Machine Shop skills

I have just (well, tomorrow afternoon) aquired the use of a fairly fully featured machine shop(lathes, mills, bandsaws, etc.). (no CnC though /ubbthreads/images/graemlins/frown.gif ) Obviously, I'm just itching to try stuff out, so I'm looking for recommendations of things to try out. I'm interested in making my own lights, particularly something involving a 5w Luxeon and 1 or 2 123A's. The question is, what is the standard optical configuration for this? should I use an NX05 and no reflective aluminium? should I leave the luxeon raw? Can anybody give me some pointers on this sort of project? I plan to model my item in solidworks before I actually go making it... Another question: any tricks for making parabolic (or mostly parabolic) reflectors on lathes that aren't CnC (but do have cutting tool speed controls and digital position readouts)? I realize that this probably isn't an issue for Luxeon mods, but if I wanted to do something parabolic... Yet another thing: Any advice on anodization? It would be much cooler to have a type 2 or 3 anodized flashlight, but how do I go about doing that myself? Is there anyone here who could help me out? General help for a newbie machinist would be nice...

Thanks,
Paul

Can't help you with the machining thing, but the most common optics in small mods seem to be NX05 and fraen LP.
Common small reflectors are McFlood and the pelican reflector. www.theledguy.com has a modified pr, the NX05 and the McFlood. I'm not sure the fraen is commonly available yet.

www.carleylamps.com has an assortment of parabolical reflectors with and without texturing.

Ahh, thanks then. I had figured on the NX05 optics because people say they seem to be a good combination between flood and throw. I guess one question is, with the NX05 (or fraen LP), do I actually need any reflector at all? The arcLS doesn't seem to have much of anything, neither (as far as I can tell) do the CMG Reactors. Is there much of any sidespill, or do those lights just not worry about the lost light? Thanks for the advice...

I appreciate the pointers, but I'm still also looking for parabolic reflector machining techniques...

If you really want to do a parabolic reflector and don't have a CNC lathe, the next best thing is to cut it out using small steps. Figure out before hand the arc you wish to describe. Divide it in to steps. eg. Start cut 1mm from circumference, cut to 2mm depth, next cut 2mm from circumference, depth 4mm and so forth, and work your way in. Note that the cut depth will probably not increase linearly for each mm farther from the circumference. You need to work that out. And keep the resolution high, ie. if your machine can handle 0.01mm increments, go for it. It will make the next part easier.

Hope you could understand that part. Let me know if you don't and I'll see if I can scan in a picture describing that procedure. Think of terracing as it applies to farming on the side of a hill.

The next part is easy, especially if you were able to make very small steps. Grab an emery cloth, rotate the work in the lathe and free hand polish until you smoothen all the steps. It actually isn't that easy because you need to make sure not to hold the cloth too long at one spot. You might end up changing the profile. And of course you need to make sure the cloth isn't pressing hard against certain parts as it might do because of its stiffness.

To get a shinier surface, just change the cloth to progressively finer grits and polish away. Good luck. Hope you understood all that.

Ahh... wow, that does sound like a good way to do it. I had thought about doing the small increments, but hadn't made the emery cloth connection. I'll have to have a go at it and see how it comes out... I think my machine can handle .01mm increments, so if I have patience, it should come out nicely... Thanks for the advice...

Still looking for information on anodization... Would it maybe be better to make the body out of stainless steel or brass instead of aluminum?

Steel at least is a bad thermal conductor. Alu is a good one. Its important since the luxeon efficiency depends on its temperature. Also, if it gets too hot it can die.

TheFire,

You will soon be having way too much fun! /ubbthreads/images/graemlins/grin.gif
(Google "anodize"- plenty of good articles.)

Larry

IIRC, brass has slightly better thermal conductivity than aluminium, but has smaller thermal capacity. So it will move heat away quickly, but not be able to store as much.

Good luck with the lathe work. It takes forever and you have to concentrate all the way. A slight slip and you might ruin all the work. And then there is the polishing phase to contend with. Same thing. Takes forever and you have to concentrate. Better do it in stages. But I have done a convex parabolic arc before and it is possible to do it that way.

Ok, Steel's out then. But aluminum is sooooo soft... Heatsinking will be an issue... For brass, I suppose I could just make more decorative heatsinking bits... I think the trick is to start with the hardest parts first (if possible) so that I'm less likely to screw those up after doing all the easy stuff. And besides, if I do screw up (and I'm using aluminum) then it's more items to test out my anodization on, right?

Let me know if you find a good anodizer that is reasonable. I just made this and want to get some kind of coating on it before it gets too tarnished. The other thing I was looking at was maybe en plating from Caswell so I could do it myself.

[ QUOTE ]
But aluminum is sooooo soft...

[/ QUOTE ]
That actually depends alot on the type or "grade" of aluminium; some are "quite" (all depends compared to what /ubbthreads/images/graemlins/smile.gif ) hard.

Olivier

Yeah... good point. I shouldn't generalize about a material... If I were to try and anodize it, I would need to experiment with a buch of different grades to figure out which ones took it well.

I found this site which seems to have a fair bit of practical information about anodizing. There are also this and this discussion and this discussion of making flat black and this personal account of anodization and this rather informative board. After some research (and especially at the last link and it's associated website) I have come to the conclusion that hardcote (type III) anodizing isn't anything too complicated. It seems that as long as you keep the temperature down (about 0 degrees C [is there any reason Liquid Nitrogen couldn't be used for this?]) and the voltage and current up (in order to combat the increased electric resistance of the forming coat) you should be able to produce type III coatings in a similar manner to type II. Now I just need to give this all a go... Does anyone else have any anodizing experience? I'm interested in anyone else's lessons and warnings... they do say look before you leap... /ubbthreads/images/graemlins/smile.gif

unnerv: that's a really cool little light you made. I'd like to be able to give something like that a shot, but I need to work on my threading skills first. It's a lot harder than it looks when the master machineist shows you how to do it... /ubbthreads/images/graemlins/icon23.gif

I've found several threads which mention Caswell's dyes, so they seem fairly popular. Also, that paintball thread suggested just regular cloth dye from the drug store, and although I don't know how well that would work, it would certainly be fairly cheap to at least try...

Do you have any recommendations, lessons learned, or just cool stories about the making of that light that I might benefit from?

I'm also a little worried about anodizing since it's supposed to change the size of the part a fair bit, as well as make it less conductive. Does anyone know how to build a mask for anodizing everything except the threads and the tail contact? I suppose it would be possible to anodize it and THEN cut the threads and re-cut a contact in the tail... the problem is that threads are SCARY to cut and a single slip will ruin a LOT of work...

TheFire,

Caswells lists "masking laquer" which is removed after processing with a solvent. This is the technique used by my previous anodizer. Corks and rubber stoppers are effective for masking the inside of tubular parts. I have built "masking racks" to save the expense of masking, which is labor inten$ive. A custom rack will also eliminate "rack marks."

BTW, don't leave any "steel" (including stainless) exposed!

Larry

Fire, the threading was not too bad, it was mostly doing the setup to make sure that everything was at proper angle etc. One trick my father showed me was to use a magnetic clamp with a dial indicator attached to the lathe in the path of the tool holder. Set the dial indicator so that when your tool reaches 50 thousandths to the bottom, the dial indicator just touches the tool holder. Then move your tool to the deepest that you want to cut the threads and set the dial to 0. When the dial hits zero you disengage the feed. I found myself over or under by about 5 thousandths which turned out fine. The pain is that it took about 6 or 7 cuts to reach the full depth on the threads for 3/4 - 32tpi, but the dial makes it quite a bit easier to make sure you stop on time. I am still a newb pretty much when it comes to the lathe, but after 20 hours of hands on training it sure becomes a lot of fun.

tvodrd: Wow, that's a great idea! I am assuming then that rubber items don't get eaten too much by the bath. Has anyone tried rubber cement (it would be easy to paint on, if it worked)? Forgive the (probably) obvious question, but what is the rack made of? If it's the same material as the parts, do you have to make another one each time you anodize something? Or can you replace just the "stem" and screw those holders into it? That does certainly look like a perfect way to anodize this sort of items, though. I shall have to look into it, even if I'm only doing a few lights...

unnerv: I spent a couple hours working on doing exterior threads today (as well as figuring out how to set the power feed for a smooth finish). I guess I have a bit of a leg up on you since the lathe (at least one of the ones available to me) has digital position readouts. Still, your suggestion for setting it is one I shall have to take into account.

For threads, how is it recommended to start and stop them? I know that it's a royal pain to try to actually pull out the tool and stop the power feed at the same time in any consistent manner, so does it work ok if the top thread on the inside and the bottom thread on the outside go all the way around? What about cutting a notch and using it to seat an o-ring? How far do your threads actually go when you're using the light? How much extra threading did you leave? How many more questions can I ask without irritating people?

What is the recommended thread pitch? Does anyone have any info on what other flashlight companies are putting on their lights? It would be fun to make an Arc thread compatible module.

What about knurling? Is there any way to do that on a lathe? If I can't come up with anything else, I may just cut a coarse set of threads one direction, then another set going the other direction, which I think might give something like a similar effect. Another option would be to just thread the exterior body. That would provide a fair bit of grip.

I think this weekend I'm going to try to set up an anodizing rig and give it a go. Maybe I'll see if I can get some liquid nitrogen to keep the temperature down and see if I can get something like a type III...

Fire, it definately sounds like you have the advantage. The machine I am working on is older than me (over 30 years old...) For my outside threads, I just pulled the feed when I hit the end. I oring'd my light, so I just cut a slot at the end of the threads. Even on the on set that does not have an oring, it came out ok to pull the feed then back out the tool. My outside threads are about a fat quarter inch deep. I went almost half inch on the inside theads so I could have a little more adjustment room. To turn on my light I need to rotate the head about 1/2 a turn with 32 tpi threads.

As far as pitch goes, I chose 32 because a finer thread does not cut as deep, so I could make the walls of the battery tube thinner. Most production lights I have seem to use coarser threads, like 20 tpi or so. If you do a search you can find most of the popular mfg thread size.

For knurling, there is actually a knurling tool that has two or three wheels with the the knurl pattern on them. That is the easiest way to go. The tool I had was dull so, I opted to use heatsink slots for decoration.

Good luck with your project

Patrick

Just a tip about anodizing...

I was interested in having some heatsinks anodized. Seems that paintballers like to have their guns dressed up. So I contacted a local paintball shop. They said they'd anodize 8 small heatsinks for $20. I wasn't particular about the color, so I just told them to throw them in with their other bits.

Chief: hmmm... that's a good idea... I'll have to look around, if I'm not satisfied with my own attempts. I'm afraid I've been bitten by the DIY bug, but that's certainly a good alternative option.