Thread created automatically to discuss a document in CD-Media.
Team 33 Arm Design Paper by: IKE
This paper discusses the concept development and engineering that went into Team 33 vertical hanger concept.
This paper discusses the concept development and engineering that went into Team 33 vertical hanger concept. The paper takes you through the initial concept evaluation phase up to the robot being ready to ship. The paper was used as a supplement when the students discussed our design with the judges.
There were some minor improvements made after ship, but that may be included in a continuous improvement paper.
This is one of the design papers I had promised to share after the Championship. Our hanger was one component of our design that helped us win the GM Industrial Design Award at the Championship this year.
We ended up using a CIM into a DeWalt transmission into a Modified Andy-Mark transmission with a VEXPro large sprocket reduction (we like to support all the FIRST sponsors). There is a picture of it in CD somewhere.
As far as the physics goes, our team has 3 engineering mentors on the team. We work with the kids to brainstorm the plus/minus of the different concepts and then we (the engineers). Most of the older High Schoolers can understand and assist with figuring out the torque and gear ratios.
Motor power * efficiency * time = change in energy (height change of CG for this one).
Transmission ratios should be figured out** speed based and torque based**. These are simple calcs and we do them with the kids. I really like making the kids do the math of those sorts of things.
(5 second hang of 120 degrees from a motor that loaded should spin at 4000 rpm. 120/5 sec = 4 rpm therefore we need a ratio around 1000:1). Some rough numbers: Dewalt is around 40:1, the toughbox gets another 5:1 and the chain drive is roughly 5:1. 5540 = 1000:1 This arm also produces about 250 ft*lbs of torque at its shoulder!
Those level of calculation we work with the kids and often by the end of the year, they can do them themselves. The conference room we borrow has a giant whiteboard wall. sometimes the engineers write, sometimes the kids write. We all do the math though. We feel that this is why FRC does not have a mentor involvement limitation rule like FTC, FLL, OCCRAā¦ It is to engage the students in these designs.
Beam bending fatigue, and the āgriptionā model (we like to make up words) are more advanced. You need basic āStaticsā from an Mech Eng. school or a good understanding of free-body diagrams (FBDs) in order to do the gription model. We like to do these calculations in order to make sure that things should work, and to figure out important parameters. This FBD was made by another engineer, and then he and I discussed the details. This helped us to understand that a longer arm added leverage to the traction patch and thus kept us from sliding down the pole. It also helped us understand the loads applied to the pole to ensure that we did not damage the surface of it (we had a 6" contact patch).
I did the beam bending calcs in order to make sure our arm would be robost enough for about 100 matches. These calculations are taught the second or third year of most Mech. Engr. programs. In 2007 we had a large arm bot that had a casastrophic arm and tower fatigue failure at the very last demo we did that year. I consider this one of my greatest successes. As Collin Chapman would say, if the racecar doesnāt fall apart after crossing the finish line, you made it too strong. Since FiM has us playing more matches, we have been leaning a little further towards inifinite life designs (Jr./Sr. year ME course work).
We still go over the harder concepts with the kids, but usually do the analysis off-line from our main meetings. This is pretty advanced stuff, and the struggle can be boring and/or misleading for most of the students. We try desperately to not be Tinkers, but to do engineering with the kids.
We also thought about this kind of hanging but we werenāt sure if the arm wonāt just fall (because of too low friction). Guess we didnāt have the knowledge to do these calculations :o
Iām sure that next year weāll use the same designing methods you used!
The motor reduction calculations and free body diagrams are things that I use all of the time when designing things (especially FIRST robots!), but it does seem like there are way to many teams that either donāt have knowledgeable mentors or students on the subject or just donāt bother doing the calculations.
Theyāre not as hard as you would think!
The stress calculations on the other hand are a little more advanced, and I doubt that very many teams actually carry them out. I have done them for a few key components that we know will act a certain way (structural arm members and axles are good examples).
Lifetime calculations are one of those things Iāve never bothered to do, since there seem to be too many factors coming into play on a competition field. I generally would just make sure that key parts have a safety factor of around 3 or so, since thatās about where the endurance limit is for most of the materials we use (as I recall, at least. I havenāt checked up on this stuff in a while).
Is there any chance you could post the excel file thatās shown in your powerpoint? Iād love to see which methods you used for all of your calculations, as they obviously have been working well for you.
Hmmmm. I will have to think about that. It is a custom tool I made and is not really āreadyā for others to poke and prod on. Plus the hardest part is the estimating of loads you will see at the joints. Most people do a good job at predicting weights, but forget about force created from accelerating masses. I wouldnāt want to be associated with a failed mechanism that some used the āIke Beam Calculatorā for.
I will recomend an excellent book.
āMachine Design: An Integrated Approachā by Robert Norton. This is a great book that goes into component design and fatigue analysis. It has a limited section on gears, but should be good enough for most FIRSTers. It is the book we used in Machine Design II at Purdue. This is a Sr. level collegiate course and requires basic understanding of Statics and Dynamics. Since it covers stress and strain, it does go to a basic enough level that a bright student could likely teach themselves. The part I like best is that it gives great examples. There are several examples for beams, cranks, poles, springs, gear-trains, cams, pin joints, ā¦
I added in a 97-2003 ppt and updated the file with a pitcure of the āfinalā product. The signs on the bottom of the bot were great for exciting the field crew and refs. I am glad some other teams started doing this as well.
I have seen some links talking about, and have received a couple of PMs regarding the 2010 arm.
A couple of words of caution. This arm and robot could rotate on the pole. This only occurred 1 time in which it resulted in a fall, but that was also a vertical pole.:ahh: ::ouch::
Happy climbing.
Thank you for posting these notes. I have been working on a similar design, and seeing your design paradigm makes me think Iām going on the right track. One quick question, however. What diameter and material did you use for your final axel of rotation. (The one with the very large sprocket on it). Testing my design yesterday, I ran into a material failure that has me worried. Lifting 180 pounds, 9 inches from the point of rotation, my 1/2" 4140 steel shaft with a 1/8" keyway failed. The shaft ultimately twisted about 120 degrees from where it started (straight!!!). Any insight would be greatly appreciated!
P.S. Amazing design and engineering on that arm! I currently have arm envy!
We also found out the hard way that your armed gripped a bit better than ours did when you āejectedā us from the pole during a match - I believe it was State quarterfinals? Jimās reaction wasā¦ memorableā¦ A bump would send us spinning, but you managed to throw us completely off the pole.
Not to take away from Killer Beeās bot at all, but here is just another example of that same method. There was discussion on this page about how it worked:
Other robots worth taking a look at are Simbots (2010) and 254 (2010). Both of these teams had systems that hung from the virtical pole but with some key differences from us. (more then one way to skin a cat)
