MtnDon's 10th Annual Old Lumens Challenge Entry - Hand Made Category

Here we go, off to a quiet start.

More to come as things take shape or at least I try to make an idea take shape.

Some days one must simply give into a crazy notion and go ahead just for the fun of it….

It’s a triangular scrap of glued-up wood, a cut-off from some old project, a 2032 cell and a 5mm led from rngwn in 1900K and 95+ CRI. A couple of minutes with a Dremel tool cutting a slot for the cell and a little sanding and we have a hand-made light.

Perhaps one of those parts may make it into my completed entry?

I love a good chuckle. It makes light - who knows if my entry will ever get that far this year!

Images are clickable to enlarge them….

My idea is more along the lines of a lantern than a conventional flashlight. But even then my idea differs from what would be expected for a typical lantern. Lanterns generally spill light in a circular pattern, all around the globe. The plan here is to have the light aimed in one direction, but not so focused as would be with reflectors or optics.

I have built or modded lights with single emitters, quite a few triples and some quads, a couple of six emitter lights and one twelve emitter light. As far as the number of led’s this will beat all previous efforts in the number of emitters, if not the number of lumens output or distance of throw. Forty emitters will be used.

The emitters are 5mm leds purchased form rngwn. They are 1900 K high CRI. They are nominally rated at 2.7 volts and 20 mA. Rangwan has a page where these may be bought. djozz tested some of this type in 3400 K and there is a report with graphs re lumens, amps, CRI and so on.

Here’s the bag they came in…

The leds will be mounted on a piece of single sided prototype perf board, 6.5 cm x 14.5 cm (2.5 inch x 5.75 inch). The copper traces are laid out to be continuous down the length but they do not connect to each other.

After a bit of diddling about with my mental concept I came up with a 5 x 8 grid of leds. The size grew out of making something of a physical size that I liked and was compatible with some of the materials I have already, as well as compromises between driver maximum amps and how that would work with the leds. The tests djozz performed indicates these leds can take much more current than 20 m, but like most or all leds there is a point when more heat is made than light. I hope I have reached a happy medium. I anticipate that mostly the light will be run on a low setting, likely no more than 20 mA per led.

Here is a representation of the 5 x 8 grid. The leds will all be wired in parallel, so there will be some jumpers at the ends of the rows.

And the other side of the perf board with led positions marked.

A close up of the perf holes which are approx 1mm in diameter on the standard 2.54mm spacing. The long leg is +.

The first row of leds are positioned to get a “feel” of what this may be.

Here’s the row with the wires bent to hold the leds in place.

Next, I need to get out to the workshop and turn on the soldering station and make a commitment.

Yes, I am excited! But we’re going away so there will be no updates for a week or more.

I’m excited, too! Interested in what you come up with for the housing.

Looking good so far Don. See you in a week!

I laid out rows of the 5mm LED’s 8 at a time. After applying liquid flux with a flux pen each wire leg of the LED was soldered to it’s copper trace. Before soldering I checked each row by jumping a 2032 cell to the pair of traces to check the polarity.
Then on to the next row of 8. After the LED’s were all soldered in place, short copper wires were used to form jumpers between all the positive lines and all the negative lines.

A further check with a 2032 cell has them all light up.

Now to build something around that.

Very cool! :sunglasses:

Looking good Don !

Today I did some wood stuff.

This is Wenge, (WHEN-ghay). It is quite hard and can be splintery. I’ve never used it before, in part because of its splintery nature and that it is expensive. I got a piece of it at a sale price a while back. So far I haven’t had any splinter issues, but then I have been watchful. I re-sawed this from some thicker stock. These pieces are just over 1/4” thick. That is a Samsung 35E 18650 cell for size comparison. This cell was purchased some time ago for a project that did not happen and has solder tabs spot welded to the terminal ends. I have two and may decide to use them both in a parallel configuration.

Wenge, like a few other exotic kinds of wood, blunts tool edges quicker than some woods. I decided it might be a good idea to cut the notches with a new 1/4” solid carbide bit in my router, rather than an older and much-used bit.

Those are called finger joints or box joints.

These will be glued together to make a box, which will hold a 18650 cell plus some other components. This is still a concept that will likely undergo some changes.

At present, the sides are 1-1/2” high. They will probably be cut down, but I am not certain how much yet. The corners will be glued with hide glue and once the glue has dried the fingertips will be sanded off. That is something I strive to NOT do to my own fingertips. That hurts.

I’ll see how this works and what it looks like with other parts I need to make to bring this to fruition.

Hey Don you must have a little practice with joinery! I could ruin a lot of wood trying to get that right.

I have had a little practice. :wink:

Nice timber the grain makes the surface look like is has ridges.

Today I glued the 4 sides together. For this, I elected to use a thin CA glue. I like Starbond CA. They have several viscosities and a few varieties in shades of brown plus black, plus accelerator and a debonder for those accidents that do occur. They can be handy for filling small knots. (Hint: Store your CA in the refrigerator. That greatly extends the shelf life.)

I could also have used a slow-setting glue such as the brown hide glue I like, but I didn’t want to wait until it was dry. That would have been tomorrow. So I first dry assembled the 4 sides and applied 2 clamps. Then I applied a few drops of thin CAto all the joints. This is as thin as water. The fingers fit tightly with no gaps, and the thin CA gets wicked into all the spaces. I hit it with the accelerator and after that had evaporated repeated a few drops of CA and more accelerator.

The pads on the clamp jaws are a HDPE. CA does not adhere to HDPE or LDPE which is handy at times. I use wax paper under the work pieces as that can help prevent stray drips from gluing your work to the workbench.

The sides of the glued up box were sanded smooth using the bench belt sander and 120, 180 and 240 grit belts.

I rummaged thru my box of padauk scraps and found a piece that was almost the needed size. A little sanding made it fit.

That is all for now. Thanks for looking. Comments or questions are welcome.


My plan is to power the led’s with a Nanjg driver using an attiny13a MCU and 4 of the 7135 regulators. The maximum output of 1400 mA split 40 ways will provide a maximum 35 mA current to each led. The driver will be flashed with Crescendo firmware.

The power will come from a single 18650 li-ion cell (because I have several). A single 14500 because it makes it possible to make this light smaller, or two because that is still small enough but has double capacity. There will also be an onboard charger with a USB-C port.

On to the next stage. I need to make a window frame, similar to what I made a few years ago for the shoji style lantern, only smaller and only one of them.

So, another dive into the wood scraps box came up with some padauk. I ripped them thinner, then cut a few grooves with the router.

Those were then ripped again to make stiles and rails for the frame.

Here they are arranged as they will be finally glued together.

I trimmed some ends and decided to assembl the outer frame one glue joint at a time. The first…

The red 90 degree angle makes assembly easier and more accuracy. I decided these joints would be done with CA. I clamped one stick (rail) to the worktable and used the guide block to align the other stick (stile) to make the 90 degree corner, and then applied a drop of CA and hit it with accelerator. (The stile is the vertical part of a window frame, the rail the horizontal stick.)

Then the second corner…

Those corners came out good…

The second rail slips right in between the ends of the two stiles.

I clamped and then glued the final 2 corners.

I used the disc sander with 220 grit sandpaper to sand the center muntin to the desired length.

It fits “just right”.

The two short pieces, (horizontal muntins), were cut and sanded to length. And I missed taking a photo of the frame in the final glued up and kind of rough and dirty looking condition. After sanding on the bench belt sander with successively finer grades of paper, 120, 180, 240, this is what the frame looks like. It still needs final edge and corner sanding with hand held sand paper.

That is all for now. Thanks for looking.

Nice work, Don, and the beautiful wood selection again! I love those plastic squares. As near as I can tell there are two main manufacturers of those and I have sets from each - remarkably every one of them is actually square-square. Reminds me, I need to trim off the long ends on a set of those minis to make them 2”x2”.

Nice pictures and documentation. :+1: Looking good so far.

So with the window frame with grid more or less complete I need to make the body of the light. Sort of a case. This will be what the padauk window frame fits into. I am using more of the wenge for this. I began with 3/4” stock. It will be trimmed down later, but at this point I am not certain what the final dimensions will be.

I am using finger joints for corners again. The same 1/4” bit, just deeper cuts to suit the thicker stock.

The corners fit together pretty well.

Next, we have the two sides and top piece with the corner joints fitted together. The padauk window frame is just sitting on top, approximately how it will be positioned. The previously made battery box is shown, as well as a piece of wenge that may get used, all while I contemplate my next steps.

I rip sawed the 3/4 inch thick stock to about a 1/2 inch thickness.

I cut a rabbet (notch) along the long edge of the wenge where the padauk window frame will be inserted. I used the same 1/4” bit I used for the cutting of the fingers.

The router bit leaves a rounded shape to the rabbet end. That gets cut out with a sharp chisel. In the close up view, looking at the side of the finger jointed end, the pattern of the wenge gives it a camouflage appearance. It is slightly difficult to see detail.

The shot with the chisel is worse. My focus was off. Sorry about that. The shiny face of the chisel adds to the confusion to my eye. But I did get the corners cut out square.

That is all for now. Thanks for having a look…

With the three wenge body pieces assembled I found, as expected, the cut recess not quite wide enough to allow the padauk window frame to be inserted. It was just off by sightly more than 1/16”. I find it better to have to trim than to want to add material. :wink: I trimmed the width using the table saw, a little bit off each side rail.

I trimmed down the thickness of the 3/4” wenge pieces. The following photo shows the fit and also shows that a small wedge of wenge chipped out of the top rail while cutting the recess. I’ll see how to deal with that later. The fingers will be trimmed later. Amputated?

I will be sanding the assembled front face. I clamped the assembly together.

Here’s the body assembly face down on the bench belt sander. This is a staged photo with the sander not running.

Just an FYI… Sanding belts clog with sanded wood particles. A rubber eraser bar cleans the belt and extends the belt life. Sander is running in this next photo.

After sanding with 120 and then a 180 belt the surfaces are smoother and at the same level. The chipped out section is less noticeable, especially after the orange sanding dust was blow away.

I hope to get some more done today if I can.