Alright. It looks like it is finished. Unless I implement a mod!

I tore it down, cleaned up the mug shell a little and took some more photos as I finished details and reassembled.

I added a spacer to the bottom of the lower module insert. This was cut from a piece of scrap walnut and fixed in place with some 3M double sided tape. The tape has a little give and will squish down when the module is secured with screws.

Pacman is gonna eat the switch…

I drilled out the mount holes and glued in some aluminum threaded inserts. I couldn’t find machine screws long enough to use a nut so took this route. Looks nicer anyhow I think.

A view of the bottom plate that is glued to the mug bottom. The center hole was used as a positioning guide and is now filled with a section of dowel, glued in. The switch boot is from one of those cheapy zoom lights. The holes with the recess will be used to secure the lower module inside the mug. The other two holes were temporary holes used in determining where / how to mount the lower module. I’ll fill them with a couple short lengths of dowel but was out of an appropriate size.

Next we have the LED and MCPCB mounted to the aluminum block. The star was filed flat and heat transfer compound was used.

A view of the heat sink mounted to the lid. The machine screws go through to the aluminum block. The unfilled outer holes will be used in assembling the upper module.

The lower module insert being secured with machine screws through the bottom. I used a length of dowel rod to help hold the assembly. Stainless steel 6-32 machine screws.

There it is with the lower module secured.

A view looking down inside the mug… lower module secured in place.

A view of the bottom plate

Another view…

Here’s the upper plate. Driver mounted to one side and the cell contact top springs on the other side. (not visible here). The threaded aluminum posts will be used to connect the two halves of the upper module.

Here we have the other side of that piece with the springs barely visible. The brass tube is the + lead from the cells to the driver.

Upper module assembled with more spacers, tube section plus fine tuning washers.

The 22 gauge wires were trimmed and soldered to the MCPCB

The test fitting of the remote phosphor dome. Two small notches were made to clear the spacer posts and another two notches to clear the LED wires. I used a hot soldering iron tip.

The completed upper module in the correct orientation…

Kind of a difficult shot with glare on the mug. This one shows the upper module half being inserted The brass tube of the upper must be inserted into the lower brass tube.

There! Success; upper tube inside lower tube… To properly fit the upper module the colors of the wood used for the two halves must be matched. I use the walnut segment as my guide. The “handmadeness” is revealed by the slightly different spacing of the cell holes around the circumference.

Next we can see when the contact springs are just barely touching the cell ends there is approximately a 1/4” gap between the lid and the rim of the mug. This can be adjusted by removing or installing washers between the aluminum spacer posts of the upper module. This will permit changing to a longer cell at anytime in the future. The lower surface of the lid has a rubber sealing ring. That combined with the heat transfer compound used between the heat sink and the lid should make it safe to have at least light rain fall on the light.

The insert assembly is pushed down and the band tightened to secure the lid.

All set to go now !!!!

It does light and moves upwards through the 6 levels of, 2% 6% 15% 25% 50% 100% with 100% being 1.4 amps.

I’ll do some dark room shots when I can

Illumination… on low; 2%

Tonight I’ll try some dark room shots with the camera on manual settings. I haven’t pursued beam shots yet so some experimentation will be required. I complicated things by leaving the tripod out of town…

Light / brightness shots. These are harder to do than one might think. Our eyes are too darn accomodating in changing what we can see in different levels of light. If the first picture is a more or less accurate rendition of how the light illuminates on level 6, the highest level using 1.4 amps then the depicted brightness of level 3 (15% or 210 mA) or level 4 (25% or 350 mA) are not bright enough.

For the record my 70 year old eyes have no problem reading normal book size print on level 3 210 mA). Level 4 is better, but in the space I was using level 5 (50% or 700 mA) is not needed.

Level 1 (28 mA) is of limited use I suppose, but thows enough light to keep me from tripping over the chair by the door way.

Anyhow, FWIW, here we go from highest level down. What the camera recorded….

First level 6 (bright, 1.4 amps)

level 5, 700 mA

level 4, 350 mA

level 3, 210 mA

level 2, 85 mA

level one, 28 mA

On level 1 my eyes can still make out the counters and items on them. So it’s not as dark as the camera makes it to appear.

I think I am going to get myself a meter, something like the LX1330B. Then I can at least make some measurements and see some figures I can maybe relate to better.

Of course that will mean another project… an integrating sphere of some kind in order to add to the usefulness and add meaning to the numbers.

The camera was on manual and the ISO was NOT on auto.

Maybe I’ll try repeating the set of images but move the lantern out of the frame as suggested. I like it. Makes me really want to find a polycarbonate mug instead of the acrylic. I actually did find a couple of sellers of polycarbonate jar top 16 ounce mugs. But the minimum order quantity is 24 mugs; $68 plus shipping. I don’t need 24….

I must thank DBSAR for inspiration. It was his lantern project with the downfire LED that helped me formulate this plan. The vessel, well that was purely from my own weird thoughts.

I do find the phosphor dome illumination to be pleasant. The dome does not have that annoying pin prick effect that many commercially made lanterns have. I do prefer lights in the 4500 -5000 K range over the warmer tones. I’m different that way. So this suits me fine. The phosphors are available in some warmer temperatures but in smaller cone / elliptical shapes. This dome was easy to fit over a standard LED star. A cone shaped phosphor with a diameter of around 15 mm and smaller would make for a more difficult mount. Just another challenge though!

I am thinking about incorporating a micro LED voltmeter, but space is a bit tight inside the mug. … thinking cap is still on…. I think it would be cool to be able to monitor the battery voltage without having to access the cells. Maybe a momentary contact switch to activate for the reading? But I also don’t want to make it more difficult to access the cells. Easy changing of the cells was a prime goal of mine. And switches mean perforations and more wires……

Yes that is the meter type. It may be possible to mortise one into the cell block. Would be cool. With a very low draw having it only on with the light on would likely be hardly noticeable. Thanks for the offer.

I decided to do some run time tests. The first one was perhaps ill planned. Running at level 4, approximately a 25% power setting of around 350 mA, according to theory. The run began with all 4 cells at 4.1 volts. Twelve (12) hours later the voltage, after a short rest, was reading 4.03 volts. I’m thinking I should have used a single cell to spped it up, although drawing 4x the current from a single cell may not correlate exactly to the lower draw on multiple cells. I don’t know; I am guessing and extrapolating from lead acid experience.

Anyhow I stopped that test and am recharging. At that rate level 4 should last a few days. I may run this again when I don’t feel so impatient.

My second bout will be on level 5, approximately 50% power, or 700 mA. Four cells or a single? Have yet to decide.

I should also do some “tailcap” current readings to see how close the actual flow is to the theoretical programmed levels. I’ll do that after the 50% run time test.

Just completed a run time test using a single 30Q, 3000 mAh cell and running the light on level 5, 50% power = 700 mA.

The cell had been recently charged. Voltage reading at the beginning was 4.20 volts.

After 3.25 hours the light was shut off, I went to bed. In the AM the cell read 3.71 volts.

After a total run time today of 5.5 hours the cell voltage read 3.22 volts after being given a 30 minute rest.

I ran it another 30 minutes, for a total 6 hours and shut off. The immediate voltage reading was 3.07. 15 minutes later the voltage had recovered to 3.10.

I ran it another hour, actually 58 minutes and picked it up and headed to the shop with it still on. On the way the lamp blinked three times. As I reached the workbench it began to blink. The voltage read 3.01 volts.

The Nanjg specs state that at 2.9 to 3 volts this is exactly what the driver should make the light do.

For the last hour run the brightness level appeared to be decreasing. Bumping to level 6 did still increase brightness. It did not appear to be as bright as before, but since I have no meter that is a purely subjective statement. I do have a meter on order though.

Anyhow at the sixth hour the brightness can still be called servicable. If we multiply the 6 hours by 4 (cells) I believe it is safe to say that we have about 24 hours os useful service from a full set of four cells, when run on level 5. Good enough for me I’d say.

I forgot to note that on level 5, the heat sink ran about 20 degrees F higher than the ambient air temperature, whereas on level 4 it ran about 8 degrees F warmer.

Everybody (here in US and Canada) have a great Labor Day weekend. Stay safe but have fun.

Well I am embarassed. I was showing the light to my son on the weekend. Suddenly I became aware that there only appeared to be three 7135 chips in place on the driver. It can be difficult to see clearly around the upper area where the mug walls turn inward and then upward again. Sure enough, close inspection found only three 7135’s. Photos before 8/27 show 4 chips! 8/27 was about when I disassebled major components as I changed some design elements. I must have lost one 7135 as I heated the upper plate to change the driver to the new plate. Or lost it during the reassembly. Too much heat and not enough attention.

I did think the run times I had obtained were impressive! Rightly so considering the light has been running at 3/4 of what I thought it was. But all to the good; 3 × 7135 do seem to be quite adequate for the lights purpose.

Rufusbduck wrote:

Doh! At least it wasn’t the mcu that fell off.

I would have noticed that a bit earlier though.

So this morning I went and measured with meter... I should have done that before and I would have realized something was amiss.

Top row is the asked for percents for the levels, 2nd row the calculated values based on 3 x 7135 = 1.05 amps. 3rd row is measured value

2% ........6% ........15% ........25% .......50% .......100%
21 mA .. 64 mA .. 158 mA .. 262 mA .. 525 mA .. 1050 mA
19 mA .. 59 mA .. 145 mA ..255 mA .. 501 mA .. 1031 mA

Now if I could make some lumens sense......