|
I came really close to buying this for the power switch for my shapeoko. http://www.oreillyauto.com/site/c/detail/CTI0/84826.oap?ck=Search_skull+switch_-1_-1&keyword=skull+switch But I'm just a handful of crimp connectors away from having limit switches installed so that's nice.
|
#
?
Sep 25, 2015 02:49
|
|
|
|
| # ? May 22, 2024 12:21 |
|
|
And the limit switches are done and a couple successful home tests have been performed. 
|
|
#
?
Sep 26, 2015 21:42
|
|
|
I'm finally making the plunge in to cnc with my sherline lathe and mill. I'm getting a cbc ready lathe from Craigslist but it needs motors and a driver box still. It is set up for nema 23 motors. I will read the thread when I get a chance, but until then can anyone recommend a driver box and give me any good links to learn what all this stuff means? Such as nema and what kind of stats I need to know about when buying gear. My goals are to turn some stuff on my lathe and use it as a mill to do some engraving of a logo in a round part.
|
|
#
?
Sep 28, 2015 01:24
|
|
|
Brekelefuw posted:I'm finally making the plunge in to cnc with my sherline lathe and mill. I'm getting a cbc ready lathe from Craigslist but it needs motors and a driver box still. The only thing the nema classification tells you is basically the mounting screw profile. NEMA 23 means the motor face is ~2.3" square, and the screw holes are some kind of standard spacing inside that 2.3" square, which you can look up diagrams/dimensions for. Within NEMA 23, you can get relatively short or long motor casings, longer generally equating to more torque, and more current consumption. Also they inductance, resistance, etc can vary even with basically exactly same sized motors depending on how they wind the coils. The majority of steppers have 1.8 step angle, or 200steps/rev, but you can sometimes find them in 0.9, 400 steps/rev. I've only really dealt with smaller nema 17 motors so far, so I can't really recommend a driver box for nema 23. It should be rated for at least as much current as whichever motor you get. You can find some pretty cheap ~$20 ones on ebay that are good for 5A or so, not sure how reliable they are though.
|
|
#
?
Sep 28, 2015 04:08
|
|
|
This place sells kits that might interest you: http://www.cncrouterparts.com/nema-23-cnc-electronics-c-48_22.html?osCsid=s0j8ov6cfihia637re08k8sdt3 Or, at the very least, you can use their kits as a guideline for sourcing your own parts.
|
|
#
?
Sep 28, 2015 04:57
|
|
|
I have been doing some program research, and Autodesk Fusion 360 really seems to be what I want to hook my Sherline in to once i get my hardware set up. What I can't seem to find is what else i need to make it from the Fusion model to the physical part. Do I draw it and do toolpaths in Fusion and then output that G-code to Mach 3 or something like that, or does Fusion also provide a direct link to work the driver box and lathe? EDIT: The stepper driver I am looking at is the PMDX 422, as it is USB and seems to be something that would work. http://www.pmdx.com/PMDX-422
|
|
#
?
Sep 30, 2015 00:18
|
|
|
Fusion 360 has a CAM mode, so you generate the g code in it and it'll output a file. Then you send that file with something like UniversalGCodeSender to your machine.
|
|
#
?
Sep 30, 2015 00:40
|
|
|
Does anyone know if there is a rule of thumb for max plunge for a given cutter diameter for plastic like the (1/2 diameter of cutter) max for metal? Got a stencil to make that might require a 1/32" mill to do and stepdowns of <= 1/64" would add a lot of time to an already long job.
|
|
#
?
Oct 1, 2015 00:03
|
|
|
Vcarve pro defaults to 2x bit diameter for end mills, but it hugely depends on material and bit specifics and feed rates and all sorts of factors that it takes years to figure out by experimentation. With that in mind, there's a tool out there that attempts to provide that wisdom, but I forget what the darn thing is called. But you give it various parameters from your setup and it guides you on DoC and feed rates and such. I'm on my phone right now so I can't effectively dig it up myself.
|
|
#
?
Oct 1, 2015 17:21
|
|
|
G-Wizard Calculator or HSM Advisor?
|
|
#
?
Oct 2, 2015 00:22
|
|
|
Hah, I should have guessed software like that exists already. Thanks you two.
|
|
#
?
Oct 2, 2015 14:50
|
|
|
I've also had pretty good luck with this web page: http://zero-divide.net/?page=fswizard Not quite as robust as the software packages seem to be, but still beats doing chip load calculations on a napkin.
|
|
#
?
Oct 3, 2015 01:37
|
|
|
I think that I will buy this to run my Sherline. People on a Sherline CNC said they used them with good results. https://www.automationtechnologiesinc.com/products-page/cnc-3_axis_4_axis_kits/nema-23-stepper-motors-4-axis-kit-3 I will use 185oz steppers though, because they also recommended not going too big with the motors. I also will get the USB Motion controller so I can run it form USB once I find a program to run the G code. I am so exited to get started on this stuff! Now I just need to figure out how to engrave on a radius so I can do serial numbers etc.
|
|
#
?
Oct 6, 2015 01:29
|
|
|
Ambrose Burnside posted:So I know I want a Taig mill, I'm just not sure who I wanna buy it from. Things I'm interested in- This is super-old and stuff but I'll answer it anyway. I've been reading the taig mailing lists for a while. DSLS from what I understand is a system to let your steppers know that something has gone wrong and they've missed a step. Unfortunately, there's not much you can do about it - it isn't able to correct the error, just note that it's there. The deepgroove is just a taig cnc-ready thing with all the stuff set up. Nobody ever talks about it, but that's can be good too. The guy who runs Soigeneris is Jeff Birt and he really seems to know what he's talking about. In particular, he made a very persuasive argument that bigger motors are not better, and that you need to match the motors to the machine and that the 280oz-in motors most people use are in fact nonoptimal. I don't know enough things to say if he's right, but no one on the list seemed to disagree significantly. If I was doing it all again, I'd go with soigeneris. rotor fucked around with this message at 21:36 on Oct 15, 2015 |
|
#
?
Oct 15, 2015 21:34
|
|
|
I went with 185oz motors over the 280oz ones because someone said the same thing to me when I asked on the Sherline CNC group. All my parts just came in the mail today. Can't wait to get it all running. Hopefully I will have it wired in the next week or so.
|
|
#
?
Oct 15, 2015 22:41
|
|
|
Similarly, I've been looking off and on for more powerful steppers for my shapeoko 2. There's plenty of NEMA 23s but usually the current draw is too high for the gshield. Found this 115 oz-in NEMA 17 yesterday though: http://www.automationdirect.com/adc/Shopping/Catalog/Motion_Control/Stepper_Systems/Stepper_Motors_-z-_Cables/STP-MTR-17060 In addition to being more powerful (I think about double over what used to come with the inventables kit) I won't have to mess with changing the mounts and such for NEMA 23 size, which is really nice. Still got to figure out just how in the gently caress I'm going to approach converting my mini mill over to cnc. No way I'm blowing $400+ on an incomplete kit at the moment.
|
|
#
?
Oct 16, 2015 22:00
|
|
|
rotor posted:This is super-old and stuff but I'll answer it anyway. It's useful anyways, thanks.
|
|
#
?
Oct 16, 2015 22:08
|
|
|
This is an interesting way to keep chips out of the ways. Guy used some sort of tensioned curtain that retracts or extends as the machine moves. https://www.youtube.com/watch?v=HK_5G3VHaak
|
|
#
?
Oct 26, 2015 21:29
|
|
|
Parts Kit posted:Guy used some sort of tensioned curtain that retracts or extends as the machine moves. It's probably originally something like this: http://www.amazon.com/Car-Window-Shade-Pack-Satisfaction/dp/B00YGJE7FK He may have replaced the vinyl to make it longer.
|
|
#
?
Oct 27, 2015 00:23
|
|
|
"Here is a Way Cover setup I made from a Window Shade I got from Lowes. Grey material is pvc shower moisture barrier material from lowes also." Smart. That's an awesome idea.
|
|
#
?
Oct 27, 2015 01:54
|
|
|
Very cool idea on the cover. Well, I got my board all wired up and hooked up my x axis. All the lights are on and Mach3 recognizes the parallel-USB driver chip etc, but when I press my arrow keys the motor doesn't move. Mach3's DRO displays the coordinates as if it is moving the motor though. Now starts the fun trouble shooting so I can get to hooking up the remaining 3 axis. Also, I have a load of questions but I will save them until I get my motors actually running.
|
|
#
?
Oct 27, 2015 03:36
|
|
|
That curtain looks clever as hell but there's a few points where his tooling can shred the sheeting because of the angle. He should have some kind of horizontal roller to keep it from sloping too shallow.
|
|
#
?
Oct 27, 2015 07:50
|
|
|
gently caress yes! reconfigured Mach3 and got both my X and Z axis going strong. Now I have to make a test run using the g code and then its chip making time!
|
|
#
?
Oct 29, 2015 03:03
|
|
|
Those style of covers are very popular in machine shops http://www.dynatect.com/protective-covers/roll-up-covers Just an example, I've never used this company.
|
|
#
?
Oct 29, 2015 18:11
|
|
|
Mofabio posted:So I'm about 1/3 of the way through converting my Grizzly G0768 8x16 lathe to cnc+VFD. I figured I'd start dropping pictures here in case there's any interest in a probably-longer-than-anticipated build log. Phew, okay. Finally hooked up the VFD. It's using a Rockwell half-horse I got for free at some seminar thingy, a $30 Baldor 3-phase motor, and way-too-expensive everything else.  Why does poo poo take so long? Because I only have a half-busted cordless drill. Cordless drill + 2.75" hole saw + 5/16" hot roll steel = lots and lots of cursing. After chewing through like 5 bits (yes I used lots of oil), I caved and took it to a machine shop. Plus side: it got done. Downside: it took a month for them to do it. Oh well, at least it was cheap. What's left? Going to direct-drive the z-axis with a G540 + 381oz stepper (re: stepper power discussion, all I did was read the G540 driver spec sheet and found the one Keling stepper that met the G540's rated inductance). Going to timing belt the x-axis (no idea how to tension the belt yet- maybe just get the girlfriend to hold it down). I have a buttload of aluminum square stock from another project, which I'm gonna use to make a custom apron that'll hold both the z-axis ballscrew car and the x-axis stepper. My goal's end of 2015. We'll see! Mofabio fucked around with this message at 22:51 on Nov 10, 2015 |
|
#
?
Nov 10, 2015 00:52
|
|
|
If anyone else is trying to bring svgs from Illustrator to Fusion 360, a scaling factor of 0.0185 seems to get fusion 360 to give you the resulting size you actually wanted. (Otherwise nice little 8x8 graphics for engraving become hundreds of inches in size)
|
|
#
?
Nov 15, 2015 20:14
|
|
|
Parts Kit posted:If anyone else is trying to bring svgs from Illustrator to Fusion 360, a scaling factor of 0.0185 seems to get fusion 360 to give you the resulting size you actually wanted. (Otherwise nice little 8x8 graphics for engraving become hundreds of inches in size) SVG is terrible for CAD because there is no way to use real world units, its all converted to pixels based on an arbitrary DPI decided by whichever software you are using.
|
|
#
?
Nov 15, 2015 21:22
|
|
|
Has anyone else made eyes at Inventables' Carvey? They should be shipping like any day now if their site is to be believed. Despite the fact that it looks overkill for my simple needs of carving names into plastic and wood I'd jump on one based on the fact that it's enclosed and junk if nothing else and the software looks pretty mature. I'm not interested in fiddling with getting or keeping a tool working as my new hobby. Has anyone had experience with a Carvey?
|
|
#
?
Nov 15, 2015 23:28
|
|
|
Not unless they're on the development team I'd guess.
|
|
#
?
Nov 16, 2015 00:19
|
|
|
peepsalot posted:SVG is terrible for CAD because there is no way to use real world units, its all converted to pixels based on an arbitrary DPI decided by whichever software you are using. Mister Sinewave posted:Has anyone else made eyes at Inventables' Carvey? They should be shipping like any day now if their site is to be believed. The other thing is the X-Carve and Shapeoko have a lot more potential in them -- you could turn them into lower power laser engravers or draw knife setups. Really just about anything assuming you can mount it to the z-axis and find a way to control it. And if a smaller home office usable machine is what you want you could go much, much further with a mini-mill conversion, albeit with the caveat that you'll have to come up with an enclosure to keep it from getting crazy messy. Pretty sure you'll spend a lot less than $2k on that too since last I checked mills are ~$600ish, conversions about $400-$500, and a tiny g setup + steppers another $250-$400. A mini mill would have the option to use R8 collets (which go from really god drat tiny to I think 3/4"?) instead of just 1/8" or 1/4" shank bits. So you'd have a fuckload more options for cutters. Also 4th axis capable thanks to the much much more generous Z-axis range on mini mills. Seriously this is way cooler than a Carvey. Eyeball guesstimate would be about 10" by 6-7" by 9-10" work area. https://www.youtube.com/watch?v=kf1awhU2UOc
|
|
#
?
Nov 16, 2015 00:45
|
|
|
So I posted in the 3D printer thread, but I'm looking to get into CNC. Where's a good place to start? So far I've found a Shapeoko 2 on eBay here. Is getting the 3 worth it? If I were to pick up the 2 on eBay, is the X-Carve worth it?
|
|
#
?
Nov 16, 2015 01:12
|
|
|
If money is an issue you can get the shapeoko 2 and install some of the revised parts that make up the X-carve, like the one piece X-axis gantry and the revised Z-axis plate. Both are in the plates and carriages section of Inventables. That said the buy it now price is not great as that's what they were selling for during the sale shortly before the x-carve was announced. Also the dremel tool knockoff is horrid and should be immediately replaced before you start cutting anything. God that thing pissed me off when I tried to use it to make some mounts for a makita router. If you got it for the opening bid it would be a respectable deal. Also you can get drag chain cheaply off ebay and microswitches from amazon. I'd suggest getting both beforehand if you went with the shapeoko as starting off with them will make life easier. Similarly getting a premade router or spindle mount from Inventables would make life a lot easier over fabricating one.
|
|
#
?
Nov 16, 2015 01:35
|
|
|
So how do i connect the switches? Is there a wiki or something about how to do these upgrades?
|
|
#
?
Nov 16, 2015 01:39
|
|
|
Yes, it's actually really easy though the wiki isn't very clear on some of it. Give me a moment and I'll dig up the link and post specifics.
|
|
#
?
Nov 16, 2015 01:48
|
|
|
Thanks. So if I get that Shapeoko 2 at the bid price, spend $150 or so on a good router and the mount, maybe $50 more on upgrades that's about $650. That's a nice price to start
|
|
#
?
Nov 16, 2015 02:01
|
|
|
Okay, here's the links you need to read. Link 1: connections https://github.com/grbl/grbl/wiki/Connecting-Grbl Link 2: configuration https://github.com/grbl/grbl/wiki/Configuring-Grbl-v0.8 (I'm assuming he has a v.08 board, there's instructions for others on other pages and reflashing it shouldn't be too hard if it's older) Link 3: summary of the possible options on the shapeoko wiki. Has some diagrams, which is nice. http://www.shapeoko.com/wiki/index.php/Home/Limit_Switches Materials needed: 1. Switches with a normally open pin. Microswitches are good and the 3 pin ones have normally open (NO) and normally closed (NC) pins. Just make sure you're using the right two pins. 2. Wire for the connections. I was lazy and bought a long (25ft?) thing of telephone cable since it's flexible and has 4 wires in it. 3. Female quick connects, spade style, if you don't want to solder to the switches' terminals. I think I needed 1/8". You can get these (and a crimper) at most auto parts stores. 4. soldering stuff 5. 3 capacitors, 0.01uF 6. Arduino "protoshield" 7. something to make mounts with, and appropriate screws/washers/nuts (probably M3). I'm pretty sure all of that can be bought on amazon. What you are going to do is use the protoshield to set up the wiring for the switches, with the capacitors in parallel, so that the switch will bring one of the three (specific!) pins on the arduino uno into a connection with ground. When the arduino notices one of those three pins going to ground it realizes (very quickly!) the machine has reached the end of its travel and shuts down to prevent breaking belts or burning up motors. Assuming the guy hasn't messed with the default config those are pins 9 (x), 10 (y), and, 11 (z). On the protoshield there will be a pad for each pin, once you've identified 9, 10, and 11 you will run a wire from that pad to the interior part of the protoshield, and then have that go in parallel with the cap and the input switch pin. The other switch pin (NC) will then lead to ground, as will the other end of the cap. There's a nice long ground rail on the protoshield so what I did was put all the terminations on that rail so I didn't go insane doing extra soldering. Ideally you'll want two switches per axis so you don't crash into the end of either direction, and their wires can just be twisted together at the arduino end for simplicity (if you use telephone cable you'll have 4 nice wires to play with so this is easy). What you have to watch out for is getting axes mixed up or having a circuit cross two axes. After everything is wired up on the protoshield correctly you just sandwich it between the arduino and the gshield. After all that stuff is done you need to make a few tweaks to firmware settings on the gshield so it knows limit switches are installed. The codes and explanation are in the second link, but here's the short version. When you open up something like Universal G Code sender there is a spot you can use to send strings of info (commands, etc) to the gshield. Do the following: $$ $16=1 $17=1 then $$ again to confirm the settings took and variables 16, 17 are both 1. The above will tell it to use limit switches and tell it you want to do a homing cycle (setting 17). With the homing cycle every time the arduino starts up it will go to the furthest positive position possible on all axes. Once it has reached all three it will stop. So you have to make sure you at the least have limit switches where it will want to go during homing (1 per axis), though 2 per axis is better in the long run for keeping the machine from loving itself to death if you mess up your CAD/CAM. After it has homed you'll have to deactivate the alarm flag that goes off when a limit switch is hit (command $X, or possibly a button in your software may do it), move it, and soft reset it, which will set that new point as the new origin (0,0,0). Make a note how much you move the axes before doing the soft reset, that way if your bit breaks and you have to shut down and change out bits you can then let it do another homing cycle, return to the former position you wrote down, then do the soft reset, and run the CAM file again. And everything should be (reasonably) lined up how it was! Way better than having to grab rear end it and hope you're not too far off. Universal G Code Sender has a position readout you can use to accomplish this, and if you have a fixture set up you can configure the gshield to always go to a (X,Y) reference point on it after homing for faster repeated operations. I haven't messed with that though. It's in the configuration link at the top if you are interested. If you have any questions on wiring, I did as well and here's where I was asking smarter people about it. http://forums.somethingawful.com/showthread.php?threadid=2734977&pagenumber=320&perpage=40#post450401538 The above stuff isn't that bad, it just takes a while to work through. What's really annoying about this whole process will be figuring out how to mount the switches and where. Google around and you'll see plenty of options, some of which you can download plans and 3d print if you've got a printer already setup. I hope that was clear, let me know if anything was confusing.
|
|
#
?
Nov 16, 2015 02:32
|
|
|
BiohazrD posted:Thanks. So if I get that Shapeoko 2 at the bid price, spend $150 or so on a good router and the mount, maybe $50 more on upgrades that's about $650. That's a nice price to start It's cool the X-Carve and Shapeoko 3 take into account some of these updates and include stuff the 2 didn't, but drat the price jumped from about 600 to about 1k. ed: oh yeah if you want to upgrade the stepper motors (the included ones are wimpy) there's a link in one of my recent posts here where I found some NEMA 17s that are about 115 in-oz, supposedly about the max power the 2 can really make use of (IIRC about 120-130 is the max). Most of the other motors (usually NEMA 23s) I've seen close to that power draw more current than the gshield can source safely, but the 17s linked don't and won't require getting new pulleys and poo poo like the 23s would. The stock motors are about half as powerful. That'll add another $150ish to your total though. Parts Kit fucked around with this message at 02:37 on Nov 16, 2015 |
|
#
?
Nov 16, 2015 02:33
|
|
|
Sweet, that should be easy enough. I have tons NC/NO switches from the absurd amount of microwaves I've ripped apart for parts. Not to mention boxes of every connector imaginable.
|
|
#
?
Nov 16, 2015 02:39
|
|
|
Parts Kit posted:Yeah, I know. But SVG is great for graphics, and sometimes I want to engrave stuff I made in illustrator so I'm kinda stuck on that front. Use PDF, dude.
|
|
#
?
Nov 16, 2015 02:47
|
|
|
|
| # ? May 22, 2024 12:21 |
|
|
How? ed: if we're talking non-cnc stuff then no duh. Also appears that Fusion 360 can't import pdfs so eh, still stuck with svg. Parts Kit fucked around with this message at 02:56 on Nov 16, 2015 |
|
#
?
Nov 16, 2015 02:52
|
|