CNC mill kits?

I am exploring the possibility of manufacturing some CNC mill kits. I have built a couple larger machines that are working out great and I feel like a scaled down version sold as a kit would do well.

I am trying to ascertain whether the 3d printer community has any interest in a milling machine. What would be important to you in terms of features, size and cost?

For what it's worth I have talked to some parts fabrication shops and it looks like I could sell variations of my design for prices ranging from about 330 to 700 dollars for a complete kit depending on features and upgrades. Do these prices look like they would be well received by the market? I have seen very few DIY kits in this price range. What CNC mill kits are being sold at this price point?

Thanks for the input. I am just looking for input at this point, not trying to sell anyone anything. After having had both for a while, I just feel that owning a mill and a 3d printer is an awesome maker combination so I want to make the mill more accessible to people.

To mill what? Heavy duty metal?

Shapeoko 2/3 and XCarve are inexpensive CNC routers / engravers

[www.shapeoko.com]

[www.inventables.com]

It would not be for milling heavy metal, but aluminum would be possible. It would mostly be for wood and plastics. Perhaps some very light cuts of mild steel would be attainable with the upgraded version that I have plans for.

I am aware of the X carve and Shapeoko. I am actually on the SO forum. I don't own either, but both seem to be very nice machines for the money. The kits that I envision would be prices considerably under those machines while not being as refined.

My taget customer is the person who spent 300 to 500 dollars on a reprap style printer and now wants to try his or her hand at subtractive machining without spending double or more what the printer cost. I'd obviously want to have someone who has never 3d printer as a customer too, but you see what I am getting at.... No frills entry level machine that is upgradeable. With all of that said, I am trying to get a grasp on what this type of customer would want and appreciate in a mill and just what else is on the market in price ranges that are roughly half what the X carve and Shapeoko machines sell for.

I would love a small, inexpensive mill that can run aluminum. Brass and steel would be awesome but aren't necessary.

The thing that's scared me away from the mills I've seen so far is that they put more effort into volume and speed, and less into precision and rigidity. If I could buy a $300 machine that could mill aluminum with respectable precision, at any speed and with more than a few cubic inches of working volume, I'd buy it in a heartbeat.

I have been looking for just that over the past few months. I decided I was just going to have to design and build one. I am looking at needing to mill aluminum, does not need to be a big build area(15x15 cm)would be good. precision would be my biggest request, it does not have to be "nuts on" but I can make bad cuts with my drill press.

Thanks for the reply. An aluminum capable machine at a 300 price point is in theory doable but much more easily achieved with a machine that ends of priced at 450 to 500 plus. The issue is mainly cutting tool quality here. Aluminum can be cut with a 35 dollar rotary tool but the depth of cut has to be fairly shallow to keep from bogging or (heaven forbid) seizing the tool. What's more, these inexpensive rotary tools are often small in size and designed with a "in the palm of your hand comfortably" ergonomic approach. This can make designing and machining a well fitting tool mount more difficult. It can be done of course, but you don't get the rigidly mounted tool that you have with a higher end tool (more on that in a sec). There are rotary tools that start to get powerful when you spend a little more money, but they are often a fixed speed which can make you have to tackle certain materials at less than ideal spindle speeds. From here, you can spend a little more money still and get an actual variable speed router that's made to fit in a router base. These usually have a good speed range (with some even having closed loop speed control), plenty of power (often approaching 1 horsepower or more), and a good cylindrical shape once the base is removed that provides a good mounting surface. Once you are at this point, you have 90 to 150 dollars in a cutting tool. These are very nice for what they are, but I wouldn't be able to afford to offer them in a complete package for anything near 300. Maybe at 450 to 500 but that's still with "no frills" elsewhere on the machine. A machine in the 300 to 350 dollar range would necessarily have a small 100 to 200 watt tool that can do the job so long as the user tempers his or her expectations and doesn't get too aggressive.

And I think you have mentioned a good point on rigidity versus speed. Its fun to brag about a 20000mm per minute feed rate and how hard of acceleration you can pull, but it's mostly a waste in terms of what is practical on a desktop hobby level machine. Its nice to have, but often just not needed. Somewhat on this note, I will be designing around NEMA 17s with perhaps 23s offered as an upgrade. The pulling force that say a 60 oz in motor can generate with the moment arm given by these small gt2 pulleys is usually pretty sufficient to pull a tool through material at a good speed and reasonable acceleration.

I would be putting the priority on rigidity. The higher priced machine that I plan on would have greater rigidity than the lower priced one, but even my base model would be built with rigidity as a goal. I have some materials and geometries in mind that will provide cost effective rigidity.

Quote
cat.farmer
I have been looking for just that over the past few months. I decided I was just going to have to design and build one. I am looking at needing to mill aluminum, does not need to be a big build area(15x15 cm)would be good. precision would be my biggest request, it does not have to be "nuts on" but I can make bad cuts with my drill press.

Thanks for adding to the discussion. Can you quantify what level of precision you are looking for? I honestly don't have aspirations of producing anything capable of producing a couple thousandths precision because affordability is much more of a priority for me than world class precision. The machine will be belt driven in the interest of affordability. With that said, a belt driven machine can do good work. I've milled aluminum for some parts I needed and gotten good useable results.

Quote
DRobs86

Thanks for adding to the discussion. Can you quantify what level of precision you are looking for? I honestly don't have aspirations of producing anything capable of producing a couple thousandths precision because affordability is much more of a priority for me than world class precision. The machine will be belt driven in the interest of affordability. With that said, a belt driven machine can do good work. I've milled aluminum for some parts I needed and gotten good useable results.

It's hard for me to say what level I am looking for. I'm not a machinist, just a hobbyist. but I would be happy with a inexpensive device that cuts a decent straight line.
As a side note, I would probably turn away from a belt driven device, the belts would stretch quickly under force and add to backlash. I know you are looking at price, but I would say lead screws are a must, for rigidity. I would think with the cheap lead screws out there you would only add a few dollars to the build cost.

I also agree with you comments on the motor, as I looked a designing my own, that was indeed the most expensive part, in fact over half the cost of building it was the motor. In my case, I was going to go to my local used tool shop and pick up something. not an option for you. I was thinking maybe of a VSR drill, and building a spindle to attach it to, that can handle the lateral forces.

I'm also not a machinist, or even an engineer for that matter. I just love to build things. I have done some calculations on GT2 belt in the past. As long as thing stay reasonable, the belt stretch is pretty modest. 9mm GT2 belt is available and it is about 60% more resistant to stretch than 6mm GT2. Even 6mm GT2 belt can be suitable for many applications. The thing is that he cutting force of a hobby level machine is fairly low. It's low enough that stretch during the snatch of hard acceleration is really what causes problems in my opinion. Ignoring friction and a few other complication factors for a moment, accelerations of 666mm/s^2 are enough to get you to 3000mm/min in 75 milliseconds and requires force that amounts to about 0.003" elongation of 9mm belt (0.005" on a 6mm belt) on a 15 lb gantry and a 15 inch span. The problem comes, in my opinion, when someone tries to pull silly accelerations without understanding the cost in terms of belt elongation and the very limited benefit in terms of saved machining time. If one were to take the same scenario and accelerate instead at 3000mm/s^2, they would get to 3000mm/min in only about 16.5ms, but the belt elongation would be about 0.013" and there is actually more elongation than the distance traveled to get to max speed. On small direction changes this ends up with a part that is probably not much worse to your eyeball, but definitely less precise. (Note: These calculations are all just really quick, back of the envelope calcs so don't hold me to them!)

I agree that lead screws are preferable and I would want them on a high end machine (well, a ball screw if I was already spending a ton of money), but the lower priced lead screw and nut combos on the market are usually not very good (some may be, though). I have two of them, one of which is installed on a machine, and the backlash on my screw driven axis is almost certainly greater than the backlash on my belt driven axes. Better screws and nuts are out there, but you are talking about changing the cost of your power transmission mechanism from a $5 or $10 per axis cost to a $25 or $35 per axis cost on even a small machine. Part of this cost is the need for good couplers as well. Those "2 dollars free shipping" from China flex couplers are really not good for a milling application in my opinion. I'm using about a $10 coupler on the screw driven axis on the machine I use.

I am by no means an expert so I'm open to the possibility that I've drawn some incorrect theoretical conclusions on belt elongation. From a practical standpoint though, I can tell you that I've made and sold over 100 wood art items over the last few months with a belt driven machine using engagement depths usually about 6.5mm for profile ops at about 1200mm/min feedrate and it does just fine. I don't sell alluminum parts, but I have made them. I actually have made aluminum parts for another machine build and slow and shallow produces good results.

Anyway, I'm not trying to sell you on a belt driven machine, I just feel like they aren't as bad as they are made out to be. I do agree that a machine driven with high quality screws and anti-backlash nuts is almost always better.

Edited 1 time(s). Last edit at 12/19/2015 02:30PM by DRobs86.

Properly designed belt driven machines can outperform other types. See this thread.

Or just use two belts in parallel, or a different kind of belt that's wider. Belts are cheap.

If it's gonna cut metal, I don't think belts are the way to go. They're cheap, but will need replacing regularly and will really mess with accuracy. Just my opinion, but if it can't be accurate to within a couple of thousandths, then it's not worth it.

Quote
seabass07
If it's gonna cut metal, I don't think belts are the way to go. They're cheap, but will need replacing regularly and will really mess with accuracy. Just my opinion, but if it can't be accurate to within a couple of thousandths, then it's not worth it.

If you want to hold 0.002 inch throughout a part, consistently and with any notion of speed you will need to be in the market for a 300,000 dollar laser cutter instead of a 300 dollar hobby cnc kit. All joking aside, 2 thou of an inch is 0.05mm. Most professional machine shops are going to tack onto your quote if you ask them to hold two thousandth because, depending on the complexity of the part, it's going to be tough to get within spec. Sure it's possible to hold such a tight tolerance milling metal, but you are going to have to be meticulous and have a heavy and rigid machine. Even if you have a great drive screw with immeasurably small backlash, you still have gantry deflection and end mill deflection to deal with. My point is that there is a huge difference in achieving a 50 micron layer height on a print and holding every feature to 50 microns when a tool has to year away hard material.

This wasn't meant to sound like a lecture. If you are making parts that need 2 thousandths tolerance more power to you. I just believe that most do not.

I never said throughout a part, but if I'm making a part that has a feature that I want to fit snugly into another or be in a precise place, I'd like that to be possible. It's not really that difficult with care. $300k will buy you way more than you need to hit a .002" tolerance. Keep in mind, we're talking about aluminum and plastics. If there is significant deflection while taking a few thou off of some aluminum, then it should be more rigid or not be advertised as good for cutting metal. Just my opinion. I'm just saying to put emphasis on rigidity if it's going to be advertised as good for cutting metal.

... I've modified several CNC-mills and other machines (XY(Z)-plotters, waterjet-cutters, ) into laser-engravers/cutters - the prices for the machines were from 3k€ to 15k€ for aluminium framed ones with 0.0125 to 0.005mm resolution or 40k€ to 100k€ for granite framed and 0.002mm resolution. The 'repeatability' of this machines is stated around 2x to 5x the resolution value.

But I too have a prototype of a 'cheapish' Isel-Miniflat, which were sold for 2k€ with stepper motors, while my prototype has servos and 500-er encoders (2000 lpr) and 16x5mm-spindles, what's a resolution of 0.0025 microns and up to 150mm/s (250mm/s max with higher servo-voltage) speed with the actual controller ... but I'm engraving with 80mm/s max to avoid mechanical oszillations/wobbling in the long Z-axis (610mm length with motor, 210mm travel).

And actually I'm developing around a small, pretty rigid CNC-machine with belts in XY and spindle in Z for different applications around paste-jetting, 3D-printing and laser-engraving -- here a (German) thread with some videos of it:
[forums.reprap.org]

The belt-driven system is much faster (up to 400mm/s) and theroretically in the accuracy range of 10 microns too, but for the elasticity of the belts the resulting accuracy is depending on the speed too -- the faster, the more 'wobbling' will occure.

I've measured up to 0.3mm 'wobbling' after direction changes with the belts at max. 200mm/s, while this was something like 0.03 mm with the spindle-driven Miniflat-prototype at max. 100mm/s -- the start/stop-speeds should be similar on both machines, so this 10x bigger 'wobble'-value for the belts should be direct comparable ...

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

Quote
seabass07
I never said throughout a part, but if I'm making a part that has a feature that I want to fit snugly into another or be in a precise place, I'd like that to be possible. It's not really that difficult with care. $300k will buy you way more than you need to hit a .002" tolerance. Keep in mind, we're talking about aluminum and plastics. If there is significant deflection while taking a few thou off of some aluminum, then it should be more rigid or not be advertised as good for cutting metal. Just my opinion. I'm just saying to put emphasis on rigidity if it's going to be advertised as good for cutting metal.

Fair enough... And the 300k laser machine was admittedly a silly half joking exaggeration.

To be completely clear, I would not claim great aluminum cutting performance on a $300 machine. A machine that cheap pretty much necessarily comes with an inexpensive rotary tool that probably has a thou or so of runout anyway. What I will say is that if you are willing to take very very small stepdowns (say 0.1 mm) then it would be possible to make a part.

VDX that is some impressive work. I translated that page from German to English so I'm sure I missed some of the technical details but I am impressed.

And it looks like you and I agree on the belt drive and have similar findings. Small width belts just should not be run too fast while the tool is engaged.

... this thread has two sides, so check the last side too for the newest posts/videos

I have yet to test the belt-driven machine for laser-engraving - here I can see direct the difference to the Miniflat with the 5mm-spindles.

Evantually we will design a different paste-jetter setup too with spindles instead of the belts for a special customer project, but this can take a while.

In the meantime I have to (hard) test it with soldering paste and different fast curing pastes for front panel manufacturing and some sorts of 3D-printing ...

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]