1st MTG2 Build

Argh! Well. Sheesh. Even the penny was wrong. Only direction is up from here. I guess I’m going to have to fall back to a variation of the Dorpmuller and Itinifni P-60 builds. The Itinifni use of a copper pipe/sleeve helped him get a good fit but I don’t intend on using the AA or brass pill bottom. Both appear pretty straight forward. I also would like to do the FET on a 105c trick as that just sounds explosively cool, but I’m not sure where exactly to put it. I need to see a picture as some of the descriptions on other threads probably make sense to people that can identify the parts of an IC. (Not me)

95% copper is technically 'brass', and I imagine under some unrealistic extreme tests you could find a measurable difference between 'brass' with 95% copper and C101/C110, but I doubt it would translate into anything meaningful in real-world applications.

FET is mounted upside down here, with the legs straightened out and then re-bent the opposite way to touch the PCB. In this pic, the outer leg is ground, inner leg is the gate (the PWM input from the MCU), and the black LED- wire attaches to the big metal pad on the upper surface of the FET (wire not shown in the pic).

I thought pennies had to be 1964 or older to be mostly copper? I will go look that up. I am no coin expert! lol

You might be surprised… even tiny changes in a material can radically affect it’s thermal conductivity. Thermal conductivity does not change linearly with to the ratio of the materials in the mix. For instance, pnly a few parts per thousand of phosphorus in a copper alloy will utterly kill its thermal conductivity.

On the other hand there are many different varieties of brass, ranging from 65% copper to 95% (pre-1982 pennies are 95%). I doubt all the varieties of brass have the same thermal properties.

EDIT: Found it:

Gilding Brass (95% copper). Looks like it has 60% of the thermal conductivity of pure copper.

thermal conductivity chart

Comfychair, that’s perfect! Thanks. The diagram is a huge help.

Are you set up to flash driver firmware? I don't know if any stock drivers will run the FET correctly, it has to use firmware written for the 9.4kHz 'phase-correct' PWM.

Hey, that's MY line!

So, like-for like test. 16mm copper MCPCB on a filed-down pre-'83 brass penny, 16mm copper MCPCB on an identical size/thickness C101 (or whatever) copper disc. Same power to both. How many lumens difference between these two materials with vastly different thermal properties?

I'm going to be bold, and predict the difference will be within the margin of error for the measuring equipment. :~

Comfychair, that’s perfect! Thanks. The diagram is a huge help. I think I can use the driver that I broke to make this work. Which would be kinda funny. Thanks.

I thought I posted this before, I used comfy’s diagram and it worked ’til I put it in the pill and pulled off the positive via completely. So I desoldered it and in the process nicked off a little chip that RMM carefully put on the spring side of the board. Never found the piece/chip.

You might want to elaborate a little… most of us probably have a rusty memory of this thread by now.

I’m assuming that you’d like to know what the missing part was so that you can get another part and put that in it’s place.

If Im not confused, hes taking some crack for back pain too so he might be less than coherent anyway lol. But I could have him confused with someone else.

The Composition of the Cent

Following is a brief chronology of the metal composition of the cent coin (penny):
• The composition was pure copper from 1793 to 1837.
• From 1837 to 1857, the cent was made of bronze (95 percent copper, and five percent tin and zinc).
• From 1857, the cent was 88 percent copper and 12 percent nickel, giving the coin a whitish appearance.
• The cent was again bronze (95 percent copper, and five percent tin and zinc) from 1864 to 1962.
(Note: In 1943, the coin’s composition was changed to zinc-coated steel. This change was only for the year 1943 and was due to the critical use of copper for the war effort. However, a limited number of copper pennies were minted that year. You can read more about the rare, collectible 1943 copper penny in “What’s So Special about the 1943 Copper Penny.”)
• In 1962, the cent’s tin content, which was quite small, was removed. That made the metal composition of the cent 95 percent copper and 5 percent zinc.
• The alloy remained 95 percent copper and 5 percent zinc until 1982, when the composition was changed to 97.5 percent zinc and 2.5 percent copper (copper-plated zinc). Cents of both compositions appeared in that year.

Thanks for the link Scotty, old or not, a fun read. Only can add that OL mentioned using hardware copper pipe and plugs when doing custom work. No idea about purity but would not be surprised if Lowes mixed the copper for max profit.

Still deciding on which host to use to avoid welding glove option :wink:

Yup, narcotic addled mind is in full effect. Back still hurts but I can attain otherwise comfortable positions. The driver that I tortured is in an empty cocoa tin waiting to donate parts to other future builds. I was just barely able to resist pulverizing it with a hammer. My reasoning at the time was that everything I did with that driver was so pathetic that should my skill or luck continue and I took a hammer to it I would likely zing a fragment into my eye and lose the eye. So, nope not looking to fix it. I just replaced it with a FET Zener driver from RMM’s site. It’s a ferocious and ludicrous hand grenade of light. After 60 seconds it should step down from turbo and it very well might but it gets way the hell too hot to hold long before that. The element of danger makes it more cool, sorta.

Illuminaria, if I were to do it again I would get copper discs from Etsy.com.

This would be the best host without a doubt.

Awesome :slight_smile:

This has me curious if there are any 2x18650 tube lights and or how difficult it would be to make an mtg2 heatsink well in a solarforce gladiator

I don’t know of any straight-up-tube lights (head is same diameter as body) that take 2x18650 or 2x18500. There are plenty of opportunities for custom work of course.

In my mind one of the most straightforward custom builds would be to make a head which attached to Solarforce extensions, similar to what dsche has done with his OrthoDogs project. Two 18650 extensions attached to a tailcap leaves space for 2x18700 with about 1 or 2 threads. That's no good, so either use shorter cells or switch to 4x123A extensions. 63.9mm vs 33.6mm of non-nesting tube, so you'd net an extra 6.6mm of battery space. Since there is 10mm or more of dead, unthreaded space on the front of an extension tube you would have that to play with as well as the 5+mm of threaded area that your pill screws into (depending on springs etc). With 15mm you could fit a lot of the light engine inside the head. 5mm for the driver area, between 12 and 20mm for the reflector, then another 1.5mm for the lens, then something to retain that stuff.

At that point you're talking about a pretty long tube light, but it's actually still shorter than a Streamlight Stinger.