17 September 2004, 17:52
Bwana-beRe: Start-up costs?
Thanks guys. It's .358, FYI. I'll check out Richard's site. Sounds like such fun, and right up my alley, but I'd need to approximate an A-Frame or the like in order to make it worth my creative while.
18 September 2004, 00:13
Bwana-beAlso, Martin, how do you control the rolling of the inside of the jacket, so that it comes out uniform? If I get you, you're just applying force on the jacket from the top, and making it lump inward. How do you do this and still leave the top half of the jacket ready for point forming?
I'm very interested. I read that Swift's original A-Frame had a small hole in the partition as well, I guess for the same reason.
18 September 2004, 05:15
MartinPottsAlso, Martin, how do you control the rolling of the inside of the jacket,
Bwana-be
The amount of copper that is rolled down is determend by the size of the top punch..
( The jacket is held by a Core seat die .And a top punch od used as i said to roll the inside of the jacket down over the botton core .how it's controled is detemend by the diamator of the top punch. if the jacket has an inside diam of .300 and you want to partion half a jacket inside lenght
you would use a top punch of .292 or somthing close the more you roll down from the inside the thiner your remaining
wall will be..
I tried to show some of the guys this through pictures
but as you can tell i am stuck trying to use a still video capture camera.
(The gray spot is the hole and the lead showing..)
Everthing is done using a swage press.
once the bottom half of the bullet's been partioned
as you can ( kinda make out ) there is a small bleed
hole . I then flux it top half of the jacket and place
the top core into the jacket and melt it down this melts the two half's together and bond's the jacket to the lead
Remmber this is MY way of doing a partioned jacket
And the word for rolling the inside's down is best decribed as Drawing the inside of the jacket down remmber the copper
is very.! soft and since it goes through reheating over and over again there is no ( lumping of the copper inside walls)
so that it comes out uniform? If I get you, you're just applying force on the jacket from the top, and making it lump inward. How do you do this and still leave the top half of the jacket ready for point forming?
----------
Once the two core's have been melted together it looks like any other normaljacketed core ready for point forming
it's just lead riding at the top of the jacket
I'm very interested. I read that Swift's original A-Frame had a small hole in the partition as well, I guess for the same reason.
--------------
Yep and really it's a better way if ya think about it
if you ever hit anything with the bullet and it mushrooms back to far ( with todays partioned bullet )you can lose
the top part of the core . the old way the two half's are melted together in the jacket through the small hole.
I left out a few things.and sorry once again for the really bad pictures . i have caught hell for them before.
O and if i did not already mention it the bottom half
of the jacket can be threaded to make the bond harder to break .
Pottsy