DIY LED Canister Light Questions

Hi all,

Another newbie looking for DIY advice. I've done some preliminary research and searches, and found just enough answers to spur a ton of new questions. Hopefully I'm posting this in the right forums; mods please feel free to move it if it's not.

I'm interested in building my own LED canister dive light. I've never done this before, have a startling lack of electrical/circuitry experience and tools, but I'm fairly handy and can follow instructions and figure things out/improvise when needed. Also, it seems like a really fun project, I might be able to save a little money (maybe not enough), and I think I have realistic expectations.

Here's what I'm looking for: a 10-15w light outputting about 700-1000 lumens, for about 4 hours. 200-300ft depth, just in case. A tight beam is key, I really don't want a flood. Something that's 6 degrees or less would be ideal. Because of that, it sounds like a single emitter would be best, though I'd be happy to go 3-4 emitters if that's what it takes. In essence, I'd like to emulate, or slightly beat, this: http://www.salvosupply.com/inc/sdetail/2057 - if I can do it for less than $400, I'd be totally stoked.

So, any advice would be greatly appreciated. Also, I've got a few specific questions stemming from the somewhat geeky talk I've been reading here.

1. Is there a general tutorial or how-to with regards to the electronics? I figure I'll need an emitter, some wires and solder, a heat sink, and maybe a driver? Or I can direct-drive but it's more strain on the LED? I'm not sure exactly how that works.

2. I'm a bit clueless about the batteries. Amps, watts, watt-hours, voltage, yikes. It's been a while since my last physics class. I figure for 4hr burn time, I'll need something that's around 12 volt, 4.5-5+ amps? I'm looking at using a NiMH pack such as http://www.batteryspace.com/index.asp?PageAction=VIEWPROD&ProdID=2219. I've also heard that Li-Ion or LiPo is a pretty dangerous chemistry for the novice/clueless, and if that's true, I'd like to not burn down my apt-should I just stick with NiMH?

3. LEDs. It sounds like there are a lot of choices and a lot of favorites out there. The Seoul P7, Cree Q5? Some have some nice aspheric lenses that allow for a super-tight square beam pattern? Will these single-emitter LEDs output enough light to equal 600-700 lumens, in a way that directly compares with a 10watt HID (roughly 500-600 lumens)?

4. What's with all these lumen ratings anyways? Is it like LCD refresh rate where manufacturers go around gaming the specs? I see links to some 10, 15, 20, 50, 100watt LED emitters with huge die arrays, but they don't seem to be used frequently in flashlights. Obviously they use a lot more power, but I figure there may be more leeway when using a canister.

5. That's about it for now. If anyone has any advice on waterproofing, that'd be great, but I think I'm going to go with cut-down maglite route, and either build one of those 3" PVC canisters, or potentially poke some holes in a premade watertight bottle like http://www.leisurepro.com/Prod/OMSC...eld=Relevance&DescSort=0&Description=on&Hit=1.
Might be a bit small for the battery pack, dunno...

Anyway, sorry for the big, rambling post.

I can start you off by telling you which emitter and optic is used in that Salvo. It's a six-die Ostar and the Fraen 35mm reflector.

http://www.mytinysun.com/online/templatemedia/all_lang/resources/Fraen_FRC_Datasheet.pdf

You could drive it with a Shark driver unless you want it to be dimmable, in which case you need a MaxFlex. I'd go with the Shark and not bother with dimming. A housing can be had from here :

http://www.miemczyk-shop.de/

the LK35N with cable gland should be ok. They can supply a battery canister too. You'd need at least a 12V 4500 MiMh battery pack or maybe a 14.8v 4800 mah Li-ion would be better. I reckon $400 should be just about do-able..
For my money, if you are taking it to 300 feet I wouldn't mess about with pvc pipe. IMHO buy a proper battery housing for around $100 and put your own pack in there.

http://cgi.ebay.co.uk/Li-Ion-Batter...64472QQcmdZViewItemQQ_trksidZp1742.m153.l1262

I don't know anything about the case that you will be putting everything in.

A single Cree XR-E R2 hits around 250 lumen max at the max recommended driver current of 1 amp(around 3.7 watts total). That is around 1000 lumen for 15 watts. Less out the front because of losses in the reflector/optic you want to use.

The Shark is probably a good driver for this and this quad optic might work for a narrow beam. I'm guessing up to 850 lumen out the front and a minimal of 700.

For batteries, a 12-14.8v battery pack with 5 aH and up should be good.

If you use the Cree XR-E Q5 instead of the R2 you will loose some output but the Q5s are easier and cheaper to get. Probably less then $200 total with all parts minus the case for the battery and the light.

I have never made something like this so I don't know the actual process.

:welcome:

OK 3 links for you of previous projects of mine over the last 2 months.

http://sods.freeforums.org/ist-led-upgrade-t507.html
http://sods.freeforums.org/diy-led-can-t448.html
http://sods.freeforums.org/here-we-go-again-t610.html

A couple of tips for a noobie.

KISS Keep It Simple S*****
I highly recommend using a P7 with DX reflector, Kai 2.8a 3 mode driver and a modified Maglite.
Try and keep the voltage just above what is required by the driver/LED. In the case of the Kai driver this is 6 x 1.2 NiMh.
You can fit 3 D size cells next to each other in a 80mm diameter PVC tube.

You will need at least 7000mah cells to get 4 hours burn time.
4 Hours is quite a bit. My 1st halogen can light lasted 1.25 hours but was good for 2 dives as you dont use it all the time esp if you are doing deep dives as most of the time is spent on deco.
My 90mm PVC cannister has been to 52 meters several times without issue. You can use PVC deeper but you need to design it right. EG double layers and reinforced end caps. 300ft (100meters) is pretty damn deep. If you can afford to that sort of diving you could probably afford to buy a commercial unit but then there is nothing wrong with over designing your light either.
I think the weakness in my can lights are probably the lenses ( at least the glass lenses I use) and maybe the glands but you can get 10 bar rated glands. Check out blue globe glands from www.tecme.de

Totally sealed PVC cannisters are easy to make and if using a reed switch the only place they can leak is the gland. If you need to get inside just cut the bottom off and replace it with a new end cap. They are cheap enough. Charge thru the light head.

Focusing a multi emmiter to a tight spot is hard and requires a big reflector. I have not seen the Fraen reflectors linked to above by Steve, they may be great but they are only 35mm which is smaller than the 50mm Maglite head. For a tight spot they also require a Ostar with no dome lens. Can you buy Ostars without dome lenses or do you need to remove them yourself? Ostars are alot more expencieve than P7s.

Good luck with your project Im sure you will learn heaps and have fun. But unless you think everything thru and design it right before you start buying parts you may not save money ( but you will have lots of parts for future projects!!).

Thank you all for your replies.

It's good to know that my benchmark Salvo is using an Ostar emitter. Is it actually accurate to say that a one-emitter OStar puts out over 600 lumens of brightness, whereas a single P7 will put out 200-300 lumens?

Packhorse - thanks also your responses - I saw your reports on SB and that helped spur me down this path. I'm not pretending I'm going down to 300ft, but my experience with 50-100m "water resistant" watches makes me want to err on the conservative side of depth ratings, especially if it's something that will be jostled about as much as a dive light.

Is there a reason other than price to go with a P7? Is there a lot more work to go with an OStar (availability of drivers, or LEDs without domes, etc.)? From the Fraen link, it sounds like an LED without a dome lens will intrinsically be better for a tight focus?

Thanks again all.

A C or D bin should easily put out 600 lumen.
But the choice of LED is only part of the equation. You need to match it with a reflector and DX and Kai supply a 50mm reflector for the P7 that fits a maglite. If you can source a reflector that fits a mag and works with a ostar then it may be a better choice although more expensive Or you could get a custom light head for your choice of reflector/optics. Driver and battery voltage is another consideration. Battery voltage may dictate cannister size and burn time too.

My 4 x Q5 aspheric blows my P7 away and the cost difference is minimal probably 20% more max. Actually it may even be cheaper. unless you want the same run times.

It really comes down to whether you want to use a Mag or a proper waterproof head...that's probably your first decision. If you go with a Mag, then I have used the DX reflector mentioned earlier with 4 x Cree Q5's and it gives a nice tight hotspot with plenty of spill too. It would be a good combination IMHO (and it's cheap). You could start off with a Mag host and then when funds allow, and you start diving deeper, you could just buy a 50mm ID waterproof housing from the place I linked to and transfer the internals across. I don't know of any 50mm diameter reflectors/optics for the Ostar but there are members here who know far more than I do about such things..

Personally I think the Mag makes a great head.
Some use the "stepped lens" design which makes it more reliable at depth but I dont and I have never had an issue even with dozens of dives bellow 40meters.
I dont necessarily see another housing as an upgrade. As mentioned before just choose a good gland and a nice thick lens. Glass is more transparent (although in water I doubt it will make much difference) and more scratch resistant. While acrylic is cheap and strong. I paid less than $2 each for acrylic lenes so when they get too scratched I just swap them out.

I would quite like to see a 4x Q5 optic in use. I doubt it would be as bright as the aspheric but it should provide more spill which you may find usefull.