|
ps they're now selling refurbished mills for $600 because people have been swapping ones with the v threads in for the ballscrew versions. At $600 it's a bit of a no brainer imo
|
#
?
Mar 12, 2017 17:21
|
|
|
|
| # ? May 20, 2024 00:44 |
|
|
rotor posted:I think they have a distributor problems. As in, they don't really have very many. My budget is flexible and I'm not trying to go completely starving-student. Just trying to stay aware of all of the tooling and holding stuff I'll also need to buy too. I was looking at this Taig, but I'm not sure how much I care about the 4th axis, and I assume he's asking too much since it's been listed for weeks. https://kansascity.craigslist.org/tls/5979180393.html More and more I'm thinking I want something larger than a mini-mill. I'd like the ability to mill most of the usable-width of a 19" equipment rack blank and the G0704 is looking pretty tempting with 18-7/8" of horizontal travel. There's a Grizzly store a few hours south, so I'm hoping to stop and look in person.
|
|
#
?
Mar 12, 2017 18:15
|
|
|
eddiewalker posted:the G0704 is looking pretty tempting with 18-7/8" of horizontal travel. Horizontal travel is misleading for two reasons: 1) you don't get full use of working across that whole length when you consider the work-holding required to fit something like that. 2) milling machines are less accurate and stable at the extreme ends of their travel, and its error is only more exaggerated on smaller hobby machines.
|
|
#
?
Mar 12, 2017 20:09
|
|
|
I'm expecting some tweaking and accuracy issues. Hopefully I'm not biting off more than I can chew. The BF20-style mills like the G0704 look so much more formidable than a mini mill side-by-side for not a whole lot more money. That's probably just about maxing out my available garage space for future-proofing.  I'm interested in the Hossmachine DVD of instructions and CAD drawings to "reprap" a CNC conversion in stages using the machine itself. Seems like a solid set of data on one place and a good way to become really familiar with the process and stretch out the initial cash outlay.
|
|
#
?
Mar 12, 2017 22:27
|
|
|
eddiewalker posted:My budget is flexible and I'm not trying to go completely starving-student. Just trying to stay aware of all of the tooling and holding stuff I'll also need to buy too. yeah that seems like way too much. I paid iirc ~1700 for the full cnc setup w/ controller & motor & stuff. quote:More and more I'm thinking I want something larger than a mini-mill. I'd like the ability to mill most of the usable-width of a 19" equipment rack blank and the G0704 is looking pretty tempting with 18-7/8" of horizontal travel. There's a Grizzly store a few hours south, so I'm hoping to stop and look in person. get as large as your shop can fit and you can afford imo. Tormachs are nice. :-) I went with the taig primarily because of it's size - my shop space is tiny.
|
|
#
?
Mar 13, 2017 06:24
|
|
|
help me get my workflow up and smooth and you're welcome to runtime on my machine *edit* Have any of you guys tried tooling around with Mach4 - is there any reason to upgrade from Mach3? CrazyLittle fucked around with this message at 18:44 on Mar 13, 2017 |
|
#
?
Mar 13, 2017 17:57
|
|
|
Well it's one bigger.
|
|
#
?
Mar 14, 2017 07:31
|
|
|
Is there an mpg or something that I can hook directly in to my motors to run them, instead of having to hook up my computer and mach 3 and my BoB etc? My lathe is being used manually and CNC, so being able to jog the motors without the rest of the setup would be great.
|
|
#
?
Mar 14, 2017 17:11
|
|
|
Sagebrush posted:Well it's one bigger. That's true. If one of the 4 Machs go bad at least I would still have 3 leftover
|
|
#
?
Mar 14, 2017 18:01
|
|
|
I'm thinking about building a soundproof (-dampening) enclosure for my Taig. Right now it's at a shop I get out to maybe once every few weeks where sound isn't an issue, but I now have the space to set it up in my apartment so if I can manage the noise I think it'll be viable. I gotta build an enclosure sooner or later, might as well be one that manages sound if it'll help me right now. An 80/20 extruded frame with MDF/acrylic panels is probably my go-to approach here, but 1/4" MDF sidewalls aren't going to muffle poo poo. What approaches to people take here? I know, broadly speaking, 1) seal the sound in like it was water and 2) use either mass or vibration isolation to manage the rest, I'm just not sure how that's worked out for other people (particularly if they used 80/20).
|
|
#
?
Mar 18, 2017 00:21
|
|
|
Ambrose Burnside posted:I'm thinking about building a soundproof (-dampening) enclosure for my Taig. Right now it's at a shop I get out to maybe once every few weeks where sound isn't an issue, but I now have the space to set it up in my apartment so if I can manage the noise I think it'll be viable. I gotta build an enclosure sooner or later, might as well be one that manages sound if it'll help me right now. For my 3D printer, I used 4" of rockwool on 4 sides wrapped in thin fabric, and dampened the steppers with cork. It's 2m from my pillow and after soundproofing, it's possible to sleep through a print. Warning though, rockwool isn't sold in small quantities. I've been trying to use up the $60 worth of batts on other projects (speakers, ground insulation for snow camping, hydroponics). You might be able to offload your excess on craigslist as well.
|
|
#
?
Mar 18, 2017 07:16
|
|
|
I think ive followed the trajectory of a lot of people in regards to "oh hey aluminum extrusions are great" *checks linear foot pricing* that's, fine, I can work with that, *actually prices out an enclosure with fasteners and hardware* well uh hmmmmm. If it's gonna be tabletop instead of a standalone workstation, which makes sense for me right now, I think an extrusion enclosure stops making a ton of sense if I'm on a budget. Frees up more money for focusing on sound-deadening to boot. I like this kind of approach: https://www.youtube.com/watch?v=Yxa8HH3t_Ds With the visible interior enclosure floating on a rubber/foam pad surrounded by some kinda rockwool-esque filler. Only big downside for me is it'd be heavy as hell, and I may have to move this thing a couple times in the next few years. For sound-deadening overkill i'm also admiring this enclosure, which ditches viewports entirely and only permits observation of the process via a webcam, which definitely sounds really silly until I price out decently-thick acrylic/polycarb and think about the webcam I already own and don't use: https://www.youtube.com/watch?v=8HL5R518NMo
|
|
#
?
Mar 19, 2017 19:25
|
|
|
Ambrose Burnside posted:I think ive followed the trajectory of a lot of people in regards to "oh hey aluminum extrusions are great" *checks linear foot pricing* that's, fine, I can work with that, *actually prices out an enclosure with fasteners and hardware* well uh hmmmmm. Haha yeah, the cost for 80/20 stuff definitely adds up fast. I recently priced out an order for two 12" extrusions and enough associated hardware to make a motor mount for my lathe and it was almost $60. This was through the 80/20 Inc Garage Sale eBay store which has mostly surplus / returned / damaged items too. Bought it anyway 
|
|
#
?
Mar 20, 2017 06:01
|
|
|
Ambrose Burnside posted:I'm thinking about building a soundproof (-dampening) enclosure for my Taig. I built a sound enclosure at work out of 1.5" extrusion and double paned 1/4" polycarb paneling. We were peaking at 85db before and got down to 75db after. Still noisy but direction was to get under 80db so we could have longer shifts in production. We use TSLOTS brand extrusion and I had panels on both the inside and outside faces. I also had aluminum angle extrusion overlapping the gaps around the access doors to block the sound paths. For our laser enclosures we think "tortured path" just like your water analogy. For home I'd use maybe 1/2" MDF over a lumber frame to make a 5 sided box and some locating features to help guide it on and off. Might be tricky dealing with chips but some planning can minimize the mess. Maybe the guide features can act as a catch tray or something.
|
|
#
?
Mar 20, 2017 06:38
|
|
|
If you stuffed rock wool between those double walls would you see a significant improvement do you reckon?
|
|
#
?
Mar 20, 2017 09:55
|
|
|
DethMarine21 posted:Haha yeah, the cost for 80/20 stuff definitely adds up fast. I recently priced out an order for two 12" extrusions and enough associated hardware to make a motor mount for my lathe and it was almost $60. This was through the 80/20 Inc Garage Sale eBay store which has mostly surplus / returned / damaged items too. Bought it anyway It definitely makes sense for that kind of thing, I'm looking at some of the 1x3 or u-channel off-brand stuff for some adjustable jigs because it beats the hell out of mapping out (and then actually machining) all kinds of finicky dowel holes and miter/t-slots. It makes a lot -less- sense for an enclosure where that expensive-rear end extrusion is putting its engineering to work as a table leg or structural member that's never actually going to need almost any of its functionality over its lifetime. In any case, I just remembered that I have a bunch of salvaged 1/16" sheet lead that's in good shape minus some tack-holes, and apparently 1/16" of lead has the same acoustic absorption as two 1/4" Plexiglas panels with an air gap, which'll probably do very nicely as an intermediate liner up in there with the rockwool between layers of MDF. Ambrose Burnside fucked around with this message at 15:35 on Mar 20, 2017 |
|
#
?
Mar 20, 2017 15:20
|
|
|
Just do some amateur carpentry and make a insulated sheetrock covered box made out of 2x4s. It won't look pretty, but it'll be quiet. Hard to see through though!
|
|
#
?
Mar 20, 2017 17:43
|
|
|
cakesmith handyman posted:If you stuffed rock wool between those double walls would you see a significant improvement do you reckon? Yes, if we didn't hit the sound goal with the double wall we were going to stuff some insulation in there. We didn't do it in the first place because it is a cleanroom and non particulating insulation is costly and it would impede the view of the machinery which needs periodic operator interaction. I only took one acoustics class in school but the math is very similar to heat transfer through walls or electricity through resistors so in my case the sound excites the polycarb, then the polycarbonate excites the air gap, and the air gap excites the outer polycarbonate and that excites the room air. Each interface costs more energy. The frame is the easiest path for the sound because it is just a piece of aluminum. I'm recommending soft lumber wood because it doesn't transmit vibration like metal will. Sound control is a huge science and you can spend a lot of time thinking about it. I recommend doing something cheap and fast to minimize your investment and start testing.
|
|
#
?
Mar 20, 2017 18:28
|
|
|
Methylethylaldehyde posted:Just do some amateur carpentry and make a insulated sheetrock covered box made out of 2x4s. It won't look pretty, but it'll be quiet. Hard to see through though! Yeah, that's the proximate plan. Mostly trying to figure out the dimensions (machine plus maximum travel plus a little wiggle room for weird parts and fixturing) and what sort of accommodations i should make to minimize possible cutting fluids from messing things up. I'm also gonna make some accommodation for cooling the spindle motor, NYCCNC's enclosure had a computer fan + duct with a baffled outlet elsewhere and it wouldn't take much extra work. Throw a metre of LED strip along the top edge and deal with a cable passthrough and i'm laughing. Ambrose Burnside fucked around with this message at 18:36 on Mar 20, 2017 |
|
#
?
Mar 20, 2017 18:33
|
|
|
Ambrose Burnside posted:Yeah, that's the proximate plan. Mostly trying to figure out the dimensions (machine plus maximum travel plus a little wiggle room for weird parts and fixturing) and what sort of accommodations i should make to minimize possible cutting fluids from messing things up. Make it 6" bigger than you think you need, because bigger than poo poo you'll need to add something and the box will be exactly half an inch too narrow. Other than that, you can do whatever you like. Sheetrock has some nice sound deadening properties, but doesn't like moisture at all. The inside I'd just use plywood, seal it with urethane, then put some moisture barrier to plastic up to keep the water off it. Fill the inside with either the rigid foam, or stuff it full of the pink panther poo poo. The front door can hang on fence gate hinges, and you can use sticky door sealant tape to make a gasket for it. The closure side can be a simple hasp or cam action latch. You can also put a window in the door made of 1/4" plexi or polycarbonate so you can see that it's failing in real time. Or you can get one of those NEST wide angle webcams or cheapo chinese equivalents and stick it inside to watch stuff if the box is opaque.
|
|
#
?
Mar 21, 2017 16:04
|
|
|
Methylethylaldehyde posted:You can also put a window in the door made of 1/4" plexi or polycarbonate so you can see that it's failing in real time. this really cuts to the quick of it for me e: but yeah, thanks- I'm making it more complicated than it has to be because I'm good at that, just need to actually build the thing and see how it shakes out before i get too worried about little details.
|
|
#
?
Mar 21, 2017 22:57
|
|
|
Ambrose Burnside posted:this really cuts to the quick of it for me It's more any CNC code you ever write is gonna gently caress something up, the trick is to figure out what that is and fix it before the cutter breaks or the part does. It's only after you get the code tuned for the tool, workholding, and machine that you can walk away and let it do it's thing.
|
|
#
?
Mar 22, 2017 01:14
|
|
|
what are you talking about, i dont make those kinds of mistakes *tries not to look at bent 1/4" plate that looks an awful lot like someone crashed a chuck into it with Haas toolroom mill-spindle-like force at full rapids*
|
|
#
?
Mar 22, 2017 03:14
|
|
|
 When I import my NC file from eCam to Mach3, the toolpaths are completely wrong and I don't know why. The main window in the image is Mach3, and the little one int he black rectangle is the eCam toolpath as it's supposed to be. I tried running it on my machine anyways just to test, but it made lots of noises and didn't move anywhere near what it was supposed to. CNC is no fun right now. I really thought I was close to getting the machine actually making chips. EDIT: eCam output using their FANUC 2 axis lathe post. Brekelefuw fucked around with this message at 03:17 on Mar 24, 2017 |
|
#
?
Mar 24, 2017 03:13
|
|
|
Brekelefuw posted:
Lathes can be confusing like that, especially in very old times. How old is the lathe youre using this one, which Fanuc control? A 1980s vintage Fanuc 6 series will have different positioning methods than a much newer one.
|
|
#
?
Mar 24, 2017 03:35
|
|
|
CarForumPoster posted:Lathes can be confusing like that, especially in very old times. How old is the lathe youre using this one, which Fanuc control? A 1980s vintage Fanuc 6 series will have different positioning methods than a much newer one. It's a Sherline running a UC-100 smooth stepper with 4 KL-4030 stepper drivers (so I can run my 4 axis mill as well as the lathe) and a C10 breakout board.
|
|
#
?
Mar 24, 2017 04:01
|
|
|
Brekelefuw posted:It's a Sherline running a UC-100 smooth stepper with 4 KL-4030 stepper drivers (so I can run my 4 axis mill as well as the lathe) and a C10 breakout board. In that case I have no idea.
|
|
#
?
Mar 24, 2017 13:57
|
|
|
On my end a lot of canned functions involving arcs freak the gently caress out when I move em from cambam to mach3, 'translating' between gcode 'dialects' seems to cause a lot of weirdness on the hobby level where things aren't nearly as streamlined as with industry-important dialects. And gcode for lathes can get much weirder/run a much wider gamut than with boring ol 3/4-axis mill programs, which would make things worse. IIRC you don't know gcode, right? You're gonna have to gently caress around with the gcode. Writing my first real program was like pullin teeth but it got me to being able to troubleshoot and modify generated programs, which is the real value. It looks like a real simple program you're trying to run, if I were in your shoes I'd consider sitting down and trying to write it by hand with the aid of gcode references and sample programs. Test it with a simulator program, maybe do some air cuts to know you have the broad strokes right. It's a huge pain in the rear end, nobody does it this way normally, but it's kinda essential for grasping what's going on under the hood. Now look at your working program and then at the generated one. What's different? What's the rhyme and reason behind what it's doing? Is there a pattern to how it's skewing your original directives? When you figure that out, you can work out what's getting lost in machine translation, and adjust for it going forward. Ambrose Burnside fucked around with this message at 17:30 on Mar 24, 2017 |
|
#
?
Mar 24, 2017 17:27
|
|
|
Ambrose Burnside posted:On my end a lot of canned functions involving arcs freak the gently caress out when I move em from cambam to mach3, 'translating' between gcode 'dialects' seems to cause a lot of weirdness on the hobby level where things aren't nearly as streamlined as with industry-important dialects. And gcode for lathes can get much weirder/run a much wider gamut than with boring ol 3/4-axis mill programs, which would make things worse. Use Fusion360, super simple, easy to do, powerful as gently caress. When you post the code, converting it from a CAM plan and model to the actual G-code, the port-processor does a ton of work to make the output code comply with whatever hosed up specific flavor of the G-code implementation they use. Every single machine has a very specific way it implements anything more complex than G0/G1, even simple arcs can have specific issues, where some controllers don't want lines longer than 40 characters, or more than 5 decimals worth of J or whatever. For shits and giggles I compared a bog-standard mach3 setup to the most special snowflake controller I could find, and it can be 110% impossible to reconcile. Canned functions are machine controller specific.
|
|
#
?
Mar 28, 2017 19:09
|
|
|
I would absolutely advise learning the major gcode commands for a CNC lathe. Probably more than half your parts can be made with just G1/G0. To start, try writing a basic turning program with different depths and feedrates for roughing and finishing.
|
|
#
?
Mar 29, 2017 17:07
|
|
|
Yep. I've been going through the code that the cam program spat out to see what the commands meant, and removing any that didn't apply to my setup. I've also got out my CNC textbooks to start exploring writing my own code for the simpler things I need to turn.
|
|
#
?
Mar 29, 2017 18:00
|
|
|
Methylethylaldehyde posted:Use Fusion360, super simple, easy to do, powerful as gently caress. When you post the code, converting it from a CAM plan and model to the actual G-code, the port-processor does a ton of work to make the output code comply with whatever hosed up specific flavor of the G-code implementation they use. Every single machine has a very specific way it implements anything more complex than G0/G1, even simple arcs can have specific issues, where some controllers don't want lines longer than 40 characters, or more than 5 decimals worth of J or whatever. Yeah, after seeing thisoldtony farting around with Fusion I'm getting into it and it's fantastic. I haven't put much time in on my Taig lately but when I get the chance I can hopefully kick Cambam to the curb.
|
|
#
?
Mar 29, 2017 19:05
|
|
|
...speaking of which, finally got some Me Time in the home shop and it feels good to be noticeably-better at this. Much less lost than I used to be and I don't rely on the software nearly so much any more. It's nice to be like, oh, I wanna do a couple simple profiling cuts, lemme just jog over to where I want to be and tab over to MDI, G91, write a couple lines of gcode and get good results faster and more reliably than i used to. That and understanding the quirks of smaller-scale milling. Earlier on I got lovely results and blamed the mill tram but a lot of the time it was just not grasping small tool rigidity and how, no, if you try to take a meaningful cut off the side of a block with the entire side surface of a 3/16" endmill, the fucky results aren't because of column tilt.
|
|
#
?
Apr 3, 2017 03:35
|
|
|
I am thinking of getting into some small scale cnc routing, and wanted to know just how rubbish the Chinese diy kits are. I don't need a large workspace, 30cm *30cm would be enough for my purposes, and it would just be wood and maybe acrylic.
|
|
#
?
May 1, 2017 07:30
|
|
|
NPR Journalizard posted:I am thinking of getting into some small scale cnc routing, and wanted to know just how rubbish the Chinese diy kits are. I don't need a large workspace, 30cm *30cm would be enough for my purposes, and it would just be wood and maybe acrylic. But I am confident in saying that if this is your first CNC experience, no matter what route you go make sure what you get is an actual kit that you put together yourself. When something goes wrong (and when you're starting out, it will) knowing how the machine is put together will save you a lot of headaches.
|
|
#
?
May 1, 2017 13:51
|
|
|
NPR Journalizard posted:I am thinking of getting into some small scale cnc routing, and wanted to know just how rubbish the Chinese diy kits are. I don't need a large workspace, 30cm *30cm would be enough for my purposes, and it would just be wood and maybe acrylic. My personal experience (6040 machine): mechanics are pretty solid, electronics were flakey. Spindle controller burnt out, and the optoisolators on the CNC parallel board were the limiting factor in speed.
|
|
#
?
May 3, 2017 01:56
|
|
|
Just ordered a gecko G540. Decided to switch to it from my current kit to help eliminate assembly error caused by my bad wiring or non understanding. Anyone interested in 4 kl-4030 stepper drivers and a c10 6 axis breakout board?
|
|
#
?
May 5, 2017 17:35
|
|
|
Can anyone recommend a good web page to read about stepper drivers? I read about Geckos and various common arduino drivers but it's just like looking at catalogues and I don't learn much. Similarly where can I learn about stepper motors and their control parameters? I'm interested in how well you can control the torque, accelerations and velocities - beyond the basic info. I intend to make a CNC router that can do aluminium and a 6 axis robotic arm soon, preferably using some of the same hardware. I've made 3D printers before and own a RF-45 type mill. For robotic arms it's quite important to control torque and acceleration to prevent 'jerky' movement, and there are equations to predict how much torque you will need (fairly horrible if you don't like vector calculus).
|
|
#
?
May 6, 2017 11:53
|
|
|
Mach 3 motor tuning has a graph you can adjust to tell the motor how to accelerate, how fast to go etc. There are some good code commands to tell the motor to either do exactly as written, or to accelerate and decelerate through the directions, which leads to rounded corners etc. Can't remember which command it is though. That might help with less jerk in a robot.
|
|
#
?
May 6, 2017 12:42
|
|
|
|
| # ? May 20, 2024 00:44 |
|
|
Mudfly posted:Can anyone recommend a good web page to read about stepper drivers? I read about Geckos and various common arduino drivers but it's just like looking at catalogues and I don't learn much. http://www.geckodrive.com/support/step-motor-basics.html
|
|
#
?
May 6, 2017 13:23
|
|