|
biracial bear for uncut posted:You can use your last hole in one setup as your first alignment hole when moving setups. Maybe one program working off one edge datum, then the sequential programs basically calling X/Y zero the hole you locate off of in the subsequent setups. Thanks for the ideas. I thought about bolting an endstop rod smaller in diameter than the holes to the table to accomplish just this; the good news is as long as the holes are spaced measuring-tape accurate down the bar there really won't be any cost to being lazy, they really only need to be accurate in Y as illustrated. Bad news is my G540 might have died in anticipation. No lights, redid the wires, checked multiple PSUs, replaced the fuse, measured the resistance across the r040, all of which check out, but nothing works. Gecko tech support blew my dick clean off: calling on a friday afternoon I had a dude walking me through probing components with a dmm over the phone within 2 minutes. It is now clear that god does not want me to have a working mill, but a positive tech support experience really lifted my spirits.
|
#
?
May 16, 2020 00:40
|
|
|
|
| # ? Apr 25, 2024 14:15 |
|
|
While I await the return of my G540, let's gently caress around: I originally intended to make a heavy table base for the plasma but thinking about the design I wanted something I could tear down. This led me to consider simply making legs out of extrusion, but I already have steel so why not just make my own legs and braces? 3x2" 11ga legs with a bracket made of 3/32" into which I made four holes scribed with calipers that line up with the t-slots on the extrusion. I don't know how to weld but I've gotten far enough with this amazon stick welder that nothing's falling apart yet. Here's how big a 2x4"x72" extrusion with a single 36" leg is compared to a Shapeoko XXL.  The grinder hides most of my crimes:  Three more legs to go. Then feet. Then braces. [edit] Two more. gently caress this is too big for my garage. 
Ambihelical Hexnut fucked around with this message at 03:02 on May 23, 2020 |
|
#
?
May 22, 2020 23:33
|
|
|
Hows the shapeoko been? Getting much use out of it? I never ended up getting one, picked up a 3D printer instead and now I've gotten used to having the garage free for the actual car after cleaning it out.
|
|
#
?
May 23, 2020 03:48
|
|
|
I've gotten a ridiculous amount of use out of it; I pre-ordered in early 2015, upgraded to XXL in 2017, ran a small sign making business for a couple years with it, and have used it to make all kinds of plaques and engravings for my day job the entire time. This is in addition to things I do for myself and the family. These days it's a pretty mature product and works very well if you get dust collection and workholding sorted. Everyone on the internet wants to pretend that it is also a mill, but it is not a mill. It's a gantry router for sheet stock.
|
|
#
?
May 23, 2020 06:07
|
|
|
Yeah I never got why people are so keen to try to use them for mills on metal etc when there are so many applications for the materials it can actually handle. The plasma seems like a good idea, it's a nice process that I haven't worked with since school and the simple plasma CNCs don't need to be as pricy as they are. I can't cut a straight line with a torch to save my life so I look forward to seeing your progress.
|
|
#
?
May 23, 2020 06:32
|
|
|
Okay after another full day of this I have four legs and four supports. Size mockup: a) I'm going to have to rearrange my garage when it comes time to assemble this thing. b) If I make it a foot taller I can park the mower underneath. Probably shouldn't.  Support bars cut. The trusty harbor freight 4x6 bandsaw takes like 30 minutes per cut on this tubing. Each one support needs two 45 degree cuts, then a piece of flat bar cut out to make a bracket for the extrusion. Then you have 8 edges and 8 corners per support which need to be sanded and given a small bevel. Then clamp it together on the table, tack one side, undo all the clamps, tack the other side, undo all the clamps, and start welding about an inch at a time alternating sides so it doesn't warp/etc. It goes much slower in real life than on youtube. Bad welding:  Hide with grindy:  Then go back over it again because you can still see pits and low areas, then grind it again. I'm extremely tired of drilling brackets by hand, can't wait for the stupid G540 to get back so I can just mill them.
|
|
#
?
May 24, 2020 01:45
|
|
|
 Mocking it up. Need to add feet plates.
|
|
#
?
May 24, 2020 18:11
|
|
|
Karia posted:Rather than indicating in the hole every time you move, you can keep your drillbit down in the hole (plunged down past the flutes so the shank is in the hole), unclamp, and use the X axis to drag the bar over. You'll probably lose a thou or two every time because the hole's going to be slightly oversize, but maybe that's acceptable, and once you know that offset you can compensate for it if you want. I was going to say this in my original post, but I know anybody with any industrial CNC machinist background would have a heart attack at the thought of using the positioning motors on a CNC to drag stock around (because they aren't designed with that in mind). Also it's just bad practice in general, you can bend or otherwise break a tool if the stock is heavy enough.
|
|
#
?
May 26, 2020 19:04
|
|
|
biracial bear for uncut posted:I was going to say this in my original post, but I know anybody with any industrial CNC machinist background would have a heart attack at the thought of using the positioning motors on a CNC to drag stock around (because they aren't designed with that in mind). I have an industrial CNC machinist background, though I've been out of it for >5 years. There are many many examples of this done in industry and lots of tooling out there for it. Its almost certainly within the design limits of the pictured machine, but if theres any question, look up the specs. That said, Karia's specific solution is likely not repeatable within a few thou as hoped. Drilling this in a vice is not likely to give the desired results, but tolerances weren't mentioned. You can bang out a fixture plate for this hat will likely be better than what you want for a few bucks and maybe 1 hour of time. Move it by hand unless you gotta do a bunch of them. You can position the below orange piece as far apart as needed, just indicate them in on some machined square feature. Support the center.
|
|
#
?
May 26, 2020 19:40
|
|
|
So was reading the wiring manual for the Fadal VMCs yesterday to try to figure out how hard it would be to wire up the special single phase transformer when I realized... the spindle just goes to current taps on the "this goes to the breaker" side of the transformer, the transformer doesn't do anything to the power going to the spindle drive. And on the single phase version it literally only hooks up 2 instead of 3 wires. Motherfucker... the only thing the transformer is there for is to make 120V to run the controls and poo poo. So long story short I'm 95% sure I can get one of these suckers and run it on the circuit that used to go to my hot tub. Now to build a shed!
|
|
#
?
May 26, 2020 23:43
|
|
|
Wow thanks for the fixturing input guys. I agree that some kind of indexable plate will be the easiest and most accurate. I don't have the stuff on hand to get beautiful precision dowels into a beautiful plate, but I also don't need to face the top of the steel so a few well placed screws could probably serve as both a repeatable edge and clamps. Still not there yet because I keep loving with my mill: today's project was getting the LMS pwm spindle controller installed so I can control my spindle from mach3/gcode instead of just manually. It is an add-on board, and curiously a replacement chip to put in a socket on the brushless motor controller. Board installed (lil guy on bottom):  Finally using up every drat pin on the G540. I wired up a 6-pin G16 for limit switches, another one for the spindle control, and a stereo minijack for the touch probe:  I don't know why Gecko decided to put the pins along the bottom requiring an immediate 90 degree bend and ensuring you have to remove the controller to change anything, but it's my main complaint. I have one more remaining thing to cram into this crowded box of electrical noise which is a 48v cooling fan and G540 heatsinks.
|
|
#
?
May 27, 2020 05:42
|
|
|
Get some roll spring pins from Mcmaster-Carr and tap them into drilled holes with a mallet. They'll work better than screws and be a good indexing surface as well as long as you have the cut facing away from the side you want to index off of.
|
|
#
?
May 27, 2020 14:30
|
|
|
biracial bear for uncut posted:I was going to say this in my original post, but I know anybody with any industrial CNC machinist background would have a heart attack at the thought of using the positioning motors on a CNC to drag stock around (because they aren't designed with that in mind). I have ~8 years of industry experience. The motors wouldn't be the concern unless they're very under-spec'd. The only thing I'd be worried about is the spindle bearings: they're not designed to take static load. You'd need to minimize the bending moment by plunging the tool down as deep as possible into the material so you're running on the shank (that would also improve accuracy since you're not locating on the flutes.) Realistically, though, the load here is honestly pretty small, easily <20 lb-f assuming the stock is well supported on both ends so it's not bending the tool with its weight (note the linear bearings being used to support the cantilevered end.) Accuracy can definitely vary, but my understanding was that that dimension wasn't critical, and if it was you could use this method to get close then indicate the holes in. The more critical dimension (distance to the edge of the material) is pretty much exclusively controlled by the vise. I'll agree making a fixture plate is a better idea, though.
|
|
#
?
May 27, 2020 18:10
|
|
|
Karia posted:I have ~8 years of industry experience. The motors wouldn't be the concern unless they're very under-spec'd. The only thing I'd be worried about is the spindle bearings: they're not designed to take static load. You'd need to minimize the bending moment by plunging the tool down as deep as possible into the material so you're running on the shank (that would also improve accuracy since you're not locating on the flutes.) Most of the CNC machines I'm familiar with are large capacity routers that machine nested parts out of huge blocks of stock, so they definitely aren't meant to move 10' x 20' x 3"+ thick sheets of stock around (you really don't want to put that kind of load on the spindle anyway, shank attached or not). Even if it isn't steel, lumber and plastics get really heavy at that size. EDIT: and yeah, when I mentioned that freakout it was from personal experience seeing the guys in the shop drill a 1/2" rod into the end of a piece of stock and then hook it onto the spindle head by jogging the spindle down to it and tightening a collet onto it before using the Z axis to lift the end and the X-axis to drag it onto the table. Some Pinko Commie fucked around with this message at 18:58 on May 27, 2020 |
|
#
?
May 27, 2020 18:56
|
|
|
biracial bear for uncut posted:Most of the CNC machines I'm familiar with are large capacity routers that machine nested parts out of huge blocks of stock, so they definitely aren't meant to move 10' x 20' x 3"+ thick sheets of stock around (you really don't want to put that kind of load on the spindle anyway, shank attached or not). Even if it isn't steel, lumber and plastics get really heavy at that size.  Yeah, I understand your hesitance! Moving well-supported segments of bar stock short distances on a low-friction bearing is one thing, but yanking around a (conservatively) 1100lb piece of wood with a router spindle using a crappy single point of support is a flat "nope." I hope that person no longer works with you. EDIT: Uh, just to clarify: I mean I hope they got fired, not that I hope that they died. Karia fucked around with this message at 19:38 on May 27, 2020 |
|
#
?
May 27, 2020 19:21
|
|
|
Yeah, they got fired, but not for that.
|
|
#
?
May 28, 2020 00:56
|
|
|
Got some aluminum plate, roll pins, and various accouterments on the way in. In the mean time, for the first time in a long time, I was able to make some cuts to test everything out. Homing, soft limits, spindle control all working great now. I found a used bimba air cylinder on ebay for $14 and the hoss machine designs for the power drawbar on archive.org, so that's the next thing this machine needs. All I wanted was an enclosure, just one enclosure, and I wouldn't give it to me. Since I didn't sing enough praises for Gecko customer support last time: they received my non-working G540, which I purchased second hand three years ago, and the original owner apparently bought in 2014, replaced a damaged board, and mailed it back to me for $0.00.
|
|
#
?
May 30, 2020 06:02
|
|
|
Ambihelical Hexnut posted:Since I didn't sing enough praises for Gecko customer support last time: they received my non-working G540, which I purchased second hand three years ago, and the original owner apparently bought in 2014, replaced a damaged board, and mailed it back to me for $0.00. That influences decisions on my part.
|
|
#
?
May 30, 2020 16:46
|
|
|
CarForumPoster posted:I have an industrial CNC machinist background, though I've been out of it for >5 years. There are many many examples of this done in industry and lots of tooling out there for it. Its almost certainly within the design limits of the pictured machine, but if theres any question, look up the specs. I ordered some stock to do it this way but got impatient today to see if I could achieve something similar with what I have on hand:  Currently there are three pins tapped in for positioning the bar in Y. The two bolts on the right are in threaded holes spaced the same as the holes we'll be drilling for the rails, so they can locate the bar in X and clamp it down. The corresponding two holes on the left are relief space for drilling the two new holes in the bar. The remaining three holes can be used for pins or clamps. So with this version I'd just run the code to drill two new holes, then unscrew, slide the bar down, and screw it in at those holes. Clamp at the bottom left corner with another screw head if necessary. I've got a much wider piece of aluminum bar coming in that should make it possible to do multiple sets of holes per pass but this is the basic concept. Think it'll work? Also: I'd never looked to thingiverse for drawer organization before but I've been burning up the 3d printer this week making bins and collet holders: 
|
|
#
?
May 31, 2020 02:11
|
|
|
gently caress it, let's roll: It worked perfectly as-is. Unscrew two screws, move it down six inches, screw in two screws, cycle start. Easy cheesy. I supported both ends with rollers and put some flat stock under the bar on the left end of the table to keep it shimmed level.
|
|
#
?
Jun 1, 2020 04:02
|
|
|
Ambihelical Hexnut posted:gently caress it, let's roll: I'm reminded of this so many times: 75% of machining is setup. Congrats on a simple, repeatable setup that accomplished the goal.
|
|
#
?
Jun 1, 2020 14:43
|
|
|
Thanks! Two y-axis rails (2x4) and one x-axis rail (2x2) bolted up. Getting the gantry mounted requires making plates to link each carriage cart and also hold upright supports for the x-axis. The carriage plates need to be 5" wide and 8.5" long, and I'll cut them from some 0.75" plate and use the mill just to drill the bolt holes since my Y-axis travel is only about 5" For the upright part I haven't decided, I could use a short length of extrusion or just mill a vertical plate. I also need to make a plate for the x axis carriage, and purchase a cheapo openbuilds z-axis for carrying the torch. On the framing side I've now got the right taps and materials to finish my legs/feet so I can work on that too. Might need more 3x2 tubing depending upon how flimsy it feels assembled.
|
|
#
?
Jun 2, 2020 07:17
|
|
|
After several further modifications my plate ended up being used for all the remaining steel plates for 45 degree brackets and table feet: Four legs, more sloppy rear end stick welding, and some screw-in adjustable feet. I need to add a second 3/8 nut on each foot to act as a jam nut for holding the height setting in position.  Time to start making the plates that will tie two linear carriages together and hold the gantry uprights:  I got some 5"x0.75" 6061 stock for this. It's almost maxing out ol' reliable.  Setting in pockets for the screw slots on the first side. These pieces are too big to mill in a single setup on my flyweight machine.  Since plasma tables are typically splashy I wanted a nicer electronics enclosure to protect everything. I got an Altelix (I think?) brand ABS enclosure with a metal mounting plate. This is my meanwell psu and leadshine mx4660 controller inside, should be just enough room for wiring and a THC.
|
|
#
?
Jun 7, 2020 04:07
|
|
|
Ambihelical Hexnut posted:After several further modifications my plate ended up being used for all the remaining steel plates for 45 degree brackets and table feet: Ooo cut me off a slice of that bar. Nothing like the thrill of hoping all your electronics fit in the enclosure!
|
|
#
?
Jun 7, 2020 04:33
|
|
|
 Aluminum wool. These bolt slots will allow each side's linear carriage to screw to the common plate, holding them together around the linear rail. I'll put an upright plate on top of this plate, into which the gantry beam will screw. Something I realized while working on this is that, because it's bigger than my travels, I can't locate my wcs origin from the edges which are only bandsaw cut. If I do, then when I flip the plate around to do the opposite side, I'll be indicating off the opposite cut end which might be many thousands different distance from the original four slots than the new one, since it's not exactly 8.500" long. No worries, I milled a 1"x1"x.125" pocket in the top center of the plate to establish a common reference which can be reached by both setups. This is an inelegant solution, but it'll work. Unfortunately I'm getting a weird USB Smoothstepper error about once an hour while running, so this is slow as gently caress. I am half done with the first of three plates.
|
|
#
?
Jun 8, 2020 04:32
|
|
|
mekilljoydammit posted:That influences decisions on my part. I am also influenced. (I'm sick of no-name crap off of fleabay and flamazon.)
|
|
#
?
Jun 8, 2020 06:06
|
|
|
Ambihelical Hexnut posted:
Does the smoothstepper send steps in realtime, or does it keep a buffer?
|
|
#
?
Jun 8, 2020 06:34
|
|
|
Trying to get CNC approachable has been a really big challenge for me at my maker space. I think that's changing. I finally got to trying out Easel today, and oh man is it great. I think folks who understand how to use the laser engravers will really be able to bridge that gap. Even Vectric VCarve is a tough learner software coming in from nothing. Easel's accessibility is really nice, and it has the key feature: simulate tool paths. If someone in your life is struggling to figure out CAM, it's a great entry point. To attest, I imported a picture and converted it to an SVG to import to Easel and engrave: https://i.imgur.com/iN4l4pd.gifv
|
|
#
?
Jun 29, 2020 07:02
|
|
|
I know double posting is in poor taste, but it's been like a week so: Does anyone know of any pendants that are compatible with Easel? The X-Pendant is promising, but the fella who makes them is on a break right now, and considering the most recent review is from August 2019, I'm not really holding my breath.
|
|
#
?
Jul 4, 2020 18:21
|
|
|
Hey dudes. I'm interested in making holes using a CNC router on the side of a box, where the surface would be about 60mm above the base. Most of the CNCs I see on Amazon have working areas of height 45mm. How would you approach this? Thank you.
|
|
#
?
Jul 6, 2020 05:36
|
|
|
I would disassemble the box. Even if you had a router with enough vertical clearance to handle it, it's going to be hard to fixture a hollow box in a way that it won't deflect under the tool forces. What is the box made of and how thick are the walls? Is it just holes, or are there slots that would require side cutting?
|
|
#
?
Jul 6, 2020 05:43
|
|
|
Thank you. ABS. ~3mm A mix of circular (Maybe drill press?) and 2 rectangular.
|
|
#
?
Jul 6, 2020 05:50
|
|
|
If it's just one box and it doesn't have to be laser-precise I would just do it by hand. Lay it out with a ruler, a scriber and a center-punch. The slots can be done with a drill and files. If it's a lot of boxes and/or it has to be laser-precise, get a mini CNC that has the correct amount of Z clearance and make some sort of fixture/spoilboard that goes inside the box, locates it precisely in place, and supports the wall to be drilled from underneath. It's a bit more setup but honestly 90% of good CNC work is having a good job setup before you turn anything on.
|
|
#
?
Jul 6, 2020 06:06
|
|
|
Thank you! Going to see what I can find. Wasn't even considering the force aspect.
|
|
#
?
Jul 6, 2020 06:23
|
|
|
Sagebrush posted:It's a bit more setup but honestly 90% of good CNC work is having a good job setup before you turn anything on. Machining in general. I spend 50% of my time figuring out the setup and fixturing for any non-standard machine work. There's a reason most machine shops have entire rooms of jigs if they do the same jobs over and over. Even my local job shop does this: a room full of soft jaws and jigs, a wall of filling cabinets full of setup sheets and prints with notes in the margins.
|
|
#
?
Jul 6, 2020 14:40
|
|
|
sharkytm posted:Machining in general. I spend 50% of my time figuring out the setup and fixturing for any non-standard machine work. There's a reason most machine shops have entire rooms of jigs if they do the same jobs over and over. Even my local job shop does this: a room full of soft jaws and jigs, a wall of filling cabinets full of setup sheets and prints with notes in the margins. Not machining, but it’s neat: I worked at a custom wire fabrication company that had literally 45 years of accumulated large wooden welding fit-up jigs on site, except instead of a dedicated room they suspended literally thousands of big plywood tooling assemblies from a sort of beef rail-esque motorized hanger system installed into the ceiling above the entire factory floor, with an extensive wire net shielding people from falling tooling. If by some insane chance a client from 1978 came back and wanted more of those racking units, you know the ones, they could pull the fit-up jig from that specific job (assuming anybody could find the tool ID after that long) and produce another. Probably a horrendous fire hazard, but yeah, illustrative of how the sunk labour of job set-up is a precious thing that you try to only have to do once if possible. Ambrose Burnside fucked around with this message at 16:45 on Jul 6, 2020 |
|
#
?
Jul 6, 2020 16:43
|
|
|
This is super fun because an ENORMOUS part of industrial grade 3D printing is for tooling, jigs, and fixtures rather than finished products. Now, you may not necessarily need to keep stuff around that long or if you lose a tiny part of your fixture, you can just reprint it instead of digging out there old drawings to remake it. Could be anything from specialized soft jaws to a stopper for your sine plate. Making stuff friggin owns, and all the ways to skin the cat make it so much fun.
|
|
#
?
Jul 6, 2020 17:57
|
|
|
NewFatMike posted:This is super fun because an ENORMOUS part of industrial grade 3D printing is for tooling, jigs, and fixtures rather than finished products. I just finished designing and printing a jig that holds an underwater connector at a 2.5° angle offset around the mounting hole because when you tighten it from the opposite side, the rubber twists a little in my previous jig and the connector ends up at a 2.5° angle. It's just enough to notice and tickle my "pictures not straight on the wall" sense. It took 30 minutes to design, 2 hours to print three of them, and I used them today to build a bunch of units. No more slight angular offset! I use FDM parts all the time as assembly holders, chassis service carriers, and regularly include them with the finished product to help my customers work on the products. The cost is negligible once the design is completed, and it's so cool to be able to make jigs without spending metal money.
|
|
#
?
Jul 7, 2020 01:19
|
|
|
I hope this is the correct thread to ask but if not could someone recommend me a thread please? I’m wondering if any goons have access to a 5 axis turn mill?
|
|
#
?
Jul 7, 2020 12:28
|
|
|
|
| # ? Apr 25, 2024 14:15 |
|
|
Bape Culture posted:I hope this is the correct thread to ask but if not could someone recommend me a thread please? Can also ask in Blacksmithing & metalwork: Anything involving penetration has to be anodized. I haven't seen anyone mention a mill turn or showcase parts that were clearly made on one but definitely plenty of knowledgeable lurkers.
|
|
#
?
Jul 7, 2020 12:37
|
|