tag:blogger.com,1999:blog-5001794558945659136.post4418169977114291812..comments2024-03-23T06:32:36.547-05:00Comments on Mark Rehorst's Tech Topics: Designing A Low Cost, Printable Worm Gear Reducer For Belt Lifted 3D Printer Z AxesUnknownnoreply@blogger.comBlogger14125tag:blogger.com,1999:blog-5001794558945659136.post-37650028027571977002021-08-11T08:12:44.081-05:002021-08-11T08:12:44.081-05:00Or just get a motor with a better gearbox. Someone...Or just get a motor with a better gearbox. Someone at the makerspace used one of these to lift the bed using belts in his printer and it prints as well as any I have ever seen.<br /><br />https://www.ebay.com/itm/324001642095?hash=item4b7000b66f:g:gvQAAOSwTh1d50kEMark Rehorst DMDhttps://www.blogger.com/profile/13616282129430791353noreply@blogger.comtag:blogger.com,1999:blog-5001794558945659136.post-44221308953210435412021-08-11T05:03:43.279-05:002021-08-11T05:03:43.279-05:00Hi Mark, First of all thank you for your effort to...Hi Mark, First of all thank you for your effort to put all data here. I designed the same z axis worm gear-pulley concept and failed as same as you do. My ratio was worse 49:1. I should have read your blog first :). I spend many hours to find the issue. I tried, bed pıd, prints at different temps, different speeds, different microstepping. None of them worked. Only layer height helped little bit. I machined gears by myself with a hobby lathe. Of course gear was not at the center as it should be. If I move my z axis 0,1mm step by step my readings are like so ; 1.4, 1.4, 1.2, 1, 0.8, 0.7, 0.7. I'm going back to lead screw... OnurDhttps://www.blogger.com/profile/04669066143755506090noreply@blogger.comtag:blogger.com,1999:blog-5001794558945659136.post-30295053537744261532020-10-26T14:09:53.569-05:002020-10-26T14:09:53.569-05:00Well, get to it! You know what they say: "The...Well, get to it! You know what they say: "The road to hell is paved with good intentions!"Mark Rehorst DMDhttps://www.blogger.com/profile/13616282129430791353noreply@blogger.comtag:blogger.com,1999:blog-5001794558945659136.post-89653453236789820932020-10-24T15:36:50.275-05:002020-10-24T15:36:50.275-05:00Hi Mark, I am impressed about your 3D printer. I h...Hi Mark, I am impressed about your 3D printer. I have the intention to start building one.Zoefkehttps://www.blogger.com/profile/17641984681744925083noreply@blogger.comtag:blogger.com,1999:blog-5001794558945659136.post-86998069784249157352019-05-11T19:12:08.668-05:002019-05-11T19:12:08.668-05:00It seems pretty tight, but it is possible that tha...It seems pretty tight, but it is possible that that is the cause of the problem with the print.Mark Rehorst DMDhttps://www.blogger.com/profile/13616282129430791353noreply@blogger.comtag:blogger.com,1999:blog-5001794558945659136.post-67795911713785077722019-05-11T06:00:19.303-05:002019-05-11T06:00:19.303-05:00Is there much clearance between the worm gear bore...Is there much clearance between the worm gear bore ID and the shaft OD?<br />A setscrew mount isn't as concentric as a press fit.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-5001794558945659136.post-40652421671879154752018-10-22T11:30:47.919-05:002018-10-22T11:30:47.919-05:00I will add a link to the models for the yellow gea...I will add a link to the models for the yellow gearbox in the test, above. It probably won't do much good unless you come up with some way to correct the positioning errors caused by the gears.<br /><br />I recently saw that someone had made a belt lifted Z axis using this gear set: https://www.ebay.com/itm/1M-20T-B5-1T-B5-Worm-Gear-1-Module-90-Angle-Set-Kit-Ratio-20-1-Wheelbase-18mm/262744570681 . It looks like they are better quality gears based on photos of the prints I saw- there were no visible Z axis artifacts. Unfortunately, both gears have 5 mm bore. It shouldn't be too hard to bore out the disc gear for an 8mm shaft.Mark Rehorst DMDhttps://www.blogger.com/profile/13616282129430791353noreply@blogger.comtag:blogger.com,1999:blog-5001794558945659136.post-31875927190334801252018-10-22T10:33:22.857-05:002018-10-22T10:33:22.857-05:00Hi Mark, I'm in building a smaller version of ...Hi Mark, I'm in building a smaller version of your UMMD, about 2/3 of UMMD's size. The idea of the belt-driven z-axis impressed me, but it would make things much easier if I had the stl's of the worm gear box (btw, I ordered exactly the components you mentioned above ;-) ). Do you think you can publish the gearbox stl's?<br />Thanks in advance<br />HermannHermannhttps://www.blogger.com/profile/16629199104379630962noreply@blogger.comtag:blogger.com,1999:blog-5001794558945659136.post-22900180434792370702018-10-03T15:59:25.140-05:002018-10-03T15:59:25.140-05:00Globoid gears are intended to increase the contact...Globoid gears are intended to increase the contact area between the gears to reduce wear. That isn't the problem here. he problem is the quality of the gears, not the shape of the teeth. Mark Rehorst DMDhttps://www.blogger.com/profile/13616282129430791353noreply@blogger.comtag:blogger.com,1999:blog-5001794558945659136.post-47415133713986244562018-10-01T16:40:24.125-05:002018-10-01T16:40:24.125-05:00Would a globoid worm gear solve the problem?Would a globoid worm gear solve the problem?Michaelhttps://www.blogger.com/profile/17672751910849825242noreply@blogger.comtag:blogger.com,1999:blog-5001794558945659136.post-29457775069171556802018-07-06T15:56:17.290-05:002018-07-06T15:56:17.290-05:00I hadn't thought about the telescope problem, ...I hadn't thought about the telescope problem, but it makes perfect sense. If I were a coder I might try to get into the controller's firmware source code and modify it to read a table to bump the Z values. Two other places to make such corrections would be in the slicer, or post-process the gcode. Unfortunately, I am not a coder, and there aren't many people interested in doing this type of Z axis who can also do the necessary coding. Maybe someday...Mark Rehorst DMDhttps://www.blogger.com/profile/13616282129430791353noreply@blogger.comtag:blogger.com,1999:blog-5001794558945659136.post-68241064977182838882018-07-06T15:51:32.247-05:002018-07-06T15:51:32.247-05:00In theory, I could measure the error as I have don...In theory, I could measure the error as I have done and write a post-processing script to run on the the gcode file that would bump all Z axis movements by the error amount based on where it is in that 2 mm repetitive cycle.Mark Rehorst DMDhttps://www.blogger.com/profile/13616282129430791353noreply@blogger.comtag:blogger.com,1999:blog-5001794558945659136.post-75701868642590138032018-05-11T12:10:06.049-05:002018-05-11T12:10:06.049-05:00This kind of problem has existed for a very long t...This kind of problem has existed for a very long time in the worm gears used for telescope positioning, since non-linearities in the worm gear's travel per rotation will show up as a speeding-up and slowing-down of the telescope's travel over very small distances. When tracking stars at high magnification, this is ruinous to long exposure photography. Two techniques have been used to deal with it when resorting to very expensive high-precision gearing is out of budget: First, Periodic Error Correction (or "PEC"), in which the actual travel of the worm gear is plotted, as you have, and this is stored in system firmware as a correction table, such that the rotation of the worm is sped up/slowed down by an offsetting amount at the appropriate points in the curve so as to flatten out the travel profile as the scope corrects for the Earth's rotation. The system just needs to understand where in the rotation profile the worm is at any given location. This can make a huge difference, and make an otherwise unusable worm far more accurate without an increase in cost. The Meade prosumer grade scopes (LX200 series) use this technique. The other technique is more common because it's generic to the specific mechanics, and is known as autoguiding. In this technique, the user focuses the telescope on the desired point, and then a secondary CCD camera chooses a secondary suitably bright "guide star" in its field of view, and issues real-time corrections to the drive motors as the guide star appears to "wobble" due to any inaccuracies in the drive train. The correction capability in this kind of technique is limited only by the pixel resolution of the camera relative to the optics, and its ability to distinguish when the guide star has perceptibly moved from the desired location.<br /><br />The first technique sounds entirely usable for 3D printing with suitable firmware modification, but the second technique less so due to its reliance on precision optics and the fact that the telescope is trying to remain focused on a fixed point as the earth moves under it.<br />Andyhttp://rosinsmoke.wordpress.comnoreply@blogger.comtag:blogger.com,1999:blog-5001794558945659136.post-68633349670845801952018-05-11T11:23:19.026-05:002018-05-11T11:23:19.026-05:00Can you calibrate the error out with your ustep re...Can you calibrate the error out with your ustep resolution? Probably would need an additional sensor to know your worm location at all times?midlifemadnesshttps://www.blogger.com/profile/13238450311905572985noreply@blogger.com