6th Annual BLF Handmade Entry (Finished!) - gchart

subscribed because its a lantern build! :slight_smile:

I like the idea (a TP4056?). I dunno if it’ll fit well into the design, but I’ll definitely keep it in mind if it looks like it’ll fit!

Nice idea, looking forward to watching the progress.

You are the man, to do this that is. :slight_smile:

Do many LEDs in parallel, so they all get a smaller bit of current, that way you get much higher lumens/W and you can get longer runtime :slight_smile:

Ingenious :slight_smile:

Love the tilt idea, though some kind of lockout would be useful. :slight_smile:

Subscribed.
Good luck!

Have any of you tried PCBWay or AllPCB for your PCBs? I’m leaning towards PCBWay for this project. I always use Oshpark but I’m looking to get ones with white solder mask. They’re both in China, but seem to have good prices, lots of options, and pretty quick turn around time.

Cant help you there. I’ve only used Oshpark.

Ok, I’m still learning about proper driver designs, so I’m going to throw my schematic and PCB out here for critique before I order it. The PCB has been uploaded to Oshpark, but I will likely be using a different service (see above).

https://oshpark.com/shared_projects/zMUnQFCd


Another note… I decided to simplify a bit and spec out a Current Regulating Diode (1ma version) in place of the LDO+resistor. I’ve never seen one used but they appear pretty basic.

Oh, another thing… While this is one board, I’m basically designing it as if they were two separate ones, the circuitry is independent of each other. One side will be used as the main board of the lantern while the other will be a slave board.

Why is there a ring around the center pad? What purpose does it serve?

Looking at your schematic. Can you describe to me in words how you expect it to function? I’m having a hard time trying to understand it. :person_facepalming:

EDIT: In particular, the switching doesn’t make sense to me. I understand the slide switch disconnects the ground and turns the light off, but what is the tilt switch supposed to accomplish?

Hey David!

  • Ring around battery pads: my plan for holding the battery in place is to have a ring of 3/4” copper pipe (perhaps 1/2” tall) attached to the top of each board with just enough of a gap that the battery can be slid into position and then be held by the spring on the neg pad. The exposed ring on the PCB will allow me to solder the copper pipe pieces in place.
  • Switches: the slide switch, as you said, will disconnect the ground and completely disable the lantern. When that’s switched on, the mercury (Hg) switch will turn the 7135 on or off.
  • Functionality: When the primary board is face up, the mercury switch will be On, enabling the 7135 and pushing 350mA through the LEDs. When you physically flip the lantern over, the mercury switch will turn the 7135 off. However, current will still be able to flow through the Current Regulating Diode (Semitec S-102T) sending 1mA to the LEDs for a dim glow (akin to the illuminated tailcaps I’m so fond of :wink: ). I am debating, however, if 1mA might be too bright. Most of my tailcaps operate at around ~0.15mA, but then again, this lantern is much larger than a tailcap.

OK, I get it now, the two power deliveries are wired parallel to the LEDs. Either just the 1mA source or that plus the 7135 will be on. :+1:

Yup!

The PCBs came in today. At this point, I’ve got all of the parts except the LEDs.

Next up: preparing the housing (probably the most difficult step for me, limited access to proper tools)

Even though my 4000K LEDs aren’t in yet, I couldn’t help but build a prototype while I wait! I have some 5000K on hand, so I used those - I just wanted something warmer for a “lantern” style device.

Differences between this light engine and what I have envisioned for the final version:

  • 5000K LEDs instead of 4000K
  • Used a 1800 Ohm resistor instead of the 1mA current regulating diode. Assuming a supply voltage of 4.2V and a LED forward voltage of 2.43V (as measured), that should give me roughly 1mA on a full battery: (4.2V - 2.5V) / 1800 Ohms = 0.00098A = 0.98mA
  • Jumped the switch pads with a wire instead of a switch (like the CRD, I only bought one switch)
  • Stranded copper wire for the conductor pillars instead of the copper rods (only bought enough for the final version)
  • Battery is only held in place by the spring tension. I plan on having a partial copper surround and then a velcro strap or something to make sure it doesn’t budge.

Dim mode, Bright mode
.

Action video:

Another tidbit… assuming a 3000 mAh battery, this should last 8.6 hours on high or 4.2 months on low/moonlight.

Nice progress. Is the battery going to be removable?

That’s pretty cool. You are certainly making some headway into the build.