What type of filament did you print with? For instance, ABS can shrink by as much as 1% from the printed size. So if that was an ABS print, it could actually be spot on. If you printed in PLA however, you may have some tweaking to do to get your prints calibrated...
It is pla, going to do a few more test prints to get it dialed in more. What is an acceptable deviation on PLA, or should it be spot on?
What I liked in your 25mm cube photo was the sharp corners. That means that your travel is spot on both X & Y. (Yeah, that single wheel you are running up the corners is indeed working!) And the surface looked smooth, a good indication for Z travel consistency. What layer height and pulleys are you using on the Z axis? As for calibrating your 25/24.8mm variation. 25/24.8=1.0080645 Assuming you are using GT2 belts & 20 Tooth pulley and therefore assuming your current setting in Default_Axis_Steps_Per_Unit is 80: 80*1.00806=80.6451
Thanks keith, that was my next question if the config file only used whole numbers, or if decimals worked. That cube was with .3 layer heights. Have been printing a few more and have tweaked to get better results. I will print another cube or two to get closer to size, but i am within .1 right now. I am running a 20 tooth on the motor and 40 tooth on the rods for the z axis, to give me a 2:1 drive.
Yes, you can use decimals for all 3 axis, if you want to get into it that deep. Bragging rights for marketability are usually at 0.05 resolution. BUT, using decimals will get you into stopping on microsteps and they can be slightly inaccurate. Steps (and half steps) are physical locations in the stepper, but microsteps are theoretical. They are controlled by current which can be affected by speed, heat, wiring, etc. So, if you start to see degradation of your sharp corners and smooth walls while setting the steps_per_unit don't assume your steps_per_unit is correct and look elsewhere. You may need to play around increasing or decreasing the decimal to find the optimum balance between accuracy and repeatability at multiple sizes. Like default perimeter width settings in Slic3r, you'll be surprised what a small difference in steps_per_unit can make in printed quality. I gotta hand it to ya Josh - this is a beautiful machine. It is clean looking. This would look proud setting in someone's living room. Please, don't slack up when you get to the wiring . Do you have a ballpark figure for what the bom will cost?
Thanks for everything Keith. I just bought a load of wire so i can ditch most of the 16ga that i used from the home store, it is a little large for most of the printer parts. Also designing a pcb holder to integrate into the case. Should be able to wire it pretty cleanly after that. I think i am sitting right at $1000 right now all in, maybe a little bit more. Will be down for a couple of days now, got a clog in my extruded and managed to break the barrel or whatever it is called between the heating block and the body. Ordered some spares and a 2nd extruder to have a spare on hand. I have actually been quite amazed with the quality of prints that it has been churning out with some of the ghetto rigging i have done to get it working to start. It is difficult to get a printer going without having one to start with. Attached are a couple pictures of an end stop bracket for my Y axis. Printed at a .2 layer height at 60mm/s w/ 80mm/s infill and 200mm/s moves. There is still much improvement to be made, but i am satisfied with it for now. Few more items on the list to design and print. Endstop brackets, extruder mount w/ fans, belt tensioner (probably the most important, getting some skipping if I excede 200mm/s moves, which I should probably just be happy with moves that fast).
Hey Josh, Love the build! I have been following TomH's build and he had also pointed me to your build. I would like to come up with a slight hybrid of ideas from both of your setups. Do you happen to have a BOM for this build?
The Notched Version keeps the idlers closer to the Vslot witch is preferable. otherwise you need to raise the other idlers higher to match with long spacers
Sorry it has been so long since updating. Have been working a ton and had some downtime due to broken parts. Got it working again today. Adam the rattling you were hearing was the shroud for the fan on my extruder. Also i am using a 1810mm belt. It is part 6R51M905060 from http://www.sdp-si.com I have made quite a few modifications in solidworks that i need to get printed out. Once i figure out if all of those work i will try to export it all.
Can you explain to me how you made a closed loop out of your z-axis gt2 belt? A normal belt clamp wouldn't work with the belt running over the pulleys right?
Nice build. As someone without a 3d printer to print custom parts, this build is something I'm very interested in, along with various ones that Keith is building. Can you put up a BOM? Or some additional information about extrusion lengths, etc? What about controller, firmwares? Final question, and one that makes me question some of the CoreXY...how does one go about replacing that Z axis belt without tearing 90% of the machine apart? My luck is that's something that breaks and ends up taking a week of work to replace when compared to things like direct drive on rods.
I would assume that you could just mark the locations of the pillow block bearings, undo the screws holding them, slide the belt under the rods put it on, and tighten them back up? Might have trouble with the tee nuts though. Probably have to get the ones that are spring loaded.
Quite a few updates since my last post. I got it all working with off the shelf parts, but was not satisfied with the quality of prints that i was getting and the reliability of the machine. I have spent quite a bit of time designing and printing new parts to replace the bought ones. So it is possible and with enough tweaking i'm sure it would create great quality prints. I did not achieve my desired print volume for many reasons (knew during the design phase it was probably not going to happen). I am sitting at 330x310x350 currently. With a little more to tweak in X & Y with some new part designs eventually. But i am happy with it for now. I also decided to locate my spool inside the cube so that is a limiting factor now (could pick up 30mm with a change in that immediately) but i like it in there and it works well. I started with a bowden setup but was very disapointed with the leaking and over extrusion (had 700mm of tube). For this size of printer it just didn't work. I tried up to 10mm of retraction with not much gain in print quality. So i decided to sacrifice some speed for a better print quality and went direct drive. I am trying a couple different sizes of motors to find the happy medium between weight and torque. Using one that is only 20mm thick right now (13in/lb), it works most of the time but there are some skipped steps. Have a slightly larger one on the way. I also added some print bite and it is hands down the best modification i have made to the printer so far. ABS is now extremely easy to print with (with no warping). I have attached a zip file with all of the parts in solidworks along with the .igs that you requested sonya. Let me know if they don't work for some reason. One change i have made that is not reflected in the CAD files is there are 4 wheels on all of the gantry plates. I tried 3 on the Y gantry, but was getting some twisting in the gantry (some was caused by a bad tensioner design that has been redone). But, i did add 4 wheels before that and it fixed the problem. Added a few pictures below of its current state. Jezter, funny you bring that up. I did have my Z drive belt break a couple weeks back so i had to replace it. It was fairly simple, removed the front 2 bed bolts and like David said lift up the pillow blocks and slide it under. But after doing that, i will be designing some mounts for the lower pillow blocks so that the pulley will be located on the bottom. Making belt changes much easier by only having the remove the bolts on the front of the bed. Next is to add the plexi enclosure, i will be doing that soon. Otherwise i run it several hours most days and it works very well. Have put close to 3 rolls of filament through it so far. Still have not been able to sit down and put together a BOM but it is on my to do list.
I know this is an old posting, but hopefully someone will be able to chime in and clear the haze for me. I see by this post a 20t to 40t is being mentioned as the used combination for the Z axis lead screws, but after downloading the Solidworks file and opening it up, it shows a 30t being used. Can anybody verify which is the correct one to use? Thanks.