New Build

[Richardmg]

Hello all!

I am looking for some rough guidance on my project. Any advice is appreciated, but specifically looking for component direction.

First off, this will be a 3 axis wood working machine with occasional aluminum.

My goals for this machine are...

1. fairly small working volume roughly 12"x18"x3"
2. As high a resolution as is reasonable to consider for wood.
3. excellent repeat-ability and rigidity.

I intend to fabricate the machine from square steel tubing and cast epoxy for rail beds. Probably a fixed gantry. Also planning on Hiwin rails and ball screws.

With this in mind, should I even consider the higher precision grades for the rails/ballscrews, or is "C" more than enough? What size rails does the reader recommend? I am thinking for such a small machine, 15mm is fine but I am all ears. As far as controllers and steppers go, I have no clue how much I need,other than the fact that I cannot afford servos or linear motors. As far as spindle goes, I know I want better precision than can be afforded by routers, but that's about it.

Thanks for reading!

Similar Threads:

• Build Thread- Router home build Australia. Solid steel build
• Need some direction to build a small machine to allow me to build a larger one
• Mint's Build Aluminum/Steel Build thread.
• Newbie - To build or not to build Router/Plasma Table

[wizard]

Hello all!

I am looking for some rough guidance on my project. Any advice is appreciated, but specifically looking for component direction.

First off, this will be a 3 axis wood working machine with occasional aluminum.

My goals for this machine are...

1. fairly small working volume roughly 12"x18"x3"
2. As high a resolution as is reasonable to consider for wood.
3. excellent repeat-ability and rigidity.

I intend to fabricate the machine from square steel tubing and cast epoxy for rail beds. Probably a fixed gantry. Also planning on Hiwin rails and ball screws.

The size works to your advantage. The only thing I'd worry about in this case is Z axis hight, you want to be absolutely certain that 3" is enough. If you add a spoil board and fixturing you can quickly have a very tight fit between the work piece and gantry. It might be good enough for what you want to do but you need to verify that in your mind.

On a small machine, especially if the gantry span is narrow, you can get away with small tubing cross section wise. Even so your gantry beam will likely need to be of a larger cross section than other frame members to separate the Y axis linear bearings enough to provide stiff support for the X axis.

With this in mind, should I even consider the higher precision grades for the rails/ballscrews, or is "C" more than enough? What size rails does the reader recommend?

I would think that C would be good enough for what you have described so far. Your budget is a factor here, if running open loop higher precision lead screws do help for precision over a given distance. This is important to understand but you need to have an idea of what your expectations are and what your budget will allow. Frankly if servos are too rich for you right now Im not sure pursuing high precision components is an ideal approach.

Do realize that ball screws are subject to wear and eventual replacement. They can also be replaced at anytime if you decide you would benefit for higher precision.

I am thinking for such a small machine, 15mm is fine but I am all ears. As far as controllers and steppers go, I have no clue how much I need,other than the fact that I cannot afford servos or linear motors. As far as spindle goes, I know I want better precision than can be afforded by routers, but that's about it.

Thanks for reading!

Nail down your machine design, size and construction material first, then work on the controls issues.

Highlight to us what you have access to machine shop wise if anything. While epoxy leveling works there are still many things that need to be done on such a machine that would benefit from being able to machine parts. The nature of the machines design impacts the need for machine work.

[Richardmg]

Thanks for the reply, wizard.

Frankly if servos are too rich for you right now Im not sure pursuing high precision components is an ideal approach.
.

Good point, I agree. I also agree that a bit more z travel is prudent.

As far as machine shop access goes, I do believe I have some limited access through the University I am a student at.

I'm working on a cad file of the rig, I'll post as soon as I have something presentable.

Oh, one other thought: how good is 80/20 stuff? Can I expect similar precision as I might get with cast epoxy?

[johncarpenter]

Richard,
I have a ISEL Techno DaVinci CNC that is 10 in by 12 in by 3.5 in. Is is a great rigid small machine but the limited Z travel is a hassle. And a fourth axis is impossible. So I am building another moving table CNC with a fixed gantry that can be lowered or raised (and then aligned). Do your self a favor and figure out a way to get more Z without losing too much rigidity.

[NIC 77]

First off, this will be a 3 axis wood working machine with occasional aluminum.

My goals for this machine are...

1. fairly small working volume roughly 12"x18"x3"

To cut a 3" piece with a 3" inch bit you need 4" of movement. The extra inch is for clearance. Now take that 3" bit out and put a 0.5" long bit in there and try to reach the same table. Now that's 6.5" of movement to have the ability to cut both half inch thick and 3" thick pieces all the way to the table using different bits. So think 7" of Z movement with 4" minimum of clearance between the bottom of the fixed part of your Z or gantry, whichever is lowest.

I intend to fabricate the machine from square steel tubing and cast epoxy for rail beds.

look through my posts in 1Jumper10's 4x6 build. I give my thoughts on cast epoxy there.

Do you have a welder? A combination of welded and bolted pieces is probably easiest.

What size tubing? There is a scrap yard near me when I can get pieces of 6" x 6" x 0.25" or 8" x 8" x 0.25" square tubing at 30 cents a pound. It makes it easier than buying full lengths. If you have a similar scrap yard near you, think about big tubes and simple instead of small tubes intricately welded or bolted together and complicated when you are designing.

Probably a fixed gantry.

Yes. For a machine with such small movements, that is a good choice.

Also planning on Hiwin rails and ball screws.

Yes to the namebrand rails and bearings. I don't know what Hiwin ball screws cost. Yes if the price is comparable to other options such as the rolled Chinese ballscrews that are everywhere. No if they are way more expensive.

For my build I am using Bosch Rexroth linear rails, so good quality and brand name, SYK motor mounts, also expensive, and Bun brand bearings. The Bun brand bearings are made in China but are good quality. I hear some of the Chinese bearing end supports can be hit and miss but I am very happy with what I got from Bun. The ones I have are C5.

I will be using rolled Chinese ballscrews with double nuts, and my build is far larger and more expensive than yours will be and the majority of my cutting will be aluminum. I think you are too worried about this. It is better to spend your money on good other things, like rails, and get adequate bearings and mounts. And if you have the equipment and knowledge to machine your own bearing supports and motor mounts, then still don't buy expensive ballscrews.

I looked on EBay for lightly used brand name ballscrews. No luck. I did notice that often smaller lengths than what I required were there, and sometimes they included end supports and even mounts and those things can be pricey. Perhaps you can get lucky. I wouldn't buy a used one without the end supports included because you don't know what the end machining is.

With this in mind, should I even consider the higher precision grades for the rails/ballscrews, or is "C" more than enough?

No, don't consider it. Also, keep in mind that the accuracy rating includes variations in movement / length, and not just local variations, and you could adjust out most of that through your software if you were able to measure how far you were off from ideal, which you probably can't.

What size rails does the reader recommend? I am thinking for such a small machine, 15mm is fine but I am all ears.

20mm for a machine your size.

As far as controllers and steppers go, I have no clue how much I need,other than the fact that I cannot afford servos or linear motors.

Don't buy expensive ballscrews and you will have more money. Steppers are good, but put some thought into what drivers and board you use. Many of the control software that is available requires you to buy the main board from them.

As far as spindle goes, I know I want better precision than can be afforded by routers, but that's about it.

This is where you will loose your precision. The lower priced spindles, and by that I mean 1 or 2kw spindles from china, (they are probably all made in China those ones, even if sold by an American company) are not more precise than a router, at least I have seen nothing to convince me of that fact. They can perform much better overall, and if you try to cut aluminum with a router, even using light cuts, you will destroy that router prematurely. I know, I've wrecked a couple of them. I've seen ads boasting Chinese spindles with less than one thousandth of an inch runout, but then when I look at the video of them testing it, there is just over .001 runout and they are measuring less than 5mm away from where the tool enters the spindle. So for a one inch bit that's .005" runout at the tip. At two inches double that. That's about as bad as the crappy Hitachi router I used to use. Perhaps someone with some hands on experience with them can give you some better info, and there may be a way to fix that, but every testing video I've seen is always the same thing and it's a joke.

So yeah, don't worry about expensive or ground ballscrews.

Thanks for reading!

[ger21]

15mm rails are more than adequate, but 20mm rails are much easier to work with, because 15mm rails use small 4mm screws.

Automation Overstock has some precision ground screws at good proces, if you can make them work for you.
discounted automation products: NSK Precision Ground & Rolled Ballscrews, NSK Ball Screws: 6mm ⌀ thru 15mm ⌀

Otherwise, I'd just use 1610 or 2010 chinese ballscrews.

For motors, I'd go with some 400 oz Nema 34's with Leadshine AM882's or similar. Low inductance motors in the 5A to 6A range, which will be capable of high rpm's.
AM882's are the best bang for the buck in stepper drives. But you get no support from Leadshine with them.

Spindle precision is almost solely dependent on runout. Unless you have a bad collet, or poorly machined taper in your router, you shouldn't see much difference between a spindle and router. But a router is quieter, has bigger bearings, and is more powerful. And they are cheap enough that a router doesn't make a lot of sense unless you are on a really tight budget.
If you're using linear bearings, you shouldn't be using a router, imo.

[louieatienza]

I think since you're building a machine for wood, not some aerospace parts, you'll be fine with standard grade. I'd use 5mm pitch since your travels are short; this will allow you to use smaller motors as well as gain resolution. I would resize your working volume that you could buy off-the-shelf parts in pre-cut lengths, which should save you some cash. 15mm is normally more than plenty stiff, though I'd have my doubts about the smaller rail against an epoxy bed; not sure if there's enough surface area under there. With the short axes you have, you may find your tube to be flat enough as-is. Also with a build of this size steppers make more sense; they're plenty powerful enough. What you get with the spindles are better and more bearings (2 at the nose) which will handle radial forces better, since these machines can move that router way faster and with more strength than a human with a handheld router. Runout is not too much a problem from my experience.