"Indie Bullet" - 1600 mW 445nm True DIY

I don't know if this is a build tutorial, seller props, or just me wanting to show everyone what I created this evening

A little while back I had Ray J Neal create some custom heatsinks for me. Since we were shipping stuff up to the great white north anyway, I let him talk me into grabbing this one-off beauty of a host as well. I'm thrilled that I did! He gave me an awesome deal on it (and threw in a whole bunch of extras too - great guy to deal with)

I say one-off, because I don't think Ray is selling this as a regular host, and there were a few excentricities (like the need to thermal epoxy the heatsink in place). But the product itself was awesome. The heatsink was beautifully machined.

Here's the fun part - I really wanted to DIY this thing into a 1600 mW beast, and I wanted to do it "indie" style

So no $46 for dual flexdrives, no super expensive batteries. What was the plan?

- 1600 mA of current
- Dual parallel LM 1117 drivers, at 800mA of current each
- 2x 16340s (OR) 3x 10440s

I wasn't sure it could be done. For one, to get my resistor combination of 1.5 Ohms, I happened to need 3x resistors. Multiplied by 2, this meant 6 resistors. Where to fit them? In the pill!

Six (6) resistors fit just about perfectly inside the pill. I needed a few extra holes on the top of it (for 4x resistor lead wires, plus 2x power leads). Thank god for flaminpyro's thin silicon wire!


With resistors securely hidden away, and both ends of the pill reattached, I thermal epoxied the LM 1117s to the top of the pill (an American penny - thanks Ray!) and soldered the resistor leads to the ICs.


Then I added a cap and diode to the combined parallel output of the drivers.


A quick check with my test load to make sure the current was set.


This part was a bit different because of the host design. An outer crown went on first. I added some low-tech electrical tape to make sure none of the components would touch the heatsink once installed. (though by my measurements they wouldn't have anyway - but better to play it safe)


Some thermal epoxy to keep the HS in place, and viola, final product!


Here it is compared to my "Guerilla 445" and a typical SurvivalLaser. I don't know if this comes across in the photo, but the "Indie Bullet" is brighter.


Batteries:
The host will fit either 3x10440s or 2x16340s. Of course it will also take 1x18650, but my driver setup won't run on 1 cell. I found that once wired to a real 445nm diode (final setup), the 2x16340s only supplied around 800mA TOTAL (ie, both drivers, final parallel build). No good. However, 3x10440s (not even freshly charged), gave an actual current (again, wired as final setup with 445nm diode) of ~1400 mA.

Why only 1400 instead of 1600 mA? I wouldn't have *expected* a ton more from 10440s, however, I was actually SEEING more on the test load setup. I may have had a test load that didn't give a good enough representation of what a real diode would be like.

It's also quite possible that freshly charged 10440s willgive me the full 1600mA instead of just 1400mA. Either way, a powerful laser.

Total cost of driver: $2.50 plus wire, solder, and sweat

+1 :beer:

Awesome 'Indie' job rhd! You took an excellent route on the driver build also, well done & congrats on the DIY

Its always nice to see some driver inginuity
:beer:

EDIT: Crap.. I cant give you rep... gotta spread out some more.
I owe you one rep..

Hey - very nice buidl! :beer:
And a great host - good work Ray!

I have never build a laser before and I want to be sure, that I'm right: one battery is not enough, because with your driver design, you need at least the required voltage for the diode. With a boost driver, the output of the battery is transformed to the required voltage, which limits the running time of the battery?

donprado said:

I want to be sure, that I'm right: one battery is not enough, because with your driver design, you need at least the required voltage for the diode. With a boost driver, the output of the battery is transformed to the required voltage, which limits the running time of the battery?

Click to expand...

Pretty much bang on

A few clarifications -

I'd mention that with my driver design (LM1117 based), you actually need at least the required voltage for the diode PLUS another ~1.5V or so for the chip's brain itself. You may also see a lot of LM317 designs (exact same circuit), and those chips need a bit more, around ~2.5V on top of the diode's required voltage.

Regarding running time for the battery -
I think you're probably correct, but I'd want to understand the context of that point a bit better. I mean, a multi-cell setup (like mine) also limits running time of the batteries. For example, if you have MORE voltage than the diode and driver need (as I would when using 3x10440s), that excess voltage is dissipated as heat. That's a big waste. In fact, my driver design here would have around 12V when using 3x10440s. That's a pretty inefficient setup, because my driver is probably having to dissipate at least 4V as heat.

My perspective on this issue, is that I don't really care. I can't say that I really understand why some people worry about battery life / runtime. I would have to question the use-case that requires a long laser life between battery changes.

IE, what are you DOING with your laser that makes runtime critical?
- Are you in a tent, etching artwork into birch bark, miles from a charging source for weeks at a time?

I live in a house, with electricity. When my batteries die, I change them
I couldn't care less whether I get 10 minutes or 100 minutes of run time from a set of batteries. To be honest, I don't even know if either figure is realistic, because I never keep my batteries in a laser long enough for them to go dead. I'm sure someone will reply with their use-case scenario that makes run-time a crucial consideration, but it isn't for me.

Donprado:
Are you gearing up for your first build?

Very nicely done!
I can see the influence of DTR's style, am I right?

You might have gotten away with smaller/less resistors than those 6 big ones,
but it looks like you had plenty of space to spare anyway.

On the voltages and current issue:
I seem to have noticed that the batteries need some "headroom" voltage to be able
to push higher currents consistently.
If your circuits voltage requirements are very close to the batterie's voltage, there's
not much tension left to deliver high currents.
Remember also the more amps you demand from the batteries, the more their voltage will sag.


One argument against short run-times:
The wear and tear on the threads and the o-ring on the tailcap every time you have to take the batteries out.
Of course if the host has been machined fairly precise, this is probably not a very
serious issue.

Hey - Thanks for the kind words

To be honest, I don't think I've actually seen DTRs builds, but I'll check them out. Is he big on the DIY drivers? (something about spending $23 per driver just conflicts with my DIYer compass!! I know they're worth it, but it just burns my soul, lol)

You're right about the resistors. In reality, I could have cut space there. But each IC needed it's own resistors, and I wanted a resistor value of 1.5 ohms. The easiest way I could get there with what I had, was to use 3x 1-ohm resistors (two of them in parallel).

I think with 3x10440 I should have lots of voltage headroom, shouldn't I ?
That gives me 10 to 12 V in total. At most, I think the diode + driver could only eat up around 8V.

Re run-time: Brilliant!
I don't know how that didn't occur to me. Especially given the number of times I've killed threads on my favorite flashlights. (IE, when used as flashlights, with frequent battery swaps!)

VERY nice job! I had forgotten all about that host until you PM'ed me last night. I would be selling these left and right in the Pro Shop, but I don't think a lot of people would want to epoxy the heatsink in to place. But, if anyone shows interest in your thread, I may have to offer it. If I do, I would like to use this thread as a build guide for potential buyers. If anyone likes that host (besides RHD and myself), shoot me a PM and I will make one. $35 and that includes the driver heatsink as well :beer:

Again, very nice build. + rep!

news: Your host has attracted attention from japan, rhd.

laserdaisuki said:

FLEX DRIVERを使用せずにビルドレーザーを制作

Click to expand...

(produced without the use of FLEX DRIVER)

I'd like to know who is their contact inside LPF... :thinking:

:beer:

I don't quite understand what happened ^ here, lol ?

Someone from japan got interested on your build. International attention, boy!

link: laserdaisuki.bbs.fc2.com (the last lines, 549).

Oh neat!

Though, not to burst my own bubble, but if you CTRL-F the URL for LPF, it turns out that this site has linked up to 25+ of our threads. I think it's some sort of blog of an avid reader of our forum Still very cool!

Yeah.. I'm not really worried but it sound strange for me.. they are stalking us.. everywhere..

very good build dude.. the beam shots are insane.

edit: uuh.. btw.. what are terminating the beams..? Looks like your curtains now have burn marks dude.

Lol, I have heavy black curtains.

And as a result of that photoshoot, they now have, not burns, but holes.

Luckily it was a double-layer of curtains. The ones in behind are solid white canvas, and weren't penetrated.

But it really sucks for me. I love those curtains. In the morning sunlight, if I'm still trying to sleep, 3 little lines of light sneak through

rhd said:

Donprado:
Are you gearing up for your first build?

Click to expand...

Yes I'm still planning the project. I have some experience in electoronics, lasers and mechanics - should be fun to combine this in a laser build:yh:

At the moment I want to figure out the pros and cons of the different drivers. The answers of you and the other guys helped a lot - THANKS :beer: