Forum Replies Created
duckParticipantNovember 20, 2018 at 10:37 amPost count: 49
Thank you for posting the video showing how the Arm Bottom broke.
I don’t know if it is helpful, but the Arm Bottom that I received from Niryo in my v2.0 kit weighed 85g. I suppose different filaments might have different weight and other properties but maybe that can serve as a reference point for the parts that you are printing.
I hope that the 40% fill part holds up better.
One more question: how are you embedding YouTube videos in your posts? When I tried to paste the EMBED text from Youtube into my thread in the Showcase section, I got a security error. Niryo’s WordPress functionality doesn’t seem to like something in that text.duckParticipantNovember 20, 2018 at 10:07 amPost count: 49
Here is a short video that I made of the first power up attempt. As you can see in the video, there were no sparks, but I did encounter an issue with the connection to the Axis 3 motor on the CAN bus, so I will attempt to diagnose the issue and update you all again soon.duckParticipantNovember 20, 2018 at 6:17 amPost count: 49
Hello again everyone,
Here are some more detailed notes about my experience assembling a Niryo One v2.0 robot. These notes are intended for several audiences: (1) people who might be interested in assembling a robot themselves (to give an idea of some of the challenges others have faced), (2) those of you who have already assembled a Niryo One robot (to give you an opportunity to share your own experiences whether they are similar or different from my own), and (3) to the staff of Niryo to help improve the assembly procedures and to make them easier for others.
The diagrams in the assembly instructions in the Google Slides ESTIMATED that the five steps of the assembly procedure should take about 4 hours. The ACTUAL time that it took me to follow the five steps, not including preparation and review of parts before actual assembly, was closer to 8 hours. Maybe I’m a slow poke!
Here’s a detailed breakdown:
STEP 1 — Assembling the Base; ESTIMATED: 40 MIN; ACTUAL: 100 MIN
STEP 2 — Assembling the Shoulder; ESTIMATED: 60 MIN; ACTUAL: 90 MIN
STEP 3 — Assembling the Elbow; ESTIMATED: 60 MIN; ACTUAL: 120 MIN
STEP 4 — Assembling the Wrist; ESTIMATED: 40 MIN; ACTUAL: 65 MIN
STEP 5 — Assembling the Drawer and Final Steps; ESTIMATED: 40 MIN; ACTUAL: 65 MIN
Building a robot is a bit like climbing a mountain. That’s partly why I’ve named the robots that I have after mountains in the area where I live that I have climbed. Let me use this analogy a bit more to provide some concrete recommendations:
- Take your time. Don’t rush.
- Preparation is key. Watch the video for a step in its entirety before you commence the assembly work on that step. Lay out only the parts that you will need for that step on the table before starting.
- Aim to complete each of the five steps in a single day. Sometimes it may not be possible to complete a step in a day. Don’t worry, that’s fine, but it’s better not to be overly ambitious. I would recommend avoiding trying to complete two or more steps on the same day even if you feel like you are on a roll. I tried to complete the last two steps in a single day. When I encountered a simple problem with the 3D printed parts in the base in Step 5, I was too tired to effectively resolve it and had to leave the finishing touches for another day.
- Make sure to get ample rest along the way. Drink fluids, make sure that you are physically comfortable.
- Consider the unboxing activity as the first step to tackle. Clear a table and lay all the components out sorted by category (electrical, mechanical, 3D printed parts, etc.). Compare what you have received with the components in the full detailed BOM on Google Drive. In my case, there were several missing and damaged components, but the staff at Niryo was very kind to send them to me promptly and it didn’t end up delaying the assembly, because I was able to start and the missing parts arrived just when I needed them.
- Post photos and questions here on the forums in the troubleshooting section if you get stuck. The chances are high that someone else faced the same issue in their assembly work and would be happy to offer suggestions; it might take a few hours or even a day or two though for someone to see your post, so be patient and try to make your post as soon as possible after you encounter a problem to allow for enough time for people to respond.
- You will need some basic tools and supplies in addition to the tools included in the kit. For example you will need a little bit of all-purpose grease (as used for bicycles and automobiles). You’ll also need some pliers, wire cutters, sand paper or files, normal screwdrivers, etc. I’ve posted a photo in the Google Drive directory mentioned below at the bottom of this post showing the extra tools I made use of during the assembly.
HANDLING ARTIFACTS FROM 3D PRINTING
The Niryo One is the first useful object or machine that I have personally encountered that is made out of 3D-printed parts. I live in an area that has many factories that produce the factory machinery to make injected plastic parts and, in general, 3D printing tends not to be particularly well regarded in my area. I have also worked on projects in this industry many years and perhaps some of the bias against 3D printing has rubbed off on me, but Niryo One is definitely changing my opinion.
That said, I am somewhat persnickety about PLA threads that droop down on overhanging surfaces and prevent easy insertion of screws and mating with other parts.
I would recommend that people do an initial careful inspection of the 3D-printed parts that they have received (or printed themselves) and remove any glaring artifacts. Some examples and additional detail.
- Small threads around the fan mount inside the base caused a lot of noise on my first robot; it sounded like a lawn mower! Cutting them away made things very quiet.
- When I was assembling the second robot (v2.0), I was confronted with a problem in Step 5 where the Drawer wouldn’t close without bulging— you can see that the Drawer isn’t completely closed at the end of Step 5 of my video; there is also a photo in the Google Drive. I ended up having to file around the semicircles on the blue PLA of the base where the black Air Routing parts are inserted on the side in order to get the Air Routing parts to insert properly and for the rail on which the Drawer sits to slide smoothly.
- I almost wish that I had the foresight to install the Air Routing parts and test sliding of the empty Drawer before completing Step 1 because it would have been alot easier to work with these parts before the full weight of all the parts of the arm had been assembled, but I understand why Niryo may have preferred to suggest that the Air Routing parts only be installed in Step 5.
- Sometimes the PLA threads don’t cause any problem and I’m fining leaving them. In many cases taking them away might lead to more problems than leaving them be would cause. Once you remove one layer sometimes you find what’s underneath is worse, so I would recommend to take a cautious and conservative approach to the removal of these artifacts. I find the threads on the overhang which form the flat top of the Base asthetically pleasing but they are in fact loose (not completely integrated into the structure of the base). On the other hand, they don’t cause any mechanical or structural problems and the texture that they create is a fun aspect of the design of the robot.
I have taken some photos of the parts that I received, the extra tools that I used, some of the 3D printing defects that I encountered and some other aspects of the assembly process. You can view them here.
I am going to stop here because this post is already too long, but I will endeavor to post some more notes as questions and queries for Niryo about some issues that I encountered during the assembly process.
Please feel free to post comments in this thread with your own experiences and/or any questions you might have.duckParticipantNovember 19, 2018 at 9:58 pmPost count: 49
To answer your question above, no, the arm does not straighten on its own on either of my robots. The weight of the full arm when assembled seems to counteract the torsion spring. I did experience the kind of behavior to which I believe you are referring in regards to the Shoulder assembly after completing Step 2 of the v2.0 robot assembly (before the Elbow and all the parts of the Arm where attached in Step 5).
You mentioned that you found something broken when you disassembled the Shoulder. Was the 3D-printed Arm Top broken inside where the torsion spring attaches? That would probably make the torsion spring float freely inside and explain the behavior you were experiencing.
Would you please post a photograph showing what was broken? I am just interested in what was causing the problem. This would probably be of use to others on the forum in diagnosing problems that might occur with their robots.duckParticipantNovember 19, 2018 at 7:32 amPost count: 49
I guess I am used to doing automatic calibrations periodically since with the older software the manual calibration method didn’t work as reliably.
Do you have a v1.x robot with a stepper on Axis 4 or a v2.0 robot with a Dynamixel XL-430 on Axis like @wodcleaner?
I assume that your robot is able to complete automatic calibration successfully. How is it behaving otherwise? Any problems or is it working as expected?duckParticipantNovember 18, 2018 at 9:12 pmPost count: 49
I am wondering if the problems that you are experiencing are related to something happening with the torsion spring inside Axis 2 (the Shoulder) becoming unseated from it’s mount in the Arm Bottom.
I was surprised at the strength of the torsion spring during the assembly of the v2.0 robot. It may have to do with the differences in the assembly procedure compared to v1.x, but I am wondering if I the torsion spring in the v1.x robot that I have is having some issues. Neither of the robots that I have is behaving as extremely as what @wodcleaner shows in his video; I really do suspect that what he is seeing is a fully unseated torsion spring.
It would probably be best to see what Edouard and others from Niryo might advise, but it might be worthwhile for you to disassemble the Shoulder to check whether the torsion spring is still seated properly.
I have had some experience disassembling the Shoulder. It is somewhat difficult to accomplish. You need to use a regular screwdriver and a putty knife to carefully coax the torsion spring out of it’s sockets in order to get the Arm out of the Shoulder.duckParticipantOctober 31, 2018 at 8:45 pmPost count: 49
Thank you for sharing the additional information and your experience with Google Drive. I edited my post above to embed direct image URLs from Google Drive. I also changed the URL for the XML code to the same format, but in my browser it downloads the XML code into a file.
If you have any additional ideas about how to share XML or Blockly code directly in a way that would be easier to share with others, please let me know.
For reference purposes, I found the following article useful in helping me understand how to construct the URLs from images on Google Drive so that I could embed the images: How to upload images for direct URL in Google Drive [elfsight]duckParticipantOctober 31, 2018 at 7:33 amPost count: 49
Thank you for your continuing support.
I suppose I should post the screenshot images and XML code on a server where I can use the image file names (instead of google drive which provides non-content-indicating links) so that I can directly embed them in a message on the forums.
Is there a way to embed XML inside a message on the forums? I tried pasting the XML code directly in the “Text” and “Visual” tabs on a post, but it is not rendered correctly.
I would appreciate any advice that you might have about this so that I can post code more easily on these forums in the future.duckParticipantOctober 31, 2018 at 7:11 amPost count: 49
Thank you for the feedback and encouragement, @Edouard-Renard.
Here are screenshots and the XML for the blockly code for this sequence.
I will be sure to keep you updated on the progress of the assembly of Katsurao’s younger brother.
duckParticipantOctober 18, 2018 at 10:10 amPost count: 49duckParticipantOctober 18, 2018 at 10:09 amPost count: 49duckParticipantAugust 13, 2018 at 9:45 pmPost count: 49
- This reply was modified 1 year, 5 months ago by duck. Reason: replaced links from Google Drive with direct URLs for images and XML code
Thank you Edouard for your thoughtful and detailed reply to some many issues.
I thought I would post some updates on several of the issues in my original post for the reference purposes in any case anyone else experiences similar situations.
1. fan in base is very noisy
This issue seemed to be related to strings of filament inside the Base, artifacts of the 3D printing from the extreme overhang inside this object.
I was able to substantially reduce noise by cutting away the strings and artifacts that I found during a teardown of the Base following the YouTube videos in the Assembly Manual.
2. “gear grinding” during calibration
3. lurching (non-smooth) movement on some axises
6. double calibration necessary to issue commands without receiving errors
These issues seem to have been resolved for the most part after the work that I have done in rebuilding Axis 2 and Axis 3. Please refer to additional detail of this work in my other posts in this section of the forum (“Niryo One calibration issues” and “plastic shavings under shoulder and motor for joint 2”).duckParticipantAugust 13, 2018 at 9:23 pmPost count: 49
Thank you for your prompt and thoughtful reply to my post.
This issue was on the back burner for me since I was repairing Axis 2 and Axis 3 as documented in my separate posts (“Niryo One calibration issues” and “plastic shavings under shoulder and motor for joint 2”) in this section of the forums.
It appears that I am still periodically running into this issue. I will endeavor to diagnose in more detail about the exact circumstances when the robot is unable to perform a particular movement.
I have reviewed the new debugging recommendations that you have posted in the Docs section of the website and started tinkering around with the advice in “Debug Niryo One motors one by one” and I will take a closer look at that. I will probably have some questions and I will either post them in this thread or start a new one as appropriate.duckParticipantAugust 13, 2018 at 8:39 pmPost count: 49
Hello Niryo community,
I would like to just give an update on the resolution of this issue.
I hope that Edouard or someone at Niryo will reply if they have additional insights or if I have made any mistakes in my understanding or explanation.
The problem was related to misalignment of the Arm Bottom and friction between the inner surface of the Arm Top and the Arm Bottom and the GT2 belt associated with Axis 2. The fix involved disassembling the Arm, sanding the inner surface of the Arm Top, replacing the GT2 belt and reassembling the Arm and re-tensioning the belt for Axis 2 by following the YouTube videos in the Assembly Manual.
I am indebted to Edouard and the team at Niryo for providing me valuable support and encouragement through this work.
I have a robot that was assembled in France before April. Edouard told me that they decided to make some modifications to the inner surface of the Arm Top when they revise the STL files so that others won’t experience similar problems (apparently no one else has reported the same issue that I have).
As I noted in my reply today in the thread “Niryo One calibration issues”, I believe that this issue may have been exacerbated by the jostling the assembled robot received during shipment from France.
For reference purposes, here are some photos that I took that show detail of the original issue. If your robot is properly assembled, you should be able to fit a playing card between the surfaces of the Arm Top and Arm Bottom above the stepper motor for Axis 2.
Thank you for your support, Edouard and Marc. Happy roboting everyone!duckParticipantAugust 13, 2018 at 8:10 pmPost count: 49
Hello Niryo community,
Forgive my much belated post on this thread to report that I was able to get my robot working once again and to try to provide a synopsis of what was wrong and what needed to be fixed.
I hope that Edouard or someone else at Niryo will also add further comments if I have missed anything or my understanding is incorrect.
Marc and Edouard took a look at the video that I posted earlier in this thread and quickly determined that the problem was related to Axis 3. To quote from Edouard’s email on July 4: “30 s : the calibration of axis 3 starts at 0:29 and is stopped at 0:59, which means timeout).”
I took apart the Arm Top and the gear train associated with Axis 3 by following the YouTube videos associated with the Assembly Manual. Essentially the fix was to re-tension the belt.
Edouard and the team in France were very helpful in giving me hints and ideas how to overcome the obstacles that I encountered. The fix for my robot involved a bit more work because I have one of the early robots that was assembled by the team in France in before April, so there were some production issues that Edouard told me they have since resolved in the robots that they have started shipping in August.
For reference to some of these issues, please refer to the separate post that I made in this section of the forum titled “plastic shavings under shoulder and motor for joint 2”.
It appears that the tension on the belts for both Axis 2 and Axis 3 were disturbed during shipment, so I would recommend that those of you who have received assembled robots (or anyone who reships the robot or takes it for a short car ride somewhere) be sure to check the tension of these belts and review the YouTube videos related to the Assembly Manual if you experience any periodic issues related to calibration.
Good luck with your robotic adventures and I look forward to seeing what the rest of you create using the Niryo One.
- This reply was modified 1 year, 7 months ago by duck.