Shapeoko Machine capabilities and limitations

Talk about all things CNC
chamnit
Posts: 376
Joined: Tue Aug 12, 2014 2:16 pm
Location: Albuquerque NM, USA
Contact:

Re: Shapeoko Machine capabilities and limitations

Post by chamnit » Wed Dec 16, 2015 11:16 pm

This is a comment on the 25lb max load on machine stiffness calculations.

There was a YouTube video, posted by Edward, showing the SO3 cutting aluminum at very high rate, using high-speed machining toolpaths. It was cutting faster than most bench top milling machines on the market. This is probably on the upper end of how hard you'd push the machine. That said, if you plug in the cut parameters into FSWizard, you get about a cutting load about 7 lbs on the machine. Not anywhere near 25 lbs.

There are a ton of additional factors that need to be accounted for, such as continuous duty limits of the spindle, tool deflection so the tool doesn't break, ideal feeds and speeds to not wear out the tool prematurely or burn/melt/shred/bend/break your job material (remember it has to take that same equal and opposite load), etc, etc. Probably the most important factor is the common usage of roughing and finishing passes when making a part. Machinists typically do a high-speed, high-removal rate roughing pass up to 0.010"-0.050" of the final dimension, and then do a final finishing pass that lightly loads the spindle and tool to keep the final part accurate.

No machine or structure is infinitely stiff. A good machine is a balance of the usage intent, smart and wise design trades, and end-user cost. At the lower end of the cost spectrum in the CNC market, the SO3 has a very nice balance of everything for its usage intent. The same can be said for its cousin the X-Carve.

Personally, I think a 9mm GT2 belt will help some users that tend to push the machine hard, but you are still limited by the force of the stepper motors and its ability to maintain position under load. The motor's stiffness contribution is probably much lower than the belt, especially in between full steps. I'm not really sure how much benefit there actually is with a 9mm belt versus a general feeling because it cuts slightly different.

WillAdams
Posts: 8488
Joined: Mon Apr 09, 2012 6:11 pm
Location: Pennsylvania --- south of the Turnpike, East of US-15
Contact:

Re: Shapeoko Machine capabilities and limitations

Post by WillAdams » Wed Dec 16, 2015 11:42 pm

I believe that a big advantage for the 9mm belt upgrade is that it reduces the need to over tension the belts and reduces the possibility of skipping a tooth on the pulleys.
Shapeoko 3XL #0006 w/Makita RT0701 Router w/0.125″ and ¼″ Elaire precision collets
Nomad 883 Pro #596 (bamboo)

sgth0mas
Posts: 101
Joined: Sat Dec 12, 2015 5:57 pm

Re: Shapeoko Machine capabilities and limitations

Post by sgth0mas » Tue Jan 12, 2016 2:22 am

Gadgetman! wrote:Max load of 25lbs sound a bit much, really. (I don't like pushing stuff to the max.)
What about using 12mm wide belts?
and limiting yourself to say, 15lbs?

Rack and Pinion has some issues with wear, of course, but I assume that you can handle that without too much trouble.

As for the 'added investments', not all need those.
The Homing probe and enclosure? Sure, homing probe and all that can be nice, particularly when doing jobs that require toolchanges, or when cutting multiple parts out of blanks mounted in a fixture. It's not needed for simple work, though. Enclosure kits are for those who can't make one themselves. We're fabricators here...
Dust shoes? sounds like a job for a SO...
Clamps? We don't need no steenkin clamps! ;-) tie downs can be made out of scraps of plywood and bolts.
Aluminium bed? That's some right fancy stuff you got there. ;-) I'd love to have one myself, but then again, my SO won't fit on it.
It's probably not all that much more stable than my torsion box, though.
(or heavier... The bl**dy box weighs in at 40Kg, at least... )

You don't NEED that Dewalt 611 either. No, really. It's just the recommended spindle for this machine at this point of time.
If you have another spindle you want to use, such as a DC spindle, and you can mount it onto the machine, go ahead.
(DC spindles tends to be quieter. You may not even need that enclosure then.)

The SO series of machines are wonderfully customizable, and can be adapted to suit lots of different tasks.

Whatever you decide, keep us posted.
Update:

I ended up getting an Axiom AutoRoute 1 (AR1). Of all the machines I viewed and researched this was the one that I strongly believe will hold the tightest tolerances in the sub $2500 price range without going to a converted mini mill (which is limited power and very limited volume). The spindle carriage on the sharks had a substantial amount of play for a $4K machine, I believe a few others mentioned this concern. The AR1 also included homing switches, an industrial controller and a hand pendent which I like a lot more than I expected. It's an exceptionally rigid machine as the base is welded steel.

However, there are some downsides to this machine. I paid 2.5 times the cost of a Shapeoko 3 for a machine and software with a decently smaller work area (12 x 12 x 3). I will also not be able to upgrade the unit for a larger build volume. I guess that should detail the weight I ended up putting on the ability to hold tighter tolerances. I eventually decided that I wanted to target closer to +/- .005" max so that played largely into the decision as well.

I still want a Shapeoko 3 and may pick one up in a year or so second hand. Or I may wait for another upgraded unit so I can get a much larger build volume and convert it to rack and pinion. But for the upcoming year, most of my projects are relatively small...I will just have to visit techshop if I need to cut larger items.

twforeman
Posts: 1347
Joined: Tue Jan 29, 2013 4:51 pm
Location: Minneapolis, MN
Contact:

Re: Shapeoko Machine capabilities and limitations

Post by twforeman » Tue Jan 12, 2016 2:54 am

That AR1 is an interesting looking machine. I like the ball screws. The bearing surface area on the "v wheels" looks a little skimpy to me, but that may not really be a big issue.

I'll be interested to see how it holds up to heavy use. Keep us posted will you? :)
Ender 3 3D Printer
ShapeOko v3 serial #0004 - upgrade thread
All of my ShapeOko related blog posts

sgth0mas
Posts: 101
Joined: Sat Dec 12, 2015 5:57 pm

Re: Shapeoko Machine capabilities and limitations

Post by sgth0mas » Tue Jan 12, 2016 3:37 am

twforeman wrote:That AR1 is an interesting looking machine. I like the ball screws. The bearing surface area on the "v wheels" looks a little skimpy to me, but that may not really be a big issue.

I'll be interested to see how it holds up to heavy use. Keep us posted will you? :)
This is my biggest concern. I really wish they would have put the prismatic guides on all axes..but I'm sure that woulda been another 10% or so added onto the cost. While it's designed exceptionally well with respect to the guides and guide wheels...the wheels are plastic so I will have to treat them as a wear item.

samc99us
Posts: 269
Joined: Mon Nov 11, 2013 5:20 pm

Re: Shapeoko Machine capabilities and limitations

Post by samc99us » Wed Jan 13, 2016 10:05 pm

northbear wrote:
sgth0mas wrote: Now, let’s look at the belt properties. The Gates data sheet lists a 1” GT2 belt to have a modulus of 18,000, and specifies proportional reduction for smaller sizes with a further 10% between ¼” and ½”. That means the 18,000 modulus becomes 4,050.

To combat this, one could increase the width of the belt to 9mm.
Another way to improve the belt is going with a 3mm pitch belt rather than using a 2mm pitch belt. 3mm pitch belt has a modules of 30,000 (for 1" wide belt). It also has a greater ultimate strength (2200 lbs vs 1200 lbs) and working tension (114 lbs vs 25 lbs) again for for 1" wide belt. As sgth0mas mentions, these values are proportionally smaller and derated for smaller width belts (see second page of second link below).

I have been happy with my 3mm pitch belts, although I have not pushed to see what the best tolerances I can get out of them would be. On the other hand, I do believe I still hold the current speed record http://www.shapeoko.com/forum/viewtopic ... =30#p21023. Of course this is probably because all of the new shapeoko 3 owners didn't know there was a shapeoko racing league :lol:

Sources of information (mostly for my own reference so I don't have to search to find them again later):
http://www.sdp-si.com/D265/PDF/D265T015.pdf
http://www.bbman.com/assets/files/pdf-l ... erties.pdf
Well, I beat you on Z-axis speeds. 240 in/min is what I am dialed in for, with anti-backlash nut. I am running a 4-start ACME leadscrew and 270 oz-in stepper motor however...

I don't think the belts are a limiting issue. I have the stock 5mm belts on my machine, all axis save z, and it is much larger than the Shapeoko 3. I am cutting aerospace tooling to under 0.1mm (0.004") accuracy. The biggest limit is structural, namely, z-axis flex about the x-axis. It is visible when plunge drilling, so I do that very very slowly. I also do roughing passes followed by finishing passes and try to keep my tool load under 10lbs, preferably under 7lbs of force to keep delfection down. Sure, certain operations would be faster on a mill but not much. Overall, I am happy with this machine and my next upgrade is basically slightly larger extrusions, 1/4" aluminum plates, 15mm belts on all axis, steel v-rail on the x-axis and possibly the same on the z-axis (also evaluating an off the shelf ballscrew z-axis with linear bearings).

-Sam
Modified Shapeoko 2: 1500mm fully supported y-axis, 1000 mm 4060 x-axis, ACME Belt Drive Z-axis, Dewalt DNP611, full aluminum t-slot table, 4x NEMA23 180 oz/in stepper motors driven by a G540

sgth0mas
Posts: 101
Joined: Sat Dec 12, 2015 5:57 pm

Re: Shapeoko Machine capabilities and limitations

Post by sgth0mas » Wed Jan 13, 2016 10:40 pm

samc99us wrote:
northbear wrote:
sgth0mas wrote: Now, let’s look at the belt properties. The Gates data sheet lists a 1” GT2 belt to have a modulus of 18,000, and specifies proportional reduction for smaller sizes with a further 10% between ¼” and ½”. That means the 18,000 modulus becomes 4,050.

To combat this, one could increase the width of the belt to 9mm.
Another way to improve the belt is going with a 3mm pitch belt rather than using a 2mm pitch belt. 3mm pitch belt has a modules of 30,000 (for 1" wide belt). It also has a greater ultimate strength (2200 lbs vs 1200 lbs) and working tension (114 lbs vs 25 lbs) again for for 1" wide belt. As sgth0mas mentions, these values are proportionally smaller and derated for smaller width belts (see second page of second link below).

I have been happy with my 3mm pitch belts, although I have not pushed to see what the best tolerances I can get out of them would be. On the other hand, I do believe I still hold the current speed record http://www.shapeoko.com/forum/viewtopic ... =30#p21023. Of course this is probably because all of the new shapeoko 3 owners didn't know there was a shapeoko racing league :lol:

Sources of information (mostly for my own reference so I don't have to search to find them again later):
http://www.sdp-si.com/D265/PDF/D265T015.pdf
http://www.bbman.com/assets/files/pdf-l ... erties.pdf
Well, I beat you on Z-axis speeds. 240 in/min is what I am dialed in for, with anti-backlash nut. I am running a 4-start ACME leadscrew and 270 oz-in stepper motor however...

I don't think the belts are a limiting issue. I have the stock 5mm belts on my machine, all axis save z, and it is much larger than the Shapeoko 3. I am cutting aerospace tooling to under 0.1mm (0.004") accuracy. The biggest limit is structural, namely, z-axis flex about the x-axis. It is visible when plunge drilling, so I do that very very slowly. I also do roughing passes followed by finishing passes and try to keep my tool load under 10lbs, preferably under 7lbs of force to keep delfection down. Sure, certain operations would be faster on a mill but not much. Overall, I am happy with this machine and my next upgrade is basically slightly larger extrusions, 1/4" aluminum plates, 15mm belts on all axis, steel v-rail on the x-axis and possibly the same on the z-axis (also evaluating an off the shelf ballscrew z-axis with linear bearings).

-Sam
At 10lbs load your belt has over .025" of stretch available. What other component in your machine deflects that much?

Fablicator
Posts: 113
Joined: Thu Jul 02, 2015 7:59 pm

Re: Shapeoko Machine capabilities and limitations

Post by Fablicator » Thu Jan 14, 2016 1:46 am

sgth0mas wrote: At 10lbs load your belt has over .025" of stretch available. What other component in your machine deflects that much?
The Z axis flexes at least .025" with 10lbs of load at the tool tip.

samc99us
Posts: 269
Joined: Mon Nov 11, 2013 5:20 pm

Re: Shapeoko Machine capabilities and limitations

Post by samc99us » Thu Jan 14, 2016 1:18 pm

A stock z-axis, yes. Mine doesn't flex that much, but the x-axis gantry twisting about the gantry itself is certainly on that order.

Also, I typically do finishing passes at much lower loading, like 1lb of cutting force or less. That would result in 0.0025in of belt stretch according to your math, though I am not sure if you computed that for the y-axis or the x-axis. I have run the FEA on the beams (you can also do straightforward analytical calculations). At the middle of the gantry you can see more than 0.025in of gantry deflection.

Honestly the Axiom AutoRoute 1 Basic machine appears less stiff than my machine and to suffer from the x-axis carriage twisting issue even more. The linear control may be better, but I doubt you will see the improvement the ballscrew brings to the table with plastic v-wheels running in extruded slots (plastic v-wheels running on v-guide rail is another matter entirely). Their higher end machines with real linear slides look much much better.
Modified Shapeoko 2: 1500mm fully supported y-axis, 1000 mm 4060 x-axis, ACME Belt Drive Z-axis, Dewalt DNP611, full aluminum t-slot table, 4x NEMA23 180 oz/in stepper motors driven by a G540

sgth0mas
Posts: 101
Joined: Sat Dec 12, 2015 5:57 pm

Re: Shapeoko Machine capabilities and limitations

Post by sgth0mas » Thu Jan 14, 2016 1:39 pm

samc99us wrote:A stock z-axis, yes. Mine doesn't flex that much, but the x-axis gantry twisting about the gantry itself is certainly on that order.

Also, I typically do finishing passes at much lower loading, like 1lb of cutting force or less. That would result in 0.0025in of belt stretch according to your math, though I am not sure if you computed that for the y-axis or the x-axis. I have run the FEA on the beams (you can also do straightforward analytical calculations). At the middle of the gantry you can see more than 0.025in of gantry deflection.

Honestly the Axiom AutoRoute 1 Basic machine appears less stiff than my machine and to suffer from the x-axis carriage twisting issue even more. The linear control may be better, but I doubt you will see the improvement the ballscrew brings to the table with plastic v-wheels running in extruded slots (plastic v-wheels running on v-guide rail is another matter entirely). Their higher end machines with real linear slides look much much better.

You do realize deflections are additive right? So belt stretch is in addition to other deflections.

Care to post up the FEA? Id love to look at your setup.

The axiom has linear guides to prevent gantry twist, and the gantry supports have a substantially higher area MOI in the loading direction than any machine with a piece of aluminum plate. Im not sure what science youre basic any of your guesses off of...can you elaborate? The SO3 has a neat way of dropping the length, but it will put a harsher moment about the wheels. Show some math and help me out so i can at least see where youre coming from

Im not worried about Positional error with the plastic wheels, im worried about wear. It will not behave differently than other guide wheel setups until its worn and requires adjustment or replacement.

Additionally, a welded frame is going to be MUCH stiffer than a screw together frame. If you truly understand solid mechanics, that should make you think differently already.

Edit: Sorry if i sound harsh, ive been in design review mode all week.

Post Reply