I have read it and that is how I started to think about fixture design, but the thing that lead me to a vise and a chuck, was the problem that each piece of round stock wasn’t identical, so cutting out a “hole” to place the stock would not have worked.
If I used a 90 degree “stop”, then the center of the stock wouldn’t be in the same place each time, so I’d have to redefine origin for every part.
As it is, I can set my origin to a point on the vise, or the chuck or the board, once. I can set the work-offset, once, and then all day I can machine a part, remove it from the chuck, insert new stock, tighten the chuck and start the next part, no redefining origins required.
The single biggest help I received was from a redditor who pointed out that there was a thing called “fixture design”. Once I heard that, it all started to come together. You design your part and then you design the fixture.
Until then I was thinking in terms of stops and clamps. Fine for some jobs, but not universally so.
No doubt when theory and reality collide when I actually start making chips, I’ll see how well this lesson holds, but so far it seems fit for purpose.
A new platform is like $38 plus shipping ordered via support@snapmaker.com, cheap enough to play with without risking warranty or being stuck without one. the one that I got as spare was the new, square style, machined and pretty level.
Check out the old books by Edward Hoffman, like Jig and Fixture Design. There are a ton of similarly-named books for woodworkers, but they tend to be for making tablesaw or router accessories. Stick with the metalworking ones, they are much more interesting.
Not to get too off topic, but I think that’s why I’m amused by people wanting ‘quick change’ beds or toolheads. Coming from a woodworking background most of the time is spent getting ready to cut or making jigs and fixtures. The actual cutting takes almost no time in comparison.
Maybe 5 minutes to change from printing to cnc isn’t a big deal.
-S
Or on the SM the actual cutting takes so long that the 5min isn’t an issue.
Obviously, you’re right.
But as it’s (for me) a hobby where I can’t always spend a lot of time on, it helps to reduce that part. + it’s a great learning opportunity to get to know the machine and its capabilities. And just fun projects.
But obviously, you have a valid point. So I’ll blame it on the need for instant gratification. Good thing I am an early millennial and not a Gen Z or it would be even worse .
(but I am still happy with my mods. That’s also a tradition to improve and modify your tools and workspace. Even in woodworking, if you haven’t built your own slightly modified workbench, you’re not considered a real woodworker. At least that’s the impression I get. But off topic indeed.)
The advantage of quick change tooling in the machining world is that it holds calibration between changes. The quickness is not so much in replacing the tool (or whatever), but in the calibration / measuring / centering / etc. that is avoided by using such fixtures.
I’m sure the argument could be made that calibration is required (or recommended) regardless. Though with the 3D build plate used as a base for all the others (which is what is usually suggested), the most painstaking calibration procedure is taken care of.
For my part, it takes a lot longer than 5 minutes to switch modes: the way my enclosure is positioned, I only have access via the front of the machine. There’s a mirror mounted to the back of the enclosure so I can see those damn toolhead screws. The bed alone takes five minutes, the toolhead another ten, except for the 3-D toolhead which has to be mounted twice in order to verify the bed-level sensor - we’re talking about a half hour there, what with booting the machine to position it over th bed and adjust to the second calibration card.
The main benefit of a quick-change toolhead, as I see it, would be a proper registration surface, so that you don’t have this sort of “whoops, it’s about half a millimeter lower than when you last assembled it, due to the play in the bolt-holes”.
As for the beds, well, why not? A solid base bed that the others bolt onto with far fewer screws would be great. The laser bed components irritate me to no end: I already had to label them L CL CR R because they are not interchangeable, and they rattle a bit once installed. Those guys are getting glued to MDF some day.
I built my own enclosure and have a door on the side of the controller and one in back. I thought it made more sense to approach from the back since that’s where the toolhead screws are. In use now I almost never use the back door except when looking to align things, to vacuum, or when I have a workpiece longer than the bed.
I use a power screwdriver that has adjustable torque to tighten all the screws. My head seems to settle in the exact same place every time if I just let it hang down, but I always run calibration every time I swap the heated bed back in.
For most laser I just use the cnc bed. If I am cutting I’ll use the laser plates. Mine rattle if I’m not careful to push them together as I tighten them, but luckily that usually stops it.
-S
.