Been a long couple weeks around here but all has turned out well. Time's been short but this morn finds me putting some of the HBC DE-tales to paper.

--decided to use the '329 ogive. I've fired this one to 600 yds varying the arrival speed and always get nice, round holes. Meaning there's no transition bug going from super to subsonic speed-- or above and below the speed of sound at around 1107 fps.

--Tom Meyers last yr figured a BC exceeding .50 for a 1.25" version of the 329 lengthened from 1.17+".

--essentially what I've done recutting the original 329 ogive with my mold is taking the bore rider and making this a front driver... with some strength. We're around .150" for that driver and something on the order of .050 or so of leade taper [.309 down to .300]. This mimics the short 'porch' of the LBT spitzer that I've sent downrange by the thousands with excellent accuracy.

--understand this recut 329 length at 1.17" is THE BULLET in my friend's 30/30 break open Topper. At 309 dia going down that 3065 bore groups run an easy inch at 100 in still air- this only visually inspecting the slugs. We've gotten some screamer clusters of 3-4 buried in one hole. The load is around 13.7 grs of 4227 and Rem 9 1/2M's-- seated to engage at closing and just AC'd.

I'll be boxing up some 329 samples today to send to our Draw-er Trk to get duplicate figures on this ogive and front driver arrangement.

Thanks to all,
Jay

Aladin:

I am an engineer that does his own CAD work (small company :-), so if you ever need a back-up CAD guy (AutoCAD 2004), drop me a line.

Resp'y,
Bob S.
rls1998@aol.com

Bob S.

No heartburn on my part - there are peanty of work/projects around to play with. We sharpen eachother's skills in so doing. (Sound like boolet casters?)

I'd love to share drawing files with you.
It's been a while since I've updated, still working with AC 2001. Not a lot of difference for what we're doing here though.

I'd like to see CASTPICS used for a repository for some of the better mould designs too.

Tim and Aladin:

I'm not trying to elbow my way in ... you notice I said "back-up".

Tim where are you in VA? I am still using A/C 2000 at home, so I have to remember to save stuff that I take home as a 2000 file.

Resp'y,
Bob S.

Bob, if you are serious, truly so, you might do some real research in coming up with a system to auto read some template/boiler boolit design(s), allow a user to modify the critical dimensions over the net (web page(s)), and generate direct machine code from a final drawing. I had conversations with a guy in Florida about a year ago, and have not had any response since. Reply on this topic so all can see. ... felix

Bob - I'm in Pulaksi (SW of Roanoke on 81). Taught AutoCAD for 13 years at the local CC.
(I took NO offense at your comment and did not think of it as horning-in. On the contrary I welcome the expansion of ideas and processes - as I'm sure Aladin does. Thanks for the extra politeness - it fits well in the character of cast booleters.)

Felix - GOOD THOUGHTS. We need to persue this.

If one has a website (and perhaps there are other options of doing this) AutoCAD has a function wherein one may post a 'locked' version of the drawing (so others don't take a full working copy of the drawing - also able to restrict who can open it). There is the capability then of the 'customer' to open the drawing and 'red-line' it and add comments.

This is commonly used in architectural firms.

Let's keep this exploration going!

Bob, Aladin, Felix, et al. -

Let me also add a few thoughts about CAD and design:

I taught my students: "Most people back up their work only AFTER the first catestrophic event."

The program "C" works so well because it's a one-person designed program (ok 2 people designed it).

I still have my first CAD drawing - and the stack of punch cards for it.

I would like to know if the folks that make moulds have CAD program options of choice that they would like to share with us. There are some out there that do CNC mould cutting that could be naturally interfaced with the right package.

I wonder if the board is willing to front such a cost for a recognized firm to do this custom work. I would do it myself, but I am no longer able physically and mentally to do it the way I want it done. My days of writing computer programs are over indeed. I even need my wife to do the checkbook these days, and that's no joke. ... felix

Doug, the CNC mold programmer at LEE, punches the physical G-code moves into his PLC to program it. He does not "import" off of CAD drawings as it is simply quicker to punch the simple program into the PLC directly.

This isn't the real time eater, anyway.

He "swings low" on the critical dimensions on purpose, cuts a trial mold, casts from it using the specified metal, then tunes the dimensions up in size (using the same mold block several times) until he hits the all the various diameters and sizes.

Then he cuts a fresh mold and casts from it to see where his boring bar deflection (this happens on fully loaded cuts) has thrown him a curve by chattering or deflecting & cutting too small.

He programs in additional rough and finishing passes and tunes the final pass G-code size to beat all his deflection and chatter issues (and to get the desired fine final surface finish).

By now he has used up 2-3 molds that aren't quite up to finished standards (I think they get sold to late folks who call LEE and ask for a special run mold "after the fact").

Having got it right, he starts the offical run.
This all holds good until:
a) a cutter dulls and cuts undersized due to deflection
b) a cutter breaks
c) he hits a bit of slag in the extruded aluminum which results in a) or b)

Then he starts the tuning process all over with a new cutter. It is all "cut and try" tuning, based strictly on the actual diameters the mold actually casts in the specified metal.

This is why these custom runs have a $100 programming fee, this whole process takes about half a day to do and there are custom nose form cutters and extra scrap to cover for as well.

I don't think you are going to get your dream out of LEE. You might try Mountain Mold over on CAST BOOLITS, he will answer you directly on this thought I would expect.

He does do "scalable design tuning" on his web site, so he is as close as anybody to what you are dreaming about.

I suspect he still has to hand-tune the first mold of a run just like LEE does to get his program completely "right".

I also expect he "swings low" on the first cuts and finishes by tuning upwards off of cast bullets, all on the same mold block so he wastes nothing - especially since most of his real custom stuff is one-off work.

Oldfeller

[ 11-23-2003, 23:07: Message edited by: Oldfeller ]

Felix -
The place to start is simply to continue to post good designs on Castpics - no additional fees needed. As interst grows follow the wave.

OF -
THanks for the play-by-play. It's good to know the sequience of events that occur after the design process.

I would love to have enough of a CNC setup to build and tune bullet designs. Not interested in production past the one or two for myself and a friend or so.

Felix:

CNC coding is not my forte. I did it once 30 years ago in the shipyard, when it was just X-Y coordinates on punch tape for the plasma cutter. That's the limit of my experience there. I did design a special sight base for a Ruger No. 1 black-powder long range rifle, looks like a regular Redfield International base, but has three dovetails stacked vertically so as you go back in range, you move the sight up a dovetail. I handed the guy a disk with an AutoCAD file on it (probably Release 10, it was a long time ago), and the three-axis CNC mill spit out the part. I have no clue how that happened, it was the "magic of science". The drawing part I can handle, though. I just do mundane things at work like blow things up, burn them up, and then design to try to keep that from happening on our aircraft carriers and 'gator freighters. :-)

Resp'y,
Bob S.

Oldfeller,
I enjoyed hearing your description of how Lee sets up to cut a mold. It sounds just as I had imagined it. Some people might think that with CNC you can just dial in whatever diameter you want and you will get it with zero tolerances --- I WISH!!!

I don't use CAD except for site illustrations (I use TurboCad and it can read/write most common graphics filetypes). The online program designs the bullet -- just numbers, but that's the language that computers speak -- and then a BASIC program takes those numbers and converts them into G-code.

For every mold that I have ever made, I have a record on the CNC diameter vs. as-cast diameter vs. alloy vs. mold material, and this is entered into a spreadsheet, giving me a huge database. The database gives a pretty good AVERAGE of what the CNC diameter should be to create the desired as-cast diameter.

The key word is AVERAGE . There is a substantial amount of random variation.

The big hitter is boring bar deflection. Some of the bars are toothpick-sized, and you better believe they bend. Of course, there are many different bars for different purposes, and each has its own personality. The smaller the caliber, the skinnier the bar, and the more it will deflect, which is why I'm taking my sweet time working my way down to the smaller calibers.

Since I don't have Lee's luxury of using the first 2 or 3 molds in a run to calibrate the program, I cut a pilot hole in each cavity, measure the pilot hole, and use that as my calibration. The pilot holes are seldom perfectly straight, due to boring bar flex, so you will get different measurements depending on how deep in the hole you measure. And measuring small holes is a bit of an art, anyway. I figure the calibration measurements are only good to 0.0005".

This has to be repeated for every boring bar that will be used for that bullet. Want a double crimp groove? That requires grinding a special bar, which will require it's own calibration pass. Want a pointy or round nose? Another special bar, and another calibration pass.

A "real" CNC machine is left turned on 24/7, in a climate controlled environment, so once it is calibrated, it is supposed to stay calibrated for a long time. My homebrew CNC tools run on PCs, and thanks to Bill Gates, if you leave them running, they'll inevitably crash after a few hours, usually while in the middle of a cavity. , so I reboot after every mold. The CNC software is supposed to remember the calibration when reboots, but in reality it loses one or two thou, so I have to re-calibrate on every mold.

After cutting the mold, the only sure way to know the diameter is to cast bullets. And of course, the as-cast diameter can depend on temperature and casting technique, so you're never quite sure how much of the variance is due to the cavity and how much is due to the so-called shrinkage (which actually has more to do with completeness of fill than with shrinkage).

If the diameter is too small, an attempt is made to recut it. Each cavity has to be dialed into position on the lathe, and this dial-in is never perfect. Because the cut will be light, the bar may chatter and rub instead of "biting" into the metal. Or then again, it might decide to bite, after all. To be safe, a very light cut is made. Then retest. Then recut. Then retest. Maybe it eventually casts within spec. If not, the mold goes into the scrap pile and you start over.

Which is a long-winded way of answering the original question about interfacing with CAD software. My computer program just goes by numbers. Drawings are only for the benefit of humans.

There is such a thing as CAM software that converts a CAD drawing into G-code, but it only works for "cookbook" machining. Believe me, cutting mold cavities is not cookbook machining.

My use of AutoCAD was laying out drawings of buildings and grounds at various penal institutions, I never got into small things. Back when I worked for RCA in the early they were working on some computer controlled production machinery for the Army tank command in Detroit, I was working on the interface for a photo type setter that was built in Germany, and the people from that group used to come over to steal ideas.

Oldfeller, Dan:

Thanks MUCH for the tutorials on how CNC mould-making works in real life. I now understand why .22's aren't too popular with the custom makers!

Dan: I suppose you recycle your "duds" into larger mould cuts - can't imagine how you would hold to your very reasonable prices if you didn't. But have you ever considered offering some of those "off-size" moulds on an "as-is" basis on a "bargain page"? I can see these CB nuts ("Have Mould, Want Rifle to Fit") going around the gunshops slugging barrels, hoping to find that perfect 343" x 337" barrel.

Just funnin', of course.

floodgate