tag:blogger.com,1999:blog-4339813531032979196.post2199860024596298584..comments2024-02-28T07:32:59.864+00:00Comments on HydraRaptor: GM17 stepper hacknopheadhttp://www.blogger.com/profile/12801535866788103677noreply@blogger.comBlogger11125tag:blogger.com,1999:blog-4339813531032979196.post-71098713298046576082009-04-25T18:20:00.000+01:002009-04-25T18:20:00.000+01:00Khiraly,
It is a combination of the things that h...Khiraly,<br /> It is a combination of the things that have worked the best so far, but not necessarily the easiest to make. I still have more things to try so it is not my final version.<br /><br />Apart from the motor change and a new motor bracket it is the same as this one: <A HREF="http://hydraraptor.blogspot.com/2009/04/dinosaur.html" REL="nofollow">hydraraptor.blogspot.com/2009/04/dinosaur</A>. The heater and nozzle or described here: <A HREF="http://hydraraptor.blogspot.com/2009/03/simply-better.html" REL="nofollow">hydraraptor.blogspot.com/2009/03/simply-better</A>.nopheadhttps://www.blogger.com/profile/12801535866788103677noreply@blogger.comtag:blogger.com,1999:blog-4339813531032979196.post-73691700951497525452009-04-25T15:45:00.000+01:002009-04-25T15:45:00.000+01:00nophead: Could you post a picture of the whole ext...nophead: Could you post a picture of the whole extruder? Where all the details can be seen.<br /><br />I think this is the latest extruder where all the improvements are compiled in.<br /><br />So I would appreciate some photos especially disassembled:)<br /><br />KhiralyAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-4339813531032979196.post-47748701647556148482009-04-24T19:28:00.000+01:002009-04-24T19:28:00.000+01:00Forrest,
So it looks like the GM17 at 5V with th...Forrest,<br /> So it looks like the GM17 at 5V with the 1:104 or 1:51 gears would have enough torque and speed for screw drive at a reasonable rate. And the shaft encoder and h-bridge give quite a cheap control solution if you have a 5V supply handy and can process the fast quadrature signals. I will put the bracket and coupler on Thingiverse so people can upgrade their extruder from GM3.<br /><br />I might be wrong but I don't think worm gears are very efficient compared to spur gears and would need an extra set of bearing as tin cans and not good with thrust.<br /><br />I was thinking of putting an meccano work gear on a NEMA17 though.nopheadhttps://www.blogger.com/profile/12801535866788103677noreply@blogger.comtag:blogger.com,1999:blog-4339813531032979196.post-76080159114127221602009-04-24T19:10:00.000+01:002009-04-24T19:10:00.000+01:00Erik,
Actually I noticed that the current drawn ...Erik,<br /> Actually I noticed that the current drawn by the stepper changes quite significantly when I load it so you could measure that to get an idea of the torque.<br /><br />Having said that I don't think the torque tells you much about what is going on at the nozzle. It is dominated by the friction at the screw and viscosity at the cold end of the barrel.<br /><br />I think you would need to measure the pressure inside the barrel to get information about flow from the nozzle.<br /><br />But I can't see why we need that. The only uncontrolled variables are: -<br /><br />The filament diameter, which we could / should measure on its way into the extruder.<br /><br />The surface temperature of the object. I don't think that alters how the plastic flows, just how it bonds.nopheadhttps://www.blogger.com/profile/12801535866788103677noreply@blogger.comtag:blogger.com,1999:blog-4339813531032979196.post-42206273801328814862009-04-24T17:47:00.000+01:002009-04-24T17:47:00.000+01:00You know, looking at this again, I am wondering wh...You know, looking at this again, I am wondering why you didn't use the worm gears that you now know how to make to do this job instead of the GM-17 gearbox?Forrest Higgshttps://www.blogger.com/profile/17208965471464716174noreply@blogger.comtag:blogger.com,1999:blog-4339813531032979196.post-59916873803789301542009-04-24T09:06:00.000+01:002009-04-24T09:06:00.000+01:00Great work! I'd love to have less moving mass and ...Great work! I'd love to have less moving mass and still stepper based extrusion. My GM3 is having a very hard time. It hasn't worn out yet, but it probably will.<br /><br />Still I wonder if going open loop with this is the right direction. You'd assume a constant extrusion rate. But maybe some feedback is desirable? If the extrusion is harder to do, it probably filled up the space between the extruder nozzle and the object. Knowing this would improve the print. You'd be able to adjust the horizontal movement speed to the extrusion rate. Perhaps this could even be automatically learned by making a test pattern and measuring the rotation of an encoder on the extruder. Start with widely spaced lines and get them closer and closer to each other until they overlap. When they start to overlap, the software can know (because of a slowdown in extruder RPM) the extrusion width for this height. The effect of extruding very low or high can then be utilized as well. The machine could also adjust the feedrate constrantly during a build to optimize build speed by maximizing the extrusion speed, not keeping it constant and.<br /><br />I know it might make things more complex, but you may get better quality and speeds in the end. What's happening with the plastic should dictate the movement, not the other way around. <br /><br />I'm not an expert at mechatronics, but this is my gut feeling on how this should work. My gut feeling may be wrong of course ;)<br /><br />Of course your print results prove that the route you're going is very successful, but I wonder if it could be even better!Erik de Bruijnhttps://www.blogger.com/profile/09433438819279256757noreply@blogger.comtag:blogger.com,1999:blog-4339813531032979196.post-77538375880399427502009-04-24T00:50:00.000+01:002009-04-24T00:50:00.000+01:00According to Dan, you are not supposed to push the...According to Dan, you are not supposed to push the GM-17 over 9v and not over 6v if it is supposed to last you a long time. BTW, he pulled his notes and it gets 84 oz-in (0.59 Nm) at 6v.Forrest Higgshttps://www.blogger.com/profile/17208965471464716174noreply@blogger.comtag:blogger.com,1999:blog-4339813531032979196.post-31165464024503517892009-04-23T19:36:00.000+01:002009-04-23T19:36:00.000+01:00GM17s are quiet when unloaded but get a lot noisie...GM17s are quiet when unloaded but get a lot noisier when extruding. I was running at 12V though to get the speed for 16mm/s as it drops a lot with load.<br /><br />I haven't looked at the RFI.<br /><br />If we stick to 6V (or 5V from PC PSU) then it would be happier, but then it is not fast enough for an M5 threaded rod extruder. Perhaps changing up to second gear would get the speed with the enough torque, but you are left with having to use the shaft encoder to regulate the speed. The motor may be more efficient but I get the impression that it is not as powerful as the 12V GM3.<br /><br />At 6V though you can use the H-bridge that comes the shaft encoder. The datasheet is in Chinese so I don't really know if it is any good.<br /><br />I prefer to use a stepper as the stop response appears much better and it uses less pins, less firmware and will last longer.<br /><br />The GM17 and its add ons and this hack make a very flexible combination for robotics though.nopheadhttps://www.blogger.com/profile/12801535866788103677noreply@blogger.comtag:blogger.com,1999:blog-4339813531032979196.post-89954726235124263362009-04-23T19:06:00.000+01:002009-04-23T19:06:00.000+01:00Dan also tells me that the motor in the GM-17 is c...Dan also tells me that the motor in the GM-17 is coreless, whatever that means. He says that it produces much less RFI as a result and is a lot quieter (I can attest to that, as I am sure that you've observed, too)Forrest Higgshttps://www.blogger.com/profile/17208965471464716174noreply@blogger.comtag:blogger.com,1999:blog-4339813531032979196.post-52238974075298874072009-04-23T19:04:00.000+01:002009-04-23T19:04:00.000+01:00BTW, the numbers for torque on the solarbotics web...BTW, the numbers for torque on the solarbotics website were wrong, as I suspected. Their webmaster is a bit of a dufus, apparently.<br /><br />Dan Gates tells me that you get 39 oz-in (0.274 Nm)at 3v and 74 oz-in (0.523 Nm) at 6v.<br /><br />Those are measured averages taken over 10 motors tested.<br /><br />Another shop that sells Solarbotics gm-17's has the right numbers.<br /><br />http://www.totalrobots.com/gear-motor-17-2281-he-offset-shaft-p-68.html?main_page=product_info&products_id=68Forrest Higgshttps://www.blogger.com/profile/17208965471464716174noreply@blogger.comtag:blogger.com,1999:blog-4339813531032979196.post-67992201685167673182009-04-23T01:51:00.000+01:002009-04-23T01:51:00.000+01:00LOL! You are impossible to keep up with, nop! :-...LOL! You are impossible to keep up with, nop! :-pForrest Higgshttps://www.blogger.com/profile/17208965471464716174noreply@blogger.com