[richardmg9]

I have wanted a cnc machine for years, and this summer is finally the time. I'm not sure if this is the correct forum as I intend to also machine metal, but this seems more active, and I will primarily work with wood.

I recently purchased 4 15mmx670mm square linear rails and 8 trucks. http://d3d71ba2asa5oz.cloudfront.net.../q11s015-1.jpg They are used. Hopefully they are still in nice condition with lots of life left! I am apprehensive about buying used rails, but new rails are monetarily out of the question. I want to build the best machine I can afford, so I figure used square rails are better than new round rails. Any thoughts on this?

The plan is for a fixed gantry. I bought these particular rails because the deal seemed too good to pass up. I originally thought I would use them for my X and Y axes, giving me a working area of ~20"x~20". I am considering using one set for my Z axis and getting some 20mm rails for the X (gantry) axis. The thought being that my Z axis will attach to my X, making the my X axis the most important to build heavy. Thoughts?

I am a financially challenged college student, and this project will be ongoing. My plan is to build a super solid "skeleton" (frame, rails, ballscrews) and continually upgrade from there. My spindle will most likely be cheap crap to start (probably my dremel. I know, I know, a dremel on a machine with square linear rails...). I have not decided if I will use nice steppers to start (~5 netwon/meter) or if I will go for something much cheaper and upgrade later.

This machine will be built similarly to this build: https://www.youtube.com/watch?v=K9UA9ZRFwWU (cad .step file attached). I want to build something very similar, just an X axis about half the size. I have access to a machine shop at the college, and intend to use cheap square steel tubing as much as possible. Similarly to the linked build, I will be milling square faces on the steel to mount my rails. I have never welded anything, and I am not sure how much help I can get, but I do have access to a tig welder at the school (this should be interesting). I do not have a bed concept that I am happy with.

Once I have the base machine running, my most pressing upgrade will be a full enclosure that is as sound proof as possible. This machine will live in an apartment. To this end, I am thinking particle board-foam-birchply sandwich, and isolating the entire machine base with a 1" thick rubber mat that I just so happen to have already. If anyone has any thoughts or resources for sound dampening, I'm all ears. Dust management is a close second. In addition, the machine will have to have at least one side no larger than 29 inches to get through apartment doors.

I recognize that this build is overkill for woodwork, but it fits the concept that's been rattling around in my head for years. I want a "do all" machine. I plan to mostly work with soft metals and wood, with the occasional steel (obviously with a nicer spindle and decent steppers/drivers). Long term plans include a 4th axis mounted to the bed, a 3d printing head, 3d scanning, and a flood coolant system similar to this build: Projects 8 coolant

Thanks for reading, I look forward to any thoughts /comments. I will post progress of the build as at happens here.

-Richard

[awerby]

For a machine intended to cut wood, the 15mm rails will be fine. Making a machine capable of cutting steel effectively is a whole other project. Aluminum might be possible if you make an exceptionally rigid machine. Use heavy steel tube for the frame (especially if you plan to mill some away), not the cheapest kind you can find. A MIG welder will work better than the TIG for putting it together. Practice on scrap before attempting to weld the parts for this machine, until your welding skills are up to the task. What will be driving the axes - screws, or rack and pinion? Dual motors on X or a single one? You might look for an actuator to use for a Z axis; they often cost less than the rails and screw they contain, and everything's already aligned. Build the frame first, and then start looking for the electric/electronic parts. Holding torque isn't the only thing to look for in a stepper motor - pay attention to the inductance as well (lower is better). Don't get the cheapest drivers available on ebay; you'll just have to throw them out and buy something else.

Forget about mounting your Dremel to this thing - you'll just kill it without making much impression on your wood. If you're really going to be using this thing in your apartment, I hope your neighbors are extremely tolerant (and preferably deaf) - these things make a lot of noise, mostly from the cutting action itself. Dust collection will be crucial, if you don't want to live like a rat in a bed of shavings. Do keep us posted on your progress.

[richardmg9]

Well, I just saw Walter's EG casting. Re-evaluating my life now. Will definitely be attempting some EG work instead of welding steel.

[louieatienza]

Look up the Wade-O design; it's a very robust machine that uses mostly off-the-shelf items, which save a lot of fabrication time.

[skubajon]

Not sure what your budget is. I've been building mine using 15mm rails. I don't plan on machining steel but will rarely and slowly.

Biggest part is not to get discouraged. You will get a ton of comments going both directions.

All in all my machine will weigh in at about 2000 lbs. Welding is easy to do with a bit of practice

Now tricky part is lining all things up. I struggled with binding because of extrusion that I bolted majority of stuff to. Mid build in ended up machining chromoly plates which sat on top on extrusion.

Take your time...take comments with a grain of salt..have fun and be creative. Can always correct or redesign your machine as you build. I know I have! Guys here call me crazy because my x is dual servo driven using 4 1000mm 15mm rails and 8 trucks just for the x. All I did was get worried that I spent a ton of extra money but it worked out great and it was my own design!

Have fun is all I'm saying. Everyone here is pretty knowledgeable and nobody means harm!! Epoxy is a great tool for assembly as well as I have learned

Sent from my SM-N910V using Tapatalk

[JSantars]

That machine is overkill. I'm not even sure it's an efficient design. There's a lot of mass to accelerate and it isn't very portable. If you don't plan on moving around very much and you have unhindered 24hr access to machine tools and welding equipment/skill then this is fine. If this is not your case, then you might want to rethink things.
I'm not trying destroy dreams or put down that builder. Heck, I watched all three vids and enjoyed them very much, but we all have access to the same resources here and routers capable of nonferrous metals at half the weight, cost, and complexity are not a rarity. Adding, don't underestimate the amount of time behind the scenes that this builder (a seasoned welder and machinist in my professional opinion(though I had concerns the way he finished those nut blocks)) went through thinking and planning.
This post isn't meant to deter you from your dreams it's meant to help you notice that the road ahead isn't as pleasant and flowery as the ukulele music and cute kid make it out to be.
If you have campus access then this is a plus. I'm confident you can find a professor that'll get into the project and may even turn it into a learning experience for machine design..like a tech club...hint hint If you can gain access to a welder that understands what you're trying to achieve then that's an even bigger bonus.
Good luck in all of your endeavors and keep us posted.

[louieatienza]

For sound dampening, look into stone wool. Cheap and effective. Homasote is one brand name. And if you cut stuff that make hot swarf fly around, this stuff will not burn. And it's impervious to water as well; it just drains out of it.

Again, look at the Wade O design. Substitute the Blanchard ground steel top for Mic-6 cast aluminum, that you can counterbore and drill yourself, and you'll save a ton of money. Build a frame of used 8020 and you'll save more. Get the Misumi stands and the steel tube and have that machined. There's guys selling aluminum t-slot table on e-bay. A prefab Z carriage and 1.5HP water-cooled spindle. All you need then are ballscrews, bearing blocks, couplings, ballnut carriers, steppers, stepper motor mounts, drives, and controller software.

[richardmg9]

Thank you for the solid advice. I like the Mic-6 stuff. Mineral wool seems like a winner for sound dampening.

[richardmg9]

You are right about the linked machine. I thought I knew a decent amount about machines and what I wanted. That was before I spend ~200 hours reading here. About the club involvement... I recently got my electrical engineer friend to work on a driver for me.

[JSantars]

He's milling steel pretty well with it though.

[richardmg9]

Ok, so I have changed my build quite a lot. I thought I had my build planned out and was just posting the concept here for troubleshooting. I have learned a lot from this forum, more than I expected, and my build is reflecting this.

I think I want to pass on steel fabrication in favor of epoxy granite. I am also now thinking a moving gantry would better suit me. I want to minimize machining requirements, and to that end I have come up with the following plan:
A flat rectangular base with both working area and gantry support area included. I plan to make the base from E/G and finish the top with self leveling epoxy. Next, I plan to make 2 gantry supports cast upside down with the feet finished in self leveling epoxy. The next step is to place the gantry supports upright and side by side on the center of the base and connect the tops with a temporary channel. This should allow me to cast self leveling epoxy on both gantry supports for a level and plumb rail support. I do plan to trim meniscus off the feet first. I plan to attach gantry supports to the bed with cast bolt anchors. I want to avoid adjustments on this connection.

For the gantry, I do not have a solid plan, but I am currently thinking a self leveling epoxy for the rail mounts again. I want the adjustments to square my gantry with my bed to be between the part connecting the gantry trucks and the gantry. ie: a single part connects 2 gantry support trucks per side, and the gantry is mounted to that part with adjustable bolts.

To summarize, I am proposing that I:
1) build a flat base with a self leveling surface
2) build gantry supports with self leveling feet
3) cast self leveling rail mounts on the gantry supports
4) use self leveling face on gantry for rail mounts
5) square gantry face to machine bed with bolts mounting gantry to it's guide trucks

That is the current general plan for the machine base. I have a few other thoughts to share:

I want both bolt holes and vacuum grid on my bed. I am thinking for part boltdown holes that I can cast them in the E/G a few mm above the surface, them cast the self leveling surface a few mm above them with some form of plug to provide an end product that is recessed. This is also the plan for the linear rail mounts. I would love some advice on how to get these sockets placed as square and vertically as possible. If a single bar of steel/aluminum with a machined face and milled threads is best, so be it, but I would prefer some method which allows me to cast "plugs".

For the vacuum bed, I was thinking that I could machine a grid into the self leveling epoxy as it is pure epoxy without the sand/gravel. Is a pure epoxy working surface a good idea? is it tough enough? can I cut a vacuum grid in it?

Speaking of general "tough enough", I want to machine any hardwood with no issue. I want aluminum machining to be as smooth as pine. Harder materials like steel would be a lovely plus as a fringe case with possible lubrication. I realize that I could make a reasonable aluminum/wood machine with 80/20, but I want to make the most capable machine I can. I plan(at the moment) to use linear servos eventually on this machine. My goal is accuracy over speed. Resolution is the parameter I wish to optimize.

A few questions I have failed to find answers to:
1) How critical is hole alignment for my linear rails? It seems as though the practice is to bolt one rail down tightly and slave the next to it with adjustments done by tapping and tightenting/losening. This leads me to believe that I do not have to get my rail mounting threads super accurate? How accurate should the mounting holes be? (I have read the THK mounting PFD) In other words: can I just measure as best as I can, then use a drill press on my MDF mould to mount my rail mounting sockets?
2) Are self leveling epoxies good enough to use the build method I have described? I think the proposed method with produce a square machine, but I am ignorant.

On to casting... I am trying to find methods that will allow for strong and dense castings. I see that aggregate grading and vibration are key issues. I do not want to buy vibrators. Spending 100$+ on tools just to make my castings is not acceptable. I have seen builds using manual tamping to get reasonable density. I like this method as it is free. That being said, I would like to make the best product I can. I had the idea of using a crappy-but-powerful second hand subwoofer. Is this a potential good idea?

Any thoughts, comments, criticisms, critiques, ect. are more than welcome. Tell me why my ideas are crap! Thank you for reading, and thank you for your comments.

[louieatienza]

Ok, so I have changed my build quite a lot. I thought I had my build planned out and was just posting the concept here for troubleshooting. I have learned a lot from this forum, more than I expected, and my build is reflecting this.

I think I want to pass on steel fabrication in favor of epoxy granite. I am also now thinking a moving gantry would better suit me. I want to minimize machining requirements, and to that end I have come up with the following plan:
A flat rectangular base with both working area and gantry support area included. I plan to make the base from E/G and finish the top with self leveling epoxy. Next, I plan to make 2 gantry supports cast upside down with the feet finished in self leveling epoxy. The next step is to place the gantry supports upright and side by side on the center of the base and connect the tops with a temporary channel. This should allow me to cast self leveling epoxy on both gantry supports for a level and plumb rail support. I do plan to trim meniscus off the feet first. I plan to attach gantry supports to the bed with cast bolt anchors. I want to avoid adjustments on this connection.

For the gantry, I do not have a solid plan, but I am currently thinking a self leveling epoxy for the rail mounts again. I want the adjustments to square my gantry with my bed to be between the part connecting the gantry trucks and the gantry. ie: a single part connects 2 gantry support trucks per side, and the gantry is mounted to that part with adjustable bolts.

To summarize, I am proposing that I:
1) build a flat base with a self leveling surface
2) build gantry supports with self leveling feet
3) cast self leveling rail mounts on the gantry supports
4) use self leveling face on gantry for rail mounts
5) square gantry face to machine bed with bolts mounting gantry to it's guide trucks

That is the current general plan for the machine base. I have a few other thoughts to share:

I want both bolt holes and vacuum grid on my bed. I am thinking for part boltdown holes that I can cast them in the E/G a few mm above the surface, them cast the self leveling surface a few mm above them with some form of plug to provide an end product that is recessed. This is also the plan for the linear rail mounts. I would love some advice on how to get these sockets placed as square and vertically as possible. If a single bar of steel/aluminum with a machined face and milled threads is best, so be it, but I would prefer some method which allows me to cast "plugs".

For the vacuum bed, I was thinking that I could machine a grid into the self leveling epoxy as it is pure epoxy without the sand/gravel. Is a pure epoxy working surface a good idea? is it tough enough? can I cut a vacuum grid in it?

Speaking of general "tough enough", I want to machine any hardwood with no issue. I want aluminum machining to be as smooth as pine. Harder materials like steel would be a lovely plus as a fringe case with possible lubrication. I realize that I could make a reasonable aluminum/wood machine with 80/20, but I want to make the most capable machine I can. I plan(at the moment) to use linear servos eventually on this machine. My goal is accuracy over speed. Resolution is the parameter I wish to optimize.

A few questions I have failed to find answers to:
1) How critical is hole alignment for my linear rails? It seems as though the practice is to bolt one rail down tightly and slave the next to it with adjustments done by tapping and tightenting/losening. This leads me to believe that I do not have to get my rail mounting threads super accurate? How accurate should the mounting holes be? (I have read the THK mounting PFD) In other words: can I just measure as best as I can, then use a drill press on my MDF mould to mount my rail mounting sockets?
2) Are self leveling epoxies good enough to use the build method I have described? I think the proposed method with produce a square machine, but I am ignorant.

On to casting... I am trying to find methods that will allow for strong and dense castings. I see that aggregate grading and vibration are key issues. I do not want to buy vibrators. Spending 100$+ on tools just to make my castings is not acceptable. I have seen builds using manual tamping to get reasonable density. I like this method as it is free. That being said, I would like to make the best product I can. I had the idea of using a crappy-but-powerful second hand subwoofer. Is this a potential good idea?

Any thoughts, comments, criticisms, critiques, ect. are more than welcome. Tell me why my ideas are crap! Thank you for reading, and thank you for your comments.

To get one thing out of the way, I don't think you'll see any benefit with linear servos on a machine this size. They're stupid expensive, even used, they're inefficient compared to rack and pinion or ballscrews, Likely need liquid cooling.

I'd forget about bolt holes and vacuum grid. If you have bolt holes, you can just make vacuum fixtures that bolt into them.

Accuracy... needs to be defined.

If you want the best casting possible you should use a mechanical vibrator. With all the money you're going to spend just on epoxy resin, why cheap out? What good are linear servos if you have a meh casting?

I'd probably re-evaluate your needs - specifically what it is exactly you want to do, so that you're better guided. Really, it's probably cheaper to buy a old 90s VMC, swap out the parts and retrofit a high speed spindle, than it is to attempt what you plan.

[richardmg9]

I have been misunderstood on a few things here. The linear servos are a possible DIY future project. I will NOT be buying linear servos. The vacuum grid is planned to be machined into the leveling epoxy surface, the threaded bolt holes are cast into the base as a second option for holddowns. The two are not related. ie: not planning to bolt down some sort of grid.

As far as rail mount accuracy goes, I define acceptable accuracy as accurate enough to avoid racking.

As far as my needs go, I don't have any. This is a machine I want to build because I want to build a machine.

[ger21]

1) How critical is hole alignment for my linear rails?

The 15mm Hiwins I have have 4.5mm holes, for 4mm bolts. So if the holes are more than 0.25mm off from the correct location, you won't be able to get the rails straight

[skubajon]

Just quick reference for the 15mm rails and 4mm bolts. If your going into extrusion buy the good t nuts. You can't get more than about 5ft lbs on them with the sheet metal nuts. Learned that lesson the hard way

Sent from my SM-N910V using Tapatalk

[louieatienza]

I have been misunderstood on a few things here. The linear servos are a possible DIY future project. I will NOT be buying linear servos. The vacuum grid is planned to be machined into the leveling epoxy surface, the threaded bolt holes are cast into the base as a second option for holddowns. The two are not related. ie: not planning to bolt down some sort of grid.

As far as rail mount accuracy goes, I define acceptable accuracy as accurate enough to avoid racking.

As far as my needs go, I don't have any. This is a machine I want to build because I want to build a machine.

Then I suppose whatever you decide to do will be all right.

[burdickjp]

As far as my needs go, I don't have any. This is a machine I want to build because I want to build a machine.

Anything worth doing is worth doing right.

[louieatienza]

Anything worth doing is worth doing right.

True, though what's "right" for wood may not necessarily be right for metals....

[G59]

It seems as though the practice is to bolt one rail down tightly and slave the next to it with adjustments done by tapping and tightenting/losening.

Hmmm.................think about that just for a second.
Knowing that all linear rails(even the master rail), can have a slight curvature in them, do you see a potential for problems?
So wha'ts the recommended and proper way to mount them?

Master rail on first and on a machined flat surface with an edge cut in it for reference. You butt the master rail to the 'trued' reference edge and then you bolt it down on to the machined surface which can be embedded in your EG or bloted to your steel gantry. That little reference edge is critical or both yours rails will follow it's innaccuracies.
Only then you can use the method you described to adjust the second rail to the master. This is the only way to ensure true parallelism between the two.

example.............

Something else you can do is buy flat ground stock of at least 2"x 0.125" by whatever length your rails are, and clamp it on your levelled surface, where you want your rail to go on and use its straight edge for reference until you have bolted the master rail down. This gives you an edge to butt against so the rail stays straight. Afterwords you can remove the ground stock.