The snapmaker 2.0 has grabbed my attention because of the build quality and included attachments. The upcoming dual extruder, multi filament and especially the 4-axis module are of particular interest. I do hope that the dual extruder and multi-filament module have multiple nozzles.
The four items I would still love to see:
A spindle for aluminum & brass. The included spindle is more than enough for wood and I don’t want to see that go away. But for those that desire aluminum I would love to see a small spindle. Snapmaker 2.0’s closest competitor ZMorph is capable of milling aluminum so I would love to see this addition to the snapmaker.
5-axis CNC - For the A350 I would love to see a 5-axis module. A module attached to the bed that can tilt and rotate.
A fiber laser - Maybe more of a long term request for when prices come down. A fiber laser can engrave & mark metals.
Enclosure improvement - Make the doors easier to open and close. They tend to get caught. Instead of acrylic which can crack make the windows polycarbonate. Finally a little more airtight to help reduce noise and catch all particles from milling.
For me the main concern for this isn’t the spindle but rather the rigidity of the machine. Thoush some have tried and succeeded in milling soft aluminum with the current spindle.
If your referring to the rotary module. The jury is still out if this a 4th axis or if it will just replace the y axis. My money is on replacing the y.
The bigger problem I foresee is what software is going to be generating those toolpaths? Luban?? Definitely not F360.
And yea, no chance as aluminum and brass at any capacity beyond something considered a hack, the machine just isn’t rigid enough. It’s been done, but the surface finish was only mediocre.
Depends on what you mean by “hack” this printer is moduler right? So if someone figured out how to buy some extra rails and assembled it with a cube design, with a xy gantry and z bed it might gain the rigidity it needs… maybe?
And giving how this printer was advertised im not sure I would call that a hack.
That’s a fair assessment. I think increasing the rigidity of the frame is less of an issue than the torque withstanding abilities of the rail internals. To me, hacking in this sense is the chasing down and fixing these problems one at a time to eek a little more performance out of something not designed to do that. I don’t say that in a bad way either, I’ll proudly admit to hacking the firmware to get true power grayscale lasering working.
Your example of stiffening the system with additional parallel components or something may very well be a viable solution. Something would have to be done with the toolhead mount, perhaps multiple parallel X axes mounted on 4 Z modules that would give a much wider lever arm to withstand the applied forces from the spindle. In addition to the existing Y axis.
I think the most bang for the buck and easiest upgrade would be to add rails for the bed to ride on. Most of the flex I find when using cnc is when it gets to the end of the y axis and is pressing down on the bed.
2020 (or 2040) extrusion would do the trick with some bearings and printed brackets
I completely agree. This doesn’t require nearly the level of additional material that the X and Z axes would require.
Linear rails and matching linear bearings are readily available these days, inexpensive and of adequate quality for this purpose. They’re all over eBay, for example. Many are direct-from-China shippers; pay attention if you care about risk and arrival time.
The type I’d suggest are round rails mounted on an aluminum extrusion base that holds up the rail. The designation “SBR16” is the kind of rail I’m talking about. I’d use a bushing-style linear bearing made of polymer. The reason is that these are somewhat self-wiping, so they’ll be lower maintenance than the ball-bearing-in-linear-race kind. I’m not seeing the bushing kind nearly as readily available as the ball-bearing style, so whatever.
Height adapters of some kind will be required. Just to be exhaustive, there are a number of locations.
On the new rail. This is the default position, since it keeps the bed low and the work area large. It won’t be possible if the rail + bearing height is larger than available.
On the linear modules. Needed if the new rails and bearings are too high.
Underneath the extrusion. This is perhaps the simplest to make, since it’s a slab of constant thickness.
Between the bearing and the bed. There’s got to be some adapter needed between the spiderweb subbase and the linear rail. If convenient, this might be the same as the height adapter.
Linear bearings expect to be operated flat. It might be necessary to tie the rails together to ensure they remain coplanar with each other.
The length of rail required is the sum of (1) the length of travel, (2) the distance between centers of the bearings, (3) the length of one bearing, and (4) twice a small allowance of “unused” rail at the end to keep the bearings intact. I’ve not done a detailed measurement, but it looks like 750 mm or 800 mm should be adequate. Better to buy the rails too long; if needed they can be cut to length with a hacksaw.
I don’t think that will be adequate long term for milling harder substances such as aluminum and brass… to me that is simply somthing that should have been in the design to start, as the bed wobbles at its extent even when just moving
Though brent makes a good point about the internal linear berings. They are not very hefty and some people have already reported them failing. Dubling up on the x rail might do the trick.
To be clear to @eh9 and @sdj544, I think it will mill aluminum as is(as some have done it) but I think it is not rigid enough or robust enough to mill aluminum regularly without drastically shortening the life of the modules. And without the need for very good cam skills.
Putting an SBR16 linear rail (or similar) on either side of a linear module will take the load off the internal linear bearings inside that linear module. The internal bearings are wheels with half-round rims that run on rods. These are OK in only one of the three planes of rotation, the one that contains the parallel rails, and even then they’re still not very enough apart. In the other two directions the resistance against torque at the nominal position is quite low, so in practice there’s some hysteresis in angular position depending on direction of load.
Doubling the x-axis rail with a sixth linear module would only address partially address one of those two rotation planes, as they would still share one axis where they’re weak. The one they share wouldn’t do nothing, since you’re increasing the distance between the rails, but the bearing style itself isn’t really up to the task. This is already the problem with the base of the A350; the bed wobbles significantly front to back even though the rail is doubled.
The downside of correcting this with linear rails is that they need to extend past the linear module on either side to resist torque. This will increase the volume of the machine, only really a problem if you want to use the enclosure that SM sells. Fixing this problem on all the axes means six rails in the three pairs. It’s a rather big project, essentially a new build with linear modules as one of the parts. On the upside, it’d be stiff enough for milling to put a 400 W (about 1/2 HP) motor on it without question, and maybe larger.
Hmmm, yes, toolhead rotation about its Y axis is still weak. I can’t think of a good solution that doesn’t involve adding another parallel rail above the existing X axis rail to give a lever arm to resist the torque using a leadscrew
All of these solutions together will add significant mass to the system and will require dropping the maximum acceleration massively, unless the additional leadscrews can compensate for the forces. On the Y table that seems unlikely as there will be additional mass without any additional driving screws.
I think in the end, it would be cheaper and easier to build a new machine that is better suited to milling then it would be to modify the SM2. One with linear rails that are better suited to withstand the forces and a frame that is more rigid over its complete range of motion.
Yes I agree with you now this is definitely into the range of hacking… if the linear rails were designed to be more robust I think it would fall into the modular range. But they just aren’t designed well enough. Not sure how snapmaker was planning on implementing there “3d party spindle” add on without running into these issues.
Maybe they were thinking of taking the approach of “let’s give them an addon they want and when it breaks their equipment they will buy more parts from us”?
I’ve wondered that regarding the proposed 8W laser. Are they going to spend more than $15 on the included eye protection if you buy that module? OD goes up by 10X, and going from 1.6W to 8W is fairly small, but it would be extremely ballsy of them to argue the provided OD1.19 goggles are adequate for the upgraded module as well with a reduced laser power of 517mW as viewed through the goggles we have now. Yet I just have a feeling they’ll do just that because I haven’t seen anything that would indicate otherwise.
Even a diffuse spot at 500 mW is only considered “generally safe” as long as you don’t stare at the dot for “many seconds”. Has that been clearly communicated? Do people know that it’s only safe as long as you never stare at the dot through the goggles?
I agree I think they will just sell the new module separately and expect the user to do the reaserch and buy appropriate PPE. Witch in my opinion would be fine as long as they put a warning on there… I wish they had not included laser safety glasses at all as I think it lulls inexperienced users into a false sense of safety.
Also im not entirely convinced that the laser they provide has been tested to see if it is truly a 1.6W output, and the same will be true for the 8W, especially because if you look into it, the strongest laser diode out there is only a 7W, all laser diodes above that use “pulsed” outputs that put extra wear on the diode or use high end optics to focus the beem to achieve an 8W rating at the point (still trying to figure out how that is achieved).
My point is that it may be safer then we think because the laser is not as powerful as they claim.
Oh and one last thing, I’m sure they will overcharge for it as well… since I can buy a complete 5.5W laser kit from Amazon for $150, witch is the price they are charging for the 1.6W module currently… cheapest 7W laser head i have seen (without a max run time) costs $250 alone (that doesn’t include laser driver, heat sink, fan, optics and mounting)…
