tag:blogger.com,1999:blog-1024848558336119865.post1852383352901640775..comments2024-03-20T02:26:18.297-04:00Comments on Sublime Layers: FSR Auto Calibration Setup for Duet ControllerSublimeLayershttp://www.blogger.com/profile/07266175443539080935noreply@blogger.comBlogger29125tag:blogger.com,1999:blog-1024848558336119865.post-48003064535547656432018-09-26T00:34:17.303-04:002018-09-26T00:34:17.303-04:00Just checking in to say thanks, and that I did thi...Just checking in to say thanks, and that I did this. It worked. I like the FSR's, my first layer was perfect.Anonymoushttps://www.blogger.com/profile/04990641274991466660noreply@blogger.comtag:blogger.com,1999:blog-1024848558336119865.post-57781434027366889942018-06-02T12:09:19.181-04:002018-06-02T12:09:19.181-04:00I wanted to thank you for this excellent read!! I ...I wanted to thank you for this excellent read!! I definitely loved every little bit of it. I have you bookmarked your site to check out the new stuff you post. <a href="http://www.kctparts.com/" rel="nofollow">doosan parts</a><br />Furkhundababy1https://www.blogger.com/profile/15781622677426062096noreply@blogger.comtag:blogger.com,1999:blog-1024848558336119865.post-18193955376116296362018-04-29T20:22:39.917-04:002018-04-29T20:22:39.917-04:00Thank you so much for this guide, I now have my Ro...Thank you so much for this guide, I now have my Rostock fully converted to a Duet Wifi + FSR system<br /><br />I also wanted to give back a little bit by sharing the spring tension bed clip I thought of, printed, and have successfully utilized.<br /><br />When I had the FSR plate cut from melamine, I noticed that the hole locations for a bed clip system might split the melamine I chose with a screw, and it made me think:<br />Maybe I dont want to keep tightening or untightening a screw in this wood if I have to remove the glass.<br /><br />So I put a spring on a wood screw, and designed a bed clip that restrains the glass with the spring in mind. I'd like to send you that file and a picture of the setup if I could, if others would find it useful that is.Anonymoushttps://www.blogger.com/profile/09505268447734180119noreply@blogger.comtag:blogger.com,1999:blog-1024848558336119865.post-33383060204001896292018-01-20T12:20:10.520-05:002018-01-20T12:20:10.520-05:00So I've dialed my height back to avoid crashin...So I've dialed my height back to avoid crashing into the bed. I'm trying to do auto calibration and have ran into two issues. I get the following error message on all probes. "Z probe already triggered at start of probing move". Also on one of the points the stepper motor slaps the z tower. I'm using a titan aero mount.Anonymoushttps://www.blogger.com/profile/10505061270045538325noreply@blogger.comtag:blogger.com,1999:blog-1024848558336119865.post-67322409295617894132017-08-23T22:18:21.452-04:002017-08-23T22:18:21.452-04:00Yes, absolutely. You use one of the PWM fan ports ...Yes, absolutely. You use one of the PWM fan ports and use M106 to enable thermostatic control. This is what's in my config.g file:<br /><br />M106 P1 T50 S255 H1 ; Set hotend heatsink FAN1 thermostatic control at 50°C<br />SublimeLayershttps://www.blogger.com/profile/07266175443539080935noreply@blogger.comtag:blogger.com,1999:blog-1024848558336119865.post-33168967707669567772017-08-23T22:15:14.556-04:002017-08-23T22:15:14.556-04:00Michael,
Is there a way to wire the "peek&qu...Michael,<br /><br />Is there a way to wire the "peek" fan similar to the RAMBo where it is on when the hot end is on and then shuts off below 50C?<br /><br />Thanks!Anonymoushttps://www.blogger.com/profile/11187496602040566157noreply@blogger.comtag:blogger.com,1999:blog-1024848558336119865.post-87245477204073594102017-08-04T22:22:06.484-04:002017-08-04T22:22:06.484-04:00Hey Michael
I just wanted to say thank you again ...Hey Michael<br /><br />I just wanted to say thank you again for your help! I was finally able to get a decently flat first layer by reducing my 13 point calibration circle to 85MM in diameter. I can now print out to almost the entire bed diameter. The last issue I'm still having is inconsistent overall Z height due to some filament oozing out of the nozzle during the startup calibration (which I'm now doing before every print). Right now I have to watch as the first layer starts and manually adjust the "Z Baby Stepping" offset up or down to get the layer just right. Do you have any suggestions for getting consistent initial calibrations?Anonymoushttps://www.blogger.com/profile/08215512985239997168noreply@blogger.comtag:blogger.com,1999:blog-1024848558336119865.post-82371483463425008572017-08-02T23:32:12.032-04:002017-08-02T23:32:12.032-04:00Feel free to email me michael.hackney@sublimelayer...Feel free to email me michael.hackney@sublimelayers.com<br /><br />However, the issue you are experiencing is due to the Rostock MAX's geometry as best we can tell. The way I deal with it is to reduce the outer probe radius by 10mm or so. The further in from the towers you probe, the better the deviations will be. I can describe why in an email.SublimeLayershttps://www.blogger.com/profile/07266175443539080935noreply@blogger.comtag:blogger.com,1999:blog-1024848558336119865.post-16209370305661765072017-08-02T23:27:53.028-04:002017-08-02T23:27:53.028-04:00Hi Michael,
Sorry for posting here again. I follo...Hi Michael,<br /><br />Sorry for posting here again. I followed your instructions to create that probe macro, and generally the probes have been pretty consistent, but I'm noticing a trend that probe points in-between the towers seem to be lower than at the towers themselves (probe test results in positive-negative-positive-negative... output, using points from your bed.g file for the Rostock-Duet conversion). My setup has the Onyx floating on your plate on the silicon pads on the FSRs mounted to the Rostock panel. For some reason getting rid of the double sided tape I had attaching your plate to the silicon pads seemed to drop my deviation to around 0.12-0.13 (odd?). Any ideas on what could be causing the problem? I feel like I should try to converse with you in a more convenient method, such as email. <br /><br />Thanks againAnonymoushttps://www.blogger.com/profile/06803024486803202385noreply@blogger.comtag:blogger.com,1999:blog-1024848558336119865.post-83806500176365610812017-08-01T20:19:55.890-04:002017-08-01T20:19:55.890-04:00Hi Michael
Thanks for the reply! When I did the H...Hi Michael<br /><br />Thanks for the reply! When I did the H value calibration, I did notice the "taco effect" to some degree. Here are my current H values:<br /><br />G30 P0 X0.00 Y135.00 Z-99999 H.004<br />G30 P1 X116.91 Y67.50 Z-99999 H.029<br />G30 P2 X116.91 Y-67.50 Z-99999 H.104<br />G30 P3 X0.00 Y-135.00 Z-99999 H-.026<br />G30 P4 X-116.91 Y-67.50 Z-99999 H.055<br />G30 P5 X-116.91 Y67.50 Z-99999 H.079<br />G30 P6 X0.00 Y67.50 Z-99999 H.001<br />G30 P7 X58.46 Y33.75 Z-99999 H-.004<br />G30 P8 X58.46 Y-33.75 Z-99999 H.126<br />G30 P9 X0.00 Y-67.50 Z-99999 H.075<br />G30 P10 X-58.46 Y-33.75 Z-99999 H.115<br />G30 P11 X-58.46 Y33.75 Z-99999 H.061<br />G30 P12 X0 Y0 Z-99999 S6<br /><br />I ran the test you suggested using S-1 and verified these findings. I'm using your bed design from trick laser. The FSRs are stuck to the Rostock V3 top plate, with one rubber pad stuck on top of each. Right now the melamine bed is just sitting on top of the rubber pads. The Onx PCB is stuck to the melamine using the recommended tape and a glass palte sits on top of that, which is currently stuck in place with Kapton tape along the edges. I'm intending to make swivel clips for the glass plate as soon as I can print something reliably. Thanks for any suggestions you might have!Anonymoushttps://www.blogger.com/profile/08215512985239997168noreply@blogger.comtag:blogger.com,1999:blog-1024848558336119865.post-71090821831538355322017-08-01T11:49:16.738-04:002017-08-01T11:49:16.738-04:00Meant to add, I doubt there's an issue with th...Meant to add, I doubt there's an issue with the plate or buildtak but the above test will shed some light on that.<br />SublimeLayershttps://www.blogger.com/profile/07266175443539080935noreply@blogger.comtag:blogger.com,1999:blog-1024848558336119865.post-82363404594448604642017-08-01T11:48:28.448-04:002017-08-01T11:48:28.448-04:00Hi Leon. Large deviations like you are seeing can ...Hi Leon. Large deviations like you are seeing can be due to a number of things. You've done the first thing, which is make sure the FSRs are not binding or restricted. The plate really helps eliminate that issue, but how are you attaching the plate to the printer and FSRs? I simply dab silicone on top of the FSR pad and press the plate in and let cure.<br /><br />Next up, make a copy of your bed.g file and save it called probetest.g as a macro. Change the last probe line from S6 to S-1 which tells RRF to run through the probing and report the individual probe positions and the calculation results but does not change the calibration settings. Do this 3 times and look for variability across each probe location. Feel free to email or message me on the forum and I'll take a look - or you could become a patron and I'm on my Slack channel ;)<br />SublimeLayershttps://www.blogger.com/profile/07266175443539080935noreply@blogger.comtag:blogger.com,1999:blog-1024848558336119865.post-34142976864750739082017-08-01T02:53:28.111-04:002017-08-01T02:53:28.111-04:00Hey Michael,
I've been following your blog/fo...Hey Michael,<br /><br />I've been following your blog/forum posts for guidance as I upgrade my Rostock Max v3 to Duet Wifi and FSRs for probing. So far, it's been going great. However, as I was running the autocal G32 I couldn't seem to get a deviation value of less than around 0.15~, which is not really close to the 0.05 you recommended, even after multiple autocals (all consistent around 0.15). There does not seem to be any obvious mechanical issues with my FSRs (using your new plate, LEDs trigger on very light taps across the entire bed). Do you think perhaps my glass bed or buildTak on top of that may have inconsistencies, or perhaps it is another issue?<br /><br />Thanks!Anonymoushttps://www.blogger.com/profile/06803024486803202385noreply@blogger.comtag:blogger.com,1999:blog-1024848558336119865.post-47566845997274388342017-07-26T14:48:11.982-04:002017-07-26T14:48:11.982-04:00Tyson, the first thing you need to do is determine...Tyson, the first thing you need to do is determine if there is variability in probes from run to run. To do this, open bed.g in DWC and copy it. Then make a new Macro and call it Probe Test. Copy the contents of Bed.g into it. Then find the last G30 line in the file and change the S6 to S-1. Here's what mine looks like:<br /><br />G30 P12 X0 Y0 Z-99999 S-1 ; center and auto-calibrate 6 factors<br /><br /><br />Save and then run this macro. It will probe and report it's results on the console without changing the calibration. Run it 3-5 times and then compare each of the probes points across all runs. They should all be remarkably close, if not, you have an issue with probe triggering. If you used my printed FSR mounts, there could be slight binding in one or more.<br /><br />The other thing to be aware of is a lot of people are observing what we call the "taco effect" on V3s. Basically the probes near a tower are low and opposite are high, so you get a wavy set of probe points for both the outer and inner radii in the probe set. If you have H values in the G30s in your macro, remove them first (only from the macro for now).SublimeLayershttps://www.blogger.com/profile/07266175443539080935noreply@blogger.comtag:blogger.com,1999:blog-1024848558336119865.post-12357647230530720202017-07-26T14:28:26.991-04:002017-07-26T14:28:26.991-04:00Hi Michael - I've been following your work in ...Hi Michael - I've been following your work in my quest to get my Rostock Max V3 into service and have completed the Duet Wifi/JohnSL FSR/E3DV6 upgrades, but can't seem to get my machine to level correctly. I always get high spots on the order of 1-2 mm in the sections of the bed between the towers. This seems to be an issue with the delta radius calibration, but no matter how many times I run the calibration routine nothing helps. I did go through the process of adjusting the H values in the bed.g file, which helped a little. Right now the delta radius the Autocal settles on is around 144.2mm. Any idea what could be going on? Thank You!Anonymoushttps://www.blogger.com/profile/08215512985239997168noreply@blogger.comtag:blogger.com,1999:blog-1024848558336119865.post-42013279299711684692017-07-01T11:45:24.081-04:002017-07-01T11:45:24.081-04:00Done! Sorry about that, I do try to tag things but...Done! Sorry about that, I do try to tag things but this one got away from me.SublimeLayershttps://www.blogger.com/profile/07266175443539080935noreply@blogger.comtag:blogger.com,1999:blog-1024848558336119865.post-62255772096142237462017-06-30T18:48:22.520-04:002017-06-30T18:48:22.520-04:00Could you add tags to this to make it easier to fi...Could you add tags to this to make it easier to find? This page was the missing puzzle piece, in conjunction with "converting the rostock to Duet Wifi" and "new FSR bed system". Thanks!Schlaunchahttps://www.blogger.com/profile/15796488699501460696noreply@blogger.comtag:blogger.com,1999:blog-1024848558336119865.post-46423228764951039652016-07-14T08:16:51.988-04:002016-07-14T08:16:51.988-04:00Thanks Mait, I'll fix the link in the text. Ch...Thanks Mait, I'll fix the link in the text. Cheers!SublimeLayershttps://www.blogger.com/profile/07266175443539080935noreply@blogger.comtag:blogger.com,1999:blog-1024848558336119865.post-91944229620674437552016-07-14T02:22:05.402-04:002016-07-14T02:22:05.402-04:00Hi,
Thank you for the writeup!
I'd like to po...Hi,<br /><br />Thank you for the writeup!<br />I'd like to point out that link to David's web-based bed.g generator is broken. Your "http://for%20my%20calibration%20above/" should be "http://escher3d.com/pages/wizards/wizardbed.php"<br /><br />Regards,<br />MaitAnonymoushttps://www.blogger.com/profile/18161457932205110014noreply@blogger.comtag:blogger.com,1999:blog-1024848558336119865.post-10491348024276420052016-03-13T22:02:49.191-04:002016-03-13T22:02:49.191-04:00AH, it sounds like you have your initial M665 &quo...AH, it sounds like you have your initial M665 "H" value set to high. Do this:<br /><br />home the printer and measure the distance from the nozzle to the bed. Subtract 5mm and then enter M665 HNNN where NNN is your height. Here is the comment in David's config.g file:<br /><br />;*** The homed height is deliberately set too high in the following - you will adjust it during calibration<br />M665 R105.6 L215.0 B85 H240 ; set delta radius, diagonal rod length, printable radius and homed height<br /><br />The behavior should be that the nozzle comes down to within a few mm of the bed, travels to your first probe point, probes, next probe point, probe, etc.SublimeLayershttps://www.blogger.com/profile/07266175443539080935noreply@blogger.comtag:blogger.com,1999:blog-1024848558336119865.post-63908416336334804182016-03-13T22:01:24.306-04:002016-03-13T22:01:24.306-04:00Nevermind. In the default gcode, I thought it said...Nevermind. In the default gcode, I thought it said to overestimate the bed height which was contrary to what you wrote...bed height minus ten.<br /><br />You were right, my bed height is roughly 160 and entering 150 in the config.g file worked perfectly. Overestimating (165) was causing the dragging during the probe.<br /><br />Thanks again, glad I found your blog.Anonymoushttps://www.blogger.com/profile/15166353960792560491noreply@blogger.comtag:blogger.com,1999:blog-1024848558336119865.post-82916424851054819812016-03-13T21:53:15.046-04:002016-03-13T21:53:15.046-04:00Did not know about the jumper for setting the John...Did not know about the jumper for setting the JohnSL board to NC, thanks for the tip. Thought it was causing homing problems (and it might have been), though appearently I had swapped my X and Z motors.<br /><br />Question though, when I run the AutoCal, the nozzle drops to the bed and pushes /drags on it through out the entire probing process. In Rich Cattel's Marlin calibration, it would probe the bed from 5mm or so before moving to the next point.Anonymoushttps://www.blogger.com/profile/15166353960792560491noreply@blogger.comtag:blogger.com,1999:blog-1024848558336119865.post-56889018132130223502016-03-01T12:24:19.114-05:002016-03-01T12:24:19.114-05:00I didn't have the H parameter in my bed.g for ...I didn't have the H parameter in my bed.g for simplicity but I've edited the example in the text above to include H0.0 that can be replaced. Basically, you insert H in the G30 line like this:<br /><br />G30 P0 X-69.28 Y-40.00 Z-99999 H0.0SublimeLayershttps://www.blogger.com/profile/07266175443539080935noreply@blogger.comtag:blogger.com,1999:blog-1024848558336119865.post-35642777706517931252016-03-01T00:38:49.862-05:002016-03-01T00:38:49.862-05:00"In bed.g set the H parameter for for each of..."In bed.g set the H parameter for for each of the probe points like this:<br />H = (trigger height measured at the point) - (trigger height at the bed center)"<br /><br />little confused, I don't see H parameter in the bed.g sample file you have posted. Where do I enter that parameter?Anonymoushttps://www.blogger.com/profile/03097871081017796100noreply@blogger.comtag:blogger.com,1999:blog-1024848558336119865.post-14223999275724656122016-02-11T17:24:36.143-05:002016-02-11T17:24:36.143-05:00Ok, I've published the mounts on Thingiverse h...Ok, I've published the mounts on Thingiverse here: http://www.thingiverse.com/thing:1335095SublimeLayershttps://www.blogger.com/profile/07266175443539080935noreply@blogger.com