start of a good project

[mastermoparman]

a little history about me before we get started. i have been a machinist for about a total of 5 monts now. i spent 3 months at a shop that was in shelton wa before they laied me off due to finacial restraints. i was one of 4 people who got it and they said they would hire us back but 4 months later it didnt happen and i ran out of money and had to move in with some family in lynnwood wa. 2 days after i move i had a job at another machine shop. that was last month. still there and loving it.

ok i have done some research and decided that doing a retro fit to a mill just wont work for me. im going to kneed something a little bigger so i wont out grow it so fast. i was thinking about building a gantry style mill but i hear they are plagued with rigidly problems. and what i want to do with it and what they are capable of seem to not fit together so i have a issue to figure out. here is the things i want to be able to do with it eventually.

-must be able to cut up to 304 ss
-48in x 24in y ??in z
-total foot print not much bigger then table
-upgradeable to auto tool change
-upgrade to 4 and 5 axis
-able to use a .5 em and a 1.5 fm
-able to make gears

now the way i figure it i have 2 choices
choice 1. i make a gantry style and use uber heavy duty parts and say screw it to the fast rapids.
or
choice 2. i build a horizontal style machine. would solve a few problems but create a few of its own.

now keep in mind this isn't for production work. it is for more or less prototype style work. as i will be only using it for my self and friends making any thing from rc car and plane parts to stuff for my 79 plymoth trail duster. what i had figured to do was build a gantry with a big ass stepper for the x and use 2 ball screws and 2 timing belts to drive them form the 1 stepper. another option was to use a rack and pinion set up. of which i have no idea on the accuracy rating of thous. i did spend a week at a fiberglass mold shop as a floor sweep were they had a komo gantry style router for making the master for the molds. it had what look like the strips you put on Manuel style machines to make them digital. im not sher how this works or if i could use it for my setup.

so any input is appropriated

[arie kabaalstra]

Hi,

First of all.. you already did a lot of thinking before even starting.. that's good

pity you lost your job, but then , you already got a new... so who still cares about a company that can't even afford employees..


Those strips you mentioned are "linear encoders" on a manual machine they show the position on a Digital ReadOut (DRO), on a CNC, they tell the Control where the axes are.. ( and show you the positions on the screen )

looking at the dimensions (and calculating them to Metric .. yup.. i'm a metric dude 1.22m x 0.61m ) we're looking on a relatively big machine... so your choice for a Gantry style machine is also my choice...

if you want to "upgrade"to an ATC later on, i'd make on axis slightly longer to mount the ATC rack with tools behind the table ( or beside it, depending on desired size )

a nice solution for driving the axis along the table is, one ballscrew on each side, with a stepper, or servo motor on each spindle ( and a Toothed belt connecting the two spindles on the opposite side, so they will allways run equal..
if you make your gantry wide enough, you can mount the milling head between the rails.

if you want a real ridgid machine, i'd use two rails above eachother along the table on each side ( it's gonna cost you some this way) or, build two "ribs" on top of which the gantry-bridge will run, so you don't get that much momentum in the gantry as such ( furthermore, the rails and spindles will be sitting high ( and dry ) above the table, in case you want to machine with coolant ( if you mean stainless steel Aisi 304 with "304 ss") those high ribs will also make great "spatter screens" to prevent your workshop from getting all messy..

looks like a great project.. show us some drawings as well....

[mastermoparman]

ok so the linear encoders are what make it so i don't have to home the machine every time i start it up ?
now i did a Google for rails for a cnc and came up with a tone of different types. do any of you have any recommendation's as to which to use? i was looking at the round style like i have seen a few other gantry style. but i was figuring that they would bow in the middle and cause performance issues if they weren't perfectly parallel. so im leaning to the flat lay down style
at the moment im going to work on selecting the style of parts before i design a actual blue print for it. im still trying to figure out how im going to make the table for it

[arie kabaalstra]

i've been thinking about this project today..

those round rails are good for PCB and light engraving machines, but not for Heavy-duty equipment..

the square-ish ball rail guides are really tough, DMG, and Bridgeport use em in their Milling machines as well.. so why even think about anything else? ( i know i didn't when designing my little VMC )

There are basically two types of these rails, with socket holes drilles through, or with threaded holes from the bottom, the latter can only be mounted from the bottom, which means you have to keep that in mind when designing, but!. their biggest advantage is, the dont get chips in the bolt-holes.

the version with the blind threaded holes is more expensive, but there is a way of avoiding chips in the holes, mount the rails on its side...

the basic rules of CNC design:
1 :keep whatever is moving as light as possible
2 :Keep whatever is moving as Ridgid as possible

the bad news, those two rules dont mix.. so you will have to find a compromise somewhere..

i see a lot of gantry machines whith the spindles just beside the table, and a moving gantry towering above it all.. why not making two fixed ribs with only a light, yet ridgid gantry on top of it?.. to put it in other words, lower the gantry design, drop the table!.. the table and the frame ain't moving ( at least, they're not supposed to ) spo who cares they weigh a ton?

about the encoders encoders or not, you still need to home your machine ( and why shouldn't you?.. some machines have soft limits, others have "hard limits".. if you home it, the machine knows where it is, and how far it can move in any direction.

[mastermoparman]

ok cool im still havening problems with the spindle design. i like the idea of mounting the motor in a fixed location and having just the head move, but im confused as to how to have this happen. i know the dayna i operated in school had this option but don't know how it worked. and it is beginning to look like im going to have to fab my own spindle if i want a tool auto changer

im working on getting my hadns on a copy of solid works 2006 atm and will start playing with it. shoud be interesting considering i have never used it before.

[arie kabaalstra]

if you want a fixed motor, and a sliding spindle you need a spline spindle, that allows both rotational and linear motion.

i'm not a fan of this construction.. remember the basic rules? number 2 applies here.. a slidin spindle means bearings far apart, loads of momentum in the spindle ( this is why on CNC machines the motor mostly moves with the entire head.

About Solidworks: do you have any experience with this kind of software?. it's not that hard to learn, but it's quite another game... compared to "good-ol" Autocad.

if you want an ATC, most likely you will have to design your own spindle, the mechanism is like that of airhose connectors, a spring pushing steel balls inward, clamping the tool by a special bolt, ah well.. i think you've seen those.. releasing the tool is done by a pneumatic cilinder pushing on the spring, releasing the balls, and thus the tool.. clamping is done by just letting the air out..

you could of course, start with designing a plain spindle, maybe even whit the clamping mechanism, but manually operated, then all you have to do later is, adding the cilinder, and the tool rack, belt or umbrella style ATC.

[mastermoparman]

well i had planed on making the motor stationary but i don't know how else i would do it. i intend on using a 3-5 hp single phase variable speed indexable motor. maby im over doing it. but if i build it strong now i wont have to worry later. do you think that coolant though spindle is worth the extra time and effort?

any way i have been doing some searching on some gantry style cnc and happen across this one in cnc gallery

i like it because it eliminates alot of the moving mass for the x axis and gives it some regidity. the gantry arm could use a revision, but the hight of it is enough to put a table mount 5 axis on it.
as for solid works i have used mastercam 8 back in highschool. and have 10 now although i find it harder to use then 8 lol.

my end goal with this project is to have a good how to and a good platform for beginners who will be able to upgrade their machine as their skill increases. basically lego cnc

[harryn]

I am not very experienced in machining, and none in CNC yet (just reading here).

The limited work I have done has been using 3 manual units (1 lathe, a small mill, and a bridgeport clone my brother has).

While anyone reading this forum can machine circles around me, for the projects I have wanted to build, the biggest limitations so far on ALL of them have to do with boring narrow deep holes. In particular, removing chips / dust when boring down vertically just is a pain, and it is very easy to mess up the work.

Since nearly all of my projects are designed with long holes (example 20mm dia x 200 - 250mm long), when I finally get a mill, it will be capable of horizontal boring, and I actually have toyed with the idea of being able to bore "up".

The other challenges I have seen are mess, smell, noise, and safety, and these really are all related. If you can, try to build your tool in a way that it is inherently enclosed. As an experienced machinist, you can probably keep your self safe no matter what. In my case, I have some very curious kids that I want to actively encourage to use a cnc machine, and live to talk about it in a positive way. The idea I am thinking of, is to use the enclosure to help stiffen the system, the same way a unibody car is stiffened by the sheet metal.