Here are a few things I wanted to make sure I made a mental note of as far as the MPCNC. These are either just matter of fact or perhaps thoughts / issues with a solution to be found. I don’t mean any criticism her AT ALL. I am blown away by the MPCNC and can’t believe I actually have one and all it can do.
The first consideration is temperature. During the bulk of the machining yesterday it was roughly -17c (just above 0 deg f). I was working right at the edge of my garage and had opened the door and turned off the heater very early in the day. I’d suspect it was a few degrees warmer than the current outdoor temperature as the snow falling to the floor from the machine was still melting so the heat sink that is the slab on concrete the floor is made of was surely combating the outside temperature.
This temperature should be about the coldest the MPCNC is exposed to. The LCD developed some serious lag on the screen display and you have to be very slow and careful that you’re not clicking “Home” instead of “Prepare” and that you’re starting the proper file. I think this is just a limitation of the LCD - I doubt the machine really has much adverse reaction to this mild cold. Today when machining the V1 Logo the temps were closer to jut below freezing and it was better. Still laggy, but better.
I do feel that the MPCNC may struggle if temps get much lower than this though which for me is only a concern if and when I have the machine out onsite at festivals (which I hope to mid January and again mid February). These two festivals have typically seen temps as low as -45c once wind chill is factored in (-49f). However, in these cases I will keep the machine in my enclosed cargo trailer and will consider bringing along a small electric heater that I can run in the trailer to try to keep the temp up closer to -15c. That should be a workable solution. Of course I’d be far prouder to have it running out for the world to see!
One other oddity as far as temperature goes…I printed up the LCD case and machined the ends out of Sapele. When I popped the end caps on it was a pretty tight fit so I never bothered searching for screws to attach the side panels as friction held them on just fine. However, the PLA must shrink considerably in the cold because those damn things would fall off if I even thought about moving the control unit around! I have also decided for this machine that I’d like to extend the ribbon cables if I can (can I?). I’d like them at least double what they are now if possible. Let me know if I can do that.
Snow accumulation. In those cases where snowfill is the desired effect for the carving, the machine is almost hands-off. You may have seen in the videos that the snow does “plump when you cook it” meaning that when the ice is ground up by the bit the resulting snow volume is greater than the void it came from. While this is good for filling in the void, there is a bit too much and if you don’t blow it out of the way as it’s accumulating the snow gathering around the collet will be heated (friction) and either drop as water to your carved surface which you may or may not want or just freeze up eventually choking the tool and compacting between the tool and the work surface to the point things start to skip. Once I start running my longer bits this should be less of an issue.
There are two mods I’d like to make if you all can help me figure it out. I’d like to add an air attachment like I’ve seen on others’ machines - I’m sure I can figure that part of the mod out. The trick is the second bit…I’d love to know if a solenoid of some sort could be hooked up and controlled so I could turn the air blast on and off in the gcode. Bonus points if I could control the volume of the air. You’ll may see in the V1 logo time lapse that if the void is to be color filled you actually WANT the snow cleared out. However, if you want the snowpack then you want to only give a puff every 30-50mm or so. The snow does pick up a very small amount of “heat” from the friction of the cutter (more on this further on) so if your intent is to color fill you MUST keep the void cleared while cutting or else it’ll pack and freeze solidly into the void - you’d need to either chisel it out or remachine it. Neither of which are good options.
About the cutters. I can’t speak from direct MPCNC experience on this yet however I have used upcut spiral bits in a Dremel in ice in the past and I saw similar things surfacing. When using the Dremel by hand more often that not the flutes would pack up with ice, freeze and then the bit would become useless. I saw the bit come out of the ice on the MPCNC a little bit packed but when the bit plunged back in the ice popped out luckily. I am worried however that eventually this will catch me. It could be a matter of spindle speed so I’ll pay close attention to that as I carve more. Still, once the drop table is opened up I’ll not be using the small bits unless I need the detail. And at that, I have long versions of the 1/8, 1/4, 3/8 and 1/2 bits which range from 2 1/2 inches in length to a full 6 inches in length. That’s cutter length…not overall length.
Surfacing is a must. I use a Granberg chainsaw mill to “sheet” the blocks of ice into panels roughly 3-5 inches thick depending on my needs. But it’s not a perfect process. From the factory the ice has a bulge to it so when I use the mill it follows this bulge which usually results in an offset of up to 1/4" thickness from one end to the other. When doing logo carves by hand this doesn’t matter. But to the MPCNC it does. If you have a 5mm difference in thickness and only a 10mm DOC there’s going to be issues somewhere somehow. I’ll pick up one of those flat 1" / 1/4" shank bits and make a surfacing pass in the future. That bit and a touch plate for tool changes will dial these logo panels right in for me. For the V1 logo I just held the flex adapter on the z axis and moved the gantry all around the surface of the ice. If it scratched anywhere I lifted it by hand a hair and kept moving it around. Once I felt I had the “lowest common denominator” z height I powered up the machine, moved the z axis by .1mm to engage the stepper and then set it to my hard 0,0 in preparation to run.
Snow on the steppers. I saw very little snow buildup on the steppers or on the gantry. And the air blowing out of the DW didn’t have any impact temperature-wise on the ice. The build up I DID see is around the bearings on the Y axis but that’s because it was on the receiving end of the air blasts. I imagine if I get an air attachment setup going then I’ll have a “common” area that this snow will blow to and I can take steps to deal with it accordingly. If the machine is as cold as the outside temp then the snow should just stay snow and can be blown off.
Fears:
- VERY cold temps - I don't know how the wiring will hold up in very very cold temps but if I adhere to the "heated trailer" plan maybe I can keep this a non-issue. I'll just try to refrain from super cold startups.
- Moisture overall - I don't know what things might be like on days where the temps are just above zero. I don't have a freezer to work in so rely on cold weather to do my work. If a job comes in during a mild spell I may have issues with the blown off snow melting on the rails and as a result getting onto the bearings. I know when I just started looking into the MPCNC I had asked and someone suggested moisture may be an issue. I think I'll just keep some ziplock baggies and duct tape on hand on case I ever need to make some quick raincoats for the steppers. Other than the top surface of the table, it's all been protected on all sides with exterior enamel paint. Once I cut the drop table out I'll waterproof the rest.
- The belts worry me a bit as temps get colder. I have some spares laying around so I could try some cold weather tests - but I don't have a force gauge. My brother in law does - maybe he can test them for me. Are they as strong as Ryan suggests they are? Anyone know how temp affects this strength?
One last thing - size. I’ve decided to swap my Y rails with two of the Y rails from the laser build which will give me 4 more inches on the Y axis. I built the table to utilize the 48" in the plywood sheet which shorted my usable x axis to about 38 inches. I miscalculated the Y though and my actual usable span is only 19 inches. Ice is 20 inches X 40 inches but it’s the 20" that would be most important to have access to - typically we have to cut an inch off each end of the 40 to give us nice uniform blocks (the manufacturing process puts dents in the ends of the ice blocks where the lifter plates freeze into them). But we often go edge to edge on the 20". That Muttart logo worked out because I was going off the ends, not the sides. I just set the ice block into the MPCNC offset by an inch or two so I knew I could cut off the edge and the pieces would join in the middles. I was quite impressed how well that worked out.
I’ll still have a very capably sized laser MPCNC - it’ll just be 24" X 20" instead of 24" X 24". What it does mean is that the MPCNC will have to be sorta torn apart. The outer rails are easy, it’s the gantry rails with the wiring in them that might be a pain. But I want to wire up endstops and a touchplate though so I guess I need to open up the wiring anyway here.
Oh - and I also discovered my MPCNC was ever so slightly out of square. I’ll rectify that when I reattach the feet to the table (which will need a 4" extension added onto but that’s OK).
So there you have it. Please feel free to offer me any advice or feedback on any or all of this. I know it’s a lot but if anyone ever follows this path I went down maybe it’ll help them. And maybe something in here is worthy of further discussion or new threads. Maybe some of the discussion crosses over from ice to other materials too…I tell you this much - I could buy a purpose built CNC machine made for ice carving. For $20,000…
No thanks. I’ve got one.
I’ll reread this once posted and edit where need be. Maybe I’ll note if I do in case anyone has already read through before I edited.