A gift to Grandpa, modding an old fashioned 2D

Hi, all.

I vividly remembered when I was as young as 6 and playing at my Grandpa's house--he had this 'stack' of heavy duty batteries in a cupboard with a very primitive voltmeter(a light buld soldiered with a wire), along with two shinny tin flashlight...and I believed that was the beginning of my flashaholism.

Recently, I had visited Grandpa and saw those two tin flashlight, good as new, sitting in his cupboard. And I was thinking to myself--what better gift than a mod that was made by myself as a tribute to someone who had started it all?

Finally, I did find some China made flashlight that resembled what he has: a 2D tin flashlight, with on/momentary/off swith, adjustable focus, candle-able...

Now, what I had in mind is a Lux III mod, powered by 2D alkalines, semi-fully regulated, and has decent runtime. And hopefully minimal alterations to the original flashlight.

Any suggestions of electronic/hardware components are welcomed!


Here are some pictures of this flashlight:

Tigerhead flashlight; purchased with less than 1 USD...


Closer look at the front end, notice the switch is a sliding one(off > momentary(by pressing red button) > on), also the reflector is orange peeled for better beam quality. Err...not to sounds like a salesman, the bezel is also anti-roll too! /ubbthreads/images/graemlins/wink.gif


Candle mode? The 'light engine' could be unscrewed as shown later.


Looking down the front end of body tube, without the 'light engine'; the metalic strip inside will connect to a rim at the back of light engine to complete the circuit.


Light engine. The inner rim is where the metalic strip touches it.


A look at the 'tailcap' which is string loaded inside the neg contact; back of the light engine again:


A D-heatsink by Yaesumofo will fit loosely inside the tube, with the tailcap spring, it will move up and down which needed to be fixed--any suggestion?(themal epoxy?)


Rising the heatsink above flush level, the reflector will rest around the raised mount in middle.
Note 1: it is completely off centered since the heatsink is loose inside the tube and currently being pushed by the metalic contact strip to one side.
Note 2: if compared to the second picture from top, you will see the filament is rising way above the reflector--maybe to better capture the light radiating from the bulb...but I do worry the light from Lux is more forward going, rather than radiating, will a Lux raised high above reflector actually create better reflection?


Battery in tube; there is about 5mm possible displacement of the batteries due to the spring at tailcap:


Tentatively, I am think of using a BB NG 400, but I am not sure about the runtime and how well it will be regulated...if anyone has any suggestion, do feel free to share! /ubbthreads/images/graemlins/smile.gif

This is great! Unfortunately I can't fully digest the info right now... still at work. /ubbthreads/images/graemlins/frown.gif

That said, I've been meaning to mod a vintage flashlight myself, and even have one sitting in the Milky labs waiting for surgery, heh heh. /ubbthreads/images/graemlins/naughty.gif Thanks for the inspiration! /ubbthreads/images/graemlins/blush.gif

Hi, Scott!

It is me who is indebted to the CPF for this inspiration. Any input from you and our fellow CPFers will truly be apreciated! /ubbthreads/images/graemlins/smile.gif


Right now I am still comtempleting what to do to get that heatsink fix(of course, it will be most ideal if I could make it 'switchable' with a connector, but it is not likely) and the proper electronics for this mod(probably TW0J, semi-full regulated, ideally 3-4+ hours of regulation).

Hi AuroraLite,

This is a super idea for a gift for your grandpa. He's going to be thrilled with it, especially because of the thought and effort you put into it.

I like to see vintage or vintage re-issues updated with LED systems. Good Luck and update us on your project.

Britt

A spacer of aluminium foil works wonders. You'd be surprised how well it conducts heat, and how tightly it can make things fit. You may want to wrap the O-sink with a couple of layers, enough to make it fit, then epoxy everything down when you are confident, or just push it into place so it is removable.

Heat should not be an issue, I have driven a Lux1 at 400mA on a much smaller heatsink and it hardly even got warm. Unless you are pushing 1 amp, you don't need that much thermal transfer to the walls of the light. The foil will take care of that for you, though. I have a Mag3X (3xLux3, 700-900mA EACH) with a foil shim and thermal management is very good. It's handling somewhere like 9W of power!

Runtime will be simply incredible. Lux3.. let's take it as a L Vf bin (though you can do MUCH better):

Output voltage 3.7 V
Output current 0.4 A
Output wattage 1.48 W

Input voltage 3 V
Converter efficiency 0.85 (estimated. Probably much better)
Input current 0.581A


Duracell's datasheets say that you'll get close to 16 hours of light at 0.75A draw. It should be pretty close to that considering that as alkaline batteries drain, voltage falls and the converter will pull more current to make up for it.

Britt:

Thanks for your kind words of encouragement! I shall definitely keep everyone posted! /ubbthreads/images/graemlins/smile.gif


Kevin:

Great info, and many thanks for the equation, I kinda learned a tiny bit of EE from our online course herein CPF. And from your equations and my understanding, I am considering about the possible use of BB NG 500 or BB NG 667(as suggested by Chop, from Shoppe forum; Thanks, Tony!) with a Lux III T**J:

Electronics
BB NG w/ T**J:
Output Voltage = 3.51V (worse case scenario)
Output Current = 0.5A or 0.667A
Power = 1.755W(BBNG 500) or 2.341W(BBNG 667)


Input Voltage = 3V
Converter efficiency 0.85 (worse case scenario)
Then, input current = 0.688A(BBNG 500) or 0.918A(BBNG 667)

Chop had given me a approximated guess for 4-5 hours for the 0.667A BBNG...will it be safe to assume the 0.5A BBNG has about 10 hours?


Furthermore, is it also safe to assume a somewhat linear relationship between brightness(if current increment positively correlate to brightness increment) and runtime for these three levels of BB NG with 2D?

I know it could be all theoritical, but at the least it could give me some approximation of reality. /ubbthreads/images/graemlins/blush.gif

Btw, Kevin, where did you get the runtime/data about the duracell? Is it calculated from the listed capacity of their D cell? Thanks in advance!


Heatsink issue
And thanks for the tip about Al foil! It is reassuring to find know it works so well! /ubbthreads/images/graemlins/grin.gif I think what I will do is to wrap Al foil around the heatsink until it is a snug fit into the body, then maybe wrap a thin layer of heat resilent tape to secure it.


Now, I am onto studying about the reflector shape and the lux position!(more pix follow!)

I went to Duracell's datasheets, selected the appropriate battery, and found a graph for the various discharge rates in amps. Then I eye the graph for a little while and say "I think that's approximately how much I should get" /ubbthreads/images/graemlins/grin.gif

I use the graph because it takes into account the non-linear capacities of alkalines at different discharge rates. Duracell's done all the hard work figuring out how long, at what amperage. It's the closest approximation that I can ask for without building and measuring the light itself.

McGizmo has some great numbers on how the brightness gains diminish as you go higher, it's been said that 500-700mA offers the best mix of brightness and power drain. You'll have to look around for those, I have no idea where they are.

Glad I could help /ubbthreads/images/graemlins/smile.gif

Hmm... I was just explaining something on brightness to a potential modding customer and fellow CPFer, and realized it's actually a little more involved than I had been thinking before. Here's what I now know (or think I know!) on brightness as well as battery life, in stream-of-consciousness mode... /ubbthreads/images/graemlins/wink.gif

First of all, battery life. What I do as a quick and usually fairly accurate estimate, is to start with the battery's capacity, multiply that by 0.7 (to account for voltage sag as the battery drains, increasing current flow demands from the regulation circuit, etc.), then divide by the current flow observed coming from the batteries on fresh cells. On this latter item, you'd need to use the EXPECTED current flow from the batteries if you haven't built the light yet! /ubbthreads/images/graemlins/ooo.gif The result gives me expected runtime in regulation, in hours.

Example: Let's suppose a modded KL1 gen2 head has current flow off a single 123 cell of 630mA. A typical 123 cell has a capacity of 1400mAh. Therefore...

1400mAh x 0.7 / 630mA = 1.56 hours, or 1 hour 33 minutes.

Brightness consideration #1: McGizmo and others are right in saying that brightness gains diminish as current flow through the Luxeon goes higher. Quite some time ago, while constructing his MR-X mod, Hotbeam posted a little chart of brightness vs. current flow through the X3T emitter in his light, and showed the percentage gains along the way. It was pretty clear that there were diminishing returns as one went along. If memory serves, doubling the current from 750mA to 1.5A only increased brightness by something like 38%.

Brightness consideration #2: the human eye perceives brightness in nonlinear fashion... my present understanding is that it's more of a logarithmic relationship. (That's log base 10 for those who care.) /ubbthreads/images/graemlins/smile.gif What does this mean? As an example, let's suppose you succeed in raising the lux reading for your light from 600 lux at one meter, to 1200 lux at one meter. Cool... doubled the brightness, right? Actually, yes... but your eyes won't think so. The percentage increase your eyes will see will be...

log(1200) / log(600) = 1.11, or roughly 11% increase.

Combining the two, that means the overall picture is actually a little WORSE than McGizmo paints... doubling the current flow through the emitter will generally NOT double the output... and the higher the current flow to start with, the smaller the output difference at all. Meanwhile, since the output hasn't even doubled, what the eye will perceive is an even smaller percentage increase in brightness. You've also increased the risk of cooking your emitter, cut battery life in half (or even worse), and shortened the lifespan of the emitter significantly even if it doesn't get fried immediately.

For anyone who's got one of MR Bulk's VIP lights, switching from medium to high brightness demonstrates the point... the brightness really doesn't increase all that much, but you've just reduced your runtime from 3 hours or so, down to maybe 20-30 minutes! /ubbthreads/images/graemlins/icon15.gif That's not meant to criticize the VIP at all... quite the contrary, IMHO it's a brilliant piece of workmanship, and one well worth having. I'm keepin' mine! /ubbthreads/images/graemlins/tongue.gif But to me, the genius of the VIP is that bright, white, super long running regulated medium brightness level, plus the fact that those other levels are there should you ever need 'em. /ubbthreads/images/graemlins/grin.gif

Final comment: I've been wrong before and might be here as well! So take this as ramblings from someone who has some clue about what he's saying, but is DEFINITELY not the final authority on the matter. Anyway, hope the info helps! /ubbthreads/images/graemlins/smile.gif

Scott's right, for the most part. The law of diminishing returns runs rampant in the world of luxeons, both in terms of actual output and perceived output.

If you want to build a good, solid mod, go with a 700mA bias.

My Grandpa gift is so blessed with your inputs and advices! That alone is a great motivation to get this done!

Kevin:
Thanks for the link, though I am a total newbie in the EE department, I will try to learn to read those charts and benefit from them!


Scott:
Great read and truly insightful! It is so ever true of you have said, I did wonder about the same thing when I was looking at my BBNG 750 and BBNG 500...honestly I could only preceive some difference, but not as much as 50% with my eyes.(also tint factor affects the result too).

If I get measurement of what current is flowing from the batteries to the sandwiches at the tailcap, will that be a usable info for your equations?(I know it is rather different battery type of AA vs D, but I don't wanna risk anything to my sandwiches too)


Tony:
Haha...you are always there to save my day, thanks so much for your input! What does it mean when you say 700ma 'bias'? What is that 'bias'?



Update
I did play with the reflector and it is so interesting to see various shapes of reflectors really do give different beamshots.

As for this reflector, when it is focused, I found it to be all spot and no spill whatsoever. And I discover in order for the Lux to be in focus, it need to be raised approximately 5mm above the flush of reflector.

Now, the headache part is finding the right material(Al stick?) to raise it to the right height. /ubbthreads/images/graemlins/icon15.gif I will consult with my fellow HK CPFer if I don't find the material soon.

Some pictures of my experiment and beamshot will be posted tomorrow(it is a little late, weee past my bed time /ubbthreads/images/graemlins/wink.gif)

On the other hand, a BBNG 500 and a Madmax board had been ordered, and is on its way to me...

I know it's not quite vintage, but for Christmas I made my stepfather an LED light by starting with a Mag 2C host, gutting it, cutting the cam off the reflector, mounting an aluminum disc at the top of the battery tube as heatsink, mounting a Luxeon on that, and driving the whole thing using (what else?) /ubbthreads/images/graemlins/wink.gif a Dorcy 1AAA boost circuit. The output current was probably only 400mA, but with that big Mag reflector it could throw the beam surprisingly far. I also applied some WriteRight film to the INSIDE of the lens to diffuse the beam just a bit, so he could see not only down the road, but the sides of the road as well. The whole thing ought to run for quite a while on a good pair of C alkalines.

Fortunately for me, he actually LIKED it! /ubbthreads/images/graemlins/party.gif /ubbthreads/images/graemlins/grin.gif

BTW, if you find that you need to raise your emitter a little farther into the reflector for optimal focus, you can go to any local hardware store and look through their aluminum screen door hardware... you can find solid aluminum thumbscrews that mount nicely in a threaded hole and/or with a bit of thermal epoxy, and once you grind the top flat, the emitter can be mounted on it. I did that with my stepfather's light, too. /ubbthreads/images/graemlins/smile.gif

Scott:
I only hope it turns out to be an ok mod, and did not become a fire hazard in my Grandpa's house by the end of this. /ubbthreads/images/graemlins/icon15.gif /ubbthreads/images/graemlins/smile.gif

Hmmm...the screen door thumbscrew sounds good, except I am a total idiot when it comes to locating this kind of stuff...do kindly check out the copper 'tube' below and let me know what you think. /ubbthreads/images/graemlins/blush.gif


Update:
Picture time!

Reflector
To learn more about the focus point on the reflector, I simply put the reflector on a tin surface which closed resembled the heatsink, and use some erase-clay to help locating the focal point of this reflector:


Here I use a Lux I PYAH(high dome), powered with 3V 2AA cells:


Surprisingly, the reflector really gives all spot and no spill whatsoever--I can't say it is the most useful beam for most occasions, but I guess this is probably what Grandpa used to since the shape of this reflector resemble his. Beamshot on white ceiling, 2+ meters away.

Left: TX1J BBNG 750, 20mm IMS reflector
Right: PYAH with 3V 2AA with the target reflector:
(Note: Either case does not have full batteries, and I don't know much about the white balance--so it could be a little off from reality)


In short, it seems like the Lux will have to be raised approximately 5mm above flush to be in focus.


'Light-engine'
I found a AA to C adapter fit almost perfectly into the hollow at the bottom of the D Hotlip. And later on I will dissect this puppy in half to served as a contact point for the positive and host the convertor board inside:



The following is a little copper tube that I found for raising the Lux to the focal point. It is a bit longer than 5mm, but I might grid it down a little bit or leave it as is since the mod is supposed to be focusable like a Mag:


How it looks when placed under the emitter(PYAH, just as demo):



Thanks for looking! Any comments on any parts used or ideas are truly welcomed! /ubbthreads/images/graemlins/smile.gif

Aurora, that copper post looks fine to me... copper is a better heat conductor than aluminum, so it definitely beats my idea! /ubbthreads/images/graemlins/yellowlaugh.gif

The one thing I would do is fill the hollow space inside with a heat conductive material. If it were my mod, the first thing I would try is stuffing crumpled aluminum foil into the opening as tightly as possible, then smoothing the aluminum that's at either end of the tube. This way you get greater surface area to place in contact with the emitter, and a better thermal path overall from emitter to heatsink. /ubbthreads/images/graemlins/smile.gif

BTW, if the beam really seems too narrow, you can cut a cicle of any sort of frosted or textured material, such as WriteRight or maybe even textured adhesive tape. That would cut down the distance the light travels and dim the beam by a few percentage points overall, but also redirect some of that hotspot to the sides in a nice, smooth dropoff of intensity as you move away from the hotspot. It IMHO makes a really beautiful (and useful!) beam.

Hi, everyone!

I am happy to report back that the Tigerhead 2D mod(a gift to my Grandpa!) had finally finished!

Many thanks to Kevin, Tony, Scott for your technical support; BC0311 and Cy for your kind words and pictures. /ubbthreads/images/graemlins/smile.gif

Special thanks to Kiu and AW, without you two, I wouldn't have a glue where to find this copper rod/headsink! /ubbthreads/images/graemlins/wink.gif


Here are some of the pictures:

After a great night-hike gathering with some HK hikers and CPF HK members /ubbthreads/images/graemlins/wink2.gif, I did get the idea to where to find and the use of 8mm copper rod as a heatsink and raise the TW0J to the right level.

Note: the red tape around the D lip is to prevent shorting, since the wall of the body does conduct electricity:



Here I use the AA to C adapters as the anode and convertor holder. Originally, I did want to use BB NG 500 but regertably, I think I might have damaged the board. Now it is hosting a Madmax WO:



Using two D-cell to dry run it and double check whether heat is an issue:



After all the assembling, test and trial with the best way to conduct electricity in this body...finally, putting it in use!

Note: The lux is TW0J, it looks yellow only because it shines on yellow wall paper... /ubbthreads/images/graemlins/tongue.gif


Another beam shot on ceiling, this picture somewhat accurately protrait the brightness and the tint:



The Tigerhead 2D lux III mod could also do candle mode and has momentary switch as well!! /ubbthreads/images/graemlins/grin.gif Hope this puppy will put a smile on the old man's face...Thank you for looking!



Next:
Another Tigerhead 2D and 2-3D(with optional extension, bigger reflector...). Close friends and families somehow had seen this mod in process, and now I am bound to make some more /ubbthreads/images/graemlins/tongue.gif I will update this post once more when the 2-3D is done for beamshot comparison. /ubbthreads/images/graemlins/blush.gif

AuroraLite,

Nice Tigerhead !! I may get one also /ubbthreads/images/graemlins/happy14.gif

Congratulations on getting your light up and running! /ubbthreads/images/graemlins/happy14.gif

The nice thing about the MadMax is that it will really drain the alkaline batteries as far as they will go. In fact, the Madmax has been run on just one battery alone.. that shows how low a voltage it can accomodate. No need to worry about wasting whatever power is left in those alkalines.

About the runtime, I just look at the Constant Current chart in the datasheets, which lists the discharge curve for various amp rates, then I approximately guess how many hours I will get. Of course, now that you have everything built, you can run a timed test which will be more accurate.

Your use of the copper rod is very resourceful, I myself was wondering whether the focus would be optimal, and your beamshots show that yes, it is.


My granddad passed on just a few months after I'd gotten my E1e (my first REAL light) and a couple of months before I'd discover CPF. I remember him showing me the old school Eveready chrome lights 15 years ago, and just as a thought, it would be fun to show him the current state of the art that I carry - the U2, lithium ion/lithium mag dioxide, variable power, and so on, compared to those Eveready lights. How times have changed..

Nice Job!!!

now consider sticking 2x 11,000 mah D cells (NMH)

AW:
For sure, I will take this one along for the next gathering if I still have it with me(or a new one...who knows)! /ubbthreads/images/graemlins/grin.gif


Kevin:
Thanks! The copper rod do help to make the focus...but after I have been playing with it for a while, I remember/understand that the properties of this reflector is unlike the SureFire or Mag, where you could stay focus once you have it(kinda straight forward beam)...it will need refocusing at various distance but one can do it manually.

Funny how everything has evolved and this does give one a glimpse into the past--maybe that's what my Grandpa is used to when he is my age. /ubbthreads/images/graemlins/smile.gif Sorry to hear about your Grandpa, but I am sure he will shine upon you from where he is now. /ubbthreads/images/graemlins/blush.gif


Cy:
Hmmm...I think Grandpa is too old to bother with rechargable now(just use and toss, thus the Madmax), but on the other hand, I could... /ubbthreads/images/graemlins/wink.gif

Update:

Hi, all. It has been a while since I mentioned a 'twin' brother for my Tigerhead mod.

And a few weeks ago, finally, I mod another Tigerhead but this time with a Turbohead. /ubbthreads/images/graemlins/evilgrin07.gif


The Tigerhead Twins

Left: Tigerhead with 'Turbo reflector', Original Tigerhead with 50mm Carlo 2 degree Optics:


Basically, I hunt down another Tigerhead 3D version, and swapped the reflector/head assembly with a normal 2D version. And then with a longer post to raise the Lux high above the reflector, a 'Turbo head' version is created /ubbthreads/images/graemlins/smile.gif


Inside the Tigerhead Turbo is the Lambda DC-DC convertor which gives about 550ma to the Lux TW0J. I test run the whole thing for an hour or two, and the output seems to be quite regulated for my naked eyes. And should I continue to run the test, I guess it will probably last many hours before it starts to drift into moon mode.


Candle mode for the Tigerhead Turbo /ubbthreads/images/graemlins/grin.gif


Beamshot, approx 1 meter away from the wall, it gives a very condensed hotspot(approx 1-2 inches) and virtually no side spill. Almost identical to the original Tigerhead, but more intense:




New optics for Tigerhead MadMax

Then I feel kinda sorry for the orignal Tigerhead, and decided I should re-mod it again with a better optics. After a little searching, I replace the original dish reflector with a 50mm Carlo 2 degree Optics. And the result is a always focused tight spot:



Less than 1 meter from the wall, with 50mm Carlo 2 degree:



Finally, Grandpa is said to come around early May--and I am sure he will probably be surprised to see his old friend with a new outlook! Right now, I am planning to give him the Tigerhead Madmax with 50mm Carlo Optics, since it would not require him to re-focus. /ubbthreads/images/graemlins/smile.gif


Thanks for reading!