Purpose built, high output platform.

[VidPro]

$$$$$$
3 LUXIII stars big 24$ ohh price shock, ok so 36$ if you want them to the human eye to be percieved 5% brighter.
a 12$ hunk of metal to put them on, and 3$ for some optics or reflectors.
why does that so quickly total $100 /ubbthreads/images/graemlins/smile.gif
oh gee i forgot the 30minuts of labor you might need to put into it.
batteries? you gotta have them anyways. the base you gotta have anyways.

drivers, or lost power things . .. try making something without any stuff that is blowing any of the power into anything but the led, works for incans, and if you do it right, it works for leds.

ok so led is cost ineffective for a POTENT model, but not that bad, what we might be paying for stuff, isnt nessisarily relating to what china will pump it out for.

isnt the majority of the wizz bang SF incandescent lights over 100$ from the get go? arent they the ones with the $400 one?

seems relative to me, my desklamp alone cost LOADS in leds, but in 2 more years it will pay for itself in Bulbs (used to go through halogens) alone. not to mention energy consumption.

if you could buy an incan 2 years ago for 2$ , why would i pay way more for an incan than a led now? heck they used to give them away.

if this is a incadescent ONLY thread, then why was it in general ?
if high-power is only defined by incandescent, then you aint seen enough led mods. why just overdrive a filament till it pops, overdrive a led to a 100hr runtime, and see what you get, its cool.
i dont think i would have accidentally posted in this thread, if it was incadescent only.
hotwires kick leds all over the place, but any good one isnt going to fit in my pocket.

 

[Codeman]

[ QUOTE ]
tvodrd said:
...
Bill has asked in a PM if I would get involved in this endeavor. I am seriously considering it! If the *design requirements* were coming out of my Marketing Department at work, we would all die of old age before the "sign-off!" /ubbthreads/images/graemlins/crackup.gif

Requirement #1: It be "Mouthable!" ?????

Larry

[/ QUOTE ]

I hope you do, Larry! /ubbthreads/images/graemlins/grin.gif

Although "mouthable" isn't as big a desired trait for me, the thought of 300-400 OTM lumens paints such a vivid picture, that I'll second the idea anyway!

OTM = out the mouth /ubbthreads/images/graemlins/crackup.gif

 

[js]

[ QUOTE ]
chrisse242 said:
[ QUOTE ]
tvodrd said:
This thread is a fascinating read! An obvious observation:

We all favor a different trade-off/balance between size, runtime, output. throw, color, and rechargeability/not.

[/ QUOTE ]

Larry mentioned something that is really important to me. A high output light that is not just made to show it of has to be ready when it's needed. For my type of use that more or less rules out nimh as power source. In many households a light of this type will only be used a few times a year. It's not an option to have it in a charger all the time, and with nimh the self discharge rate will probably make it useless once it's needed. So I'd like to see either lithium primaries or li-ion as power source. I'd like to see the Saft D-Cells as optional power surce.
Regarding size: In my type of use this is not a carry-light so anything goes. 3D-size would be ok.
Runtime should at least be an hour with stable output on a surefire M6-level or above.
I'm with bwaites regarding incans for better colour, and I'd favour a light without additional led's, I always carry some kind of dimmable led and I'd like this light to be focused on high output.

That's it for me...

Chrisse

[/ QUOTE ]

Just a few words on the NiMH vs. Li-ion and Saft D cells thing:

The self-discharge rate of NiMH is bad relative to other chemistries, but a pack will sit just fine for a month or 6 weeks, and still retain much of its runtime. It's not a big deal (for me anyway) to do a monthly or bi-monthly maintenance charge on a light that just sits around.

And if we're talking about such a light, why not go with non-rechargeable CR123's instead? They have a 10 year shelf-life. Can't beat that.

But, yes, Li-ion's are certainly awesome, and are great for many applications, but let's talk about a few cost issues, which Ginseng has brought up before. Very soon now, you will not even be able to buy bare Li-ion cells PERIOD. You won't even be able to SHIP them! It will be illegal! So all Li-ion's will have to be Pila types with full built in protection circuitry. OR, they will have to be built into packs with integrated protection circuitry. And doing this involves a mandatory $6,000 safety testing buy-in plus the cost of destroying 30 packs!

AND, I happen to be one of the few people who actually knows about how much the Saft VL34480 cells actually cost. Even in quantity, you are looking at $30 or more per cell. And, they won't sell them to you, only to their battery distributors and only for the purpose of them making up a Li-ion pack, which means the $6000 safety testing.

Every so often these cells get brought up again as some kind of solution or option.

Dream on. Dream on, my friends.

I talked with an engineer at Saft about these, and he confirmed what Ginseng has already said, that these cells are viewed by Saft as potentially very dangerous devices, akin to small bombs, and they keep a very tight hold on them and will only sell them for OEM use in battery packs.

No possible chance of any CPF project using them. Believe me, I know.

 

[VidPro]

yup they are dangerous.
i have some of them, and they are awesome, best thing since 9000ma ni-mhys.
but shaft is going to be selling them to the military, and medical people, soo, some Proper application of them is totally possible.

they told me they can pack em up with protection, and sell them for about 90$ a pair. the problem is that the "standard" flashlight doesnt easily hold all the extra bits and pieces, and neither are most of them designed to protect from high compression, like a drop.

i solved most of that in my simple mag mod, but people would have to sign one long arse disclaimer before i would sell it to them /ubbthreads/images/graemlins/smile.gif
they could instead take the same mod, toss in 11,500 ni-mhys, or a bunch of AAs and get close to the same capacity , with more self discharge.

but for myself i am tired of self discharge, and re-charging.

are we FAR from applying huge li-ions? i doubt it, they use them in cars, wheelchairs, and other high need battery applications. they just need to make a flashlight "model" through a real demand.

look at it this way, were not sticking them on our ear, next to our temples, while driving down the road 70Mph /ubbthreads/images/graemlins/cool.gif

 

[bwaites]

Just as a short term panacea here, there MAY be a possibility of a smaller form factor LiIon protected cell in the near future.

Jim, with all thats going on this week I haven't had time to talk with you or Wilkey, but I'll call this weekend and discuss it.

It will NOT be D cell sized though!

Bill

 

[js]

VidPro,

Yeah, they would have sold me some samples with protection circuitry added on, but according to what I was told, the extra bits would have made them not fit for my application.

That would also rule them out for use in the M6, as was once suggested. In fact, even bare, I believe that they would be pretty close to too long (for two) to work well.

Bill, I'm working days this weekend, but feel free to call any night.

 

[Ginseng]

Bill,

I'll be around.

Wilkey

 

[Tweek]

I've seen a lot of positive comments about the double-barrel battery form factor in this thread, which gave me a thought. Howabout a rounded square barrel? Such a shape would easily accomodate 8 cells in a 2x2x2 layout. One could easily use AA's, or even 123's or 17670 cells without exceeding the diameter of a Mag 2D. If you didn't have a lot of wasted space between the batteries and the bulb (aka Mag switch), 12 AA's crammed into the volume of a Mag2D is quite doable, and with todays 2500+mAh AA cells, that's quite a bit of energy on tap.

I think something like this would feel really nice in metal, I tried it with a paper towel tube as a prototype. Even if I didn't have my fingers all the way around, it still felt nice and secure, and comfortable.

The only issue I could see being a problem is if focusability would be needed. Fitting a round head on a square tube isn't that hard, but having to turn it afterwards would be complicated.

If you wanted to, you could get really custom and go with a squared off head to match the tube... /ubbthreads/images/graemlins/grin.gif

Chris

 

[andrewwynn]

look at http://rouse.com/mm 'rounded square'.. that is 4x1 layout of AAs but i've been planning to make a 2x2 or 2x1 18650 design with a pair or 3 lux3s up front.. it will be far more compact than the mag type lights.. just been too busy with the other projects to actually produce a model.

-awr

 

[Ginseng]

Tweek,

Do you remember the Energizer flat double barrel lights? These were small, low performing lights that employed the DB concept. IIRC, they came in 4, 6 and 8 cell AA setups. Clearing out some space inside the barrels allowed one to put CR123 cells in there instead. They had decent but not unlimited modding potential due to the weak switch-head region. They had extruded aluminum body tubes and if high volumes were required, that would be the way to go.

I think a four-barrel makes sense from a compactness standpoint. It's not the most efficient packing but would result in a grippable cross section. 4x of SF123 would be pretty fat in a tube but not bad at all in a square section.

In fact, a three stack body might not be bad looking either.

Larry,
Mouthable is definitely a requirement for my 24/7 EDC. I predominantly use the SF L4 gripped in my mouth.

Wilkey

PS. Does anyone remember who made the mod called "Snake eyes" some time back? It had two Lux IIRC.

 

[bwaites]

Well, a 3 stack 123 body would be the diameter of the Surefire M6, or roughly the size of a Tigerlight.

Thats too big for me!

How about a side by side case big enough for Pila size cells?

That might be "mouthable", but I'm thinking the better solution would be a small rubber based endcap with a bite block type holder, made out of mouthguard type material!

Bill

 

[andrewwynn]

Mr Bulk made a side-by-side light it might have been the snake eyes you speak of..

the MM will easily be bitable if it had the rubber cap on the end like you speak. It won't have the throw of a bigger light but it will surely output a lot of light.

-awr

 

[Ginseng]

Bill,

An M6 is a bit fat for me too, no doubt. But try this, grab three 123's or 3 of your A cells and see how that feels. Without the large tubular bore sweeping out the major diameter, three 123s is actually a pretty comfy grab even in my small hands. But I do hear what you're saying and a flat profile might be a nice option. Still, there's something very classic about the simple cylinder.

Andrew,

Thanks for the reminder about the Snake Eyes. On bitability, I think that a light need not be designed expressly for biting, just that its design not make the occasional bite hold overly painful or awkward.

Wilkey

 

[Icebreak]

Here are some unorganized observations, thoughts and stuff I've heard:

Defining the goal is important and I think that's what this thread is about?

This might be a challenge but ultimately the team that does the work will make the decisions. Surely one component of the design goal would be that the light outperforms existing flashlights.

EDC capability is a little different for everybody. For EDC I'll give up degrees of runtime in favor of small size and large output parameters.

Power source seems to be a challenge. I'm thinking the gurus may be on to something uncommon to be revealed later?

Heat is an important part of the incandescent process. Too much heat can be bad for connections, switches and the batteries closest to the bulb. A reflector with a cylindrically shaped outside diameter or an OD shape that conforms somewhat to the inside diameter of the head can be very effective at sinking unnecessary, excess heat into the bezel and body.

Point source is no longer something I just read about. It exists. Projecting 1000 lumens out of a 20mm reflector with a Welch Allyn sized filament has purpose and is a darn fine thing to observe but projecting 300 lumens out of a 30mm reflector with a Surefire sized filament is going to out throw the former significantly.

.10 mm laterally = 100 yds. axially.
.10 mm axially = 1 mi. laterally...eventually.

Good luck guys. I know your experiments can be expensive and your work time consuming. I appreciate it and will be watching.

--------

- Jeff

 

[bwaites]

Icebreak,

Very coherent observations and thoughts!

I think your first statement is VERY important.

Defining the goal IS what this thread is about. I can come up with about 30 lights I'd like to have, but the truth is that they would fit ONLY what I want, not what most people want!

We need some guidance as to what most people are looking for and when I talk to most of my non-flashaholic friends, they say... "Half the size and twice as bright as a Maglite". Then I say, "Which Maglite?" and they say, "There is more than one?"!

So what I was looking for was the general feeling of what people wanted, because I recognize that when too many hands are in the soup, you may end up with stew! Put another way, "The camel is a horse designed by a committee."

So I don't want finished ideas, I wanted a general feeling.

I'm starting to get the feeling that not many people actually KNOW what they really want, but they have ideas of what they think they want, and thats ok too!!

Bill

 

[HarryN]

I will not claim an expertise in hot wire, but I do have a few observations culled from reading, bench experience, and some rather painful and expensive build projects

1) Cells Failing / Splitting
- I suspect that the battery packs for high powered lights are being too tightly packed, without adequate means of heat removal.
- There is a tendency for pack builders to wrap up the packs with lots of kapton tape, which is a relatively poor conductor. While this is good for electrical protection, perhaps a thermally conductive tape, such as Al tape would be a benefit if properly applied.
- Virtually every cell will start putting out substantial heat from internal resistance. That heat needs to be conducted out to the body, not held in.
- There are some experiences out there with very substantial "events" with high powered cells which are best kept well isolated from the end user. These requires thick walls.

For this reason, I have been working on an Al body which has two, completely independent "bores" for cells. This is not the lightest design, but it has the potential to rapidly carry battery generated thermal energy out to the surface. It is also built with battery event containment in mind.

2) Round lights are cheaper to build, but not necessarily the most ergonomic solutions.

- It is not hard to see that there is a practical limit to how large a tube can be and fit in the hand.
- It is also not hard to see that drills and lathes are good at making round things.
- There are challenges and costs related to making side by side battery setups, but it does allow for a larger than normal size to be held in the hand (since it is now flattened)

3) You need to build from the perspective of "power desired" and "realistic cell capacity" , not "desired size".

- While it can be somewhat disputed, it is challenging to run normal 123 cells substantially beyond 1.5 amps safely, and 1 amp is really a nice place to run them.
- 3 amps is getting toward the upper end of many Ni based setups, although there are exceptions.

For my setup, that means with 8 cells, I am roughly limited to 8 x 2.5 V x 1.5 A = 30 watts. With NiCd, it is roughly the same power, but the output voltage is lower.

Since this body can also hold 4 x 18650s, similar power levels are also available from it that way.

I remain amazed at the Wilkey lights and their apparent ability to reach out toward 100 watts. This continues to humble my own work.

4) Incans bulbs do not appear to output Visible Lumens using the simple formulas being applied for over driving them.

- I know I am on thin ice here, as the incan masters are MUCH more experienced in this area, but the data being presented does not seem to support the mathmatical extrapolations being made.
- There certainly is more light coming from these overdriven beasts, but the improvement seen is not as great as predicted, indicating that the losses due to IR vs visible emission MIGHT not be a constant at higher drive currents.

To me, this implies that one of the real challenges of building a better incan light is proper voltage regulation. To get more power from cell packs, and the optimum drive voltages, it is necessary to disconnect the two from each other, and use a regulated supply.

In my case, I have purchased some 30 watt, constant voltage supplies (not pwms). These allow wide voltage input, relatively lower cell currents, and the ability to drive the bulbs at a relatively optimal drive voltage.

This is why I have set up the body with space for a 30 watt, 12 volt power supply. (also matches the cells) It is officially rated at 2500ma, but not yet tested.

Remaining to be done, build up or buy a head for this unit.

If anyone is interested, I will try to take some pics and post them. (keeping in mind these are protos)

No way will this design hit the 100 watts level, but some of the concepts might be useful.






-

 

[andrewwynn]

I really love the 2x2 layout.. 18650 or 18500 x 4 would be a great set up.. smart thinking regarding the thermal heat management of the batteries.. I would love a double-barrel design that has a pair of 20MM IMS reflectors up front.

-awr

 

[Ginseng]

Wow,

Icey, cutting right to the heart of it. This really is about defining purposes and I think our job here is to get as many thoughts and impressions voiced so that the designers can assemble a design spec that hits a sweet spot that may be unvoiced...exactly as Bill says.

Harry,

I think that I may be biased or predisposed to incandescents. Perhaps this is a reflection of my own weaknesses in things electrical.Your points have always been cogent and I think I should look back through our correspondence with a critical eye. You have proposed some fantastic ideas and perhaps a moderate but intelligent middle ground is not a bad place to be. Heck, even a hot wire maniac might be expected to mellow in his old age.

Wilkey

 

[js]

HarryN,

Kaptom tape is perhaps the BEST electrically insulating layer that is ALSO as heat conducting as possible for such a product. Also it is very strong and thus can be very thin--which increases heat conduction. It is better than shrink wrap. This is part of the reason that the RC community people use it. The other reason is that it can hold up to very intense heat and not deform or melt.

Anyway, one could obviously NOT use Aluminum tape for this application! Because it would electrically short cells in a stack and between stacks!

One certainly should think about heat dissipation, to be sure, and, for example, in the USL battery packs, this is why there is no shrink wrap around the overall pack, because it would add yet another layer that would resist heat flow. But, really, battery pack generated heat is not much of an issue with high current NiMH or NiCd cells because they have such low internal resistance. For example, any of the high rate A sized NiMH cells typically have 6-10 mOhm of resistance. That's .006 to .01 ohms. It would take a 3 or 4 amp draw to even drop a tenth of a watt of power dissipation inside them.

I have absolutely no idea where you came up with 3 amps as the upper limit to Ni based applications! The USL or Aurora is pushing 8 or 9 or 10 amps, depending on the lamp used. And many RC applications are drawing 20 or even 30 amps from NiMH cells. The GP 1100 2/3A which I use in the M6-R packs, for example, are quite capable of delivering 20 amps.

As for cells blowing up or leaking or what-not, NiCd and NiMH are almost immune to exploding, from what I have seen. They may heat up and leak and be destroyed, but they rarely, if ever, explode, even when short circuited. This is one of the reasons I like them so much, over, say, Li-ion cells. That, and their ability to handle very high draws. 100 Watts? Oh yes.

As for incan bulbs and re-rating formulas, they are reasonably accurate as long as you stay fairly close to the confirmed design ratings. The phenomenon you are talking about is due to the physiology of the eye. It takes at least a 50 percent increase in lumens to make a noticeable brightness difference to the eye. And it really takes a doubling of lumens to get to the point of feeling that you have a significantly brighter light on your hands. Thus brightness does not scale linearly with power. A 100 watt light is not 10 times as "bright" as a 10 watt light.

Voltage regulation is very important for getting the most out of incans, but it is not really necessary, especially if you aren't that concerned with the occaisonal instaflash and short lamp lifetimes. The USL or Tiger11, for example, are driven so hard that they are more efficient at the end of cycle than almost any production light is at the beginining of cycle.

 

[Codeman]

[ QUOTE ]
js said:
...
A 100 watt light is not 10 times as "bright" as a 10 watt light.
...

[/ QUOTE ]

I don't know why, but I seem to need reminders about this from time to time.

There are far greater minds at work in this thread than mine when it comes to practical design ideas, so I'll go back to

/ubbthreads/images/graemlins/popcorn.gif