Much educational CNCing today

I did something that took quite a bit of effort (since I had to figure out a few things).

My FIL has a knife without a handle. Asked me to make a handle. The way I chose to do it, was to make a handle out of two symmetrical halves, so that a profile of the tang is removed from both halves. They would "close around the tang" and mate together.

The tang is a little curvy, and I wanted to replicate its shape, both for the heck of it and for snug fit.

I measured the profile every 1/4 inch, using a electronic edge finder as a probe, kind of. (I do have a probe, but have not had a chance to hook it up, more on this later). Then coded it in G code. For the other half, I had to invert the coordinates, so as to get a mirror image.

The bottom line is that after a few mistakes as usual, the halves aligned perfectly and I could press fit the tang between them. Then I used some EDM tubing to make two makeshift rivets.

Considering that my FIL abuses this knife, I will also add a few screws to hold the halves together. These rivets would not "do it".

Alternatively, I may weld them together around the perimeter.

i
Reply to
Ignoramus4078
Loading thread data ...

A flatbed scanner would do a nice job capturing the tang shape (with a ruler in the scan for scale) so you could pull it into your CAD software as a background image to trace over.

Reply to
Pete C.

Well, what I should really do, is to learn to use my probe and install a audio jack outlet for mine.

i
Reply to
Ignoramus24925

Probe is fine for complex 3D surfaces, but a flatbed scanner is far faster for 2D stuff that will fit on the flatbed.

Reply to
Pete C.

Well, I still need to learn to use the probe.

i
Reply to
Ignoramus24925

...

Just so you know there's more than one approach, I like a probe for 2D shapes. You can go right from points to Gcode. I don't own a decent scanner, just a three function printer, so I've never really tried that route.

For 3D, I really like my laser displacement sensor. About 100 times faster than a probe. But I do need "the kid" to turn the point cloud into a program.

Karl

Reply to
Karl Townsend

Exactly.

Karl, do you have some example on the web of those laser sensors?

i
Reply to
Ignoramus24925

Or use a digital camera and Corel Draw and Corel trace then export it as a .dxf file. Works a treat for plasma cutting tables. Steve

Reply to
Up North

do a search on ebay for laser displacement.

Here's my controller:

formatting link
looks like the head isn't included in this offer.

It pays to shop ebay a while. The same units go from $20 to $1000+

my unit puts out 4 - 20 miliamps measuring a distance within +/- .0015 inches. You take data on the fly. For most stuff, sample once every

0.050 inches and move the machine at about 5ipm. if you're really into detail, I've seen scans of a coin and you can read the writing in the point cloud.

rhino is the best software for dealing with point clouds from what little I know.

Karl

Reply to
Karl Townsend

formatting link

I used to write software dealing with point clouds. That was for a defense contractor, and the objective was to recognize incoming warheads and determine several characteristics.

I will try to read up on laser sensors, right now I cannot tell a good one from a bad one, or a complete from incomplete one. It sounds extremely interesting, compared to a touch probe.

i
Reply to
Ignoramus24925

formatting link
>

Which side of the ocean?

I am the Sword of my Family and the Shield of my Nation. If sent, I will crush everything you have built, burn everything you love, and kill every one of you. (Hebrew quote)

Reply to
Gunner Asch

formatting link
>>

The "wrong" side of the ocean.

i

Reply to
Ignoramus24925

formatting link
>>>

The bottom? LOL

Reply to
John R. Carroll

formatting link
>>>

So did you manage to recognize an incoming Minuteman II or Peacekeeper?

Gunner

I am the Sword of my Family and the Shield of my Nation. If sent, I will crush everything you have built, burn everything you love, and kill every one of you. (Hebrew quote)

Reply to
Gunner Asch

formatting link
>>>>

Well, it worked pretty decently in simulation.

i
Reply to
Ignoramus20906

Do they nutate much?

jsw

Reply to
Jim Wilkins

Umm, if he told us, he'd have to kill us! Thank goodness they only got to check this out in simulation. the ONLY people who really know are on Kwajalein, and they certainly aren't talking. They get a couple Minutemen dummy warheads a year incoming.

Jon

Reply to
Jon Elson

formatting link
>>>>>

Cool!

Gunner

I am the Sword of my Family and the Shield of my Nation. If sent, I will crush everything you have built, burn everything you love, and kill every one of you. (Hebrew quote)

Reply to
Gunner Asch

Hmm ... sounds like a program which I heard of to develop software to do image-recognition on aircraft. It was one of those neural net things, and they tried training it on available photos.

End result was something *very* good at telling a sharp photo from a blurry one -- given that the training for our aircraft was done from publicity stills from the aircraft manufacturers, while the ones for the "other" aircraft were from spy work -- and tended to be not nearly as good photographically. :-)

So -- who knows what your thing would have done in practice. :-)

Enjoy, DoN.

Reply to
DoN. Nichols

Our people also worked a whole lot on this in the star wars programs, they went far enough to send up rockets with seeker heads that would try to discern the decoys from the dummy reentry vehicles. What worked best was to observe the rate of cooling of the objects as they flew through space after separating from the bus. The dummys (weight the same as real warheads in reentry vehicles) stayed warm, the decoys (something like balloons) cooled more quickly. They also did some kinetic kill intercepts, some worked, quite a number didn't.

We were allowed a certain number of test missiles a year under the START treaties, and tried to get as much info out of those as possible. So, while the site at Kwajalein recorded the impacts of the RV's in the ocean, they also fired a variety of seeker and interceptor missiles at the incoming test shots and also played many laser tag games and radar experiments. Much of that is still secret stuff, but some thin details have been let out. Kwajalein is supposed to be the most secret and secure US site in the world.

Jon

Reply to
Jon Elson

PolyTech Forum website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.