At the Feburary 2019 meeting we are planning to build some club robots. This thread will be about that project. Please feel free to give feedback or ideas about the project.
So far we have:
This is from Stuart’s Google plus post.
Josh, Jason, myself, and others have been talking about making inexpensive robot kits for the club to help people get started building robots. I’ve made a rough but functional prototype based on some ideas being discussed. The linked video shows it remotely controlled over WiFi on carpet with depleted batteries.
video: https://youtu.be/wB69J8r-doU
code: https://github.com/sta97/Robomo-Kit-2018
I made a spreadsheet with the materials costs of my robot. As of right now it doesn’t account for time, 3D printed parts, wood, and the screws. I assume that the extra costs of those parts would parts would add about $5 and the time would cost about $10 if we cared to account for it.
https://docs.google.com/spreadsheets/d/1XHghHxV6hZgd_78cU6qECwgayxG-LDFlCrABOsWYgeY/edit?usp=sharing
I think that might work, though I’m a bit wary, as it doesn’t look very sturdy. I’m also concerned that it might restrict the chassis to only be manufactured by certain tools like a laser cutter unless we simplified it.
The main thing you want is for a laser cutter to be capable of making the custom parts, right? Does it have to be 100% of custom parts?
If that is what you are thinking you would like, maybe you should just buy the completed kits. I don’t see the benefit of trying to duplicate what is so broadly available and cheap if that is the real goal. That said, I do not think that is in the best interest of the club.
Maybe we should step back a moment and clearly define the goal of the club creating a basic robot. If it is only to reach the cheapest basic bot, not to be confused with inexpensive, then stop wasting time and buy some of the junk kits and be over it (save the dime and don’t burden me with one of those). If it is to come up with a cost effective platform that is durable enough to experiment with multiple club projects, then let’s allow for the process to educate the club and enjoy making something that has at least some level of originality. What will it be?
I know you’re not asking me, but I’d like to give my opinion.
I personally would like the club to have its own original design, even if it’s not perfect. Replicating an existing design is not something I want to do, as I personally enjoy the process of developing a robot from scratch for a particular purpose.
Durable yet inexpensive robots would be great, but from my perspective, I’m unsure how we’re going to hit the extremely low budget without using parts that have a few issues. If you could help by giving us some examples of inexpensive yet good quality parts, that would be great.
I am starting to seriously consider kits. Unless someone has something concrete in mind for an original. I don’t have the ability to create one. I’m not sure why it would be bad for the club. We have serious time and budget constraints. This is what I’ve been looking at.
I do have something in mind. I am planning to finish fully documenting my design after I take in to account the feedback I got at the meeting. I’ll start posting what I have once I’ve made some progress on that.
Unfortunately I can’t put any time in this week until I’ve finished all my assignments for this semester.
I would like to try a prototype with continuous rotation servos, but for the moment I’m going to focus on TT Motors because they are currently working in the design.
I also plan on trying a ball caster to see its effectiveness on my carpet.
Here is a list of my current planned changes:
It will be designed in mind such that if someone takes just a rectangle of wood, only drills holes, and does not put slots in it, they could still create a functional robot (perhaps minus encoders).
Because Fusion 360 is unusual for CAD in that it’s cloud based saving, I’ll post a viewable link to my design rather than the files themselves.
For full documentation the first step will be finishing the CAD assembly so every part is included. Second step will be documenting the electronics wiring with a visual breadboard layout tool. Third step will be creating a Google Docs document with assembly instructions with pictures.
Edit:
It will also be designed in mind such that you will not have to use wood, but could use any flat material.
Here’s the deal on the kits. I am going to buy them if we don’t create our own. I don’t have the ability to create a base myself, so someone will have to step up and make a design. It would also need to be done by the January meeting to give time to order, print or cut parts by February. I didn’t intend for this to be the ultimate robot design, just some thing fun to put together. That could be done in the time allotted in the meeting. Also, providing some experience with a micro controller.
So I’m fine with either option.
Let me know if you care.
0 voters
Maybe add a power switch to your spreadsheet?
Josh, have you looked over the spreadsheet enough to agree with the electrical components and their costs?
I have a design for a chassis I would consider posting, but it would probably put the total per robot at about $36 each, based on the costs listed in the spreadsheet and motors/wheels from Zagros.
The prices look right to me.
$36 would be in range of what we could buy. I just looked at Zagros’s site and they seem reasonable.
https://www.zagrosrobotics.com/shop/item.aspx?itemid=763
Same price as TT motors except the wheels are extra. If they give us the 10% discount it would be closer.
I was thinking about using an AA battery box with a power switch built-in. The spreadsheet simply reflects the costs on the robot I built plus some hypothetical robots. I’ll update the spreadsheet with AA battery boxes.
This is the sort of thing I’m thinking about trying: http://a.co/d/cBYavN7
I use those 4AA with switch for some of my kits. They work alright for robot power until you need more juice. Then you have to get a switch anyway.
I have a pile of big slide switches from my first generation Gobbit kits if the club wants some. They would need leads soldered.
Here is a rough draft of a 6" x 8" base I was considering within the $36 I mentioned before.
You can see a 3D model here: https://a360.co/2Qn7eBn
The cheap ball casters on all of the super cheap kits are not actually made to work in the “ball down” orientation. They tend to drag most of the time, or start to drag after some debris is picked up from the floor.
Also, with rough surface, such as concrete, they will get scarred and no longer roll. The novel and over built solution I came up with is a BIG ball caster using a golf ball. I have not made a prototype yet, but tests are promising that it would be big enough and very durable for bumpy and abrasive surfaces indoor and out (outside I would probably upgrade to RC style tires like the ones I have on my 4 wheel bases).
The model of the caster I made shows a folding PVC configuration that would be CNC milled from one piece of 6mm expanded pvc and folded around the ball and nylon spacers/rollers. It could also be easily converted to a Laser cut or hand cut version form wood or acrylic. Even better, it could be remodeled and 3D printed (CNC milling as shown would be significantly faster to fabricate but would require a few more screws in assembly than 3D print).
I am pretty busy this month. If the club has interest in this or a similar version, please vote in the poll and comment on revisions that may be desirable.
Wow, that’s really cool and it’s better than what I was planning in terms of what most people in the club want.
I think the golf ball will work out well.
I’ll keep working on my design just for the sake of finishing it.
If you don’t mind, I’d like to incorporate the golf ball caster idea into my robot design.
EDIT:
As is, that design looks like it could be 3D printed.
Sure you can use it. There will be some small tweaks and adjustments if this bot is chosen in moving forward, but it will most likely be similar in size.
I don’t think I would try printing it as is. A single part with different roller slots and maybe a little smaller footprint would be a few of the changes for a printed version. Unless someone with a printer wanted to take on the task, I would stick with my initial general plan if I am to cut the parts.
I really like the golf ball idea. My only question about that is whether the dimples of the golf ball would “catch” on the “jaws” holding the ball, causing the robot to jerk. Of course, if that were the case, there are other equally inexpensive types of balls that could be used.
