I have not completed a sketch of the internal construction of the switch, but since there has already been so much discussion of my various options, I thought I should at least post a sketch of the exterior and explain how the switch actually functions. The stock switch is a momentary push switch that can be locked “On” by operating a “slide” to hold it in the down position. Pushing the momentary switch lowers a conductive “lever” down at the front of the flashlight in such a way that it makes contact with the negative terminal of the existing 3V incandescent bulb assembly. One thought that has occurred to me is to keep this existing mechanical arrangement, and simply not run any current through it. If the negative terminal of the battery holder were isolated from the flashlight body, I think I could (possibly?) run a good wire (or copper braid) up along side the D-Cell adapters and somehow attach it to the existing “lever” such that it will make contact with the negative terminal of my custom Driver/Pill arrangement, when the stock button is depressed.
Also, I have been trying to obtain some definitive information on Eneloops, and it appears to be pretty certain that Eneloop Pros can handle 5A, or even 7A, if necessay. Please take a look at this chart from HKJ’s battery comparator:
So, ideally, I was hoping to keep the entire stock switch assembly (so in the unlikely event that I ever decided to “restore” this flashlight, it could be done). But, even in the worst case, I must retain the current external switch, as far a appearance. I feel it would completely defeat the purpose of my project, if a modern switch were visible on the exterior of the completed flashlight. (The “fun” here is in being incognito!)
Therefore, I am wondering whether or not I should fundamentally rethink what I am attempting to do. Perhaps it is overly ambitious of me to attempt to drive a 6V MT-G2 at 5A or 7A? Am I not simply making my task a lot harder? Just how much brighter is the MT-G2 than an XM-L2, if were to attempt 5A or even 7A in the MT-G” in comparison to a “tame” 3A into the XM-L2?
Having decided (firmly?) not to use any modern optics and simply direct a single LED at the center of the original glass aspheric lens, it seems to be that my choice is between the two emitters previously mentioned and pictured below:
Am I wrong? Is it going to be “worth the extra effort” to use a MT-G2? All comments and advice will be greatly appreciated.
If you use multiple emitters with that glass lens that came with the light, the lens will be change the angle of each beam. If you have 4 emitters you’ll likely end up with 4 separate beams.
Yes, the MT-G2 is worth it but it's not really extra effort when you remember you're
1. Greatly simplifying the optics. Trying to make a triple reflector or TIR work well under the original lens will be frustrating. Plus you save the cost of this component.
2. Relaxing driver accommodations. A buck driver is very often tall due to a toroid inductor, often need 10mm space.
3. Simplifying power. Since you like 6 (?) AA's in series anyway, the MT-G2's Vf plays into your hands.
4. Making it more elegant. There are flashlights with 50 to 100 or more 5mm LEDs. They will light up a room, but nobody would call it an elegant design.
I think you might possibly be able to use an e-switch FET driver and use the stock switch to control it. Use the slide in off and mom. as a lockout, and the momentary button to signal the driver.
Regarding the ridged shell, you might want to consider not making the pill a very tight fit, so you could fill the space with something that conforms and fills the ridges. 18sixfifty has a thread on using JBWeld mixed with silicon carbide (lapidary grit) s a potting compound. I think it would make a great space-filling material with good heat transfer properties. You would avoid making high-tolerance parts by hand, and eliminate the ridged shell to pill air spaces. Depends if there's a good way to wire B- to the driver in the available space.
Thanks for your statements. It further confirms what I “thought” I was hearing from others, but your statements are very clear. I think I have now settled on the MT-G2, once and for all.
I was hoping not to have to do anything “permanent to this old flashlight. Of course, if it comes down to it, I will certainly go with a non-reversible custom build, but I feel it might be nice (and more ”elegant” as you say), if the custom parts were all removable. Ideally, I could just take them out and put the old parts back in.
Is your proposed JBWeld removable? (In my experience it is not.) I was thinking about going for a fairly tight fit, and attempting to fill any gaps with something like the following material:
EDIT: Photo deleted. The product depicted was not genuine copper wool. It was “fake” copper colored mylar.
This is just an idea. I am not certain whether such “copper wool” would be practical to “stuff” as tightly as possible between the “pill/heat sink” and the flashlight body (hopefully filling the threads). And, I am not even certain yet whether or not this item being sold on eBay is even real copper metal (I am asking that question of them), but I do know such material does exist out there.
I addition, I am planning to build what I believe to be a very heavy duty copper pill assembly. Considering the very short runtime that I can expect, I am kind of hoping that the pill/heat sink can temporarily absorb enough heat to keep the LED safe, without the need for all that much immediate heat dissipation through the flashlight body, and out into the external world. But, that may simply be a pipe dream. I guess I won’t really know that until I get a chance to test the assembly and see just how hot it gets and how quickly.
Unfortunately, I still haven’t been able to locate a good source of small quantities of genuine copper wool. But, the following product, aluminum wool, is readily available on eBay and should still have fairly good thermal properties. (Half as good as pure copper, but twice as good as brass.)
This product is apparently “real aluminum” (EN AW 1050A alloy). This may be a crazy idea as a method of improving the fit between a heat sink and an irregularly shaped flashlight body. I can already foresee potential problems with “stray” strands shorting something out.
If your light engine is a snug slip fit on the inside of the ribs, how could you possibly pack a meaningful amount of wool into the troughs of the ridges?
The aluminum wool you’ve pictured is clearly very coarse, much moreso than steel wool for example. In fact, it doesn’t look wooly at all… it looks like a bunch of parallel fibers strips stuffed into a bag!
I’d guess that the performance of anything other than the tightest packing of this stuff would be exceeded by a solid pour of silicone. You can load silicone with powdered copper, aluminum, or silicon carbide if you want. I don’t know how much you can get in there and still have it pour nicely, but I imagine a 10-30% mix would still pour.
Thanks for your comments. That is precisely why I posted the photo, so that others could see and easily understand what I was thinking. (Obviously, not a very good “thought” but I felt it might be worth consideration.)
I have no experience or knowledge of the silicone you mention. Can it be removed?
Yes, it can be removed. It is often used as a sealant and remains pliable. That said… I’m not sure that you can count on getting a snug light engine which was slathered in silicone out without damaging the host. Hopefully others with more experience in the matter may chime in.
Silicone is often used a window sealant & for similar tasks. You almost certainly have run into Silicone Caulk before.
The problem, (well, one of the problems), with a press-fit pill is that the steel tube may not be strong enough. I had a couple of ribbed steel flashlights ('60's-'70's era) split open by alkaline cells that swelled inside. Is your light made out of a solid tube or sheet metal rolled into a tube with a seam?
If there is enough space for stuffing material, maybe these could work? Should definitely be reversible, at least.
How original does it have to be? What’s a little JB Weld inside the ridges going to hurt? I think the JB Weld mixed with lapidary is going to give you the best heat conductivity through the shell that you can get. But, maybe instead of filling it between the shell and pill like caulk, you could smear some in the shell first. Push it into the ridges only enough to fill them up and make the inside of the tube straight through. After it sets up, bore or sand it out to fit the pill, since you still want the pill to contact the actual metal shell as much as possible. That way, the JB Weld isn’t actually adhered to the pill, making future disassembly easier. If you wanted to go back to “original”, would the JB Weld filling the ridges really be a problem?