Register a SA Forums Account here!
JOINING THE SA FORUMS WILL REMOVE THIS BIG AD, THE ANNOYING UNDERLINED ADS, AND STUPID INTERSTITIAL ADS!!!

You can: log in, read the tech support FAQ, or request your lost password. This dumb message (and those ads) will appear on every screen until you register! Get rid of this crap by registering your own SA Forums Account and joining roughly 150,000 Goons, for the one-time price of $9.95! We charge money because it costs us $3,400 per month for bandwidth bills alone, and since we don't believe in shoving popup ads to our registered users, we try to make the money back through forum registrations.
«40 »
  • Post
  • Reply
rotor
Jun 11, 2001

Official Carrier
of the Neil Bush Torch

 
 
 
 
teh butts


i just set it for whatever and hope nothing breaks.

Adbot
ADBOT LOVES YOU

rotor
Jun 11, 2001

Official Carrier
of the Neil Bush Torch

 
 
 
 
teh butts


so far lots of stuff has broken

echinopsis
Apr 13, 2004



College Slice

whatever doesn't kill you

rotor
Jun 11, 2001

Official Carrier
of the Neil Bush Torch

 
 
 
 
teh butts


CarForumPoster posted:

I shall explain it then. 1% of cutter diameter per tooth is a good rule-of-thumb starting feed rate for milling aluminum.

Power consumption is essentially a function of MRR and the way to decrease your cutter load effectively is to use constant cutter engagement angles. You don't want to run the cutter on center line nor do you want to slot if you can avoid it so your "good" cutter step over ranges tend to be 10-35%, 65-85% with the ideal being 70%. With <50% you get chip thinning effects and should increase feed but on your baby mill at 1% cutter/tooth (e.g. a 1/4" cutter would be fed at .0025/tooth or for a 2 flute .005/rev) you'll be ok staying in the 10-30% step over range even if your effective cutting thickness isn't the full 1%.

for serious tho this is good advise, thanks

CarForumPoster
Jun 26, 2013


You are welcome.

ante
Apr 9, 2005

No... Not without incident.

Designing a CNC for making PCBs on no real budget with no real idea what I'm doing


Those bottom rails slot into a drill press table I can probably convert for a Z axis.

VictualSquid
Feb 29, 2012

Gently enveloping the target with indiscriminate love.


Substrate Mills are pretty crazy stuff.

kafkasgoldfish
Jan 25, 2006

God is the sweat running down his back...

ante posted:

Designing a CNC for making PCBs on no real budget with no real idea what I'm doing


Those bottom rails slot into a drill press table I can probably convert for a Z axis.

Setting up parallel guide rails can be pretty tricky (at least for hobbyist 3d printers). It looks like you're just going to capture the rods in holes in the angles on each end, but getting four holes that allow for two perfectly parallel rods is easier said than done. Maybe you could include some adjustability there?

my ideas are terrible: #1 capture the rod ends in small plates that can float in oversized holes with nuts and bolts. #2 what if you relied on something else to join the two square tubes and then split the angles on each end into thirds so that each pieces that could be adjusted/shimmed individually?

Sagebrush
Feb 26, 2012





Gravy Boat 2k

gently caress. I forgot one parameter in my CAM software that always gets me and now I have two 5-hour jobs that were cut improperly by 50 thousandths. Fffffffffff

It's recoverable, but drat

ante
Apr 9, 2005

No... Not without incident.

kafkasgoldfish posted:

Setting up parallel guide rails can be pretty tricky (at least for hobbyist 3d printers). It looks like you're just going to capture the rods in holes in the angles on each end, but getting four holes that allow for two perfectly parallel rods is easier said than done. Maybe you could include some adjustability there?

my ideas are terrible: #1 capture the rod ends in small plates that can float in oversized holes with nuts and bolts. #2 what if you relied on something else to join the two square tubes and then split the angles on each end into thirds so that each pieces that could be adjusted/shimmed individually?

I was going to maybe add a couple square tubes in the centre to join the square tubes/increase rigidity instead of just relying on the angles.


You've got a good point with the holes in the angles. I was thinking of maybe using my drill press to go right through both sides of some square tube, then cut that into two angles. That will still result in mis-centred holes on opposite sides of the face, though.

Hrm.


Drilling long-ways through rectangular tube and then splitting it might work. I don't like that, though. What are my odds in being able to make a jig out of wood and keeping the holes pretty accurate?


Splitting the pieces shouldn't be a problem. I can put another piece of angle behind them to keep the rods in, it'll be able to keep them square, too.

I dunno. I'll think about it. Tomorrow I should have a revised version. Amateur machine accuracy is so difficult.

echinopsis
Apr 13, 2004



College Slice

Sagebrush posted:

gently caress. I forgot one parameter in my CAM software that always gets me and now I have two 5-hour jobs that were cut improperly by 50 thousandths. Fffffffffff

It's recoverable, but drat

this is why you should never do anything that matters

kafkasgoldfish
Jan 25, 2006

God is the sweat running down his back...

ante posted:

I was going to maybe add a couple square tubes in the centre to join the square tubes/increase rigidity instead of just relying on the angles.


You've got a good point with the holes in the angles. I was thinking of maybe using my drill press to go right through both sides of some square tube, then cut that into two angles. That will still result in mis-centred holes on opposite sides of the face, though.

Hrm.


Drilling long-ways through rectangular tube and then splitting it might work. I don't like that, though. What are my odds in being able to make a jig out of wood and keeping the holes pretty accurate?


Splitting the pieces shouldn't be a problem. I can put another piece of angle behind them to keep the rods in, it'll be able to keep them square, too.

I dunno. I'll think about it. Tomorrow I should have a revised version. Amateur machine accuracy is so difficult.

One other thing, you can probably leave one end of the threaded rod hanging instead of seating it in the far plate. Depending on how much $$ you put into the rod and how well it's seated against the stepper, it may not be perfectly straight and one end might have some wobble (even a tiny amount could cause some binding).

For the platform, are you going to use linear bearings?

edit: You might consider using using aluminum extrusion from someone like Misumi. It would simplify the assembly significantly.

kafkasgoldfish fucked around with this message at 01:10 on Nov 2, 2013

ante
Apr 9, 2005

No... Not without incident.

I'm planning on spending almost no money on this if I can help it.
All of the linear motion stuff is salvaged from a bunch of free printers/scanners I got. Some 13" rods and brass bushings. I can carve out a pillow block for the bushings out of some 3/8" acrylic or something.


Aluminum extrusion would save me some hole-drilling in the square members, at the cost of some stiffness. So I might do that for a prototype, I'd ideally like to throw this together as fast as possible.

So I've got this, now:


I added some squares in the middle, and I'll just use some threaded rod through the whole assembly and torqued down really tight to keep everything stiff.

The linear rods are held by holes in the inner angles, and kept from slipping out by a solid/undrilled outer angle.

I'm really not sure about the rod straightness solution I had going in my head last night, though. I'll do some googling for good ways to make it pretty accurate, but maybe I'll have to do some loving around when it's all built. Oversize the holes, fill them with hot glue and slide the platform back and forth a whole bunch? I dunno.

I'll work on the motor assembly and coupling next.

CarForumPoster
Jun 26, 2013


ante posted:

I'm planning on spending almost no money on this if I can help it.
All of the linear motion stuff is salvaged from a bunch of free printers/scanners I got. Some 13" rods and brass bushings. I can carve out a pillow block for the bushings out of some 3/8" acrylic or something.


Aluminum extrusion would save me some hole-drilling in the square members, at the cost of some stiffness. So I might do that for a prototype, I'd ideally like to throw this together as fast as possible.

So I've got this, now:


I added some squares in the middle, and I'll just use some threaded rod through the whole assembly and torqued down really tight to keep everything stiff.

The linear rods are held by holes in the inner angles, and kept from slipping out by a solid/undrilled outer angle.

I'm really not sure about the rod straightness solution I had going in my head last night, though. I'll do some googling for good ways to make it pretty accurate, but maybe I'll have to do some loving around when it's all built. Oversize the holes, fill them with hot glue and slide the platform back and forth a whole bunch? I dunno.

I'll work on the motor assembly and coupling next.

This seems like it has a lot of extra parts, why not just make the table follow the square tubing and the lead screw pass through the cross members?

ante
Apr 9, 2005

No... Not without incident.

Seems to me that the linear rods are necessary for accuracy. I could lower the platform like you say, but then there's nothing stopping the side-to-side motion or the up-and-down. Square tubing isn't exactly built to tight tolerances, so bearings wouldn't go over the surfaces perfectly smoothly.


Or am I overlooking something?

CarForumPoster
Jun 26, 2013


Unless you're ordering ground rods, which might be a good idea theyre not too much more expensive, theyre basically extruded in the same way. Dont expect your straightness per foot to be great with rod either. If you could get thick walled tubing (or just solid, this is the base of a machine) weld it all together and then just square it up on a bridgeport you'll find massive improvements in all these regards.


Keep in mind weldings not going to leave anything square so unless you line bore the holes for the control rods after welding they'll be pretty far off as well. This is why eliminating them and machining a control surface is probably the best idea. Do you know anyone with a mill?

ante
Apr 9, 2005

No... Not without incident.

I've got a bunch of rods salvaged from printers and scanners and no access to a welder or mill.

A mill would be pretty keen

CarForumPoster
Jun 26, 2013


ante posted:

I've got a bunch of rods salvaged from printers and scanners and no access to a welder or mill.

A mill would be pretty keen

Good news is if the rods have ends that are mounted together to make a continuous rail theyre likely pretty good tolerance wise. I.E. if its all one assembly, hooray.

kafkasgoldfish
Jan 25, 2006

God is the sweat running down his back...

CarForumPoster posted:

Unless you're ordering ground rods, which might be a good idea theyre not too much more expensive, theyre basically extruded in the same way. Dont expect your straightness per foot to be great with rod either. If you could get thick walled tubing (or just solid, this is the base of a machine) weld it all together and then just square it up on a bridgeport you'll find massive improvements in all these regards.


Keep in mind weldings not going to leave anything square so unless you line bore the holes for the control rods after welding they'll be pretty far off as well. This is why eliminating them and machining a control surface is probably the best idea. Do you know anyone with a mill?

This is just a pcb mill, seems like solid tubing, milling and welding are a bit over the top. He could build the frame out of dimensional oak 2x2s and cheesy galvanized brackets from THD it'd be more than rigid enough.

Also, what is the scale of this thing? If you're using scrap components from printers and the like, none of those rods can be much more than say 12-16"?


p.s. misumi extrusions are like $1/10cm, are totally sufficient strength wise and they'll cut them to length for free. last time I'll mention it.

CarForumPoster
Jun 26, 2013


My concerns weren't of rigidity, they were of accuracy and eliminating parts.

Sagebrush
Feb 26, 2012





Gravy Boat 2k

Yeah, while rigidity isn't a huge deal for PCB milling, you definitely need precision. Complex circuits have spacings on the order of 0.005"; even for hobbyist use you need something that can hit 10 thousandths repeatedly, and .001 or better is ideal.

Bad Munki
Nov 4, 2008

We're all mad here.



I'm all about CNCs and such, I've built one myself and have plans to build a much bigger/better one, I think they're awesome and fun and...if your goal is to make PCBs, why not just chemically etch them? It's pretty darn easy and requires minimal investment. You don't even need your own laser printer, you can get a pack of toner transfer sheets and get it printed at kinko's or wherever. Get some TRF foil to go with, and you can even make solder masks and silkscreens.

Sagebrush
Feb 26, 2012





Gravy Boat 2k

For me -- I'm all about dry processes. I try to avoid things that require large volumes of liquid and toxic chemicals as much as possible. Mills make a lot of dust and chips, but cleaning that up is a lot more pleasant than cleaning up a quart of spilled ferric chloride. I already have a CNC mill, though ... if I didn't, and I was only interested in making PCBs, I don't think I'd build one just for that purpose.

Bad Munki
Nov 4, 2008

We're all mad here.



That's fair, I was just curious what the motivation was. I've done muriatic acid + hydrogen peroxide etching, and man, it's a cinch. You don't need a ton, either: put a few tablespoons of etchant in a zip log bag with your board and you're golden. Ziplocs withstand the etchant just fine, and make it so you need a lot less. Anyhow, carry on, just wanted to chime in there.

For what it's worth, the mill I currently have was actually designed for doing circuit boards. I liked the design, but wanted a bigger envelope (designed with a 6x6 envelope which I expanded to 12x12) so now of course the tolerance is awful as you approach the edges.

My current plan is a Joe's 4x4 built as a 2x3, and when I finally have enough shop space, I'll partially rebuild it as a 4x8 so I can drop sheet goods on there.

ante
Apr 9, 2005

No... Not without incident.

kafkasgoldfish posted:

This is just a pcb mill, seems like solid tubing, milling and welding are a bit over the top. He could build the frame out of dimensional oak 2x2s and cheesy galvanized brackets from THD it'd be more than rigid enough.

Also, what is the scale of this thing? If you're using scrap components from printers and the like, none of those rods can be much more than say 12-16"?


p.s. misumi extrusions are like $1/10cm, are totally sufficient strength wise and they'll cut them to length for free. last time I'll mention it.

Yeah, the rods are all around 13".

On this one axis with a 5" table, so I should be able to get 7". That's plenty for most of the PCBs I'd want to build.

I'm in Canada, so shipping makes Misumi not really worth it (although they have an awesome system). I can and probably will get some extruded rails locally.



And for making PCBs, I've used the UV exposure method, and I found dialing in all the timings to be an enormous pain in the hole. At the time, I was working a stupid amount of hours, so my time was more valuable than just sending it out. Ideally, I'd like to slap this together really quickly, and have all of the errors at least repeatable enough that I can correct for most of them in software.

Basically I want to be able to design something and then press "print" and have a physical copy in a few minutes. I don't know how realistic that is, but it's good mech engineering practice in the meantime.

cakesmith handyman
Jul 22, 2007

Pip-Pip old chap! Last one in is a rotten egg what what.



I don't know if this is the best place to ask but my brother wants to build a wood mill for his little cabinetry business. He needs a 1500x500x100mm envelope, Any suggestions on where to start? he'll need to cut all types of wood but nothing harder.

sixide
Oct 25, 2004


ante posted:

Seems to me that the linear rods are necessary for accuracy. I could lower the platform like you say, but then there's nothing stopping the side-to-side motion or the up-and-down. Square tubing isn't exactly built to tight tolerances, so bearings wouldn't go over the surfaces perfectly smoothly.


Or am I overlooking something?

The rods seem pretty dodgy. I don't think you'll get the rigidity you're looking for, especially in the center of the range. There's also no simple way for a hobbyist to "true" the bearing surfaces. Much easier to use a flat surface and lap the ways against a reference surface (sheet glass is a great poor man's surface plate) before lapping the carriage against the ways.

CarForumPoster
Jun 26, 2013


Cakefool posted:

I don't know if this is the best place to ask but my brother wants to build a wood mill for his little cabinetry business. He needs a 1500x500x100mm envelope, Any suggestions on where to start? he'll need to cut all types of wood but nothing harder.

Id start with googling and ebaying a CNC router kit. I bet china makes one and someone imports it. Since tolerances are non existent for wood a lot of corners can be cut that cant be for aluminum.

Id assume there are formulas for calculating the horse power you need for a spindle for milling wood similar to aluminum/steel/etc. Add a decent fudge factor to that because router bits are not ground to the same quality nor are they replaced as often as metal cutting bits.

Sagebrush
Feb 26, 2012





Gravy Boat 2k

CarForumPoster posted:

Since tolerances are non existent for wood a lot of corners can be cut that cant be for aluminum.

No, no, corner cutting depends on the tool radius.

rotor
Jun 11, 2001

Official Carrier
of the Neil Bush Torch

 
 
 
 
teh butts


Sagebrush posted:

No, no, corner cutting depends on the tool radius.

it's a fine line between clever and stupid

Sagebrush
Feb 26, 2012





Gravy Boat 2k

I like to keep one foot on either side, riding it with my balls.

CarForumPoster
Jun 26, 2013


Sagebrush posted:

No, no, corner cutting depends on the tool radius.



Loving Africa Chaps
Dec 3, 2007


We had not left it yet, but when I would wake in the night, I would lie, listening, homesick for it already.



I'd like to make a small cnc mill. The parts I'd be playing with would be small, like 100mm x 150mm x 40mm. Most of it would be plastics but I'd like to be able to some light brass and aluminium. Most of the designs I've seen have much bigger areas and it seems silly to sacrifice rigidity and accuracy for space I'd never use. Anyone got any links to any examples of small mills?

rotor
Jun 11, 2001

Official Carrier
of the Neil Bush Torch

 
 
 
 
teh butts


Loving Africa Chaps posted:

I'd like to make a small cnc mill. The parts I'd be playing with would be small, like 100mm x 150mm x 40mm. Most of it would be plastics but I'd like to be able to some light brass and aluminium. Most of the designs I've seen have much bigger areas and it seems silly to sacrifice rigidity and accuracy for space I'd never use. Anyone got any links to any examples of small mills?

what's your budget?

Loving Africa Chaps
Dec 3, 2007


We had not left it yet, but when I would wake in the night, I would lie, listening, homesick for it already.



rotor posted:

what's your budget?

Ideally about 500 or less. I've got a dialysis machine I can pull stepper motors from and MY GIRLFRIEND works in special effects so has a few machinist friends I can probably blag some materials from so I guess most of that budget can go to ball screws, controller and spindle.

My rough idea is that by keeping the build area as small as possible it might make getting a steel or aluminium plate to use as a base for a gantry style design less expensive

Loving Africa Chaps fucked around with this message at 15:19 on Nov 10, 2013

Sagebrush
Feb 26, 2012





Gravy Boat 2k

Cross posting from Cycle Asylum, I made some rubber parts for my motorcycle by CNC'ing a mold and casting them from liquid resins:

Sagebrush posted:

Finally finished a project I've been working on for some time.

So I like the look of those rubber knee grips that a lot of older motorcycles had, and which are still around on modern Triumphs and Enfields and such.



The very earliest versions of my bike, a Honda CL350, and its precedessor, the 305, came with rubber knee grips like those above. The 1971 model has a differently shaped tank and did not come with grips. When I look on eBay, I can find lots of grips for different models (tons for the S90 for instance), which wouldn't fit, and a few NOS grips from things like CB450s that might fit okay but they aren't really designed for my tank shape and they're usually $150+ a set. There are also generic ones available, but they're usually pretty ugly or designed for a peanut tank; and finally there are zillions of types of grip that say Triumph or Norton or whatever on them but I can't do that, can I

So I decided to make my own.

1: doodle some concepts and lay out the general design on a photo of the side of the gas tank.


2 (not shown): print out the drawing at various scales, cut out paper representations of the grip, and figure out the proper size, since you don't know if what you're drawing will actually look right in real life. Record the scale factors you need to use when modeling the part.

3: import concept to CAD program of your choice and, using the concept drawing for scale, model the actual part in three dimensions.


4: invert this 3D model, creating negative geometry that can be used as a mold.


5: cut this mold (and a mirrored version for the other side of the tank) on your CNC machine. Make the logo insert with your laser cutter because you haven't got an end mill small enough to cut all the details you need.



6: clean and prepare mold with releasing agents.


7: fill with 2-part liquid urethane rubber


8: after 16 hours, demold parts


9: wash off all curing oils and traces of releasing agents


10: nice rubber parts!


11: clean gas tank and attach parts with heavy-duty silicone glue, apply moderate clamping force for 24 hours


12: done


Now you have the only rubber knee grips designed specifically for a 1971 CL350 in the entire world!

The modeling material is RenShape, in case anyone was wondering. It's a phenomenally awesome prototyping board that is also phenomenally expensive.

Bad Munki
Nov 4, 2008

We're all mad here.



Those are fantastic.

Elston Gunn
Apr 15, 2005



That is so cool. I really regret selling my '71 CL350.

cakesmith handyman
Jul 22, 2007

Pip-Pip old chap! Last one in is a rotten egg what what.



That's incredible.

Adbot
ADBOT LOVES YOU

Seat Safety Switch
May 27, 2008

MY RELIGION IS THE SMALL BLOCK V8 AND COMMANDMENTS ONE THROUGH TEN ARE NEVER LIFT.



Pillbug

That gives me so many ideas for other crap I could make. So cool.

  • 1
  • 2
  • 3
  • 4
  • 5
  • Post
  • Reply
«40 »