@Kenneth Wiersema set the channel purpose: Cad the robot/communication about Cad
@andrew_peterson has joined the channel
@parker_hutchinson has joined the channel
@rose_bandrowski has joined the channel
Here are the rules for the game: https://docs.google.com/document/d/1zj_22Bq5r8r8qUdUnMUC3clL5ifV9iZSvIFbYBoAxzU/edit?usp=sharing
Bill of materials will be coming later, probably in the next two days
I am not clear on Bill of materials ? For what ? Supplies you need ?
For what the designers can use in the robots, aka the inventory of the Tetrix kits.
@rosebandrowski has renamed the channel from "cadteam" to "cadteam"
And has anyone played with FRCSim? I'm looking for some way to get familiar with the robot code with something that I can deploy to so I can test that I am doing close to the right thing.
@randygroves: check with @mrosen on this - he spent a substantial amount of time trying to get the sim environment running and found it to be a giant time sync with no payback. As I understand it, there are two competing solutions - the other from autodesk. We briefly looked at this last year and, again, we found no joy. In all cases a certain amount of time is required to define the robot model. This is where the cadteam would be involved. This autodesk solution appears to be getting more continuous investment and iirc relies on unity and ros (both non-trivial installs). One other related topic, the the python environment has a way of simulating a robot (only in the sense of running code and producing outputs), I wrote up some notes on this here:
And yes! it would be great if programmers could debug on a virtual robot, so I always encourage people to investigate the state of affairs.
We haven’t looked into solid works, and porting Helios into inventor for synthesis was aggravating enough, without trying to send it to a different manufacture’s program. I’d say stick with synthesis, as that has autodesk interface, and autodesk is the company we’ve mostly used cad programs from. I’m currently against changing the CAD program unless someone can give me a really good reason why another might be better.
So no, we don’t and I’d just prefer to be having to use only one other cad program to output a robot for simulation rather than add a new and more unfamiliar one to my plate
Fair enough!! I'm looking for a way to start understanding programming the RoboRio without a RoboRio. The FRCSim was one of the first things that bubbled to the top of the pile. It sounds like master Rosen might have already traversed some of this territory.
And being very interested in Python, I will take a look at RobotPy. Looks like there are more teams that are starting to use Python. I realize that this would be a major shift, so not to be taken lightly. @dana_batali made some very interesting points - especially about the complexity of Java and Eclipse.
Synthesis is what is currently being looked at, but I can't really speak to software side of things, I just did the cad for it.
K - thanks. I started looking at that as well.
It certainly has a nice robot simulator - I've been driving the Helios robot around in Synthesis.
cool! i would love to see a demo next meeting if that's an easy thing to do.
I'd say talk to Chris about it, he's been looking at synthesis for driving team practice. I think all you need is a joystick (you can't really run the robot well from arrow keys) and the Cad file and it should work. Buttons might be mapped differently though
I've been using a XBox controller on a PC. You can remap keys - though I haven't done it yet.
@randy_groves so if you have it running, does this mean you can connect a new robot to it for debugging? When you run the Helios robot are you running our helios code? Other questions come to mind: how would our PID code work - eg: how do sensors work?
I can drive it without the Helios programming, I think the code exporter is just for autonomous codes
So I feel like you aren't really driving it - in the sense that our code may remap buttons, remap accelleration curves for drivability, etc?
No - I'm just loading the CAD definition that Kenneth made, and using the Synthesis app to do tele-op driving. I don't know at this point what code possibilities are.
Yes, you can move the robot, and no, it doesn't have any of the extras given by the code. And I don't have anything to use outside of keyboard to drive it, so I never played around with it.
So this is probably a key point for all involved, especially for driving practice... Without running our code what you are really doing is perhaps getting a sense of the field, different viewpoints, etc. just not really learning the nuances of a particular robot's driving experience, etc.. It will be interesting to see how far down this road we find it "affordable" to go.
Well, I don't think anyone has tried to upload the code and run it yet, might be side programming thing to play around with, if there's time
this roadmap shows all the things that aren't yet implemented.
I would get Declan involve in this conversation. @declanfreemangleason @Kenneth WiersemaKenneth please invite Declan to this group.
@declan_freemangleason has joined the channel
So - I'm trying to get the 'Run User Code' option in Synthesis working. After finding that I needed to hard wire JAVA_HOME in the Makefile.java file - per https://forums.autodesk.com/t5/bxd-synthesis-forum/running-java-project-in-code-viewer-make-error-127/m-p/7350959#M109 (I'll detail that later), I now have the following error:
src\org\usfirst\frc\team4915\steamworks\commands\ArcadeDriveCommand.java:6: error: package com.ctre.CANTalon does not exist import com.ctre.CANTalon.TalonControlMode
I seem to remember some instructions on the CANTalon code, but I don't remember where they are.
Ok, I went through this entire process
CANTalon isn't supported, but I have some dummy code written at https://github.com/pietroglyph/2017-STEAMworks/tree/controlmanager-sim that should work.
@declan_freemangleason Thanks - I realized after that I'd seen your note about trying this before. We're moving into our new house over the next 3-4 days, so I may not get back to it for a while.
You might also need this Makefile. You can replace the one under %programfiles(x86)%\Autodesk\Synthesis\SynthesisDrive\HELBuildTool. The key line is 42, where you add things to the classpath.
Once you do that you will need to add the above jar files into %programfiles(x86)%\Autodesk\Synthesis\SynthesisDrive\jars.
@declan_freemangleason Excellent! That will help a lot!
When things calm down, I'll try and document this as I go.
I will note that this doesn't actually run in anything but disabled mode, and the only way I think you can get around that is by using the NI FRC Driverstation software (which I think requires a license key). I'm not sure which mentor has the most experience with that... @riyadth?
@riyadth Yeah - that was going to be one of my next questions - is a license key available?
Could you guys see about moving this to programming, because unless you want a cad model for a simulation software, this isn’t exactly on topic with the channel.
@Kenneth Wiersema Yes - it's definitely straying off the CAD topic.
@declanfreemangleason @randygroves: installing the driverstation software on a pc is straightforward. Yes, you do need a license key which comes with the kit of parts each year. For now, it's possible for randy to use last years' number (which, I believe is either in the programming tote or in coach's hands)
For the meeting Saturday, here's the list of how many parts can be used. Naming of parts can be found here, or ask me what something is. http://troop503.info/robotics/wp-content/uploads/2011/05/TETRIXPartsIdentification.pdf https://docs.google.com/spreadsheets/d/1Kx9TqgTsclc5glKLYtGr_Ytteigif658AULwNPROT3s/edit?usp=sharing
@Sean_Williams has joined the channel
Which CAD program does the team use? Autodesk Fusion 360? (Is there a "Spartronics CAD 101" guide somewhere?)
Yes, we use Fusion 360, I don't think there is a guide anywhere though.
Thanks. And I recommend you think about setting up a document to help get people started. The team can add to it year after year, and it will help more and more each year.
Possibly, but Fusion has a lot of good documentation on how to use the program, and frankly CAD is really learned best by doing projects and figuring out how to use it as you go. You can’t really learn it by reading and watching videos, it’s much more a doing thing.
Understood, but realize that I couldn't even find a document that said which CAD package was being used. You should have a document saying where to get the software, where to find the documents/tutorials, and a list of projects to get started with. That's CAD 101.
Ideally other (non-CAD team) members could grab the document and learn a few things about CAD. Maybe they'll join the team next year.
If any one is interested, there was a fusion 360 update a few days ago, here's a link to a blog post on it. Nothing that really effects us, other than the existence of a browser type interface. (I didn't know about it) https://www.autodesk.com/products/fusion-360/blog/december-5-2017-update-whats-new/
Distribution of CAD people across subteams: @andrewpeterson and @SeanWilliams are working on the Climber @parkerhutchinson and @David Ard are working on the Scissor lift @kaedricholt is working with the Intake
I'm doing non-specific sub-team things and helping when there is a need.
@Kenneth Wiersema I have a feeling you’ll be doing a lot of general file management as well. We want to be sure that everyone is using the latest version of files every day to ensure all systems integrate as desired.
Ex: We don’t want to climber to interfere with the intake system but they may be connected and want to take up similar locations on the robot.
We must remember to put the latest versions of files on the cloud at an agreed upon interval, preferably daily but it depends on how often you make changes to designs. Might need to be multiple times in a meeting.
@whobbs1496 @David Ard @parker_hutchinson Rather than make the scissor arm a rigid body so things don't move, let's get the joints set correctly. This will allow us to manipulate the model and see the extension of arms and overall space allocations with the pistons/actuators in motion. This should also give us the ability to determine space and extension restrictions when placing cubes at various heights.One area of concern is the intake/grabber and its extension past the robot at various positions. A 3D model that moves is a big help in this regard.
I can help out with that, as I’ve got it to work it a different design. Also, I’d like to have some idea of what the intake group is doing please, as there are only two pistons in the folder, and I haven’t seen a drawing either
Comments after seeing the assembly and part creation process in CAD last night: Rather than drag parts back and forth to get placements, there is an option to create joints to origin planes. As long as most parts are created at the origin (or in reference to these planes), this is extremely useful. Ex: the center planes of the scissor lift will align with the center planes of the chassis. David and I changed the scissor lift bottom rails to reflect this. When the rails are inserted into the scissor lift assembly, parts will be symmetric about the XY plane. If parts are created in reference to known locations (like an origin), we can create symmetric joints and mirror parts and bodies about the center planes. Then the center/origin is really the center of everything and the ground/baseline/top plane (XZ) is really the ground/bottom of the robot. It will make the end "full robot" model more intuitive.
@Kenneth Wiersema if you message kaedric or rose they should have a drawing you can look at
Some of the dimensions are off visually, use the numbers not the drawing. It’s a sketch sorry
I was going to wait for permission, but I can start now. Also is the scissor lift getting directly mounted to the chassis, or is there a platform underneath it?
There will most likely be a platform, as I don't believe it can be directly mounted on.
How are the pieces going to be mounted together? I don't see any kind of bar to connect them
We have not designed that yet. We'll be able to get that done next meeting quickly. It looks like someone might have already done some work on the joints.
I got the scissor lift kind of working. I'll explain the issue at the meeting
I’ve recently discovered motion/sliding joints. Should be useful with the scissor lift and the intake since both will rotate/slide.
The scissor lift uses rotational just for almost all of it. I was trying to get the pin slot to work, but there were issues with how the base was set up
@channel I want all Cad designs of the primary modules done by the beginning of the meeting Friday. @andrewpeterson @SeanWilliams get the basic climber stuff designed. We don’t need the gear box ratio for the cad. @kaedric_holt get the harvester done ASAP. That’s the final big space issue.
Climber currently is stuck working around the base of the scissor lift. Which I’ll have done by tomorrow night
The Base of the scissor lift is finished, but it's still flexible on what exactly it is. Currently the scissor lift is nearly at the right height (0.175 in to high, issues). Also the grabber is mounted on the scissor lift, but the flipper system still needs to be cadded. Here's the current base design, so Climber will need to work around this (adding structure is fine, just tell me what you want)
Question: why is the center support inside the chassis vs on top? Because of the harvester? If it moves to the top, we can shorten the vertical support tubes by ~2” and save a bit of weight (not that weight is much of a concern with aluminum tubing).
We can start adding cylinders as sub teams decide on placement and reserve the space needed for the strokes. Also add the air tanks if we know the size. If they’re in the model and people can see their size, it will make determining placement easier.
I was talking to Paul and he suggested this idea of the keel and tree parts. Also, I’d say it reduces materials used. I’d like to keep it on the bottom for now to provide more space for the harvester. And the stabilizing bars are more easily mounted like this
I see that we can post our CAD models to The Blue Alliance. I don't know if we want to, but I thought I'd mention it. (I'm proud of our CAD effort this year! I'd love to show it off to the world.) https://www.thebluealliance.com/team/4915/2018
There’s some polishing that needs to be done, and I’d like to get it up to date with the modern robot. I haven’t had enough spare time with me in a good enough mood to finish it. I’ll probably have it done next week.
@Ethan Rininger has joined the channel
@Jack Scheiderman has joined the channel
@Merrill_Keating has joined the channel
@emerson_nicholas has joined the channel
@Stephen Reinhardt has joined the channel
Here's the presentation from the meeting today-I'll probably be adding more things as the meetings progress. https://docs.google.com/presentation/d/1UxIY1-GUM6EtWENmHqHHd9WqygbBHJsJmmn6K93fkDk/edit?usp=sharing
@Ulysses Glanzrock has joined the channel
@channel Here's the homework that I promised from the meeting yesterday. The link is to a google drive folder that contains the photos of all of the drawings that were produced by mechanics at the last meeting. I'll posting a list of who were in those groups later for you guys to ask questions to. Take the drawings that are in the folders with your names and then start designing a module off of those dimensions. Some of you are going to have more work on figuring out what an actual design would be, as some are just drawings of the prototypes that they made. You are to design a module that would made made specifically to be placed on a robot during the last half of build season and would be a fully realized module. A completed designed is due on the 11/28 meeting, as I'd rather be critiquing a complete module that a half realized thing. Please ask me if you have question about anything CAD related
@channel Copying this from engineering, but all cad people should read this: It's a far better explanation of how things fit together and assemble in FRC than what I've been trying to explain. This should be required reading for all CAD people. https://www.chiefdelphi.com/forums/showthread.php?threadid=167193
I ran across this article on 3D printing gears & seems like a good one to file away here: https://engineerdog.com/2017/01/07/a-practical-guide-to-fdm-3d-printing-gears/ @paul_vibrans FYI
Hello CAD Subteam, I hope all of you have had a wonderful Thanksgiving. The other captains and I have decided that we would like to put together extra meeting dates in the Mechanics shop to help get things done before Build Season. The Mechanics, Electrical and CAD subteams all have tasks that need to be finished. All of these meetings are optional. Below are the dates: Monday 11/26 1:45-4:00 Monday 12/3 1:45-4:00 Wednesday 12/5 3:10-5:00 Monday 12/10 1:45-4:00 Please fill out the survey to whether or not you can come to the meetings by Sunday at 12:00 P.M. (Noon). Please fill out the survey even if you are not planning to come to the meetings. Here is the link to the survey: https://goo.gl/forms/OD3h6wo5xbVugC8L2 If you filled out the link in the email I sent out, please do not fill out the survey again. Thank you, Cruz Strom
@Mark Tarlton I've seen a few posts of CNC'd belly pans on CD, like this one: https://www.chiefdelphi.com/media/photos/46711? I wonder if we should consider using the xCarve to make a sweet belly pan for this year's bot. Nice way to trick out the kitbot chassis, and belly pans can be pretty valuable anyway.
*Thread Reply:* You could do that out of polycarbonate or plywood easy enough. Aluminum would take quite a while I expect. You'd need to keep a close eye on it while cutting in case something went wrong. Rule of thumb 2.5" of material cut per minute for 1/8" thick aluminum.
*Thread Reply:* lot of teams use Baltic birch plywood belly pans, though polycarb may be a sleeker choice
@chrisrin #electronic_pneumatics #engineering ..If we're thinking about fancy belly pans, we may want to think about improving functionality as well. For example, a design that can be more easily removed or rotated for easy access to the control systems. It would require careful planning to route the electrical and pneumatic lines so that it can be pivoted, or with quick disconnects, removed entirely. I remember hearing during competitions last year that better access to the components would have been nice. This could be an area where some experimentation now could be useful.
As Chris Stanley also mentioned, here's a really solid youtube series for learning Fusion 360. It starts with 2D sketching and focuses on fundamentals like deliberate position & dimension and use of constraints. By the 13th lesson you're ready to design a helical gearbox. https://www.youtube.com/watch?time_continue=534&v=y5tp4QXciK4
*Thread Reply:* One more recommendation: lesson 6, which is a long one (around an hour) but worthwhile because it covers parametric design, the use of orthogonal planes, and a few other F360 features in the creation of a hinged box. Good one. https://www.youtube.com/watch?v=VBJq2uf6Pt0
For anyone who is interested, there will be some CNC classes at the BARN over winter break to finish a project that we were working on over the summer. The dates are 12/21 from 3 to 5 pm, 12/23 2 to 4 pm, 12/27 7 to 9 pm and maybe 12/30 4:30 to 6:30. This is not mandatory by any means, but if you’re interested in seeing how cad is applied to machining and how what you make is applied, look into showing up to one of the classes. Please reply to this message to say what dates you’ll be able to make.
*Thread Reply:* I'll be there 12/21 ... I'll try to make the others if possible
*Thread Reply:* I will be able to show up today. I need to figure out if I can make the other dates.
*Thread Reply:* @Jack Scheiderman Now that BARN has the class posted, please signup for the one today, and the next one if you can. There’s a $10 fee per class, as well. https://bainbridgebarn.wildapricot.org/event-3176546
*Thread Reply:* @Kenneth Wiersema the link is telling me it starts at 7. Is that the correct time?
*Thread Reply:* It’s not the correct time, as otherwise there’s a class conflict, that’s why I set it to 3-5 instead of what we used to do from 7-9.
*Thread Reply:* Chris and I are at the barn now. It was a data entry mistake
*Thread Reply:* @Kenneth Wiersema Thanks. Also my dad was wondering if he could come with. Is that ok?
Fusion360 training program for FIRST: https://twitter.com/autodeskedu/status/1075951306446008320?s=21
I have tried to download the zip folder for this years field, however my computer is unable to open this zip folder once it is done downloading. Could someone else try so that we can have the field in the Fusion 360 shared folder.
I’ll be doing that tonight, don’t worry about it
Got the CAD downloaded. The problem with the zip folder was that it was password protected. I found a downloaded step file of it on GrabCad. The synthesis file has been released as well.
@Kenneth Wiersema Is the password the same for the game manual?
Tried, and no, I think it's in relation to the solidworks passwords through first. We have a copy that isn't the greatest though, I've found a few parts missing, but nothing overly critical. It's in the 2019 Deepspace folder
On the SpartronicsCAD folder, it should be right next to 2018
The missing parts seem to be a problem of the original solidworks files. Some says that they'll be fixing the issues sometime soon
@Kenneth Wiersema @dana_batali Sent me a link of 3d printed mounts for the picam, could be useful for a potential redesign of the vision platform (Especially #2 and #7). @Mark Tarlton Would it be a expensive or otherwise outlandish proposition to get these printed as a proof-of-concept? (https://all3dp.com/raspberry-pi-camera-cases-mounts/)
*Thread Reply:* If we don’t have time to print these/you want them soon I’m happy to get them done in PLA on my printer by Sunday.
*Thread Reply:* It's good that we're hitting our first 3D printing workflow scenario... I think it would be good to establish a workflow of prototyping with the Makerbot ABS printer and then Onyx once final. However, if the picam mounts are already proven from Thingiverse or something (I'm sure there are some out there), then they might be small enough to just go ahead and print in Onyx straight away. I'm also willing to print things in PLA on my printer here at my house.
*Thread Reply:* You’re right—we should exercise the team’s equipment to prototype (for practice), or go ahead and print on the Markforged.
*Thread Reply:* Fantastic! Mainly I was shooting these out there focusing more on prototyping, seeing what I could do with a few whipped-up mounts, so probabby the Makerbot is going to be the best bet. We'll touch base during tomorrows meeting.
*Thread Reply:* We can prep the printer bed tomorrow and launch the job when you're ready. Kenneth and Dana both know how to prepare the models once we're ready for the good stuff.
*Thread Reply:* It might be good to write down a few things for the CAD team about how to use the Makerbot ... setup process, slicer profile to use, etc. And do the same for the Markforged, and maybe do a bit of a training session along with that
*Thread Reply:* Quick update from a conversation with Mark on Friday: Mark has taken the teams Makerbot home until further notice, so I would request @declan_freemangleason to give these prints a shot. For reference I am talking about #2 and #7 on that list. Thanks! 🙂
*Thread Reply:* @Darwin Clark Could you point me to that list?
*Thread Reply:* Apologies, it wasn't in the original message, I added it.
*Thread Reply:* To clarify the printer Mark took home is not the Makerbot from the robot room. We still have the Makerbot in the robot room.
Not sure the best place to share these 3D printable transmission parts for prototyping with aluminum tube https://www.chiefdelphi.com/t/team-5254-3d-printed-prototyping-resources/335599
My CAD file of the robot at the end of the meeting 20 Jan 2019.
After some research; This is what I came up with for a potential redesign for the platform. After some digging around for 3d prints, I came across a really good base, located here: https://www.thingiverse.com/thing:3129822. The case has mounting holes in the front, which line up with the mounting holes in the PiCam PCB. It also has small raised poles, which can be mounted to the respective holes on the underbelly of the Raspberry Pi PCB. The main disadvantage of this design is the lack of microSD card access. If a card were to get corrupted during a match (unlikely) and we had to replace it (very, very unlikely) we would have a hard time. What I'm proposing for the design, is we take a simple LED ring mount and modify it. For reference, I'm thinking of something like this here: https://www.thingiverse.com/thing:2243361. We can modify the LED ring mount to have two versions: one with a larder diameter, and another with a smaller diameter. These sizes would correspond with LED ring sizes we would order. To bring the two prints together, we can add mounting holes on the angled sizes of the larger Raspberry Pi mount, and on respective locations on the LED ring mount. Collage image here: https://docs.google.com/drawings/d/1osOsIf9mwq_7XO6KCrMEpRm17B0faCgpwb1v24fWrMQ/edit?usp=sharing @Mark Tarlton @Kenneth Wiersema.
@Kenneth Wiersema, I know you're already working on a master CAD. Is it possible to share the location so we can refer to it as needed (understanding it is not done and aspects are not final or just plain wrong)? Thx
Look under Actual robot CAD folder and find the full robot cad-the only difference between it and the space allocation is that the chassis matches all of the modifications that Paul did, and the spaces cut out.
*Thread Reply:* I wonder if there is a permissions problem... I don't see an Actual robot CAD folder. Folders I see under Spartronics are zChrisrin, MTarlton, Vex Pro, The best robot, Logo, etc. There are a total of 21 folders and 2 files (Tap jig & driver station), but I'm not seeing the folder you're talking about.
*Thread Reply:* This is much more I didn’t have you on the new cad project-I had to remake the folder when will’s frogrock account expired, so the new one is the SpartronicsCAD project. I’ve been adding people to it from needs based, as I’m cleaning out a lot of the team members who were gone. If none of your files reference things outside of them, you can move them over without issues, otherwise you’re going to have to break the links in the files.
@Harrison Gilmore has joined the channel
No you can take that off if you want
If you can, move the small hole away from the larger one
@Jack Scheiderman Print list from Chute folder in F360: IMU Mount (q 2), Rail Distance Sensor (q2), Pressure gauge mount (q1) , Support Frame BLock ( 4 pieces per block, 2 sets needed), Tank Bracket (3 pieces, 1 set) , Scoop (q2),
I decided to try to modify a Thingiverse mecanum wheel design yesterday using F360. Here’s the original: https://www.thingiverse.com/thing:1454685. I learned how to convert the STL files to a modifiable format using this 6-minute video: https://youtu.be/gbnObQ0fmiY. It’s pretty easy but not an obvious workflow. The result of my work is pictured; I think it turned out ok. The plastic roller axles are the weakest part, so instead of male-female press-fit roller pairs, I think metal axles and all female rollers would be better. I also think nylon or Onyx might be good to use for the hubs, and a grippy paint-on coating would be a good enhancement to the roller surfaces. The cool thing is... we can make mecanums any size we want. Used on an intake, we could probably go quite a bit smaller than 4” to save weight & space. I weighed these 4” all-PLA (20% infill) ones: 2.85oz per wheel.
Creality Ender 3 ($230) + another $30 for stiffer bed leveling springs and a borosilicate glass bed. I would not recommend it for the team - a bit too high maintenance, though the two upgrades did help a lot with reliability.
FYI - Fusion has been behaving strangely since the last update. Models load with pieces everywhere that are normally assembled using joints. As soon as I try to move them, all the pieces fly back together!
That's been happening for a long time now, the scissor lift on themis used to always be floating on the side before I tried to move it, and it would spring back.
Ahh... I never saw it spring back before. The hatch grabber CAD is pretty much finished. I probably won't do the elastic linkage. Also the real grabber is finished as well and ready for testing if you guys get to it.
Hey! I was looking around as usual and found a couple CAD-related things to share. One of the very best FRC teams at packaging and sharing resources is team 3847 Spectrum located in Houston. And part of that sharing is a robust CAD library - here's a link: https://workbench.grabcad.com/workbench/projects/gczMZ2vZJYlzPJlob9wznnuhi1uqUhYIRx23KRz0l5Gq_g#/space/gcmegFZmFlpCU6pC0lOikiembdYGx3tphi11JfHJ0THJkK/folder/1357747
Spectrum, like the majority of FRC teams, uses Solidworks as their CAD platform. So here is Lars Christensen covering how to import and manipulate Solidworks models in Fusion 360: https://www.youtube.com/watch?v=mbW-T_WJ0uU
Lots of teams share Solidworks CAD files of their robots and mechanisms like this (though Spectrum's library has a bunch more than just their robot models). It might be valuable to get several examples from other teams of mechanisms (elevators, arms, etc.) we'd like to learn about into F360 so we can look closely at how the teams are designing and putting them together.
Another couple amazing CAD resources: 1) The Blue Alliance grabcad has 121 models you can load into a 3D viewer and spin & zoom all you like. And you could also download most of them and import to F360 if you like. Link: https://grabcad.com/the.blue.alliance.design.hub-1/models?page=1 2) The main Blue Alliance website allows teams to upload their CADs. If you use the Advanced Team Search, you can get a list of the teams that uploaded CADs each year. Note: 2018 and 2019 searches appear to be broken currently (I emailed them). But there are probably ~200 models of robots from 2008 - 2017 available. Link to 2017 search: https://www.thebluealliance.com/advanced_team_search?year=2017&seed=0&playoff_level=0&cad_model=on
*Thread Reply:* There are some inspiringly elegant designs (look simple / probably aren't all that simple). Seniors and mentors around in 2016 will appreciate this one by Big Bad Bob (FRC 319). Link to Grabcad so you can spin/zoom: https://grabcad.com/library/2016-319-big-bad-bob-1
*Thread Reply:* I found a work around if you're interested in reviewing robot CADs on Blue Alliance from 2018 and 2019. Do a Google search using [blue alliance "1 cad model" 2019] or [...2018]
Tried a couple more of these quasi-animatable 2D mechanism sketches. It’s all about taming the constraints. Here’s a fun one https://youtu.be/jCf7Iujh9ms If anyone’s interested in how to do these, happy to discuss
Couple 3D printed prototyping systems I think I shared before, but parking links here again because if we plan to design prototypes, then these pieces might be worth incorporating to reduce assembly time. We could even print some inventory of these to avoid the printing time during season.
PVC pipe prototyping system by 3847 (Spectrum): https://www.chiefdelphi.com/t/protopipe-spectrum-3847/182073 Prototyping system for use with 1x1 and 1x2 tube by 5254 (HYPE): https://www.chiefdelphi.com/t/team-5254-3d-printed-prototyping-resources/335599
@channel I just spent about 50 minutes organizing everything in the 'Downloaded Parts' folder of Spartronics CAD. Now, I need help renaming the files, during break I will post a naming format for everything. Please, I will need help with this because there is a LOT of stuff.
*Thread Reply:* Lucas, thanks for initiating this. We need to establish a common CAD workflow, understood and practiced by all CAD'ers on the team, and using a consistent COTS part library is one of the pieces we're likely to need.
*Thread Reply:* Agreed. After naming things, I would also like to start adding things that aren't already in there like all of the motors we can use, all gearboxes we can use, etc.
Here's the naming format I have so far: Gears: Number of teeth - hole type & Diameter
Flat tooth belt pulleys: # of teeth - type and diameter of hole, width of pulley
Sprockets: # of teeth - Type and diameter of hole - Type of chain used
Why not use the naming/numbering scheme of of company that makes or sells these items? For instance 25B32 is a sprocket for #25 chain with 32 teeth and a Type B hub. Add the bore size and shape as suffixes.
It would be worthwhile to have a conversation about how to organize the folder structure to support efficient CAD workflow for our Fusion 360 designers in the future. And along with that, naming conventions. I asked around about F360 workflow here: https://www.chiefdelphi.com/t/fusion-360-frc-in-2019/357519. One mentor from a team that has used F360 for years provided some good insights - see post 20. To summarize, their structure for each competition year is... 2019 Destination Deep Space Main Robot Model Primary Sub-Assembly 1 Model Primary Sub-Assembly 2 Model Half Game Field Model (including all game pieces & field elements) Folder - Unmodified COTS (with category folders under it) Folder - Parts (includes modified COTS and fully custom robot parts used in sub-assemblies) Folder - Old (old/abandoned designs - put old stuff here to keep active folders cleaner)
I'm not saying we have to use this same structure, though it looks pretty good to me, but I am saying a little planning could help us be more efficient. Thanks again, Lucas, for being proactive.
*Thread Reply:* One folder I would add to the above would hold 2D Geometry Analysis Models
Also, because apparently folders cannot be moved (only files can), I think the Unmodified COTS folder would be better situated one level up, not specific to a year. It would be a pain to move or reconstruct COTS folder structure under the new game folder each year.
One thing to keep in mind: There is currently no space limit on F360 cloud storage, so if we think it would be more efficient to have separate folder structures for each sub-assembly, and then under each sub-assembly folder have a COTS folder holding the parts used for the sub-assembly duplicated from the main COTS folder, we could do that.
Here's a Google sheet with a proposed F360 folder structure based on the above. We could collaborate on design there (feel free to add comments), and then implement the structure to our repository once we're satisfied & aligned. https://docs.google.com/spreadsheets/d/115OliPtIzgE3IAtr-aYOq9T0ZYlLVRdTptK55rsF89s/edit?usp=sharing
@paul_vibrans Do different companies use the same naming scheme?
*Thread Reply:* For roller chain sprockets made to ANSI standards naming is common across all manufacturers. You can check this by doing an Internet search for a sprocket part number and see how many different vendors are listed. For toothed belts and pulleys it is best to use the numbers assigned by the company that first developed that particular tooth shape. The belts and pulleys sold by VEX were initially created by Gates Rubber Co., now just Gates, with the trade name PowerGrip HTD. Just to check, do an Internet search for some representative Gates catalog numbers to see if equivalent items from other manufacturers show up.
@channel I also need help resetting all of the materials in CAD from steel to whatever the part is made out of (motors are exempt).
@chrisrin that is the filepath I was thinking about using. I wanted to separate the COTS with broken and unbroken links and distribute the ones with unbroken links between modules so its easier for the CAD people can access them more easily.
Hey Lucas, I like that lucidchart tool! I registered for it and I'm sure I'll find some uses for it. Here's the folder structure I proposed using the same tool, so it is easier to look at them side-by-side. https://www.lucidchart.com/invitations/accept/94762030-75c5-49d7-b0e4-4927da2c05a0
It'll be good to discuss how workflow will work across the CAD'ers on the design team, and then decide. I think it'd be informative to maybe rewind our minds to January, when we decided to go for the modified Everybot lowbot robot concept.
Finally organized my F360 Youtube links into a playlist. Many, many bases are covered here, but the thing to remember is... You will learn MUCH more by trying the techniques out in F360 yourself than by just watching the videos. Hope this is helpful to some of you. Playlist link: https://www.youtube.com/playlist?list=PLsiEJAZgdF8IsQdt_jM2uhTfBQ-Ez-P1f
@channel I'm creating a series of robotics-centered CAD projects for myself, starting with the simplest things I can come up with and moving up gradually from there. I'm wondering if any of you are interested in trying to do the projects as well to develop your F360 skills
The first project, for example, is modelling an L bracket to be used in a later project for attaching a piece of aluminum tube vertically to the kitbot chassis. Other projects include...
Anyone interested in playing around with these F360 projects & assembly puzzles this summer? Depending on interest, I may also schedule a meet-up 1 night a week, either in-person or virtual (like a Google hangout).
BTW, this is CAD focused. It does not replace the full engineering design stuff Paul is teaching. However, it could be complementary... Learning some more CAD skills should only help as you work to complete your bigger mechanism project with Paul & Mark.
Let me know if you're interested - just reply here or send DM. Thanks!
@chrisrin On a different subject related to gearboxes, I noticed that you used a WCP SS Gearbox. You might be interested in the modified version I am using on my Versa-Frame arm robot project. I turned the twin-motor version of what you used into a single speed low profile gearbox.
The gearbox: https://a360.co/2QNKOWR
The robot: https://a360.co/2Wx34sW
*Thread Reply:* Very cool - when I get a chance I'll take a look. For my little CAD practice assembly project, which gearbox I chose didn't matter too much (though I did pick a single-speed rather than shifting gearbox on purpose).
*Thread Reply:* I finally got a chance to look at your design, Lucas. Looks nice (especially the color scheme 🙂)! Is that versachassis in addition to the versaframe for the arm? I kind of like that option from the perspective of fast design & assembly. Citrus Circuits uses almost exclusively versaframe for primary structural elements (and as Cruz shared they've been to Einsteins 7 years in a row). I also like the West Coast Drive, though I'm not sure about the odd wheel spacing (I get that the omnis should mitigate - just not sure about it). I wonder if the arm drive motors/gearboxes could go back further to shorten the chain/belt run you'll have to do. Or are they there for weight distribution? I do worry a bit about COM being far back when arm is high. Have you assessed risk of back tipping (maybe need ballast in front corners)? Great to see your progress - thanks for sharing.
*Thread Reply:* The arm superstructure is made from versa-chassis square and rectangular tubing. I did this and not 1x2 and 1x1 channel because the superstructure should be the strongest part of the system with minimal flex. I truly love the versa materials after doing this because of exactly what you said: It's quick and easy to put things together in the programme, and I can move where the gearbox is. The coulson wheels are oddly placed because I wanted the rotation point to be as far forwards as possible so that the end of the arm doesn't move much when the driver is making small heading adjustments. And the gearboxes are also acting as counterbalance. When it comes to tipping, I don't see that being a huge problem because there is no drop-centre, however, I do plan on designing a kickstand into the back of the arm. It's an idea I stole from 1684 Chimera's 2018 robot. (Video for reference): https://www.youtube.com/watch?v=fbNrJPOwdbc
@Harper Nalley I recommend DMing risho900 to get access to the Fusion 360 parts library just published. I got access; looks great https://www.chiefdelphi.com/t/fusion-360-parts-library/359191
@chrisrin thanks for the heads-up, just sent him a message
I've been chatting with the Rishi, the mentor from team 6897, and since there are relatively few teams using Fusion 360 we thought it might be worth a try to have a general Slack workspace for those teams to share and collaborate. Here' s a link if interested: https://join.slack.com/t/frcfusion360/shared_invite/enQtNjY1NTI4NjI4NjE1LWRmZTcxOTc2OTgxODdkMjBlNjQ3OGY2MjRjMjE0OGU1NGNlNTgyOTFmMGZkY2EwOGNkZjNkOWRiNDlhZjNkZWM
Are students interested in casually doing a sequence of robot-centric CAD designs this summer, starting with the very simplest parts? I'm doing this for my own learning, and it would be fun to have company.
Here are the first few robotics CAD projects I have in mind: 1) flat bearing plate with 1 1/8" hole 2) L bracket suitable for connecting 2x1 tube to the kitbot chassis 3) gussets for connecting pieces of tube together
Link to 3D view of those kinds of parts: https://a360.co/2KE7d8X Subsequent projects would use those parts in simple assemblies. And keep going from there. If you've never even set up Fusion 360 or if you tried it briefly a while back and don't remember much, here's a training series specifically for FRC people that starts at the very beginning (loading the software): https://academy.autodesk.com/course/133777/fusion-360-first-robotics-competitions. I'm also glad to meet up at Starbucks or T&C or wherever to work on this stuff. The more we all develop our skills, the more we can do next season. Thanks & enjoy summer!
If anyone has experienced issues with F360, apparently an update made mid-to-late June caused joints to stop working correctly. They corrected the issue around July 4th, so all clear to continue your summer projects in F360.
I finished the Artemis proof of concept for an arm robot (I may add a gas spring and remove one of the NEOs at some point, but I'm not doing that now). Goal of the project Make a design for an arm that can be used from year to year and is capable of supporting significant increases and decreases in the gear ratio. Therefor the design must balance weight, strength, rigidity, customizability, and ease of manufacture. -How this was achieved --NEO motors are 1/3 the weight of CIMs --All tubes and channel are made using stock VersaFrame material --Vertical posts use large 2" gussets at the bottom --Arm uses two VersaFrame Channels, not tubes. (VF channels also have many holes drilled into them) --None of the sprockets or belt pulleys are either the maximum or the minimum diametre. --VersaPlanetary gearbox has an integrated encoder available (not shown) --Main arm pivot uses .75mm diametre tube as the axle. --A belt is used to power the arm from the transfer box to the arm pivot. Reason being that a belt is lighter. --The gearbox driving the arm can support between 1 and 4 motors. Things of note -the X-bracing on the arm and arm superstructure is lightweight. However, Simpler bracing designs can be used. -ArmaBot 90 gearbox used at the wrist in stead of a VersaPlanetary 90 because it is smaller and can fit between the channels. -Primary Drivetrain Gearboxes I retrofitted the WCP DS Gearbox and made it into a single speed with motors that stick out over the wheelwell.
Click link to view CAD: https://a360.co/2Wx34sW
*Thread Reply:* What do the rules say about manufacturing stuff during the preseason? I thought all custom stuff needed to be done during build season
*Thread Reply:* I'll check the rules when I'm finished writing the post. However, I am not aware of any rule that forbids making parts during preseason.
*Thread Reply:* I cannot find anything that says all parts on the robot must be manufactured during build season.
*Thread Reply:* However, you are allowed to use components taken directly off of an older robot
*Thread Reply:* See R15 from this year. I believe the components taken off of older robot are primarily COTS devices like motors, pneumatics, radios, etc.
*Thread Reply:* Oh, I see. Well, then, I am going to have to redesign the bracing.
*Thread Reply:* I suggest getting the design right to begin with and then we can look at fabrication strategies.
*Thread Reply:* What I mean by redesigning the bracing, I mean coming up with a design that doesn't require a CNC.
*Thread Reply:* We do have access to a CNC. And for designs that are complete early, we can also look outside the team for fabrication. So there are always options.
*Thread Reply:* Okay, then I'll hold off on a redesign
*Thread Reply:* It might be helpful to look at photographs of the Eiffel Tower to see how riveted (or bolted) structures go together. Mostly you will see that there are few if any tubes used as one can't get to the back side to set the rivets (or tighten the nuts). Also go to a baseball game and don't watch the Mariners lose, but look at how the structure is put together with lots of high strength bolts connecting angles, channels and H-beams, along with some welds. I don't recall many I-beams, which are like H-beams but with narrower flanges.
*Thread Reply:* Can't you rivet through the whole thing?
*Thread Reply:* No you can't rivet through the whole thing because the tube will squeeze and there will not be any grip to add friction force to the joint. The joint will be loose as if connected by pins in clearance holes.
*Thread Reply:* I've seen teams that rivet something on through the whole part. And the demonstration image at the top right of this page shows it being done: https://www.vexrobotics.com/vexpro/versaframe
*Thread Reply:* regarding rivets and the versaframe page. The big picture on top shows gussets riveted to the surface of a tube. The rivets themselves do not go all the way through the tube. When you look at it, it will look as if the rivets go all the way through, since both sides look identical, but if you looked down the tube, you'd see that they don't go all the way through.
Wikipedia has a detailed article on rivets. The section on "Joint Analysis" is interesting in why you shouldn't mix screws and rivets in a joint.
*Thread Reply:* I just watched a video on how exactly a rivet works (not entirely sure why it took me so long). I don't quite see the problem with using rivets with tubes especially considering how often they are used in FRC with tubes.
*Thread Reply:* Rivets are surface fasteners. They work fine with tubes. You rivet gussets to any surface of the tube. You just can't have rivets that go all the way through the tube.
*Thread Reply:* I understand, what I mean is I don't understand Paul's reservations on using tubes.
*Thread Reply:* Because you were talking about rivets that go THROUGH the tube. That doesn't work. Riveting to a surface of the tube will work fine because it is two solid surfaces .. and not a hollow object.
*Thread Reply:* And regarding Paul's comments about rivets and tubes, we use pop rivets which are a 'blind' fastener -- meaning you can fasten them with access from only one side of the work. Most other types of rivets require access from BOTH sides so they don't work well with tubes because the back side is not easily accessible.
*Thread Reply:* Ohhhh, now I understand, got it. Thanks!
There’s an annual CADathon getting ready to start up. Essentially, the organizers design an FRC game & then designers have a period of time to complete and submit an entry. Aside from the competition, the games (there are links to past ones in the thread) are an opportunity to think about strategy, practice 2D geometry analysis, etc. https://www.chiefdelphi.com/t/7th-bi-annual-f4-cadathon/360628/8
*Thread Reply:* In case anyone's interested, here's the game manual for the CADathon game Finite Charge 😉 https://docs.google.com/document/d/1g8P1UC77BryxtbUC6nRTKv08BYhNE1Bf5nysUrM13kU/edit
*Thread Reply:* The 65 CADathon entries will be shared on a FUN episode at 5:30 today. See the latest post in this thread: https://www.chiefdelphi.com/t/7th-bi-annual-f4-cadathon/360628/31
@Kenneth Wiersema It must be about time for you to head out. Thanks for all you did as student CAD leader and captain over the years, and good luck this year. Question for you: As far as importing Paul's Rhino CAD work, is it just a matter of a Step file import (something team members could learn from a video), or is there more to it? If you already explained this to someone, great! Just wanted to check on it with summer drawing to a close. Thanks.
@Kenneth Wiersema -- please correct or elaborate --
My experience is that pulling in a .stp file is easy. the issue is that the model is 'flat' .. that is lots of bodies but no structure, no joints, no user-friendly names, models of COTS components (motors, pneumatics, gears, etc) that were imported by Paul are mixed in with custom parts, etc. It's up to the person importing then to add all of this information and structure and then join the pieces together into rigid components where appropriate. And if the model gets changed by Paul, then the process needs to be repeated because there's no easy way to update the imported model with just the changes.. so it all has to be done again.
@chrisrin, @Mark Tarlton is basically right with the process, but I through together a bit more detailed explanation in this document https://docs.google.com/document/d/1TXoH2HHLaHaIiSdBtG_AVMALUCXq9HBg5kvadXNL4WY/edit?usp=sharing Let me know if the wording is a little confusing or unclear but I have it more or less setup as a step by step process to be followed through with. If you want a good example, look at what I did for the CHAOS Cargo intake, as that's the closest to what I consider to be the cleanest file conversion that was done.
Thanks for the document .. I guess I need to learn enough inventor to help with imports.
You don't really have to learn it yourself, just make certain someone is around with a copy and knows how to import and export a file-also if anyone's still interested in trying to get synthesis to work they're going to need a copy. (My opinion-Too much work and trouble for a pretty mediocre result)
*Thread Reply:* ok, I've got a copy installed on my desktop. I can at least experiment with import/export. thanks for the tips!
Thanks Kenneth. From searching around a little, it seems like the autoCAD DWG file format may also be worth looking into as a conversion conduit... apparently more information makes it over. https://forums.autodesk.com/t5/fusion-360-design-validate/exporting-from-rhinoceros-to-fusion-360/td-p/5972714
As far as Synthesis, they’re expanding Java support, so it could be more useful to programming than the past, but I tend to agree generally about not trying to shoehorn our robot in there... many other critical things ahead of that priority wise
The 259to1 Paradox gear was created in Rhino and converted to AutoCAD 2007 format. See how Fusion 360 handles an AutoCAD input. I hope Autodesk has made their products capable of talking to each other.
DWG files from AutoCAD and other applications can contain data that Fusion 360 cannot read. Fusion 360 can only import DWG files containing 2D sketch geometry or 3D solid bodies. For AutoCAD drawings - open the file in AutoCAD and purge the file, making sure to remove XREFs and images.
*Thread Reply:* @paul_vibrans, did you use the necessary Rhino DWG export options described in the forum post shared earlier? If we can get the DWG thing to work, it will be interesting to compare with STEP.
*Thread Reply:* fwiw: I was able to upload the file to Fusion 360. I then tried to import the file into an empty project and 10 minutes later, I'm still waiting. Doesn't bode well, methinks.
*Thread Reply:* I used the export feature in Rhino. There was no text. The parts were mostly polysurfaces with a few stray curves that did not get hidden.
*Thread Reply:* I will send a .stp version. I appears that Fusion has trouble with polysurfaces. There must be some difference between a solid and a polysurface that Rhino handles seamlessly and AutoCAD, and therefore Fusion, does not.
*Thread Reply:* I was successful (finally) in getting it to both import and then open the file. Import was slow. It crashed Fusion the first time I tried to open it.
At first glance, it came through as sketches but there were no "body" or "components" . So, I'm seeing outlines using lines and curves but no surfaces or solids.
The structure and part labels came through.
*Thread Reply:* The Rhino DWG export recommendation in the forum post I shared was to assign surfaces to solids for the purpose of import to F360. I’m not sure if that will help, but could be worth a try.
Update: I missed Paul’s comment that the export he sees in Rhino did not include the options described in the forum post. Maybe it’s part of a later version of Rhino than Paul’s, and in that case I agree Step may be the best choice then. Thanks.
After multiple tries, I can get Rhino to make drawings that AutoCAD will read but with some parts skipped. The first pass that I sent had a warning, but no skipped parts. I can't get back to the original settings that I had in Rhino to duplicate that. It looks like .stp is the only practical way to go to get to Fusion.
This is the .stp version of the Rhino 259to1 Paradox reducer.
That came through with all the solids and surfaces but no part labels. everything just shows up as a generic "Body" with no name. With a little work, it can be used, though
pro tip: rather than upload big CAD files to slack (where we are space-constrained), it might be advisable to upload to google docs, then place a link to the file in slack. If this is gibberish so be it 🙂. Happy to provide pointers as needed.
*Thread Reply:* I am challenged when it comes to Google Docs. If it requires a Google account of some sort, I am downright resistive.
*Thread Reply:* "resistance is futile" = we are all part of the borg
*Thread Reply:* The real solution is to send the information as individual CAD files for each part directly by e-mail to the person who will use them. Fusion can handle individual parts better than assemblies. Each part file comes with a part name. Any part that is bigger than will fit on a 3 1/4" floppy disk is too complex for us to build anyway.
*Thread Reply:* Some email programs limit attachment size, but if it works, that's perfect. Ultimately Fusion360 has cloud storage so once we get parts to those who will use them, individuals can share the files within the CAD team through the Fusion cloud.
*Thread Reply:* I remember when the "cloud" was called GE Computer Services at the end of a 120 baud telephone line with an acoustic coupler. I hated it when the line went dead half way through entering a Basic program. Cards and a machine I could see were much more reliable.
*Thread Reply:* well I remember when cloud referred to fuzzy white things in the sky 😉
Here's is THE best explanation of bodies vs. components that I have found: https://youtu.be/46UNmpQdbVc
The whole series is pretty clear and instructive, though some lessons are more in the direction of hobbiest 3D printing than FRC. But the more you use the software, however you use it, the better you will get. I'm about halfway through this series of 30 lessons.
A key gleaning is that it is OK to just use bodies for simple objects, but for more complex things with assemblies and sub-assemblies of parts, it is recommended to create components before sketching and creating the body or bodies that will compose the component. And always name a component right when it is created to keep organized. Finally, before modeling at all, it's good to conceive as much of a plan (sequence of intended steps) as you can muster with what you know at the time.
@channel This is a really great video, and if you have CAD, please watch this and his other content.
In the video he talks about the two main ways of building something in CAD: Top-down (When you design an entire assembly in one file) and bottom-up (When you design each part of an assembly in their own files and put them together in an assembly). Fusion is focused more on the former. While top-down is great, there is a drawback that he didn't mention. Because of the way Fusion utilizes the CPU (Command Processing Unit, a.k.a the processor), you will start to notice the programme is really slowing down even though your CPU usage isn't anywhere near 100%. The reason why isn't important (but if you want to buy a CPU for running Fusion, get the one with the highest base clock frequency, not the most cores), but the end effect is when you have a large number of parts in an assembly, your performance will tank. The way I get around this is through something I call middle-out. With middle out, I have a single master assembly and several sub-assemblies. This significantly improves performance and decreases design time intervals. Below is a link to photos of how I organized my elevator robot: https://drive.google.com/drive/folders/17cIvkExIGkakike2wZABNsOFSYRCGZiy?usp=sharing
From that same CAD learning video series, these are also great in my opinion. Here is a set of 7 great Fusion 360 videos totaling less than 2 hours that will get you on your way (but you can’t just watch them - you need to run F360 & do what the presenter does)... Tutorial for Absolute Beginners (34:31): https://youtu.be/qvrHuaHhqHI Includes new UI New UI vs. Old UI (14:26): https://youtu.be/tuZzHcmFtg4 For those of you used to the old UI, this will probably be helpful Sketch Constraints (12:52): https://youtu.be/BGwBZJ14KHQ In just under 13 minutes, all of the major sketch constraints are clearly and concisely covered. I can't imagine a better coverage of the topic in that short a time. How & Why to Fully Constrain Sketches (15:01): https://youtu.be/C11L136U0vQ A simple part sketch is used to cover the topic of the title as well as elegantly introduce the topic of design intent. And the presenter also provides some useful tips at the end for designing to reduce impact from changed dimensions. Components vs. Bodies (10:46): https://youtu.be/46UNmpQdbVc I shared this one earlier, but including again so this is a complete set. Assemblies & Joints (14:48): https://youtu.be/t41QmQszcbE This is something we use a lot when designing mechanisms. How to do 2D Part Drawings (16:08): https://youtu.be/L0IwfH-9Fss This is definitely a skill our team needs to develop, and as you'll see, it is not that bad. The trickiest part is knowing which features of a part to be manufactured require precision & how much precision. I’m thinking Paul and Chris Stanley can help teach that.
The other videos in his Learn Fusion 360 in 30 days are really good too, and by walking through the creation of a bunch of different models, there are many things covered that will be good tools in your toolbox. Here' s a link to the whole series: https://www.youtube.com/playlist?list=PLrZ2zKOtC_-DR2ZkMaK3YthYLErPxCnT-
@paul_vibrans Is this telling me that the uncoloured areas are unnecessary? https://drive.google.com/file/d/18GW9w9eZ5VhA5PIob5TnB-h5o0FRJ8oH/view?usp=sharing
No, I think it is telling you that the material is too thin for the applied
*Thread Reply:* What do you mean by that?
*Thread Reply:* Sorry, I clipped the sentence off. The word is "load". The metal will buckle before it yields when the thickness is small compared to its span. It all depends on the ratio of thickness to span and the modulus of elasticity of the material. I am not sure exactly how you have the analysis program set up. Does the warning go away if you use a smaller load?
*Thread Reply:* I changed the load from 2 40N to 2 10N loads, then set the Mass Ratio to 50% and the size and location of the blank/dark blue area didn't change. However the intensity of the colours decreased. You can find the file at SpartronicsCAD/zMetal Testing/Shape Optimization.
*Thread Reply:* I need to know more about the program to understand the blank area. The blank area appears to be where there is only tensile stress.
This is what it looks like with the mass ratio at 100%:
I keep forgetting to upload this: https://a360.co/30AXzYP
This is the elevator lift concept I worked on over the summer. Athena uses channel rather than rectangle tubing to minimize the weight and overall footprint of the system. There are a couple of edits that I will make when I have the time: 1) The yellow side rails on the chassis will be changed to .1" thick 2) The yellow square tube supports for the elevator will be changed to either thin aluminum or thin steel round tube and they will attach at the top of the elevator's first stage. 3) I will add encoders to the lifting versa planetary gearboxes 4) I will add in redundant distance-reading potentiometres to each stage
*Thread Reply:* I've been looking at your design.. Very nice! You obviously put a lot of thought and effort into your work. I'm looking forward to studying it more closely and also having you take us through it sometime.
I went ahead and saved the copy of this design under the "Design Class for 2019" folder. We may want to review it in the upcoming design class. This location should make it easier to find. If you do any more work on your design, you may want to update that folder with your latest work.
I also took the liberty of renaming your folder to "Harper Nalley" -- I hope that's ok with you.
I am also working on two additional projects: The first is called Hyperion and is a scissor lift concept that uses motors to lift. https://a360.co/2p9k7Tc
The second, Aegaeon (named for one of the Greek hundred-handed-ones), is an intake/manipulator in the style of the one we used on Themis at districts. https://a360.co/2BLgSUw
@Harrison Paul Cate (Electronics/Pneumatics) has joined the channel
@David Scheiderman has joined the channel
@Mark Tarlton @paulvibrans @coachchee @Jack Scheiderman @CruzStrom Here's a page with links to the three Markforged printers: https://markforged.com/products/composite/#desktop
Here are the prices with the sponsorship deal. • We have the Onyx One, which costs $3500 (no sponsorship discount on that one). • The next step up is the Onyx Pro, which adds single-strand fiberglass & will cost us $6000 (after $2000 discount). • Finally, the Mark Two will cost $7500 (after $6000 discount). The secondary materials supported by the Mark Two include fiberglass, kevlar, HSHT (high strength high temp) fiberglass, and carbon fiber. A spool (or possibly two) of each material is included. Here's a link to a datasheet on properties of the materials: http://static.markforged.com/downloads/composites-data-sheet.pdf
The other part of the sponsorship deal is 50% off on up to 10 spools of material.
Which should we get? Either the Onyx Pro or the Mark Two. It's pretty tempting to try the Mark Two since the sponsorship discounting makes it so the prices are not that different, and we might find some uses for the ultra-strong carbon fiber. But Paul & Mark, if you think fiberglass is enough, then we can save the $1500. I think the biggest benefit from the second printer will actually be double the throughput printing basic Onyx filament. Either of the better models will print Onyx on its own.
We paid $5000 for our Onyx machine so I don't have a problem with spending $7500 for a Mark Two I if we have the money.
Those prices are pre-tax of course 🙂. I agree we should go for the Mark Two. Reading the materials datasheet, I could see there being distinct applications of fiberglass (least expensive option that is stronger than Onyx), kevlar (known for high durability), and carbon (top of the line strength and stiffness).