Looking for some advice on materials to use for a Primo.
Over the past couple years I’ve move towards traditional hand tool woodworking and away from machinery. Sadly, that means my LR2 doesn’t see much action and actual gets in the way of working on longer boards on my bench.
Thinking about converting my LR2 into a Primo to make functional parts and hardware. This would probably be a 24" square Primo or under. I’ve got 3D printers modified to print engineering materials but really want to start designing and milling aluminum parts for projects as well. For that reason I’d like to make the Primo as rigid as possible.
Couple of questions:
Would a super rigid plastic material make a difference for the Primo over PLA? Really don’t want to use PLA not because I don’t like it but because my printers aren’t really designed to print it well. I was thinking 3DXTech ezPC+CF or a PC blend but I’m not sure if the extra rigidity will make a difference. Could also do PA6+CF if there needs to be a little more “give” to the parts.
Also, would mounting the MPCNC to granite be beneficial? I know a place that has tons of 2" thick countertop offcuts for really cheap and was thinking about bolting the MPCNC to it for rigidity.
PLA is the filament of choice because it is more rigid than PETG, ABS, or PLA+. Those materials are not as brittle as PLA, but also not as rigid.
Beyond those basic materials, I don’t have any personal experience. If you are printing with something as rigid as PLA, and there isn’t any significant shrinkage of the parts, and you can do similar prints (w.r.t. overhangs, etc), then you have a winning material.
The two benefits I know about for the granite are: flatness and rigidity.
I don’t think flatness will help much. The important thing isn’t that the table is flat. It is important that the table is parallel to the way the bit moves in the gantry. So surfacing a spoilboard will make the new top of the spoil board closer to parallel to the gantry than starting flat.
The weight or rigidity could be a benefit. But I have not seen the table flex being a serious contributor for any projects in the forums. The loads are just not high enough to deflect the table before something else goes wrong. Maybe the weight could reduce chatter?
How would you hold down the material to the granite? How would you avoid destroying the bits? You would need a spoil board anyway, right?
I would be willing to try it. But I wouldn’t spend $300 on it or bend over backwards to make it work.
Only thing I have to add is that I wouldn’t expect any trouble milling alunimum on a primo that size, even with run of the mill pla.
Lots of us have cut lots of aluminum. You’ll also note that we don’t even all prefer the same strategies to milling it. It can be frustrating to start cutting alunimum and have problems, then get contradictory advice but I mention it because it speaks to the performance of the primo that you don’t have to do it a specific way.
I will say that when pushing really hard or even just too deep, the entire core seems to flex away from the cut. Whether it’s the bearing holders or the gantry tubes or something in the core flexing I never figured out, but before spending money on fancy plastic, it might be worth assembling, pushing on tbe parts, and seeing what you can see.
Ok. So rigidity is the key factor I should be looking for in a material other than PLA. Would PA6+CF be a viable option? Checks the box on rigidity. I don’t THINK material creep would be a problem on this machine. If PLA is being used on the motor mounts I don’t think PA would creep at all.
Not entirely sure on a hold down situation. Is/could there be a way to keep the tool head from crashing into the marble in case of failure? Like a hard stop on the Z?
Also I noticed on the Z there is a spring coupler on the lead screw. Is it a bad idea to get a stepper with an integrated lead screw?
How would you cut through material if you couldn’t cut into the spoil board?
The size of the bit is variable. So the perfect Z hardstop is variable.
The machines usually have a sacrificial top made of mdf or plywood. The but is happy to dive into that to cut the perimeter of parts or if there is a mistake.
I see what you mean. Going to use the same stainless tube I used for the LR2. I figured with those plus the short length it would be a bit more rigid. Don’t think I’ll skim on bearing either. Been burnt a few times by bad bearings.
The problem with printing parts with PLA and then seeing if I need to upgrade is that I can’t really print PLA great. My machines are just not setup to do that. I could go PETG+CF to save some money I guess.
The weight of the router holds that down. So there is no backlash. The motor axle and the leadscrew should be touching before you tighten the grub screws.
An integrated leadscrew could be a problem if: Things aren’t perfectly aligned. Or if you need to cut the leadscrew (it is easier if the motor isn’t attached).
I’m not saying there wouldn’t be a spoilboard. Just a permanent stop that would not let the bit dive all the way through the spoilboard into the marble.
If you intend to always use the same bit, or bit type, this is possible, but I have some long bits that are more than an inch longer than the usual ones I use. The usual ones have about 1/2" of cutting edge but I have long ones with 1.5" cutting length, and more length above the cutting area. (Generally worse for the LR, but no real difference on the Primo, rigidity wise.) It does take away some machine movement room, though. A permanent stop that works for the short bits won’t help with the long ones.
Just a warning that your best bet is to try to make sure the CAM is OK.
I suppose a solution may be (on a Primo) to have different length blocks in the Z axis for.different bits, but then it’s up to you to make sure the correct one is installed.
One thing that I’ve found helpful when trying to get my head around rigidity questions is to think about whether the item deflects if you grab the toolhead by hand and pull it around.
I have my MPCNC mounted on a 40mm triboard work surface. There is zero chance that the rigidity of toolhead to Z to core to XY tubes to trucks to frame to corners to legs is remotely rigid enough to cause it do deflect at all. If I take the screws out of one corner, I can get it to lift that corner up slightly, but only enough that it can be held down hard to the table with light pressure. That light pressure is what the table sees while cutting. Also everything else in that chain is going to be so flexible that having a more damped bed won’t do anything.
Personally, I think granite may cause you more issues than it solves (harder to mount things too, primarily, but also heavier and harder to move etc.). I’d be focusing attention elsewhere.
My approach was to leave my tubes long and plan to shrink/expand my MPCNC according to the work I need to do. At 300mm/12" square, it’s immensely more rigid than it was at 600mm/24" square. If rigidity is the primary concern, I’d sooner invest the effort into making the MPCNC the smallest size possible and precisely indexing/probing the work to cut smaller sections at a time or similar.
In that case, maybe start with the stiffest think you can print for the core and related parts, and maybe the motor mounts. The feet and corners are really good and I can’t imagine them causing you problems unless you print them in ninja flex. At least, that’s what I’d do.
Yeah I guess you’re right. Have different blocks for different shaft lengths. Also I’d probably want some kind of setup jig for each length shaft to make sure it’s getting inserted into the collet the same depth every time.
Yeah. I’ll never tell anyone how to spend their money, but some of the machines here look great. Mine look like they were thrown together with whatever was laying around (they mostly were, lol). I do run mine a lot harder than a lot of folks are willing to, and they hold up great.