Flashlight

A replacement for a broken flashlight body. Uses the mill and the lathe.

I generally carry around a small flashlight, which has proven itself very valuable when poking around the insides of computers. I recently discovered a Xenon lamp flashlight that is significantly brighter than the Maglight I have had for a while. Unfortunately, the thing broke.

The reason the end broke is that the inner bore of the plastic body is off center, as you can see in the photo above. It has made the handle quite thin on one side.

I was looking at that handle, and thought “I can make one of these!”

So I started out with a piece of 3/4 inch diameter aluminum rod about 6 or 7 inches long.

 
The broken piece was only a quarter of an inch long, so it wasn’t much of a problem. I decided that I could modify the design a bit, but some parts had to be exact. That small |T| indicates roughly where the taper will be.

 

 
I can’t do much with this rod at this point, it simply doesn’t have enough support. You can see how long it is in comparison with the diameter. Time for the live center and the centering drill.

I could finish the end just a bit, but it took some very light cuts to smooth the end.

 
Ready to drill the center hole. The adjustable toolpost allows very nice centering of the lathe tool for finishing, and the camlock tailstock is a real convenience. I’ll drill the next hole 1/4 or so inches deep. It needed to be about 27/64, really. That’s the size of a AAA battery. This uses two in series.

 
Basic machining here: I’m taking a truing cut down to the beginning of the taper. That will serve as a reminder of how far to do the machining. I have a bit of slop, so if I run over by a few thousandths, it won’t matter. In fact, only one length and the screw threads are really critical.

 
The piece after the truing cut is done.

 
Reality check time. The taper is in the right place, the handle is longer than needed, but does take into account the center drill and live support hole. The diameter has been turned down to the distance across the points of the octagonal handle, not the distance across the flats. Thicker was better. It’s about 0.600 inches.

 
The center hole is bored as deeply as it can get on the lathe, which is about half of what it needs to be. I could have continued with finishing the hole, but I decided to knurl the handle first. I left about half an inch at the base, and went up to within 1/4 inch of the taper. Here’s the knurling tool ready to start.

 
I used the power feed set on fine, with the lathe on low speed to knurl. I adjusted the tension on the scissors knurling tool so that the knurling pattern was distinct across the width of the tool. I then engaged the power feed and the knurling tool crept across the handle.

 
Here’s the maximum extent. The larger part is actually the more difficult part to machine, but that’s OK. When the lathe reached the end of its travel, I stopped it, reversed the carriage feed, tightened the scissors a bit, and let it go in reverse. The lathe speed was probably about 20 RPM at the maximum.

 
Here we are, completely knurled.

 
Another reality check. The knurling covers most of the handle, and I have enough aluminum at the end to take care of all the machining I need.

 
I’ve put the body in a 5C collet fixture to mill off the center drill hole. I could have used the lathe, but the larger part does not fit in the center bore. This holds the body more securely, and the large part still fits in the collet body. The collet jaws won’t mar the finish.

 
I’ve removed the hole, and the end is quite flat. I then put it back into the lathe, large end first (the way it was) and very carefully beveled the sharp edge. I also then rounded it with a file, so there’s no danger of getting cut. The rounded edge is also much more durable.

 
Here’s the body, reversed, with the first truing cut taken. I have some room here for slop in truing the stock, but it doesn’t matter as long as I have enough aluminum for the walls and to thread. I may not cut that taper, since there’s another possible design.

 
I’ve turned the threaded part down to the proper diameter for the section before the thread. The large chunk in the middle will be the taper, or perhaps not, as I wish. The handle barely fits in the chuck, so perhaps knurling could have waited, but it did fit.

 
Oh, and before it escapes, the center hole was finished by drilling down the existing hole. At the very bottom, I drilled 1/8 inch, then brought the drill up to clean away the chips, then repeated that. The fixture is mounted on a base plate that automatically aligns the collet fixture with the mill.

 
The next item is to cut the slot for the O-ring. That seals the interior of the flashlight from water and contaminants. I’m using a cutoff tool to cut a slot. I’m also just following the dimensions here as needed.

 
Setup and ready to thread, at least to start. I did some threading with the tool positioned at a 90 degree angle to the work without the compound set at 29.5 degrees. This started to work, but was producing some rough threads.

 

Ready for threading, the autostop is installed above, and turns off the motor when the end of the thread groove has been reached. You won’t thread into a shoulder this way. The thread was 18 TPI, and 0.550 in diameter, if I remember it properly. It was also a bit square on the top of the threads.
Next, I tried the proper method, offsetting the compound by 29.5 degrees and setting the tool with my (non-existent, it’ll be on my shopping list) threading gauge. This produced better results, but the aluminum (6061) did gall a little.

I used a little tapfree fluid, and that helped some. However, there was a problem. As the compound is advanced, the end point of the thread changes slightly, and the autostop sensor needs to be reset. The next time, I think I’ll set it a bit short, then let the progression work as needed.

Most of the junk from the threading was on the top, so I did trim it off to the right diameter, making the tops square. The bottoms are a V groove as expected. If the setup is right, then even threading aluminum is easy enough. It will produce clean threads. Sorry, but I was so busy that I didn’t have time for pictures. But….

 
Here’s the reality check, showing an undamaged plastic body, and the replacement aluminum one. Everything is done except the taper and the flats.

 
Now I could have tried cutting the flats on an angle, although it would have been a real pain to do. Instead (and here you can see the threads. The front is cut down to clear the lens.) I decided to leave them flat. The original flashlight has an octagonal shape, and I duplicated this to match the molding of the lens. I milled the 90 degree flats first, then the 45 degree ones. It’s on the 5C collet spin fixture.

 
The flashlight body and the replacement. You can see the flats milled on the body at 45 degree angles.

 
The original (I had a spare unbroken one) and the replacement. The replacement has new batteries.

The final product, a bit photoedited to eliminate spots on the background. The non-knurled parts were polished with a buffing wheel on a grinder, which eliminated the sharp edges on the no-roll part of the octagon. This worked.

 

In drilling for the batteries, I had to leave a conical base to the hole bored in the body. This would have either a spring or some sort of mechanism to make a good electrical contact to the bottom of the battery. The lens presses the light assembly down into the barrel and makes contact that way.

What I did was to make a small conical section of aluminum that was turned so it barely fit down the barrel. The top was roughened for a contact, but it was flat. This was press-fit into the barrel so it will not come out. This makes a nice flat contact on the bottom end of the battery.

The original body had a bronze spring that made contact on the bottom and ran up the inside of the barrel. Since this is all aluminum, I don’t need it.

 

 

The machining steps were first to decide on a layout, then cut the bar to size. Next, the end was faced and center drilled. With the bar supported, the handle was turned to size and then knurled. The handle was left long to account for the countersunk hole.

Next, the handle was reversed and the hole for the batteries was drilled, then finished on the mill because of depth travel problems in the tailstock. Only goes out so far. However, nothing else had to be absolutely concentric with this, so any loss of precision was OK.

The top was turned down to the outside diameters needed, threaded, and trimmed to length. The O-ring was installed in the gap, and flats were cut on the mill with the spin indexer.

Overall, a nice project with no nervous-making operations. Well within the capabilities of the lathe and mill. Knurling and flat cutting were nice extras that helped the design. This took about a day and a half, with a bit of head scratching about what to do next.