/diy/ - Do It Yourself

bump

>I have looked around, but I haven't found any good resources.
How hard have you looked around? Checked amazon for highly reviewed machining books? Checked youtube for machining videos? Visited or toured a machine shop?

There's lots of good stuff out there.
Check out machinery's handbook, or millwrights and mechanics guide.

http://www.youtube.com/watch?v=U99asuDT97I

>>326279
Forgot to mention, check out the other tutorial videos by Glacern Machine Tools: http://www.youtube.com/user/GlacernMachineTools?feature=watch

They're actually really interesting, and educational. Even if you're not interested in machining or milling stuff at all.

>>326247

you can fabricate a lot of metal stuff from stock sizes and shapes, a hacksaw, vise, drill press, and standard fasteners.

Seriously, you don't always need to form shapes. Boxes, brackets, mounting things, etc.

By-hand or machined, you will need to think in terms of what materials can do. experience helps of course but you get that by just starting to make shit.

a lot of it is simply knowledge of materials, what they can do, and what the limits are.

Sorry for my late reply!

>>326273
>>326279
>>326281
>>326283

Thanks for the help! I'll look at all of those.

Don't buy new tiny crap. Get old cnc stuff for cheap, just keep your eye out and don't get ripped off. Make sure you see it under power before you buy. If you dont have three phase dont worry, you can convert two to three fairly simply.

bumpity

I would consider a Mill/Drill, Lathe, Bench Grinder, and 4x6 Bandsaw to be the core of a hobby machine shop. If you really want CNC I'd recommend using a manual machine for a few weeks just to get a feel for things.

You can certainly get very far with just a bandsaw and a drill press.

Its possible to get a good deal on used stuff but it is very hard to make that call as a complete beginner.

>>326747
Why use a manual machine? I'm just curious, I want to know what benefits it has over CNC.

Don't jump into this without actually learning to use the machines first. Contact local colleges and see if they run courses - the instructor would also probably be happy to help with selection of used equipment (don't buy new unless you are willing to fork out megabucks - new cheap/moderate stuff is designed to wear out, old stuff is far better)

>>326781
Alright, thanks. I'll look into local community colleges.

>>326776
professional metal worker here with a masters in mechanical maintenance engineering. And in my personal opinion the manual machines give you a more personal feeling, so at least start out with the manual ones to get a deeper feel for how the metal reacts. Also try to get a small basis in metallurgy.

>>326805

Don't forget that a multi thousand dollar investment is a waste if he is only making one or two x by y by z blocks with one or two holes in them.

>>326805

Thanks for the info. I'll look into metallurgy as well.

>>326815
Metallurgy is quite easy you just gotta get down the basis like wolfram has the highest heat tolerance and so forth, also know that the diameter of your drill bit and the material you use, can have a huuuge effect on how many rotations per minute you should have, so that you don't overheat and or destroy your piece.

>>326247
Check out The School of Automotive Machinists in Houston. The website is samracing com. It is mostly a school on how to build and machine motors but the machining is universal. They have a CNC program as well as a block and cylinder head course.

>>326776
Even with CNC you need to know about cutting speeds, feedrates, how to prepare a stock peice for machining, how to mount a peice securely, etc.
I spent a year on my seig x3 before converting it to CNC.

Oh and if CNC interests you, make sure you figure out the CAD CAM Gcode CNC software issus before making any investments. I didn't and it sucked.

And get a lathe too. Lathes are fun.

>>326950
>make sure you figure out the CAD CAM Gcode CNC software issus before making any investments

Thanks, I will.

bump

Visit the library. Even old books from the 50s and 60s on machining have basically the same and still applicable information about basic machining knowledge and practices.
>>If you really want CNC I'd recommend using a manual machine for a few weeks just to get a feel for things.
THIS.
If you don't learn manual first you're VERY likely to completely fuck up a very expensive CNC machine in short order. It won't do any of your thinking for you.

You need to know the basics of feed rates, end mill types and wear, facing ops, leveling and truing your equipment, and many more things than I can list off the top of my head. Getting started with a manual knee mill would be a good idea.

Knowing how to read and edit G-Code is also very important if you intend to accomplish anything with CNC. Even the G-Code that expensive CAM software outputs needs to be proofed and often times edited by hand before you run it.

>>326247

learn to work with your hands. machine tools do not do your thinking for you. if you cannot work up solutions in yr head and on paper, you can't fabricate them with any tool, hand, machine or CAD/CAM.

Seriously, learn to fabricate stuff with hand tools; a good bench, a vise, hacksaw, drill press, a good ruler and combo head.

if you CANT do things by hand, first, the rest is a waste to even think about.

Machining is a lot more difficult than people assume. Its not just about moving an endmill through the workpiece, you need to know how fast to do it, what angles to approach from, how much you can cut in one pass etc...

When using a drill press you innately know how fast you can drill by how it feels.
On a CNC machine you type in a number and that's what the machine does. If you go too fast the drill might fold or bind. Go too slow and the drill will rub and burn.

Its like riding a bike but via remote control. Without tactile feedback you need to understand your machine or small problems will go undetected and become serious.

If you don't have a basic understanding of how a machine operates you are going to make a lot of expensive mistakes.

Without exaggerating, if you type an extra zero or miss one line of g-code, your machine could start drilling holes in itself.

>>326247
This book and a pile of scrap if you're a true DIYer

>>327618
>>often times edited by hand before you run it.
why? What goes wrong?

2nd on the community college.

The problem with metal machining is that it's not a cheap thing to get into. Especially if you want to stay away from the "crap". Oh yeah, the machinery also takes up a lot of room. Even the small ones can be pretty big.

Oh also those "maker" spaces usually command ridiculous prices and doesn't make a lot of sense to you if you're learning and not just going to jump in and do your projects. I mean if you already know how to do say 80% of everything and just live in a tiny apartment and need access to the tools that's different. But being a newbie and paying the monthly dues + any classes they give doesn't make sense with community colleges around. And after taking the classes I think some colleges have a drop-in rate to work on projects even if you're not taking a class but are/have been a student...idk you have to check.

>>327741
>why? What goes wrong?
Proofing the output of the software is always a good idea to make sure that you didn't screw up Entry/Exit paths, the tool definitions are accurate to what's in the machine and what material you're cutting, the length of the G-Code isn't too much for you machine to handle (some have limited memory), and being able to manually enter code and parse together multiple small programs into a whole program is useful.

Being able to edit the G-Code in notepad and editing it at the machine are both invaluable skills to have.

Pic related is a part made at work in a HAAS VF1 with an indexing head (end plate and bolt of the custom collet is missing). It's a program that machines one row of cavities, manually indexes over, then repeats the program, and so on until all of the rows are done.

>>327741
>why? What goes wrong?
Proofing the output of the software is always a good idea to make sure that you didn't screw up Entry/Exit paths, the tool definitions are accurate to what's in the machine and what material you're cutting, the length of the G-Code isn't too much for you machine to handle (some have limited memory), and being able to manually enter code and parse together multiple small programs into a whole program is useful.

Being able to edit the G-Code in notepad and editing it at the machine are both invaluable skills to have.

Pic related is a part made at work in a HAAS VF1 with an indexing head (end plate and bolt of the custom collet is missing). It's a program that machines one row of cavities, manually indexes over, then repeats the program, and so on until all of the rows are done.

>>327805
>It's a program that machines one row of cavities, manually indexes over, then repeats the program, and so on until all of the rows are done.
isn't that only considered 3.5-axis? ;)
do something with 4-axis-simultaneous cuts

>>327815
>do something with 4-axis-simultaneous cuts
We have not yet needed to do anything more strenuous than axis substitution cutting.

Take a wild guess as to how this part was made. The outside was machined. The inside was not.

>>327815
>do something with 4-axis-simultaneous cuts
We have not yet needed to do anything more strenuous than axis substitution cutting.

Take a wild guess as to how this part was made. The outside was machined. The inside was not.

>>327863
In suspenders - Do tell

>>327863
"Pinched" the rows together from top to bottom with some heavy preasurre tool and repeated for each row while rotating at set speed? (never machined before, so thats just a guess)

>>In suspenders - Do tell
The previous image is only a slight clue.
We machine a "mold" piece out of aluminum that will determine the inside geometry. The mold is then place in a can and filled with metal powder.
The can gets welded closed and stuck in a Hot Isostatic Press and consolidated to make a fully dense part.

Then the outside is machined (indexed and aligned so that the inside pattern matches the outside). Then the inside is bored out and the remaining mold segments are removed.

Pic related is another part geometry made using the same process.

>>327866
>>327878
So, I thought it was a P/M process. What is the the finished density compared to drawn barstock?
even the best P/M's don't approach 1:1: do they ?

>>327903
>>even the best P/M's don't approach 1:1: do they ?
Obviously depends on the metal, but provided you can reach the temperatures and pressures needed you can get up to 98% density.

The consolidated material will have different machining characteristics and mechanical properties than a forged or extruded billet. Based on many factors like grain size, homogeneity, thoroughness of the degassing process, and so on.

Good resources you say:

http://thepiratebay DOT se/torrent/5926862

Check the comments for related DVDs.

I work in the upstairs bit of one of these places. Quite honestly they do have a lot of skill but I have no idea why someone couldn't just take that ability and apply it to something like oil rig work and get double the salary at a fraction of the working hours.

Rapid prototyping machines can now laser fusion steel and brass powders. A Public Domain effort of this machine technology Is Named " RepRap " and concetrates on a standardized design plastic extrusion machine. Some Users have extended plans into the actual metal part fabrication.

HTH

>>328015
... and occasionally lose a foot, lol.

i worked, once!, on a Mobil mudboat being used as survey ship for a jackup oil rig placement off the coast of nova scotia. (I was running side-scan sonar and sub-bottom profiling gear). unbelievably fucking dangerous work. mine wasn't particularly -- i just climbed up an outside, exposed metal ladder from the deck up two decks in pitching seas. literally no one would notice you went overboard until the next shift change. (it was 6hrs work, 6hrs off, 24/7 for two weeks). drunk as fuck mean asshole irish crew. food was great. i didn't shit for like 4 days due to pitching ship. slept in subdeck next to clanking chains or something.

mudboats are basically a flat football field, 6 feet off the sea, a 3-foot railing, and a towering infrastructure at the front. we had a remotely operated winch pulling our fish gear in and out. to me, it was the BIG WINCH. But when we got it out to the mudboat, it shrunk down to a tiny toy. these monster grunty guys flopped it around, and welded it straight to the deck like you or i would masking tape a postcard to the fridge.

fucking terrifying place. all the nova scotians hated us too, cuz we were out of towners and they hated the oil co's for some reason. all fishers i guess.

and th epay sucked, cuz i was a young (19, 20) shit-ass electronic tech working for the subcontractor. glad i did it, never again.

>>326379
>Get old cnc stuff for cheap

no, don't do this. It's a terrible idea

I worked on a Laser CNC machine, cuts 3 Meter sheets of metal about 25mm thick, fantastic results with good CAD operators.

>>326379
>Don't buy new tiny crap. Get old cnc stuff for cheap, just keep your eye out and don't get ripped off. Make sure you see it under power before you buy.
Some people don't have room for full-size machines
also a machine can have a bunch of problems that aren't apparent to a newbie--especially one with a CNC setup

if you have the room and have someone with the experience, it is true that you can find old full-size (manual) machines for what you will pay for a new little one.

if you are limited to small machines, Jet and Grizzly prolly have the best reputations in the USA. Harbor Freight/Northern Tool probably the worst