|
mekilljoydammit posted:OK new frontrunner is Deskproto; even the free version seems to do what I want, and it seems like its toolpath generation is a couple orders of magnitude faster. I started out with Deskproto on my Taig and it is pretty good. Another one to check out is Vectric Vcarve. It is great for 2.5D stuff and I used it with my cnc router. Might be a bit more than you want to spend though. I am on BobCad now for the trochoirdal tool path generation that allow you to get full use out of smaller machines. And also for 4 axis work on my machining center.
|
#
¿
Apr 13, 2021 00:33
|
|
|
|
| # ¿ May 8, 2024 14:36 |
|
|
NewFatMike posted:Mach 3 just kinda fuckin sucks, huh? LinuxCNC with Mesa electronics FPGA hardware step generation. I was running the PCIe FPGA cards on my lathe and Novakon mill. Now I am running the ethernet FPGA card on the VK45 machining center. I also ran it with the software step generation 10 years ago but I haven't tried it recently. It is pretty bullet proof for hobby level machines. I hit a few corner cases that were not well tested when running it on a machining center with probing. PathPilot is a GUI for LinuxCNC running on the Mesa cards.
|
|
#
¿
Apr 16, 2021 03:33
|
|
|
NewFatMike posted:Oh lol, I meant about putting together a mesa + LinuxCNC build. I'm sure I can find something serviceable with those keywords, but it never hurts to ask. The LinuxCNC forums are quite good for many of those details. Here is a starting point: https://forum.linuxcnc.org/27-driver-boards/37141-mesa-card-basics There are many options though depending on what you are using and trying to control.
|
|
#
¿
Apr 16, 2021 13:58
|
|
|
Made a bad noise on my mill. https://www.youtube.com/watch?v=wUzFzcxWWHA Ran a 5/8" tap at 5x the correct feed rate, bottomed out the T&C tool holder and snapped the tap. Luckily the Z axis drive faulted and shut everything down before more than the tap was damaged. Turns out I found a bug in my spindle gear change code and linuxcnc thought the machine was still in high gear. Should have done an air cut to verify that tapping was working correctly but that is what I get for just sending it. Managed to get the tap back out but it was a pain as it had been pressed down into the bore. The thread pretty much had to be re cut backwards to get the tap out which is a pain when the drive square is sheared off.
|
|
#
¿
May 3, 2021 21:26
|
|
|
I have had a few crashes with this machine as I find bugs in the control configuration or my programming. This is the stressful part of retrofitting a machining center. The power of the machine requires much more respect than a hobby class machine. The x and y servos on this machine are 1.5kW, more than the spindle motor on any of my old machines. First crash was due to the G43 tool offset being cleared after a fault causing the z axis to rapid into the part. I now set up my tool offsets (negative values) so that clearing them causes the axis to move away from the work rather than towards it. This was a fun one because I was milling a hole in a 2" diameter ball bearing and the z axis hit hard enough to press the ball into the vice and dent the vice. The vice had a raised ring around the dent that had to be stoned out. Next one was crashing the renishaw probe stylus into the work. Luckily it was a ceramic shaft stylus so it just exploded all over rather than damaging the probe. This one was how I discovered that there was a bug in LinuxCNC where it would not stop when the probe input trips on non-probe moves. It was supposed to stop, but didn't and I found out the hard way. I submitted a patch to fix this and it should be in the next version. Then there is the tap:  It is hard to see in the picture but there is a nice spiral fracture running from the drive end all the way to the flutes. All these crashes are why I have not rebuilt the spindle yet. I have the bearing for it but I want to get it running stable to minimize the risk I will immediately crash brand new bearings  Anyway, this is what I was making:    Both parts were made on the mill and really should have been done on a lathe but where would the fun be in that? It is a spherical bearing mount for a jib crane I am working on. It was the first time I had run a thread mill and it turned out quite well. A 2.95" 18TPI
|
|
#
¿
May 7, 2021 16:35
|
|
|
Rad-daddio posted:I've been printing out test pieces for the various conversion bits before I make them out of aluminum, but I might keep the encoder holder as is if it holds up. The monitor mount is a repurposed mini dish mount, and the keyboard/mouse platform is made from some sheet metal parts that I biffed years ago. After a lot of tinkering, I finally got the z axis ballscrew and motor mount how I wanted them. The electronics will be mounted in a case on the backside of the toolbox. I'm still up in the air on the x axis, because I want to do gang tooling but my instinct is telling me to keep it simple and just manually change tools for this project. idk... Very nice so far. I am still running the 3d printed mount for the spindle encoder on my lathe and it has been running for 5 years now. There is no real stress on that part so plastic should be fine. On the tool holder front, I also had plans/dreams of gang tooling or a turret but have never taken the plunge. Running a wedge style quick change tool post works fine for the very limited runs of parts I make. Changing the tools is repeatable enough that you can swap them without touching off. On the enclosure front, one thing I learned is that you have to be really careful to insure there are overlaps on any openings or joints. Otherwise it will leak. Mist coolant might be better for this, but running flood coolant is like having a pressure washer hose flopping around inside the enclosure. My lathe today:  It is a 10x22 lathe stripped down and put in an enclosure. Most of the time I use it as a semi-auto lathe with conversational cycles.
|
|
#
¿
May 7, 2021 19:24
|
|
|
Rad-daddio posted:Nice setup! I am using Gmoccapy as the UI. The monitor is an old monoprice resistive touch screen. The resistive screen works quite well even if your hands are covered in coolant.
|
|
#
¿
May 8, 2021 22:06
|
|
|
Rad-daddio posted:Does anyone know the specific tradeoffs/benefits for running a timing belt and a pulley on your stepper motors? I'm going that route for my lathe x axis to make to make toe motor position more compact, but from what I've read online it allows for greater accuracy if you run a smaller pulley on the motor. The biggest thing to watch for is speed. A belt reduction will reduce your max axis speed and with a stepper it may be too low to be acceptable. It does increase torque but that is not much of a concern with appropriately sized steppers. The other issue is the increase in backlash and compliance of the axis. The belt is not infinitely stiff and will increase deflection of your axis. With good design this is not an issue but it is something to look at. You will need to provision some tensioning method as well. I use an app on my phone to check belt tension. You hit the belt and the app measures the resulting tone and will let you know if the frequency is in the right range. On my lathe I use a 4:1 belt reduction but I am running DC servos with a 3000rpm rated speed. With the 4:1 reduction and 5mm pitch ball screw, I have a max rapid speed of 140ipm. With the plain ways on the lathe this is about as fast as you would want to go. With a stepper, the 750rpm ball screw speed would be achievable with direct drive and maybe a 2:1 reduction. Since I have a 1250lpr encoder, the theoretical resolution is 101600 steps/inch or 1e-5 inches. In reality, the backlash in the belt, ball screw, and ways plus the belt stretch and machine deflection limit the effective resolution to around 0.0001". If I was using steppers, I would aim to direct drive the ball screw. The closed loop steppers would also be a good idea to increase the real resolution of the stepper.
|
|
#
¿
May 14, 2021 16:08
|
|
|
Rad-daddio posted:Thanks for the feedback! Did you get a GT2 or GT3 type belt? Those are the only ones that I would use for a motion system due to the nearly zero backlash. They also have consistent motion through the tooth engagement unlike the trapezoidal style belts.
|
|
#
¿
May 15, 2021 00:45
|
|
|
Dominoes posted:Hey dudes. Can I use a 3018 to cut 0.5mm-thick aluminum sheets? Any special considerations or bits? Most of the info I find is either about engraving, or cutting 3d shapes. Is there a way to do this in SolidWorks? It seems to be more for 3d shapes as well. I have a Python script that helps me make 2d G-code, but this has the same profile as an existing 3d model I'm using, and it has some curves that I don't know if I could replicate using the script. Looks like it should be possible. From what I could find, the 3018 has a 10krpm 60W spindle motor. Using that, HSMAdvisor says that you could run a two flute 1/8" end mill slotting at about 30ipm in 0.5mm aluminum and not exceed your spindle power. You will need to dial that feed rate down (likely to 5 to 10ipm) as the 3018 is not very stiff. For what you are doing, look for a 1/8" two flute high speed steel uncoated endmill. Idealy specified for aluminum use. Carbide would be a second choice as it won't be as sharp as HSS unless you get into the high end. You will need to run lubricant or chip welding will clog your endmill in short order. When I used mill aluminum on my cnc router, I would give it a few squirts of WD40 as it was cutting and that seemed to do the trick. I can't really help with the CAM side. I use BobCad for my CAM at the moment but that is not a low cost/free option. The other system I used to use is Vectric Vcarve and that was a powerful 2.5D system but again not free. I know there are dxf to gcode programs out there but I haven't looked in a long time.
|
|
#
¿
Jun 5, 2021 04:49
|
|
|
Dominoes posted:Looks like I can buy some of that in filament form for $69. LoL! 3/5 stars on fake Amazon reviews. I have an enclosure for printing ABS. What's the word? Unless you have a printer with a heated chamber that can extrude at 380c and have a bed temp of 140c, the word is lol. I have experimented with high temp filaments like this and even on a heavily modified printer that can hit the extrusion and bed temps they are pretty much unprintable without the heated chamber.
|
|
#
¿
Jun 6, 2021 15:56
|
|
|
NewFatMike posted:MasterCAM knowers, is there a special spiral op or G or M code? I fielded a customer support call from a guy asking why his spirals weren't being posted as arcs, and I was just thinking "... Because they aren't arcs?" I assume you mean a spiral in the work plane and not a helical interpolation. The only thing I can think of would be if they were expecting NURBS output? That would create a program with a similar look to a helical interpolation with a bunch of g02.
|
|
#
¿
Jul 9, 2021 02:09
|
|
|
the spyder posted:I would like to buy a large cnc router sometime in the next year. My main use case is wood signage, furniture, and some light plastic/aluminum use. At that size class you might want to watch the machinery auctions close to you. You can probably pick up an industrial unit with vacuum table and a HSK63F or ISO30 spindle for less than the price base price of one of the hobby class machines. Bidspotter is the site I use to find the local auctions but once you have been to a few the local auctioneers will start to notify you directly.
|
|
#
¿
Jul 25, 2021 14:36
|
|
|
NewFatMike posted:They'll probably chuckle and watch. They ask to see proof of your $5000000 comprehensive liability insurance coverage and then chuckle and watch. Big equipment at auctions usually have a rigging fee noted in the auction listing. This is using the rigger the auctioneer has arranged for the sale. Sometimes you can use your own rigger but this is not usually the case. The rigging fee only includes loading the equipment onto your truck at the site. It does not include transport or unloading at your end. My experience is that you can get the rigger to transport and unload at your site for an additional fee. The couple of times I have done that, it worked out to roughly the auction rigging fee again. This is for smaller stuff that would fit on a flatbed with a forklift and a local move though. Smaller items that you can't lift and need a forklift to put in your truck will also usually be charged a "lift fee" at the site by the rigger. Typically $25-50 cash per lift so factor that in to your purchase decisions as well. Since this fee goes into the riggers beer fund, they are usually very happy to help. I always like watching the price that really big equipment goes for. Equipment the size of house will go for peanuts because it costs so much to move.
|
|
#
¿
Jul 29, 2021 19:02
|
|
|
mekilljoydammit posted:This is hitting me right in the feels... I'm watching a surplus equipment auction, and there's some stuff I'd definitely bid on for near-future-desires... but I know any real rigging company would probably not be willing to go along with my "put this in a barn" plan. I don't think they would have any problem with it to be honest. I had a 12000lb machine moved into my residential garage and the head rigger thanked me for giving him an interesting challenge. They did have to put down 8'x24' 1" thick steel plates to keep their forklift/machine from sinking through my driveway (asphalt) but it was otherwise fairly easy. Note that they had those plates on hand as this is a regular thing to move equipment across surfaces that can't support the point loads. As long as you tell them where it is going and what the conditions are they can handle it. Now if the machine sinks through your floor when they put it down, that is on you. The rigging for that machine cost $3000cad. It included pulling it out through the wall of the building it was in, transporting it 200km, and moving it into my shop through a door that had 1" of clearance on any side. It took 3 guys, two heavy flatbed trucks, and a 20000lb capacity forklift. Total time was about 5hours. I thought I got a bargain all things considered. The guy I bought the machine from suggested hiring a guy with a flatbed to winch it out for $1000. I can't imagine that would have ended well. In keeping with the moving being more expensive than the machine, the machine only cost $2000 and that included a pile of tooling. Here is a timelapse video moving the mill into my shop. The bald guy in the blue jacket is me https://www.youtube.com/watch?v=fhU-SnGCfjc
|
|
#
¿
Jul 30, 2021 05:45
|
|
|
Just finished making a fixture plate for my mill. It is in the same style as the Saunders Machine Works ones but they don't make one in the size I wanted. The plate is 10"x16"x1.5" Holes are drilled and tapped 1/2NC with a 0.25" deep 0.5005" counter bore. The counter bore allows alignment dowels to be used for repeating setups. Getting the bores right took quite a few tries in a test piece to get the boring head dialed in. 
|
|
#
¿
Oct 26, 2021 05:16
|
|
|
insta posted:That's what peering directly into the holes while blasting them with shop air is for! Hey, no peeking in my shop! They do make a wonderful screeching whistle when you blow shop air over them though. I was thinking of making an injection mold to make plugs but then I found that saunders machine works has plugs available so I am ordering them from there. Right now, I am just taking the plate off and dumping it out after a job is done.
|
|
#
¿
Oct 26, 2021 21:14
|
|
|
Well, after getting a face full of chips more often than I really wanted I ended up printing plugs for the fixture plate I made.   They actually turned out better than I expected and seem to be watertight when screwed into the plate. No coolant seems to make its way into the holes now and it can be blown or washed off with no issues. They are printed in PETG so hopefully won't dissolve in the coolant.
|
|
#
¿
Nov 27, 2021 23:31
|
|
|
Harvey Baldman posted:
I am confident in saying that any machine than claims to do both laser and router would suck at one or both of those jobs. The requirements are too orthogonal to make this combo viable unless cubic dollars were spent. Don’t buy any of the hobby machines for a business unless your time is free. One suggestion is to get a machine that has a tool holder spindle rather than just a collet spindle nose. This will make your life so much easier doing multi tool jobs. Even without a tool changer it will make the machine much more usable. I think HSK63F and ISO30 are the common ones but not really my area. For reference the job I did for you used 12 tools but I could have got that down to 6 if I had to. Note that the general rule of thumb is to budget the price of the machine again to get everything you need to get running. Tool holders, endmills, spoil boards, power hookup, measuring tools all need to be purchased and add up quickly. With a router it is probably a bit less than for a machining center but you still need to look at the total cost. The other thing to budget for is CAM software. The type of art jobs you are doing will run into the wall with any of the free options I am aware of. Aspire would probably be a good choice but not one I have personally used. I use Vcarve for artsy things but it is really only 2.5d. BobCAD is what I used for the job I did for you but that is likely overkill and is more suited for mechanical CAM. There are lots of other options out there and quite a bit will depend on your existing CAD flow.
|
|
#
¿
Dec 17, 2021 05:27
|
|
|
Trebuchet King posted:So I’ve been using mostly Onsrud bits at work but I’m trying to find something that might decently v-groove aluminum sheets up to 1/4” without getting totally spent in a jiffy. What exactly are you trying to do? Decorative curved cuts, beveling edges/chamfering, or long straight grooves? If you can tolerate a flat bottom a bit with a flat on the end will work much better as the surface speed will not be zero. If you need a sharp bottom then an engraving bit will be your best bet. Your feeds will be slow though as the surface speed at the bottom is still very low. If you are doing long straight grooves then a 90deg head with a v groove cutter would be the best bet.
|
|
#
¿
Dec 25, 2021 05:13
|
|
|
Trebuchet King posted:Yeah, that's more what I mean--I'm not looking for a specific brand or model, I'm looking for "an appropriate cutter" as you put it. Everything in the catalogs I've got at work are for acrylic, and googling has just gotten me links to the tools you use on ACM to put the folding grooves in it. Just doing surface engraves would be easy, we do that often enough. Our other departments are convinced the easiest and/or fastest and/or most cost-effective way to make parts for these signs in a package we committed to is to put 1/4" deep v-grooves in 1/4" aluminum and while I protested that my hands are kind of tied now. Okay, that is good info. If you are cutting 1/4" deep in a 1/4" plate, I assume you are cutting through the plate? Can you do a first pass with a standard square endmill and then bevel the edge? This will let you use a multi flute cutter with enough of a flat on the end you will have good tool life and can run reasonable feed rates. If you must groove to full depth then I would run a roughing pass or two with square end mills and finish with the v-groove bit. I don't think you would be able to use anything other than a single flute engraver in this case as that is the only way to get the chips out of the tip of the tool. Multiflute tools will not have enough chip clearance around the tip of the tool and will chip weld and break. For aluminum specific coatings, you would be looking at ZrN, TiB2, Amorphus Diamond (DLC), CVD diamond, or PCD diamond in increasing order of cost. I use ZrN mostly as I am typically machining extruded 6061 and this is lower in silicon and less abrasive on the tooling. I am also not doing high production. Do not use any coating with aluminum in it (AlTiN or the like) as you will have massive chip welding problems as the aluminum likes to stick together. Bright finish is better than any of the non-aluminum specific coatings. As Sagebrush mentioned, coolant is essential but a MQL (minimum quantity lubrication) system may work well if flood is not an option. I ran a Trico MQL system on my router when cutting aluminum and it worked quite well. The only issues I had were with cuts like you are proposing where the air blast had a hard time clearing the chips.
|
|
#
¿
Dec 27, 2021 03:24
|
|
|
Wanderless posted:It is a CNC lathe so it isn't unheard of, but it is pretty terrible. It lacks the rigidity to do much of anything--a .100 depth of cut in 6061 with appropriate speeds/feeds/tooling caused it to bog down, probably because it seems to have way more flex than any similarly-sized lathe I've ever used. Free taper with every cut! 0.100 depth of cut on 6061 would be above the rated spindle HP on the tormach with any feed you would want to use with carbide. HSMAdvisor calculates about 0.004 to 0.005"/rev to get the spindle power under 3HP and at that point you are rubbing more than cutting. Not defending the tormach at all but that does not seem like a realistic cut on that machine. I know on my wet noodle CNC converted 10x22 lathe I have to use ridiculously light cuts (0.02" DOC is typical). It only has a 1HP spindle but the machine can't take much more power than that. Which is why I have a Hardinge HLV waiting for me to get to doing a CNC conversion. NewFatMike posted:I don’t think that APSX can utilize a chuck other than that bar feed, can it? Might be worth asking them but I don’t want to get bludgeoned with infinite emails from them. From the looks of it, I don't think there is any way to use anything other than the bar feed on the ASPX. The bar itself is the drive as the spindle motor is attached to the back of the bar. It is a true swiss style machine so there is no real spindle. The 300W spindle motor seems really anemic to me even for a max of 0.75" stock. The spindle motor is also the C axis so you have fairly limited torque when milling and I couldn't see anything about a spindle (bar?) brake. I was tempted by the machine anyway but the 0.001" accuracy spec is pretty poor for most of my use cases. I would need an order of magnitude better than that but then I am back to a real swiss machine at $lots.
|
|
#
¿
Feb 18, 2022 02:27
|
|
|
A Wizard of Goatse posted:Thank you, this is exactly what I needed. I'd like to get it at least robust enough to handle deep cuts on hardwoods - ability to work metal would be nice in a perfect world, but I expect that's a long ways off and being able to mill out mold positives to spec is nearly as good for my needs. I already have access to a laser cutter that does shallow engravings on flat surfaces perfectly well, what i can use is something that can handle deeper shapes, undercuts, and making proper three-dimensional components. In what seems to be a common theme for home brew projects, the motion components are vast overkill for the frame. The ball screws and linear guides are massive and the ballscrews have anti-backlash nuts, someone spent a fair bit on this. As others have said, the frame is a wet noodle and will not give good performance no matter the quality of the rest of the system. One area I am concerned about is the spacing (or lack there of) of the guide blocks for the gantry. The gantry is subject to racking forces when cutting off the center of the x axis. The blocks need to be spaced further apart to resist these forces without binding or allowing the axis to go out of square. I suspect this was done to maximize the y axis travel but at the cost of poor performance. The same thing seems to have been done on the Z axis and I can't see the X axis blocks. You have the basis of a great machine using those motion components. I would suggest looking at commercial router frames to see how they are built and base a design off one of those. I don't think this frame is salvageable without a complete redesign.
|
|
#
¿
May 22, 2022 04:06
|
|
|
A Wizard of Goatse posted:Trying to do the math on how much force I can expect to put on the gantry if I try to carve various dumb things is quickly reminding me of just how bad my mech eng grades were. Now that I'm really looking at it, what's the reasoning for putting the X axis in line with/under the table in CNC routers? It's drat near universal but it seems like moving it above the bed closer to inline with the Y would significantly cut the leverage on it, and get it out of the exact spot where the most poo poo is guaranteed to fall on it The main reason is ease of access to the work surface. When you look at room size gantry mills they are often built the way you describe but you drive the work piece in with a forklift. Note that to get good stiffness you end up needing walls not just posts to support the gantry and it really constrains access. There is a reason that the gantry style mills usually have fairly limited Z travel and that is to keep the gantry supports reasonable. NewFatMike posted:Are aluminum extrusions used in a lot of CNC routers vulnerable to the same issues as the erector set looking build? Sounds like they would be from the posts earlier up. Yes, the hobby class machines built out of 8020 will have the same issues. That being said, with good design, attention to load paths, and picking appropriate section sizes, these can work quite well. It is just MUCH more expensive to build a machine this way than using welded steel. The only reason I can see to go with aluminum extrusion is to avoid welding. I think the Langmuir is a hands down winner based on the mechanical design. The unsupported rod guides on the APSX is a terrible design for a mill. There is no way that they will be stiff enough to compare in any way with the MR-1. Both of them have exposed screws and ways which is something to watch for. A very useful and quite approachable paper on machine design is here: https://my.mech.utah.edu/~bamberg/research/PrinciplesOfRapidMachineDesign/Principles of Rapid Machine Design.pdf It talks about a bunch of techniques for welded frame machine design and construction.
|
|
#
¿
May 22, 2022 20:15
|
|
|
meatpimp posted:Okay, I've been slow playing this guy for the past two months. He's willing to take $950 and deliver to my door (he's 2 hours away), so he's hungry. It does not look like a terrible design. The main weak point I see is 3d printed bearing blocks and motor mounts. I would think that would compromise the rigidity of the system. Personally I would fabricate steel motor mounts and ball screw risers. The design would be easy to achieve with basic fabrication and welding skills. I like the fact it uses steel tube instead of aluminum extrusions. This will increase the capability of the machine quite a bit. One thing to watch is the x axis seems fairly long in this kit and you might want to increase the size of the gantry tube to improve performance. If you are willing to put in the time to build it, I don't think you will find a better deal on those components. They are not hiwin parts but should be serviceable in this application. If I was in the market this would be a serious contender. The main thing is what you want to do with it? This is a good wood/plastic router that can do light aluminum milling. Steel is technically possible but not practical.
|
|
#
¿
Jun 22, 2022 03:10
|
|
|
Rescue Toaster posted:A proper dust 'extractor' with HEPA filtering is like $500-600 which seems like a lot for a basement CNC but... I don't know. I feel like my regular shop vac, even if I added a dust collector would probably burn up running constantly for an hour+ at a time. And I really don't want powdered MDF in the air in my house. I don't really have a way to setup for an external vent, either. I ran a cnc router with a shopvac and it worked fine for 8+hour run times. The big thing is to get a cyclone to get most of the dust before it gets to the shopvac filter. I used one of these https://www.clearvuecyclones.com/product/cv06-mini-body-only/ and I almost never had to change the filter on my shopvac. That being said, this was in a separate airspace from my house and I would NOT consider this sufficient if I was running in my basement. A shopvac does not have enough airflow to get all of the dust from a cnc router. It will get the big stuff but much of the fine particles will escape. I also would not consider running without a HEPA filter if I was recirculating the air especially in my house.
|
|
#
¿
Sep 9, 2022 15:20
|
|
|
Comatoast posted:I’ll be the voice of dissent here: if you don’t have $10k to spend on a kit-router (like a Shopbot), then just don’t get one. The hardest part of cnc’ing is the vector design work. Concentrate on producing fantastic designs and pay your local sign shop to run their router. The real question is what they want the machine for and what the expectations are. If the goal is a new hobby of fiddling with little CNC machines and making a few parts once in a while then a ShapeOko would be fine. If production or making parts without a fuss is the goal then 10k is the absolute minimum. The kits are in a bit of an odd spot. They can be very capable but are still firmly in the "hobby is building machines" and not "hobby is machining parts" category.
|
|
#
¿
Apr 1, 2023 22:21
|
|
|
Bad Munki posted:I had previously built a Joe’s 4x4 hybrid. I see they’ve gone through a couple generations of the EVO line now. Anyone have any notion of how that stacks up these days? My hybrid never made it back to the floor after my last move, but I’d sure like to get it rolling again, and if an upgrade is a sensible option, I’d be cool with that. It looks like a lot has changed but I assume my steppers and driver and racks and such would all still apply. Nice. I built a couple of 2'x4' Joes2006 machines 15+ years ago. One is still going strong at a buddy's place. The other got given away after I ended up using it as a table for a few years. Never got around to putting it back together after I moved. From the looks of things, you could build a hybrid evo from your current machine. The evo looks pretty much in line with the low end of the aluminum extrusion based professional machines. The main weakness I see is the angle iron linear ways. They aren't going to stand up to production use but for hobby work it should be okay. If you were starting from scratch I don't know that I would go with the joes approach. With the cost of linear ways being as low as they are, it is really hard to justify going with v-bearings/angle iron. I think it is a bit of false economy when you look at the overall cost of the machine.
|
|
#
¿
Apr 4, 2023 01:37
|
|
|
Zero VGS posted:I need to make a design to 2.5D CNC route some acrylic (like, some cuts all the through, and some routes to half-thickness), but these awful UIs confound me. With all this brouhaha over AI lately, is there any way I can describe my design to an AI and have it model something in InkScape or whatever. Like take dictation from me while I watch it draw and let me undo steps if it misunderstands what I was going for? I have made parts before by just describing the measurements to a fellow human who actually knows how to use SolidWorks or something, but using that poo poo just doesn't click with me You would be better off just describing your requirements to the local sign shop and having them do the art as well as the cutting/manufacturing. If you can't get your head around vector drawing programs then running a CNC machine is not for you. The vector drawing program is the easiest of the tools to use. After you have the vectors, you need to load them in a CAM (computer aided manufacturing) program than converts the vector into machine instructions. CAM programs are not known for their user friendliness. You also need to understand tool selection, tool path strategies, speeds and feeds calculations, and work holding. AI is great for creating a starting point but it is not at the level you can use it without heavy oversight/editing. It great at producing answers that appear correct but are actually subtly wrong (see the finger problem in image AI). Current image AI also does not draw anything the way a human would. There is no "undoing steps" because the whole image is generated at once through the statistical weighting network. You get your final result by running with additional random seeds or by changing your prompt.
|
|
#
¿
Apr 5, 2023 14:34
|
|
|
When buying an old mill, it is a good idea to check the drawbar tension. I finally got a gauge and found my CAT40 spindle only had 500lbs of drawbar tension. Somewhat less than the 2300lbs-f typically recommended. This gave me an excuse to finally rebuild the spindle as the bearings have been making noise for the last while. The first 3/4 of the spring stack came out fine and looked good. Then I poured out the rest:  Uh, I think that might be a problem. 30 of the 120 springs were shattered. I am surprised it held tension at all. The scale was used to count how many springs were in the pile of broken ones. The manual helpfully listed the springs as "spring stack: quantity 1". Luckily they are a standard size and readily available...
|
|
#
¿
Jul 31, 2023 13:37
|
|
|
Yes, that is a great series. I did watch that and used the tips quite a bit. Luckily the spindle in my machine is pretty straightforward. The telling thing is that prior to the rebuild I could run all day at max rpm and the spindle would not get warm. Now I see a temperature rise in line with expectations.
|
|
#
¿
Aug 1, 2023 16:02
|
|
|
hark posted:I wanna put files into a machine and make it cut stuff out of wood/plastic/aluminum if possible, but the first 2 for sure. What do you mean by "put files into a machine"? CNC machines are not like 3d printers where the slicer does all the heavy lifting. You need some form of CAM (or to write gcode yourself). You also have to select the tools to use (endmills/drills/etc), speeds and feeds, and tool path strategy. I apologize if you are already aware of this but it is common stumbling block. For a machine, it depends on what you want to do. If you want something that just works off the shelf, I recommend the Taig tools CNC mills. They are bullet proof in their class and won't hold you back. They are are also very cheap for what you get. I started with one of these and did quite a bit of work on steel and aluminum. Most of the cheap machines available are toys at best. It is technically possible to make parts using wood or plastic, they will still be an exercise in frustration. The open builds stuff is fine if your hobby is building machines. Not so much if you just want to use it to do something.
|
|
#
¿
Dec 14, 2023 03:32
|
|
|
Problematic Soup posted:I was wondering if this thing can do rigid tapping. I definitely plan on using the threadmilling feature if possible, but I have some ideas for some parts with very fine screws that are probably too small to threadmill, and I would appreciate the controls still being able to handle tapping the hard way. It does not look like it supports rigid tapping. From what I saw when I looked into it, the control does not get the spindle encoder feedback. The encoder just goes to the drive. I also saw something that it wouldn't reverse the spindle but that seems odd? If you can reverse the spindle then you should be able to get a tension-compression tap holder and use that for small holes even without rigid tapping. I use tension-compression holders with rigid tapping on my mill just to minimize breakage. Problematic Soup posted:I also was curious if similarly this thing can handle using a boring head. The website shows a boring head being used so that should be fine.
|
|
#
¿
Jan 11, 2024 23:30
|
|
|
I wouldn't restrict yourself to single phase machines. At that budget level, a PhasePerfect digital converter is the way to go. I run a PhasePerfect PT030 in my home shop and it is rock solid. Add a 460 or 575V step up transformer and you have industrial power in your shop. Most of my home shop runs 575V 3phase because used 575V/600V stuff is way cheaper than 220V. Rotary phase converters are too noisy and the idle power is too high so I wouldn't go that way. I had one for a while but got rid of it when the PhasePerfect came up at auction.
|
|
#
¿
Feb 4, 2024 04:54
|
|
|
I have always wanted a 5 axis setup for milling. So I finally built one. The main parts were a Sumitomo F1C-A 35 cycloidal reducer and a Yaskawa 750W sigma 7 servo. Both were things I had acquired through auction winnings so had minimal cost. Start with some blocks of steel:  Load onto the mill:  Ready for roughing:  Roughing complete:  Post stress relief heat treat:  Finishing Op 1 complete:  Finishing Op 2 complete:  Finishing Op3 complete:  Lower cover Op1:  Lower cover Op2 roughing:  Lower cover Op2 finished:  Assembly in progress:  Belt assembly:  Bottom cover on:  All done and moving (slowly for initial testing): https://i.imgur.com/2NnPydt.mp4 The real test will be when I start milling something with it. Indicating the table surface shows that it is square to around 0.0008"/6" which is well within my expectations. Most of the work I plan to do on it is in the range of 1" cube so that level of accuracy should be fine.
|
|
#
¿
Mar 4, 2024 21:56
|
|
|
CarForumPoster posted:This is fantastic! You still running LinuxCNC? How much was involved controller and wiring wise in getting the 4th and 5th axis able to make parts? Yes, still running LinuxCNC. It is not too much harder to add the 4th and 5th axis on the basic 3. The 7i77 Mesa Electronics IO card has 6 analog interfaces, 5 for the axis and one for the spindle. I do need to add another 7i84 IO expander because I am completely out and can't finish setting up the 5th axis brake. Here is a shot of part of the main control cabinet with the interface boards. There is a sub panel for the tool changer as well.  CarForumPoster posted:Also, did you send out for the heat treat or DIY? I did the heat treat myself. Picked up a kiln on kijiji for next to nothing and added a PID controller. Works great for bigger parts but it takes forever to heat up. Sagebrush posted:i am reporting this post for obviously not being "hobby" anything I believe that everyone should have a 7ton machining centre in their house  The funny thing is that everyone asks me what I make with these tools and the answer is always more tools. I built the 5th axis so I could experiment with making rollerdrive style reducers to make a more dedicated 5 axis machine. I still have a way to go until I get to Dan Gelbart's level though.
|
|
#
¿
Mar 5, 2024 04:41
|
|
|
ryanrs posted:What kind of steel did you use? Is it going to get some kind of surface finish? The bulk of it is just 1018 mild steel. Nothing fancy. The table is 4140HT as I wanted it to be a bit harder. The plan is to paint it with some macropoxy 646 but that will depend on me getting around to it. Really, it would be fine as is as the coolant has quite a bit of corrosion inhibitors in it. CarForumPoster posted:You’re where I want to be next year when I get the Fadal 4020 in here. I even have a 4th axis but no card, plan is a Linux CNC retrofit at some point after I get the mechanicals dialed in. I started with a Hitachi-Sekei VK45 that was mechanically okay but the control was partially dead. The x axis would run away as the controller output to the drive was stuck at +15V. The tool changer didn't work because a bunch of the SSRs were burnt out. And the spindle drive was dead. I got it for pretty much the scrap metal price so it was worth a shot. In my case, there was a 4th axis installed in the machine but it was sold before I got it. I ended up getting the original 4th axis a year or so ago as the guy that bought it never got it running. That is the one shown with the new 5th axis. Why a 30" class machine was sold with a 320mm rotary table is a mystery to me. Here you can see the original servo drives for XYZA and the new drive shoehorned in on the right. They are all analog drives so really easy to use. If they ever die, I will replace them with either DMM or Yaskawa depending on what I can get. 
|
|
#
¿
Mar 5, 2024 06:26
|
|
|
Yes, I am looking for a dovetail vise like the 5th axis ones. Like most of machining, workholding is 99% of the challenge. The problem is they go for almost new prices at auction so I haven't found one for a price I want to pay yet. I will make some fixtures for some of my initial testing. An ER40 collet chuck will probably be first as that gets me an easy way to hold small round stock. The mechanical setup of the older machines can be amazing. On my machine, the dynamics are not too great as you have 2+ton moving Y+Z axis. But testing with a laser interferometer shows the axis positioning linearity is around 5um over the full travel. After calibration, gauge blocks measure to <1um using the touch probe. Having 24" of daylight under the spindle really allows quite a few setups that would normally require a bigger machine. It is has 30x18x18" travel but the table is 44" wide.
|
|
#
¿
Mar 5, 2024 16:40
|
|
|
Methylethylaldehyde posted:Hell, a before and after of the machine's capabilities would also be interesting. There's a ton of old iron with controls about as powerful as a graphing calculator that a control rip and replace could bring to nearly modern performance. As someone that has done exactly this I just want to point out just how much time and money are involved. I took a 1988 Hitachi-Sekei VK45 mill that was mechanically good and converted it to LinuxCNC. It took about a year of pretty much full time work and I have about $15k in the machine right now. Very cheap for the capability but if I charged my hourly rate for the time involved I could have bought a new 5 axis DMG. Unless you have a pile of the same machine to run an assembly line, there is no way to do a retrofit that you could sell for a profit. There is too much reverse engineering, debugging, and tuning to make it viable unless your time is worth nothing. Some points: I had the wiring diagram and full maintenance manuals, without these don't bother. Beyond levelling the machine, there really is not much mechanically that is reasonably repairable. I replaced the spindle bearings and two axis of ball screw support bearings as these are cheap ($2k for a set ebay specials). If the ballscrews or ways are bad, scrap the machine unless you want to spend $10k+ for each part. The spindle drive was dead and Yaskawa quoted $15k to replace it with something modern. A repair shop quoted $5k with no guarantees. I went with a VFD for $1k but getting spindle orientation to work was a huge pain in the rear end. Old analog drives are easy to retrofit and don't necessarily need to be replaced. More modern machines with proprietary interfaces would need replacing the drives as well. Replacing the servo drives is quite doable but budget $1k/axis minimum. Rigging usually costs more than the purchase price of old iron. LinuxCNC is not actually a good choice for getting the best performance from these old machines. LinuxCNC does not have a jerk (or higher order) limited planner and that is one of the things that makes modern controls so much more capable. Unfortunately, I don't think there are any other viable options at the hobby level right now. If I was doing this commercially (ie planned to sell the machine) I would probably go with a Siemens Sinumerik control. Adaptive milling on old iron: https://www.youtube.com/watch?v=bV75vmiyxKw
|
|
#
¿
Mar 12, 2024 04:16
|
|
|
|
| # ¿ May 8, 2024 14:36 |
|
|
CarForumPoster posted:Yea you may well be right. I might give both a try, we'll see once the compressor and what not are here. Kinda continuing on this thought, I'm surprised I don't see people running a loop of copper pipe that they thermally strap to the air tank. Use the air tank as a big radiator. I'll most likely get a blasting cabinet eventually, I really like the "Apple" look of sand blasted + anodized aluminum. There isn't much point in doing so. The air will cool in the receiver anyway and if you are running the compressor hard, the receiver will heat up. I have a refrigerated air dryer and it doesn't help much for intermittent use. Most of the water still seems to settle out in the receiver judging by the amount that comes out of the drain. Once I start running the air spindle/air blast/other high volume user then the air dryer helps. I actually leave it off most of the time.
|
|
#
¿
Apr 26, 2024 01:40
|
|