Please help me make a frugal caving light

Flashlight Modding and DIY Parts
1

Hi there,

I’d like to make myself a frugal simple 3D printed caving light.

I’m new to actual flashlight electronics though rather than just buying drop ins and custom torches and Zebralights and Malkoff and OLight etc

Will 3D print the light housing and battery case and spot weld and shrink wrap my own battery pack out of either 18650/26650/18350… In parallel so 3.7 volt options only. Simple optics and a simple clicky switch on the top or side. Wire exiting then around to back of helmet and battery pack etc.

What I’d like is

A frugal driver - basically I want the high to be running at 1000-1400mA so I get nice long run times on high from a 6000mA battery pack etc and it doesn’t heat up dangerously in a 3D printed case…

Probably a linear driver as simple to start with and running things in parallel…
Pref heat and polarity protection (although as using JST connectors this is really irrelevant an issue)

While I’d quite like eventually 2 LEDs - one spot, one flood, perhaps running off one switch although could be 2 switches. I will compromise with a single to start with or a board with 3 on.

Would like
Warm/Neutral - 4000k or 5000k
High CRI
No strobes, no hidden modes etc.
3 or 4 modes - ideally something like 1-5, 33, 66, 100
So it’s something like 50mA, 350mA, 650mA, 1000mA (possibly 1400mA total if 2 LEDs and the mid mode would drop too)

I’m thinking XP-L or SST-20 but I’m also not sure about XHP ones and I don’t really have any knowledge of many new ones as don’t buy a lot of new flashlights anymore. Quite a few new LEDs seem to have emerged in the last few years when I haven’t been so active on the forums.

Happy to have 20mm or plus boards.

What I’d mainly like some help with is finding a decent driver that doesn’t have strobes/SOS and is frugal - too many I have been trawling through on AliExpress are all about pumping massive amounts of mA…
And I don’t want to have to tap 8 times and count blinks to set modes.

Also the 7135 drivers all seems to have % groups I don’t like. So is there anything I’m missing that hits the spot?

An A17DD-L FET + 1 driver?

I’d even be happy with a constant single mode light running 800mA or some such to start a build off with.

Thanks

2

Of all the places NOT to take a ‘frugal’ 3d printed head light, home made by a first time builder well that would be in a cave lol! Last place on earth you’d want it failing or breaking.

Not the sort of place to cut corners I’d imagine safety wise. Though I admittedly know nothing about caving, I think it’s obvious that you would want a VERY reliable robust light. 3d pinted may be the excact opposite as 3d prints are generally brittle and could easily be smashed on a cave roof or wall or even dropped. I wouldn’t even consider doing it that way it myself, but hey, it’s up to you lol!
If you’d like more info YuvalS has built a good few head lights and could probably offer you some advice if you ask him, check out his profile and builds - https://budgetlightforum.com/user/29597/my-threads

3

There are no issues what so ever with home made lights in caves. No serious caver does not have several backups, it really shouldn’t matter if a light or two stop working. I’ve had lights fail on me in abandoned mines, I just whipped out another one.

+1 on YuvalS, he’s built his own caving headlight from scratch and heavily modified another (good old classic Petzl Duo).

4

Yup, I would eventually use a metal housing but as a test one a 3D printed one would be fine.

I’ve been caving a good while, have backup lights etc too.

And also have a Petzl Duo with one of the CustomDuo Simple inserts in too but fancied something built as a simple light from the ground up - there’s very little in the frugal/cheap space of caving lights - either a Fenix head torch etc, a Duo, and then expensive custom lights… I was half wondering if something with a GoPro mount and a simple (more or less a Convoy S2 etc type light) in a simple housing would fill a niche.

I did not know about YuvalS so going to go through his posts now. Thanks

5

As G0OSE and Mike said, I have some experience with building caving lights but have not yet printed one.

As long as you have 2 reliable backup lights and enough spare batteries for your backup lights, I think you can go caving with your light without any fear. Worst case scenario you will use the backup lights.

My thought about your Idea:

  1. INHO regular 18650 cells are better than 1P2S packs since you compatible your friends/partners batteries and your backup lights. Also you can use a single cell to reduce wight on short trips.
  2. If all your components can provide 1000lm don't limit your light to 1.4A. Keep this option for a short turbo mode
  3. Choose a LED with high Vf to reduce driver heat
  4. I have no experience with 3D printing but as far as I understand not all materials are durable and waterproof so obviously choose a proper material.
  5. Use an aluminum heatsink (can be bought on Ali Ebay etc) for the back of the light to enable heat dissipation.

Regarding the driver:

  1. The simplest option will be DualFlex by TaskLED, I also used this driver for my first lights, It was developed for caving lights and even some commercial caving light such as Phaethon use it. I stopped using it since it has limited flexibility and does not support all the option I wanted.
  2. Some Aduril2 drivers support 2 channels for tint mixing so I believe it can be used for flood+spot channels but I havn't tries it yet and personally I think Andurail is way to complicated for a caving light
  3. With some technical abilities you can modified 2*105C drivers to work with a single switch and control 2 channels. This is my currently preferable method and I have some FW in my GitHub supporting this drivers combination

You can see my projects on my signature and welcome to ask more questions

6

Get some cheap 2 channel headlamp like Sofirn D25LR and modd it. You can print bezel to fit custom TIR optic. And check this out

7

Is caving and frugal really a good combination?

8

It looks like your jumping into deep water without getting your toes wet first. Start small 3D print improvements to existing lights that have the functions you want. Good start 3D print slip over sleeves for armoring up the lights to improve water proofing, drop resistance and just general getting dinged up on hard use. 3D printed water proof battery holders. 3D print bezel standoffs and caps to protect lenses. You could do quite a lot to get good lights into great lights for caving or as some of us say around here “Dog walking lights” After that then build something great. Be sure to show off your builds here What did you mod today? Toss those improved lights into a pool to see how well your extras worked or just drop them a few times to see if that armor is true. Welcome aboard wski!

9

When it comes to lights, cheap is no problem. When I’m caving or down in abandoned mines I much rather prefer six $10 lights than one $100 light.

10

3d printing is not very strong…

i would not take one caving…

you can get a decent light for $40

11

Yeah, “frugal” and “caving” would be like Musk or Branson wanting to go into space and buying all their mission-critical equipment on AliExpress.

12

No, not really… I don’t know how many sets of backups for mission-critical pieces of equipment Musk or Branson require, but when I go caving or down in abandoned mines I can have a backpack full of lights if I want to. At the most I think I’ve had 7 because I was testing them for photography. If even 4 of those had failed I wouldn’t have been the least concerned for my safety (I would be for my light building skills though). Is it really the same for space flight?

13

Ah sorry, some confusion, by frugal I don’t mean bodged together out of cheap materials… I mean frugal as in it sips battery power i.e. it is a lower current draw/lumen than most… And is simple and uses common parts and simple mechanisms etc. So I might design it around a Go Pro mount and without lots of modes and features like battery level checking.

Thanks Quadrupel about those links, they look really interesting

14

I think this is an important feature in itself, but even more important is a low battery warning which is basically the same mechanism so I would not omit it

15

If they’re printed but aren’t quite waterproof, then a backpack full of ’em (hardly “frugal”) might still take in water and be compromised.

And you can get seriously dead in a cave if you can’t see where you’re going and get stuck somewhere.

That dewd is still entombed in the Nutty Putty Cave after getting stuck like a clump of hair in a P-trap.

No, ain’t saying he got stuck there because his lights went out, but I wouldn’t want to try to crawl my way out like a nekkid mole-rat if my lights did go out.

Then again, those caves which have narrow parts named things like “The Birth Canal”, you couldn’t force me in there even at gunpoint.

16

Then generally… I think it was Zebralight?… has the rep for extremely efficient drivers.

I don’t have any ZLs so can’t quite attest to that; just parrotting what I heard.

I’m not really A Headlight Guy, so I use the same RJ02s, D25s, etc., and that’s it. They Work, so that’s good enough for me.

17

Not all caves involve water. I pack mine watertight regardless. A backpack full might be an exaggeration, but still I’d prefer four $10 lights over one $100 light while caving.

Of coarse. I’ve been down a tube head first where you have one arm in front, the other behind, or you simply don’t fit. The problem for me was I had the wrong arm in front. When the tube turned a little it took me about 10 minutes just to move half a meter because my shoulder kept getting jammed. Without any light I would have been screwed but cavers today don’t do this type of stuff alone. Had my headlight and hand held both failed, and I’d lost the pack I was shoving in front of me, another caver could have been able to get a light to me or light up the passage way in front of me until it widened.

Cavers bring spare batteries. My philosophy is to bring each spare 18650 cell in it’s own light, at least if weight is not a concern. The second photo in my signature link requires over 150 meters of abseiling and then ascending the rope on the way up. I’m a little more weight conscious on those kind of trips, but still, you climb up with a hall-bag attached to the bottom of the rope and hall it up once you’re at the top.

Anyhow, sorry to wski for all this off topic clutter. I’ll give it a rest now… I can at least add that there are drivers available that can be re-programmed to mode groups that you like. Most firmware development here is done with ATtiny microprocessors. I don’t how which drivers are currently available as I only use my own designs, but for sure someone can point you in the right direction if you are willing to learn a little about firmware replacement. That’s how I fell into BLF to begin with, I searched for a way to get rid of annoying blinky modes on lights and ended up here. I fell down the rabbit hole… and didn’t have enough lights to save myself :smiley:

18

For longest runtime and lowest heat (especially at low current) I’d recommend making a 2s battery pack and using a buck driver.

19

Just found this post on BLF search option...
https://budgetlightforum.com/t/-/41975

20

I’ve printed lighs, and other things, i build lights from a scratch. your task does not seem to be a complicated one, you do not really need a 3d printer either. i would not use one for this purpose. fdm prints are not durable or strong enough for it. A shelf queen light or a light you barely use, yea 3d printed will work great, but cave lights idk. i’d use something stronger and durable.
Fat +1 driver would be a bad idea, since it is basically unregulated direct drive in all modes except 350ma mode. this is exactly what you do not want. for more runtime you do not want linear regulation, but since you want to build it as cheap as possible, simple 7135 driver with 4 chips is what you want.