first ones I’ve made… probably ever. The depths are near-perfect, but the lap dimensions don’t match as well as they should on the diagonal ones.
The latch is a steel rod and the handle for it is a repurposed bottle stopper. The magnet at the end will hold against a wall-mounted steel plate when the gate is open.
Our friends Aaron and Cory are having their baby Aaaron today. They asked me to make a gate for the top of their stairs, in anticipation of an active child. They have several cats so they of course need open access through the gate. Aaron took the measurements and made a sketch. Yesterday I planed some cedar left over from my Kitchissippi chairs to a consistent thickness. Today I cut out the frame and slat pieces. As you can see, I cut the central diagonal slat slightly short, so I had to stretch it. I used a scrap of Walnut. It’s a feature! Tomorrow I plan to cut the half lap corner joints and glue together the first half of the frame. The slats will be centred in the frame, so they will be cut and lapped next. That’s where the “feature” comes in!
There’s been a lot of work between the last Solaris progress photo and this one. Most of it involved sanding, not one of my favourite activities. Many of the parts have been assembled, such as the lantern and frames. In the lower photo is the completed pendulum. It’s a little askew, but only the distance of a pendulum’s centre of mass to the pivot is important, not its’ straightness or mass.
Last Thursday evening I participated in the Kingston Wood Artisans’ Zoom meeting. The featured speaker was David Bedrosian, a very well known engineer and woodworker based in Waterloo. His talk was about the use of machinists’ gauges in the woodworking shop. One in particular got my attention – using a cheap digital caliper as a depth gauge on a drill press. I happened to have a no-name caliper I had never used. I made a wooden block to act as a carrier clamped to the spindle. I cut off the unused jaw of the caliper and chiselled a recess in the wood block to support the remaining jaw. The body of the caliper is attached to the drill press frame with a 3/4″ super magnet, and the jaw is epoxied into the wood block.
It works great. My concern that the magnet wouldn’t hold the caliper body firmly enough appears to have been a non-issue. This will vastly simplify accurate depth drilling. Thanks David!
Today’s machining took about 3.5 hours. These were all the hardwood half and quarter inch thick parts. One of the clock hands broke while I was removing flash. It’s in the small table clamp. Next is many hours of detailed sanding.
About 6 hours of machine time, I have the plywood and 3/4″ stock parts cut. They will require considerable hand work to clean up, sand, and drill. I had the CNC make marker spots for the drill locations, but for the most part did not have it drill the holes except where locations are critical. Tomorrow I’ll do the remaining parts. Then over the coming week I’ll do the manual work while power is at mid and peak rates.
Several months ago I purchased Clayton Boyer’s plans for his “Solaris” clock. It’s a followup to the “Woodworkers’ Hygrometer” that I built in 2020. I got the metal parts a couple of weeks ago, and this morning I flattened the hardwood planks. Tomorrow I’ll dimension them, then start the CNC work over the weekend. This is my first clock build.
This is the view today from my Door Into Summer. Soon I will be able to sit on the adjacent deck and enjoy the green version of this.
Scraps can find unexpected value. I’ve been redesigning the CNC portion of Fiddle #9 and realized that the wood blanks need to have a consistent reference height. That’s a problem, since they are both tapered side to side. So I’ve glued together most of the thick maple off-cuts I had lying around. I’ll run them through the jointer/planer tomorrow to get the piece to a consistent 25 or so mm thickness, then cut out a series of reference blocks. I’ll glue those to the edges of the blanks, which will not only make mounting easier, but give me a consistent height reference outside the cut areas.
Terry’s salt cellar is ready. We decided to apply the Livos finish to the inside as well, since it is food-safe. The inside has 3 coats, the outside has 4. A final buffing this morning and it’s ready for use!
A friend who now lives far away asked if I could repurpose an earlier piece as a salt cellar. She uses high-quality flake salt in her gourmet cooking and needs a dry but attractive storage vessel. I designed this piece based on the Kdat series. I enlarged the lower cavity to 150 ml with a smaller upper cavity. This determined the overall size, which is about 190 L x 90 W x 100 H mm. It has a single outer coat of Livos and will get a couple more.
Kdat2 is done and looking for a good home. Overall size is 200 mm long, 75 mm at the widest, 90 mm at the thickest. Weight is 360 g. There are 3 coats of coloured Livos finishing oil, and 2 clear coats. The magnetic latch works very well. Both the inside and outside have sensual shapes that are clearer to the hands than to the eyes. The inner shapes follow the lines of the outsides, but have their own distinct characters. The Livos finish is EU Food Safe, and will develop a nice patina through handling over the years.
Jayne was very generous on my birthday. Several items from my Lee Valley wish list arrived. This miniature vise was on the list as a potential “self-indulgence”. After 2 days work making the maple jaws and fitting everything, it’s now installed at the end of my large assembly table. For scale, the jaws are 20 cm in length.
This one will be called “Kdat2”. The Kdatlyno are an alien race in Larry Niven’s “Known Space” series, known for their touch sculptures.
The inset is a knob which actuates a magnetic latch. Although it would be possible to open the sculpture without using the knob, it would be difficult and likely to damage the seam. I’ve applied a first coat of coloured Livos finishing oil. It showed some areas which need some additional sanding, which I will tackle tomorrow. Then I’ll apply several more coats of finishing oil.
I tried to sell my old U-shaped office desk on Kijiji. Out of 100+ views, there were 4 responses. 3 dropped out when I told them where I am. The 4th low-balled me by half of my already low asking price. So yesterday Jayne helped me bring it to the shop where I partially disassembled it and rebuilt it. Here’s the result. It has wheels and bracing, and 2 shelves. The upper one uses the hardware from the keyboard shelf and the base of the U. The other is the cross brace from the left U arm. There are only 4 panels left over, which I will use for other shop projects. And yes, it’s heavy and solid!
I was sufficiently pleased with my 3D test item that I made 2 more. The red one was #2, the blue one #3. The red one is a minor modification of the first one, with additional bumps and protrusions. #3 is a separate design from a similar outline. It has a dorsal ridge and a ventral groove. They both are very tactile. Both are held together by embedded magnets and have a small cavity. The blue one will be listed on Etsy and in the online store here.
This was a project for my friends at Nine45 Designs. It’s a commission with a very general guideline of “Cribbage board with PEI”. I took an existing PEI GIS file and generated the track markers in Vectric VCarve Pro, then used the “Copy along vectors” tool to generate the peg holes. Mitchell and Liana will pour green epoxy in the PEI area, sand smooth, and finish with their special Board Butter. I am sure the new owner will be thrilled.
Jayne reminded me a few days ago that her mother’s birthday is imminent. So last night I designed a simple 3D form, almost on a whim. This morning I put an offcut from the thermo-kinetic sculpture on the CNC and cut out the pieces. I had to resize to fit the materials, but otherwise there was no problem. At least, until I noticed that the 3D Roughing on the second part was cutting through to the inside cavity. I finished everything anyway, then took a while to analyze the error. I suspect that I didn’t fully recalculate all the toolpaths before saving them. In any case, the second attempt was successful.
There was quite a bit of sanding, some of it by hand. The hollows are still not as smooth as the convex areas. In any case, there are now 2 coats of Livos oil on the “good” pieces. The “mistake” piece will join my learning gallery.
This is a set of files from Michael Tyler on the Vectric site. It took most of 2 afternoons to cut the parts. The wheels and features are two-sided. I used pieces of pallet wood that my neighbour Grant gave me earlier this year. I have no idea what the woods are. Being pallet wood, there are many imperfections, which I left where they occurred. The wood is unfinished. My intent is for Grant or Melissa to finish them if they choose, then give them to their 2 kids as presents from all of us.
Christmas presents! Both designs are courtesy of Vectric. The “Sporktulas” are by Tim Sway, and the “Catch-all Tray” is by Todd, a regular contributor to the “Project of the Month” at Vectric. With those complete, my Christmas “shopping” is almost done.
Here is the parent, and the 4 descendants. Each of the smaller Cold Traps is numbered and signed. I think the official name of this limited edition is “Cold Trap DISPSY x/4” where x<=4.
The limited edition of the smaller Cold Trap sculpture, “DISPSY”, is ready. Each has been tested and set for random light play. The colouring on each is different in detail, but all with the concept of yellow for areas that might be permanently in the sun, blue for possible permanently shaded areas holding water. Green and orange are randomly applied, signifying possible mineral deposits. Tomorrow I’ll take presentation photos.
In this morning’s news was an item that the presence of solid CO2 has been confirmed in some extreme cold trap areas, mainly in the Amundsen and Haworth craters. Both are featured on these sculptures. Amundsen is just to the right of the green area in the lower right image above.
Livos finishing oil makes the Walnut quite a bit darker, producing a rich contrast with the Maple. These were quite a bit more work that I anticipated, and they could certainly be a better fit, but overall I’m satisfied.
The second iteration is much better fitted. This time I resized the drawer outlines instead of offsetting them. And the resize was a fraction of a millimetre inward. Then I sanded as lightly as possible, just to remove tool marks. This one is usable, ready for the Livos finish.
The final epoxy coat went on the DISPSYs today, along with random colour inclusions. There’s no obvious difference from the previous photo. After that was done I tackled a small project based on a recent post on the Vectric forum. Member “zaxis” posted about CNC versions of band saw boxes that allow design options not otherwise possible. I examined her/his file, then designed my own. I missed a critical factor though. Doing a pattern offset doesn’t accurately reproduce the shape, it loosens the curves. I didn’t realize that until I fitted the drawers in their cavities. The fits are very loose. Maybe I can cushion them with felt runners. but the shape still isn’t exact.
The other issue is that a single box took all afternoon. This is not something I plan to make in volume!
The second black tinted epoxy layer made a vast difference. I added some red sparkle as well, which will give the finish a texture and red highlights. The third layer will have random coloured areas in some of the craters and highlands.
Each light cylinder has 1250 mm of colour-chasing LED ribbon, activated by a controller in the base. Each has been tested before final assembly.
The first layer of black tinted epoxy always soaks in to the point where you think that maybe more tint would have helped. It probably wouldn’t. The variations will even out with additional coats. I had hoped to apply a second coat a day later, but the surfaces were still quite tacky. As it turned out, Tuesday was very busy with other priorities. The original “Cold Trap” will be on display at the Nine 45 booth at the Toronto Cottage Show / Seasons Christmas Show at the International Centre 12-14 November.
I decided to make some smaller versions of “Cold Trap”. The image is from Clementine, the same as for “Oh My Darling!” but otherwise these will be a limited edition of the “Cold Trap” style. The bases are 350 mm diameter. Each will have a single light cylinder with a total sculpture height of 490 mm. Yes, they will be “49ers”. Over the past few days I’ve machined the undersides, then the top sides in 3D. I received the LEDs and other components this past week, so now have everything needed to complete this limited edition. Each of the 4 will be numbered and signed. They will be priced very attractively, just in time for Solstice/Saturnalia/Christmas/Hannukah/Kwanzaa/Festivus/Why not? shopping!
The “Erbschaft/Nahrung” Thermo-Kinetic sculpture project has moved to a dedicated page here.
Jayne asked for this for her birthday. I fit it in while waiting for epoxy to cure on the sculpture parts. The top and base were cut out on the CNC, as were the holes for the dowels. The shaft and details were done on my old machinist’s lathe. The top and bottom are Maple, the shaft is Roasted Maple. The shaft has a buffed-on layer of turner’s polish, then everything had a single coat of Livos Natural Oil finish.
Yes, she liked it!
A few days ago my neighbour Albert called to tell me that the Deli he owns with his wife Sandy is to be used as a movie location, and if I could make a 70’s style sign. He provided me with a couple of photos. I took them and created my own version using Affinity Photo and VCarve Pro. This is the result.
There are now 4 urn (or memory) boxes available. The green tinted one is at a retail establishment. I have added one with a blue tint. They are all in the 175-225 lb capacity (for ashes from a person). I am pursuing an opportunity with a retirees organization which will hopefully lead to ongoing work.
I keep making more urn boxes. These are all in the 175-225 lb capacity range. They all have magnetic releases. The one in the upper left has two magnets hidden in the front, they are released with anything ferrous. Canadian quarters work well. There is a retaining lip at the other side. The other two have four magnets each. The one in the front has a light green tint added to the finishing oil. The Spalted Maple took up the tint more than the Sugar Maple of the walls. The box in the back left has a top of lovely Tiger Maple. The corners are either FingerMaker CNC-cut finger joints or mitre joints reinforced with thin Walnut splines.
With both “Cold Trap” and “Oh My Darling” finished, I have gone back to a few smaller projects. Some time ago I did a proof of concept of a Tensegrity sculpture I called “Cat Tails” at Jayne’s suggestion. This morning I completed a nicer version, which I call “Cat Tails 2”. The bottom block is leftover Ash. The top block is leftover epoxy from the last few coats on the two LED sculptures. It is quite massive, so I was concerned about finding fishing line strong enough to support the weight as well as the tensional loads. The best I found is rated at 30 lb (14 kg). So far nothing has broken. The guitar tuners make adjustments easy.
Over time the epoxy “cat tails” have distorted slightly, putting the top off-centre. I doubt I’ll use epoxy for stressed components in any ongoing work.
Note the fly for scale.