Anyone worked out good workflow for testing whether their stock is actually in and running true in the jaws of the rotary tool? I get that most people won’t care that much, but I’ll be doing some pieces where “not much” is machined off, and if the stock isn’t fairly accurately true, it won’t come out correctly.
Similarly (and once it’s running true), any developed a good operation for facing the stock?
Runout gauge seems obvious choice.
After making sure my stock is fully seated in the chuck, I’ll usually quickly make a cylinder STL the size of the smallest diameter I measured across it. (I’ve used 50mm dowel and it’s never exactly round). I’ll load that into Luban as a surfacing operation first, then remeasure afterwards and use that for my actual carve.
Yes… but how to get it to rotate “just for fun”. And it won’t help with facing.
Do you face the end with that as well? And if so, how?
Finally had some time to get back to this. Using a dial indicator and just “manually pressing the button” to rotate 5 degrees over and over, I’ve been able to minimize runout (more below) and to face the ends, I’ve had to provide over-length the original stock and add operations at the end to essentially part both ends of the stock. Tedious, but it kinda works.
HOWEVER! It brings up a major issue. The chuck itself has 0.4mm of runout (or out of concentricity) Confirmed and tested with stock that I’ve turned in a real lathe and is round 0.01mm. No matter what I do, I can’t get that 0.4mm out. If you’re doing a wood figurine it probably doesn’t matter, but when trying to do precision parts (preferably press fit, but I don’t think the Snapmaker would ever be capable of that),
I also found that the jaw faces are not flat/parallel. The angle gets worse with more tightening of the chuck (stock goes through the jaws, so it should be held all along the jaws), meaning that they can’t hold what little concentricity they have! The tailstock should be the answer, but as it’s only held by two screws, it splays out whenever any sort of force is applied and makes the runout even worse.
Picked up a (used) rotary module recently. First thing I plan to do with it is to replace the three-jaw with an independent four-jaw. After that, make a bump tool to align the stock before tightening the chuck. A bump tool is basically just a ball-bearing mounted on the end of a shaft, usually held in a lathe toolholder: the bearing rides on the work and as you feed the bump tool in. the workpiece (usually rotated by hand) gets more concentric. A dial indicator is used to determine progress. I haven’t decided yet how to make this work on the snapmaker: a magentic base for the bump tool might work, or one that bolts to the CNC bed, but the feed will be a little more complicated. Might be better just to use a dowel pin in the CNC tool collet.
For the non-parallel vise jaws, consider getting a copper pipe coupler from the hardware store - these are basically an inch or two of copper pipe. Saw the entire length of the coupler. Put the coupler around the workpiece, then when you crank down the vise jaws the copper should deform under the jaws while the workpiece remains true. You can also try to true the vise jaws yourself, if you have a lathe and a dremel.
Hey. Let me know how you go finding a 4-jaw to fit. As for the bump tool - the allow used for the snapmaker is non-metallic… found that out with the dial indicator…makes life harder - so plan for bolt, OR bolt a steel plate so you can then use magnetic. Rotating is hard… best I’ve found so far is to labouriously press the 5 degree rotate button. I thought of creating a job with a 20cm OD so the tool sits out the way, but there’s no way to start/stop.
Thanks @edf - Yeah - not sure on facing the jaws yet - there seems to be enough (too much) slack in the mechanism, so either some copper pieces or similar might be the go (not sure the jaws can put enough pressure on the copper to deform suitably, like in a “real” chuck - maybe I can put them, pressure and tighten, then loosen and run a reamer through to get some degree of parellelism before closing the jaw on work stock?. I think your idea of a new chuck might be the best (either a real/better 3jaw or 4jaw). My lack of knowledge of the different chuck faceplates limits me, so let me know when you find what faceplate they’ve used (if they’ve used a standard!), and/or adapter plates you’ve used, and your recommendations when you get there. 
When I make my new tailstock holder, I’ll post the plans. The live centre seems OK, just the holder and bolting to the base seems “under thought”.
Also see my new post - Rotary Tool concentricity/runout issues with the chuck itself - #2 by Doogs75 - for more issues that I’ve run into…
Waiting for this one to arrive:
https://www.amazon.com/gp/product/B07RKW9QLV
I’ll post an update when it does. My A350 is in pieces - I am working on a linear rail mod, and keep getting interrupted (work, travel, and now no power to the machine shop for a week following the sudden arrival of three feet of snow) but should be together by the time it arrives.
I have an indicator mount on the 3D toolhead that I can replicate for the others. I also made 1/8" steel end plates that will fit on any toolhead, and will probably put those on the CNC head. For general-audience use, I was thinking of something like thin (1/4 or less) MDF on top of the heated bed. The magnet of the dial indicator base should work through that, though I haven’t tried it. There’s always magnet-impregnated MDF, which I believe I have seen on a few posts here.
For rotation, there is probably a way to do a slow-RPM rotate with Gcode. There definitely isn’t a thru-spindle to put a hand-crank on
EDIT: Thinking on it over coffee, probably the best way to go would be a toolhead-mounted bump tool. This could be held in the spindle, or attached to a dial indicator/accessory mount on the toolhead. Gcode would run the rotary attachment at a low rpm for a few rotations, then the user would jog the toolhead to bump the bearing against the workpiece. Sounds … slow.
@edf - cool! I might look at their tail-stocks as well… probably easier than trying to machine one to “pretty damn square” myself.
Came across this last night:
https://www.thingiverse.com/thing:5372172
The model might be of use in drafting up modifications.
@edf -looks like snapmaker lifted ronin’s design almost exactly! LOL. Will certainly save measurements, etc.
heh well I guess Ronin didn’t want to pay five hundred bucks for one, and who can blame him
I like his use of a more powerful motor so that a heavier chuck can be used.
Finished most of my parts last night - need to mill clearance on one piece, then drill holes in the SM base for linear rails, then I can finall assembly the thing and put it back to work. Quite a frankenmaker at this point.
1 Like