AER-Lights
Newly Enlightened
- Joined
- Feb 23, 2019
- Messages
- 91
Hi folks.
I have been lurking for quite a while but i figure it's time to chime in.
Being a soft flashaholic (I know some of you guys are pretty hardcore) but you know…
Starting out as a five year old kid screaming his heart out because he wasn't allowed every flashlight he laid his eyes on and the fact that the feeling hasn't quite gone away the better part of forty years later
My current collection isn't impressive but it does hold of a couple of SF Cs and Es etc. But… being a home machinist i guess it was inevitable.
Spending part of my vacation i finally got around to building my own flashlight from scratch and here's my story and I'll try to keep it short so you won't get pictures and explanations of every step.
Design :
Design is tricky by nature. I mean it's got to look right, be functional , and leave enough options to be interesting. The majority happened over time in my head and i initially figured I'd do a take on a P60 type host as it would leave me a lot of open options and i could just buy a proper drop drop-in but as with so many things in live the plans changed as the design progressed.
I ended up with an in-line, clean, single RC123, mule with the major difference being that it's built in steel.
So… Short and Fat
The material choice was tricky but i ended up deciding on steel as it will provide ample surface treatment options and will provide a hard substrate for whatever treatment i end up deciding on. That and the fact that I like steel. Having said that it would have made my life considerably easier if I had chosen aluminium or brass.
I had also planned to use a mcclicky switch but as i live in Europe and i forgot about the switch in the planning phase it's currently built using a chinesium switch. (It takes forever to get stuff from the states and it will cost and arm and a leg in shipping and import fees)
And i can always change it later once they arrive.
Here's the result.
Once you start breaking it down quite a few components go into one of these things.
Build:
As with all machining jobs all start out with raw stock.
Here's my "shipment". (They missed a couple of things but there's enough here to get started.)
Cut to length and faced.
I started out with the body:
First problem to overcome is boring a 17mm hole through a piece of steel on a hobby grade lathe.
It did take a bit of fiddling with feeds and speeds but eventually I got it.
Next up was turning the outside and the threading.
Keeping true to most P60 style hosts as well as Dons Mcclicky i decided to go with 20TPI but in metric sizes. I know… but i am European J
so… M20-1.25 it was. No this is not a standard thread but that's what i needed.
Next up was the head:
The head turned out to be easier that i expected and pretty much just happened.
My biggest concern was if the threads would match up so a test was needed immediately.
fortunately they fit exactly the way i intended. It's not the tightest thread in the world but it matches up with what you'd find on an SF, Solarforce, Ultrafire so well within expectations.
As my dog ate the pictures of cutting the flats on the head you'll have to do without them but that was done on a mill using a dividing head.
Next up was the tailcap. The initial design in cad was made on the assumption that I'd have the mcclicky switch available which i didn't so it was time for a new iteration using the chinesium switch.
this in turn meant that I had to do the switch holder (a piece of brass that screws into the tailcap and holds the rubber boot in place as well as providing a proper seal between the boot and the bottom end of the tailcap)
In the end it took a couple of tries to get the "headspace" right. Eg. the switch holder and the bottom end of the tailcap in order to get a proper seal on the boot.
Almost there
All that needs to be done now is the Crystal and the pill/"reflector". This being a mule it doesn't really have a reflector but more of a cover to make it look a bit better than just the bare PCB.
Yet again I had to go through a couple of iterations to get the spacing right but as with so many things it can be fixed by putting in the work.
The Crystal/glass is basically a piece of 2mm polycarbonate cut and sanded to size so not much to it.
At this point the pill is built from a old CREE XR-E and the wrong driver as I'm yet again waiting for parts.
All of the above brings us to the result.
One fully functional short and heavy steel flashlight as designed.
A size comparison for good measure. (note that the picture skews the actual size. The E2E has a body diameter of 21mm whereas the short and heavy is 22mm)
Considerations:
1. This being steel "in the white" it will rust over time so I'm planning on Nickel electroplating it. This also ties back to the loose fit on the threads. There needs to be enough slop to allow for the plating.
2. I'll update the pill with a latest generation LED and driver when they show up.
3. I'll need to update the tailcap with a mcclicky.
Reading through this post it might seem that this was a walk in the park. Well it wasn't J
I added an additional picture of bits and pieces that didn't work, was out of specification or otherwise rejected. And these are just the parts that didn't get cut down on the lathe immediately.
Well… If you've made it this far i respectfully thank you for your time.
BTW. I would welcome any comments, questions and or inspirational ideas.
May you live long and prosper V.
Stig.
I have been lurking for quite a while but i figure it's time to chime in.
Being a soft flashaholic (I know some of you guys are pretty hardcore) but you know…
Starting out as a five year old kid screaming his heart out because he wasn't allowed every flashlight he laid his eyes on and the fact that the feeling hasn't quite gone away the better part of forty years later
My current collection isn't impressive but it does hold of a couple of SF Cs and Es etc. But… being a home machinist i guess it was inevitable.
Spending part of my vacation i finally got around to building my own flashlight from scratch and here's my story and I'll try to keep it short so you won't get pictures and explanations of every step.
Design :
Design is tricky by nature. I mean it's got to look right, be functional , and leave enough options to be interesting. The majority happened over time in my head and i initially figured I'd do a take on a P60 type host as it would leave me a lot of open options and i could just buy a proper drop drop-in but as with so many things in live the plans changed as the design progressed.
I ended up with an in-line, clean, single RC123, mule with the major difference being that it's built in steel.
So… Short and Fat
The material choice was tricky but i ended up deciding on steel as it will provide ample surface treatment options and will provide a hard substrate for whatever treatment i end up deciding on. That and the fact that I like steel. Having said that it would have made my life considerably easier if I had chosen aluminium or brass.
I had also planned to use a mcclicky switch but as i live in Europe and i forgot about the switch in the planning phase it's currently built using a chinesium switch. (It takes forever to get stuff from the states and it will cost and arm and a leg in shipping and import fees)
And i can always change it later once they arrive.
Here's the result.
Once you start breaking it down quite a few components go into one of these things.
Build:
As with all machining jobs all start out with raw stock.
Here's my "shipment". (They missed a couple of things but there's enough here to get started.)
Cut to length and faced.
I started out with the body:
First problem to overcome is boring a 17mm hole through a piece of steel on a hobby grade lathe.
It did take a bit of fiddling with feeds and speeds but eventually I got it.
Next up was turning the outside and the threading.
Keeping true to most P60 style hosts as well as Dons Mcclicky i decided to go with 20TPI but in metric sizes. I know… but i am European J
so… M20-1.25 it was. No this is not a standard thread but that's what i needed.
Next up was the head:
The head turned out to be easier that i expected and pretty much just happened.
My biggest concern was if the threads would match up so a test was needed immediately.
fortunately they fit exactly the way i intended. It's not the tightest thread in the world but it matches up with what you'd find on an SF, Solarforce, Ultrafire so well within expectations.
As my dog ate the pictures of cutting the flats on the head you'll have to do without them but that was done on a mill using a dividing head.
Next up was the tailcap. The initial design in cad was made on the assumption that I'd have the mcclicky switch available which i didn't so it was time for a new iteration using the chinesium switch.
this in turn meant that I had to do the switch holder (a piece of brass that screws into the tailcap and holds the rubber boot in place as well as providing a proper seal between the boot and the bottom end of the tailcap)
In the end it took a couple of tries to get the "headspace" right. Eg. the switch holder and the bottom end of the tailcap in order to get a proper seal on the boot.
Almost there
All that needs to be done now is the Crystal and the pill/"reflector". This being a mule it doesn't really have a reflector but more of a cover to make it look a bit better than just the bare PCB.
Yet again I had to go through a couple of iterations to get the spacing right but as with so many things it can be fixed by putting in the work.
The Crystal/glass is basically a piece of 2mm polycarbonate cut and sanded to size so not much to it.
At this point the pill is built from a old CREE XR-E and the wrong driver as I'm yet again waiting for parts.
All of the above brings us to the result.
One fully functional short and heavy steel flashlight as designed.
A size comparison for good measure. (note that the picture skews the actual size. The E2E has a body diameter of 21mm whereas the short and heavy is 22mm)
Considerations:
1. This being steel "in the white" it will rust over time so I'm planning on Nickel electroplating it. This also ties back to the loose fit on the threads. There needs to be enough slop to allow for the plating.
2. I'll update the pill with a latest generation LED and driver when they show up.
3. I'll need to update the tailcap with a mcclicky.
Reading through this post it might seem that this was a walk in the park. Well it wasn't J
I added an additional picture of bits and pieces that didn't work, was out of specification or otherwise rejected. And these are just the parts that didn't get cut down on the lathe immediately.
Well… If you've made it this far i respectfully thank you for your time.
BTW. I would welcome any comments, questions and or inspirational ideas.
May you live long and prosper V.
Stig.