Help with a dive torch

[deepbluediver]

Hi all,

I recently bought a second hand underwater kinetics uk600r off the net and was really eager to recieve it. I was thinking, great a nice wide beam to see underwater with. As soon as it came i put the batteries in and turned her on. The results were extremely dissapointing. I tried it against a uk c8 light which was nearly three times as bright was a beam amost double the width despite being half the size. I noticed that there is a bit of paint flaking at the base of the reflector near the bulb. How much would this affect it? Also It runs on a 6v rechargeable 4 ampere hours (whatever that means!) lantern battery. Now i'm new to this buisness of modding torches and i want to get it as bright as i possibly can. The only problem is i have a small budget and not very much knowledge in this field. Would it be possible to upgrade it to led or stick with halogen. Can somebody please help me as i dont want to waste all the money i have just spent on this torch

Cheers,
deepblue

[Darker View]

I can not answer any specific questions on the 600, I do not have one. I can clear up the bit about ampere-hours. Ah are a measure of how much current is available from the battery and for how long. In theory it works like this... If the lamp takes one amp, it will run for four hours with that battery. If it takes 500mA (1/2A) it would run for eight hours. In practice it is a bit more complicated, but that will work for a first explanation.

[Klem]

Plastic housings limit the amount of heat you can generate which translates into less light possible, however you can still build something that will outshine what you currently have. Plastic acts as an insulator to the perfect heat-sink that is the surrounding water. You could design a thermal path from inside to outside using a metal object poking through the body and waterproofed (e.g. one or more bolts connected directly to the emitter).

If you don't want to get too involved and happy to forgo squeezing the maximmum amount of light available from that housing you could buy an off-the-shelf emitter in the 6V range and with some cutting, gluing, soldering, affix it to the inside. Depending on how hot it gets you might need to fortify the existing head-sink of the emitter to avoid damaging the LED's or the battery.

[Packhorse]

Hate to put some one off DIYing but you are probably far better off throwing it away and getting a much smaller and brighter light.

[deepbluediver]

Hi all,

Thanks for replying so quickly, also thanks 'darker view' for clearing that up and 'packhorse' i cant afford to do that. Also i dont really want to go drilling holes in my torch as i dont feel confident in my waterproofing skills. My question is though how hard would it be to upgrade it to LED and how would i go about that and if i can't do that what is the brightest halogen i can put in there? All i know about the current halogen thats in there at the moment is that its a bi pin socket and the lantern battery is a pj996 6v. What can i do to get it nice and bright without drilling holes in the body

Any help would be much appreciated,
James

[DIWdiver]

I usually agree with Packhorse, and this is no exception. With very few exceptions, DIY is mutually exclusive with "save money". By the time you buy lamp and socket (or worse yet, emitter, driver, heatsink), reflector, glue, wire, a tool or two, you're going to end up with a disappointing light that cost more than a good one. If you're not into it for the joy of building, learning, and growing, you're just going to be disappointed.

[Packhorse]

OK then if you must.

I recommend you dont use to high a power LED. Something like a XPG should be OK. Use a heat sink with lots of surface area.
You have the opportunity to use quite a wide reflector but I am guessing you will not have much depth.
A wide deep reflector will give you a nice tight beam. A smaller reflector will give you a wider flood beam.
Beam choice is personal preference. I prefer a tight beam. Especially when using a "low power" LED which will not really supply a huge amount of lumen.

A XPG P60 drop in from DX would be a good start. But its small form in behind such a large lens will make for a big light with average output.

[lucca brassi]

[350xfire]

There are many lights out there for less than $100 (Even $70) that are better than anything you will end up with out of the UK DYI... Just a thought.

[deepbluediver]

Hi Guys,

Thanks again for replying, also thanks to lucca brassi for that diagram. I have a sample XPG from cree components on its way. Looks like i am going to have to have a shot at modifying the torch. It Shouldn't cost to much if the answers to these questions are what i want to hear. Would a piece of bright mild or mild steel or aluminium work for the actual heatsink?
And would araldite be ok to use as a water sealant ( my only worry is would it start to break down with the heat from the led?)

I also heard that led's need a resistor, now do i need a resistor for it or would there be one on the sample led coming? Also what is the driver? is it the pcb? And what battery would i need? ( i need to get a new one whatever happens because the other one died last night )

And this may be asking too much but has anyone got any spare thermal paste they would be willing to donate?

I am also going to need some help on the reflector aswell. Would the current one suffice? ( i can upload a photo of it if need be)

Sorry for all the questions,
James

[350xfire]

Driver and LED will cost you about $20... All that hacking will probably get you a flooded light... In the end, I think you know where I'm headed... Just sayin'

[deepbluediver]

Driver and LED will cost you about $20... All that hacking will probably get you a flooded light... In the end, I think you know where I'm headed... Just sayin'

Yeah thats what i was thinking but i got conned out of 40 quid so you know i'm just trying to see if i can salvage it :L

[deepbluediver]

Hi guys,

Also how hot would the xp-g get?

Cheers,
James

[DIWdiver]

Aluminum for the heatsink. Way better than steel and easier to work with.

[Packhorse]

Hi guys,

Also how hot would the xp-g get?

Cheers,
James

It shouldnt get hot at all. Thats the whole idea of using a heat sink.
But if the heat sink cant transfer that heat to the water then it will just slowly heat up and cook the LED.
In reality it will dissipate heat to the air inside the torch and that will dissipate heat to the plastic body and out to the water. This is a very inefficient method but given a large enough surface area heat sink a single XPG should be OK.

I guess we shouldnt mention any thing about throwing good money after bad.

[Klem]

The guaranteed quick-fix solution in your shoes would be to dump that torch and buy an off-the-shelf brighter one. However...I would counsel to get over the dollars spent on that original housing.

Having a go turning it into a brigher LED torch... DIW is right, if you add up the cost of components and put a price on your time it is cheaper to ditch and buy. 350 and Packhorse are also right; that housing without cutting a thermal path through to the surrounding water will never permit an output that this forum would agree is impressive. Lucca's idea of having a disk of aluminium fixed to the inside by say 4 bolts solves two problems; thermal path and how to affix the disk in the right spot. You would need to consider how to recharge those batteries behind the disk and your o-ring system would have to be pretty good to prevent water seeping in if you plan to constantly unscrew those bolts to recharge. Four potential new points of ingress increases the probability that flooding will occur.

You could permanently fix that disk in place, and If deciding this I would use something that cures flexible, like 'Sikaflex' marine polyeurathane. Araldite would also work but it cures hard and brittle and I imagine the glue will eventually crack with all the expansion and contraction, and banging around of multiple dives. Remember, that housing is plastic and it will contract, expand and flex to a greater extent than metal. Even submarines made of titanium have only half the usable life of a surface ships for that reason. On the inside have a plug that runs past the disk to the front so you can plug it in to recharge.

Point is...you'll get satisfaction and learn a new skill if you try an fix it. Put a price on that and compare it against the difference between DIY, and 'ditch and buy'. Up to you.

[lucca brassi]

You would need to consider how to recharge those batteries behind the disk and your o-ring system would have to be pretty good to prevent water seeping in if you plan to constantly unscrew those bolts to recharge

.

+1 agree , for recharging you could use wet contacts , but you need space inside

Araldite would also work but it cures hard and brittle and I imagine will eventually crack with all the expansion and contraction, and banging around of multiple dives.

+1

after all that could become at end more costly than light ...
-------------------------------------------------------------------------------------
Sketch that I have drew ,presented what I would do with little Halcyon 10w HID primary light using original POM housing
-------------------------------------------------------------------------------------

Common problem with thin plastic is that after producing when are all openings designed by mould is enclosure very strong.
If yo made holes on non reinforced place you could break down force structure that work on enclosure ....result is destroyed or leaking housing.

[deepbluediver]

Hi guys,

Thanks for all that info, This is going to be harder than i thought for another reason. Looking at where everything fits in the torch i don't think i would be able to put the bolts in there. So i think thats the heatsink idea gone down the pan.

So i have two options:

Can i get away with putting the LED in and using the air and body for the heatsink?

Or am i just going to have to just replace the halogen thats in there?

Cheers,

James

[Packhorse]

Can i get away with putting the LED in and using the air and body for the heatsink?

Yes. Should last you up to 5 seconds.
Mind you if you use a P60 drop in you may get up to a few minutes.

If you want something that will last a full dive you need to get a decent heat sink.

[deepbluediver]

Yes. Should last you up to 5 seconds.
Mind you if you use a P60 drop in you may get up to a few minutes.

If you want something that will last a full dive you need to get a decent heat sink.

So can i put a heatsink onto the led and have it dissapate heat into the air in the torch because that is really the only option

[350xfire]

So can i put a heatsink onto the led and have it dissapate heat into the air in the torch because that is really the only option

Air is NOT a heatsink! You need an aluminum (or some heat dissipating metal) and a path to the water. Otherwise you will reach a point of heat saturation and temperature will begin to rise beyond the LED rated limit. By the time you re-do your light you could have easily bought something like an Intova and be better off. That light has a magnetic switch and really compact.

[Packhorse]

I think I have already said this at least twice in this thread.

If you make the heat sink surface area large enough and only us a XPG or less then you will probably get away with it. But I am unsure how big of a heat sink you will be able to fit. And in the end you will have a very bulky light.

Since I have repeated that I will also repeat this.

You are far better off throwing it away ( along with your black and white TV, Atari, Commodore 64, Motorola Brick and AM radio) and buying something smaller that will be as powerful if not more so.

The Intovas like 350 suggested are a good little torch( I like the spot beam ones not the flood).

The W200's are a good torch although I replace all the O rings and also check that the "pill" is not pushed forward into the lens when its turned on. This can un seat the lens from the sealing O ring ( Only been a problem on a couple of W200's I have seen and I have had dozens of them to modify)

The W300 is a step above the W200. But its also a step bigger and heavier ( not as big as the UK600 though) and takes 2 D cells ( I'd recommend NiMh).

The PALIGHT BG-QS88 is also a light worthy of looking at. There are several versions and you can get it as a kit with 26650 batteries and charger or just the torch. It also takes 18650's.

All these lights can be had for $50us or less.
Modifying your UK600 will cost you at least half that. It may end up costing you a lot more.

[350xfire]

YES!!!

[wquiles]

YES!!!

+1

Pack's advice is spot-on

Will

[Doc Ed]

If you post photos of what you currently have, maybe we can get our collective brains together and think of a solution for you... One thought that came to mind was that since the battery was apparently shot to begin with, this may have been the reason for the poor performance of the light.

There are some other threads on this board related to others using a UK body with LEDs. One that comes to mind was written by Rodex99 entitled cooling tested in uk housing where the bottom line was:

I think there is a clear deduction that you cannot run a LED in this kind of housing above a certain point for long without over-heating. Exactly what that point is I am not quite sure.

So it'll be a trade-off between the performance desired and the degree in which you want to modify your light, as well as the attendant expenses involved.

That said, I need to add this as well: I'm relatively new here to DIY-ing lights as well. The other posters have been doing this much longer and definitely know what they're talking about. And one common thing they say is that to DIY this light properly will cost you some money. You've stated that you want to try to save your initial investment of 40 quid - an amount I wouldn't pay for an old UK600, when they can be had for 15 quid (but then again I'd have spent the money on something else). Considering you have to replace the battery of the light already anyway, you really need to determine whether or not you wish to continue with this project in order to save money. If it was for the learning and the fun of DIY, then by all means, proceed. Otherwise, get a new light.

[DIWdiver]

Air is NOT a heatsink!

Technically, that's true in this case. The water is your heatsink, and everything else is just a path for the heat to flow from the LED to the water. But if operating above water, the air outside would be your heatsink. The 'sink' is the point into which you can dump any amount of heat without increasing the temperature.

In practice however, the final solid surface that radiates heat into the infinite sink is what we usually refer to as the 'heatsink'. So the finned aluminum block on top of the CPU or GPU in your PC is called a heatsink, even though that's really not technically accurate. To be honest, 'radiate' isn't really accurate either, since most of the heat transfer is by convection, not radiation.

Assuming you can't or don't want to cut through the case (a choice which I applaud, by the way), the best you can do is to transfer the heat from the LED to the air inside the case as efficiently as possible. From there the heat would transfer through the case to the water. I would think you might get away with as much as 5 watts this way, so an underdriven XM-L might also be a possibility. I would suggest an aluminum disk that just fits inside the case. If you could attach some fins or blades to the disk to help carry the heat down the length of the case, that would be even better. The LED should be attached as directly as possible to the disk with thermal epoxy, or with screws and thermal compound (which doesn't harden).

If you got bare emitter sample(s), then you're going to have some trouble connecting to the electrodes, which are on the bottom. It's generally easier to work with emitters that are pre-mounted on stars, which are small aluminum circuit boards which can be screwed or glued to your heatsink, and have larger electrodes on top, making connections much easier.

[DIWdiver]

I also heard that led's need a resistor, now do i need a resistor for it or would there be one on the sample led coming? Also what is the driver? is it the pcb? And what battery would i need?

There will NOT be a resistor on the LED sample.

The rest of your question is both huge and very important to being able to control the light output, LED lifetime, and battery burn time. A thorough explaination is not going to be achieved in a single post, or even in several, so I won't even attempt it. But maybe I can simplify it enough to be helpful.

In a very simple view, a battery can be considered to be a voltage source. Ideally, it will give you any amount of current at a fixed voltage until it is fully discharged.
But of course no battery really does that. The more current you draw, the lower the voltage is. The voltage also drops as the battery discharges. A fully charged Lithium cell is 4.1-4.2V, and you can discharge it to 3.5, 3.2, 2.7, 2.5V, or even lower. The more deeply you discharge it the more energy you get per charge, but the fewer charges the battery will last.

In a very simple view, the LED can be considered to be a fixed voltage load. Ideally, it will absorb any amount of current at a fixed voltage. By the way, the amount of current through the LED is the primary factor in how much light it puts out.

So if you connect a voltage source to a voltage load, what happens? If the source voltage is higher than the load voltage, an infinite current will flow. Clearly this can't happen. A college professor of mine used to say that this would cause an 'anihilation wave' to propogate from the event at the speed of light. Not something I relish experiencing. If the source voltage is less then the load voltage, no current flows. This is pretty boring, as it looks a lot like unconnected parts laying on the bench. If the voltage source and voltage load are exactly the same, the current could be anything from boring to anihilation. I want something more reliable.

So what do we do? This is where the driver comes in. The driver's job is to control the current in the LED when the battery and LED voltages don't match.

A resistor is a really crude form of driver. If you wire battery, resistor, and LED into a loop (and get the polarity correct), the current in the loop will be determined by the Battery voltage, LED voltage, and resistor value according to the following equation:

Current = (battery voltage - LED voltage) / resistance

You can see from this that as the battery voltage changes, the current will change, which means the light output will change. It will start dropping as soon as you turn on the light and continue dropping until you turn it off.

A better driver would be a variable resistor, that automatically adjusts to keep the current the same no matter what the voltages do. A linear driver is essentially this.

Even better (in most cases) is a switching driver. If the voltages are much different (like 3.4V LED and 6-8V battery), a resistor or linear driver is very inefficient. A good swtitcher can handle the voltage difference, regulate the current, and maintain high efficiency.

I am also going to need some help on the reflector a swell. Would the current one suffice? ( i can upload a photo of it if need be)

The current reflector will work poorly with an LED, even after you to modify it. The reason for this is that it was designed to catch light coming out the bulb sideways, backward, and slightly forward, and focus that into the 'spot'. Light that comes out more forward will not hit the reflector, and will spread wide to form the 'spill'. An LED has mostly 'more forward' light, so only a little will hit the reflector and form a very weak spot.
For an LED you need a much deeper, narrower reflector, or a lens of some sort.

[Klem]

If you still insist on upgrading that existing housing but are dead against building a more efficient thermal path to the outside then once again I stress...you will not be able to maximise the potential of that torch. Your light output will be limited to the point where the inside turns into an oven and cooks everything; roast LED, driver and slow-stewed batteries.

It can be done but only by limiting the power draw.

Here's a similar project to yours that underwhelmed me two years ago;

Using a Princeton Tec Miniwave which operates from 4 C Cells and a single halogen and reflector. Swapped out for a brighter LED emitter. Replacement is a Terralux 12Watt 3*XR-E giving 600lumens, but also changed the carbon cells to Li-ion to cope with the higher power draw. Note, there's more metal and a ring of nuts behind the emitters in an attempt to draw as much heat away from the LED's and delay the heating of the emitter as possible. But in the end where does that heat go? Into the batteries (which are also producing heat due to internal resistance) and back to the LED. It had a very short life as I gave it away to a mate who has all the dive gear but doesn't dive. For its size, bulk and thermal inefficiency it was otherwise destined for the bin. I hope you realise I risk my questionable reputation by even showing it to you! BTW you can put the same emitter in a converted 2D Maglite; less bulky, more robust and far more thermal-efficient for a modest increase in cost.

Princeton Tec no longer produces the old halogen Miniwave but have upgraded the housing with LEDs and call it the 'Shockwave'. The point being, the most this manufacturer believes they can squeeze out of this plastic self-contained configuration is 370 lumens.
http://princetontec.com/?q=miniwave-led

At roughly 100lumens/Watt I'd say DIW is right on the money...500 lumens max out of your slightly larger housing and without a more efficient thermal path to the outside.


FOOTNOTE: I see that Terralux have upgraded their old 700lumen TLE-300 Maglite upgrade to a 1000 lumen version
http://www.terraluxportable.com/product/tle-310-m-ex/

[deepbluediver]

If you post photos of what you currently have, maybe we can get our collective brains together and think of a solution for you... One thought that came to mind was that since the battery was apparently shot to begin with, this may have been the reason for the poor performance of the light.

There are some other threads on this board related to others using a UK body with LEDs. One that comes to mind was written by Rodex99 entitled cooling tested in uk housing where the bottom line was:



So it'll be a trade-off between the performance desired and the degree in which you want to modify your light, as well as the attendant expenses involved.

That said, I need to add this as well: I'm relatively new here to DIY-ing lights as well. The other posters have been doing this much longer and definitely know what they're talking about. And one common thing they say is that to DIY this light properly will cost you some money. You've stated that you want to try to save your initial investment of 40 quid - an amount I wouldn't pay for an old UK600, when they can be had for 15 quid (but then again I'd have spent the money on something else). Considering you have to replace the battery of the light already anyway, you really need to determine whether or not you wish to continue with this project in order to save money. If it was for the learning and the fun of DIY, then by all means, proceed. Otherwise, get a new light.

I thought that i was getting a good deal with two reachargeable batteries and a charger and a torch. he said they were in working order so i thought why not. I really regret this decision

[Doc Ed]

Maybe you can still get your money back - considering the batteries are actually dead and all. Might be the best option.

I'm sure we've all been in a similar situation like yours at one time or another, and sometimes it gets difficult to see that the most reasonable thing to do is sometimes the hardest - ie moving on.

You could keep the housing and use it for something else down the line... say like a small video camera