Video Transcript – Team Invert Senior Capstone Project
[Music]
Team INVERT Showcase
Chris Barba, Trevor Rush, Nathan Sheehan, Amanda White
I’d like to kick us off. All right guys let’s get things going. I just want to say hello everybody. Thank you for joining us this afternoon. We appreciate you taking your time to watch this. We’re team invert. We’re going to be going over our senior design capstone project with you. My name is Trevor Rush. Hear are my teammates with me. Chris Barba, Nathan Sheehan, and Amanda White. We ask that you save any questions you have until the end where we’re going to have a 10-minute Q A.
So getting into it, House of Design is a robotics integration company located nearby in Nampa Idaho. They primarily use ABB robots in manufacturing lines. These robots can range in size, weight, and functionality. House of Design solicited us to build a robot invert device that is capable of mounting the robots both on the wall and or the ceiling to more efficiently perform their daily tasks.
So moving on to some of the goals of the project. Our number one priority throughout the entirety of this thing has been safety. We’ve kept that in mind with every decision we’ve made throughout this whole process. Currently there’s not a safe and efficient way to invert these robots. Companies like House of Design have accomplished this task in the past so that’s required many people, tools, and putting people in high risk scenarios.
We came up with an innovative design that allows for this task to be accomplished with only three people and in a very safe manner. This device needs to be able to elevate the robot and rotate it incrementally that stops at 90 degrees for wall mounting and 180 degrees for ceiling mounting. This device also needed to be marketable so that House of Design can both utilize it and sell it to other manufacturing companies.
So all of this was to be accomplished within a prototyping budget of two thousand dollars while minimizing the final product up. It’s been a long process to get to the position that we are in today. We started looking at similar devices that are already on the market such as the hydraulic drum dumper and the turn and load shown here on the right. There are a lot of things that do similar you know ideas but they don’t really fit the criteria that we need to make this a successful project. So we’re trying to accomplish something that is yet to be done.
So going into some of our original ideas of design on the left here is our very first attempt in creating a robot inversion device. The main idea of this one was to have protection for the robot you know worst case scenario of it being dropped due to failure. We ended up moving away from this design due to the excessive weight, cost, and interference caused by the large size.
Main component of this design and the next iteration we had was the universal mounting plate shown on the right. This was originally thought to be connected to the very bottom of the robot to hold it in the cage and be removed after the mounting process was completed but again we found issues due to the interference it caused and the plate could only be suitable for two or three robots which does not allow for the compatibility of many robots that we were trying to achieve.
Some other iterations that we had are shown here. These were designed and then analyzed. Again we did limit the amount of steel we used which made the weight a lot more manageable for the forklift to pick up. We played around with ideas such as moving the center of mass chain drive systems, hydraulics, and gears. Each of these designs that we came up with were tested and found to have mechanical flaws. More safety concerns that led us to continue working and helped us solidify the design of our final product which Chris will go over next.
So each of the ABB robots comes in a upright and locked position. This is about their general parameter that you see here. Just sort of limited our ability to do certain things with them. One of the robots that was in our initial prescription was the ABB 6700 which weighed approximately 3500 pounds. Unfortunately due to the height difference in the center of mass and the overall height of this robot and the weight it pushed that center of mass far enough from the mass of the forklift which pros a potential to flip the robot or flip the forklift. This width of that robot has also exceeded the width of the forks into internal diameter or a distance there. So we come down to the next manageable one which was the 4600 which is close to a thousand pounds. It’s width fit nicely between support allowing us to use that as sort of our base point and the height and mass of this made the distance between the center mass of the robot and the mass of the force that’s much more manageable. And we no longer face the issue of having to flip the robot or excuse me flip these forklifts with the one that they had on hand. Next slide please.
So this is our full design here on the on the left. As you can see we try to keep the center of mass about three or four inches below the center axis drive, thus limiting the amount of torque that we would need on for the motors, lowering the cost of that. We’ve gone with some initial different types of motors that’s Trevor covered – hydraulics and things like that. This ended up being a electronic versus a worm drive servo motor it’s non-back drivable. This allowed us to rotate it slowly into place without fear of it backlashing and coming down in case of loss of power. Some of the safety features that we incorporated this is obviously HSN steel components as much as we could to make it stable and then there’s forklift locks at the rear of the fork extension which is an OSHA standard to prevent the load from sliding off. There is also a robot mounting plate position. Each robot has a different base diameter and bolting pattern so unfortunately each robot would need to have its own mounting plate design in order to locate the center of gravity of the robot and to accommodate the bolt pattern so they would be used in this. But they are easily attached we have four bolts on either side to the main assembly. There’s also a stabilizing rod in the middle in case it contains some sort of lateral load such as jarring from the force being driven or any such thing. So that sort of really helps stiffen up the load there.
Next slide please.
We also try to maintain seeing how this robot fit within the forks of the forklift we try to maintain it to the entirely within the structure while it was rotating. The only time it is not is in the center picture where it is just starting to rotate. But even then it is just barely above the ground so we figure they can raise it up just a few inches to get past the rotation and after that it stays well within the bounds of the equipment, thus limiting the need to raise it high or having a height load in order to rotate this around. They can position it almost completely and then just go ahead and install it without having to position it completely and then raise the load to a tight [inaudible]. Amanda?
I’m gonna take over here actually and um just touching on photos right here obviously this is looks kind of scary and obviously it is when you’re lifting thousands of pounds of very expensive machinery so next we’re going to go on the testing extensive testing and iteration. We had to put into this to get to the point we are currently. So as you can see we went into FEA or a finite element analysis. We had to pick a point that we thought was going to be the critical point of when the most failure could happen. As you can see we have that drop down sitting at 90 degrees because that’s going to create the highest torque. It’s going to create the most torsion in those beams. It’s going to be the most likely point where we’re going to have critical failure. As you can see on the … oh starting off first, the beam down the middle here with the forces acting on it and the plate is attached to is representing the robot attached to the mounting plate. We made an assumption that the robot attached to the plate is one solid rigid structure and there will be no failures throughout that point so we use that assumption to over design this large metal plate you
see connecting the two sides of the drop down and the two going up. The force at the end of the tube is representing the center of mass of the robot of where the force would be acting. The second force a little farther in is going to be representing the gravity. And we have also components you can slightly see the sideways perpendicular arrow to that center beam that’s going to add a component of gyration due to forklift driving and other components. For FEA we noticed that we had two critical sections in this point. First one to talk about is section A of where this two by six two beam meets the foot mounting plate and there could be two possible failures as you can see the discoloration in the factor of safety gets very low down as we get down here which could either lead to a possible failure in the structure of the tube or tearing or failure of the weld from the tube to the foot. And so those were two of the things that we would like to look into to prove that this device and tool is actually a viable answer.
On the right we have section B which is where those bolts are actually going to be attached to the robot mounting plate. As you can see we have some large red flare-ups on the outside of the bolt and on the inside. As i will refer to in the next slide we have our vertical tear out probability which is going to be parallel through the bolts and then you have the horizontal tear out which is actually going to be acting in compression. And now this looks like a bunch of numbers but basically this is the factor of safeties that we calculated for those two different critical points. According to OSHA standards a factor of safety of five is what you want to achieve. Starting over here in the top left with the critical point A which was the tube to the foot. We have our moments in the X and the moment in the Y created from the robot. As you can see moment Y creates a low effect of safety. It’s pushing close to that five but we still have good leeway room to work with and this these numbers are also backed up with our FEA. Going down below that one we have the weld failure that starts to drop a little bit below the tube failure and you can also see that moment Y is creating that problem which would be expected. Moving over to the right we were looking at different types of bolts. We had a grade five and grade eight as you can see that factor safety is ridiculously high and when it came down to it the cost difference didn’t make a big difference going with grade eight or grade five. It really did not make a big difference rather than a peace of mind and just the strength and something that you didn’t have to worry about in the back of your head because you knew there was no chance of that breaking. And then the lowest factor of safety we have is going to be the vertical tear out for the bolts. As you can see that was the thin piece of steel that was in between each bolt hole and could have a possibility of tearing when it’s at that 90 degrees location and drops down to 7.5 which is still above our five that we need for our OSHA standards. So all these are lining up with our FEA and our green sheet calculations.
And the next thing we need to prove is that we did this all right. And moving on to our testing fixtures, our first testing fixture is going to be as seen. This horizontal beam is going to represent one side of our drop down and the circled critical area is the part I talked about before where the tube is welded to the plate whether it’s the tube or the weld. We have another vertical beam where the force will be applied and this is going to be in a fashion where we will utilize the civil structure in the workshop and we will apply the force up we use the applied force and we’ll create the resultant portion that would be stimulated as if a robot was sitting in it. We’ll be able to push it break in and record the force at when the either tube or weld breaks and those should correlate to the numbers we found in the previous slide proving that this is a viable option.
Next to our second text fixture is going to be testing those the bolt locations as you can see circled in green. Once again on the right we have the arm that’s going to represent the drop down of our actual device. And then we have another beam identical on the other side that’s going to be creating the torsion along those plates. We have both of them being applied towards each other so we get that same torsion moment or the same tear out effects in the both vertical and horizontal components of the plate that you would see if this device was being used with the robot. And so both of these test fixtures should test the points that we found to be most critical and once we finish our testing that Amanda will cover on we will be able to tell if this is going to be a viable design to do what House of Design needs to do. And I think that the three points of satisfaction of making sure that we hit what we need to do will prove it viable. And now we can move on to Amanda.
Thank you. So with our testing to be conducted in the next coming days and as our project comes to an end it is important for us to step back and reflect on all the things that we’ve accomplished. And so the goal of the project was to create a mechanism that would be able to support, lift, and rotate robots of varying weights and sizes to be mounted onto walls, ceilings, and any angle in between in order to aid House of Design engineers in the installation of robots. And so throughout our time working on the project our team went through several iterations of design changes in order to create a final design which would support the lifting and rotation of ABB robots ranging from about a hundred to a thousand pounds. So our testing goals include two assemblies which are to test the tube weld bolt and plate strength of critical areas in our design.
Overall the project gave our team more experience and a better understanding of the design process and the commitment and responsibility it takes to work on long-term projects for a customer. And we are very grateful to have had the opportunity to work on this project that taught us so much and has had an impact on our customer and their potential customers. And so looking to the future of the project in the next coming days as I said we’ll be completing our fabrication testing and analysis and then we’ll be documenting this information and supplying that to our sponsors. And so as a semester and our time working on the project comes to an end we’ll be compiling all design iterations, finite element analysis, green sheet work, and data to hand over to House of Design and then with this information they can either move forward working on our design or it might be passed to another team to work on it in the future. And so if that is the case the next major milestone to hit would be to build a complete prototype, test that prototype, and then look for ways to improve the design or manufacturing of it.
As for overall ways to improve the design that we can see right now which would be most important to House of Design would be to spend the time to design a mechanism that would not only be able to support a wider range of robots but also to design a mechanism that supports robots from various companies which have different mounting and connection ports. And so with that we will move into our Q&A portion of the presentation. Free to unmute yourselves and ask questions otherwise I will be monitoring the chat.
Well I’m curious about how covid impacted your team in terms of being able to accomplish what you did.
Uh like this on the phone. That’s for sure a lot of hours doing things like this getting together and you know using the whole whiteboard thing is very helpful. We haven’t really been able to do that but we’ve been able to kind of overcome that and come together and like we said a lot of time is what it took and that’s what we’ve come up with.
I think the hardest point it comes down to is when you come to fabrication and all of that and that may have some point is why we’re still in the testing phase of where we actually are due to workshop availability. Just availability of reaching out to everyone getting parts it just it might have delayed our project slightly but in no way did it affect the quality that I think that we produced.
What advice do you have for us the next semester in terms of how to how to do this? What would you do the same? What would you do differently? You know, how would you go about this in this kind of environment for the next semester?
I can answer that. I think for the next semester I would suggest any of the big goals like senior design and everything related to that, it would be good to make sure that everyone is aware of like important deadlines ahead of time so that with everything going on that they can work to make sure things get done at a specific time. As for like going in and fabricating in the engineering innovation studio I think it’s kind of rough because we don’t know what’s going to happen. We just need to be able to adapt and be flexible and try to make things work.
Leaving room for anticipation knowing that you could have problems. Moving deadlines you would probably think could be at the end of the month maybe at the beginning of the month assuming there could be problems. Giving room to allow um different groups to need an extra week so they don’t feel behind. Even though it may feel like dates are coming up quicker it’ll give you a more um I think it’ll give you a better sense of completion and just give you more time to work on the project and kind of give you a little kick in the butt to get going.
What i’m hearing you say is that the penalty of the unexpected is higher in this kind of environment. So you’ve got to have extra effort on being on schedule because you got to allow for the fact that that schedule may change. Is that what I’m hearing?
And sometimes it comes down to the simplest things like I was fabricating the other day and uh they ran out of welding wire and it wasn’t due to [inaudible] fault. He had ordered it months ago, it’s just that Norco hasn’t made the delivery. So now I’m stick welding and not good at it so it takes longer so I mean it just who knows what’s going to come up due to shortage of being able to just run around like a year ago, you know.
Yeah okay, yeah.
So I have a I have a question about the motors, right? You got a worm drive motor on each side. I assume they’re just gonna be working in parallel. Have you thought about controlling them and and worried about how they might fight each other? How are there any way to coordinate their their action?
So I did a lot of looking into the motors and picking out what we needed. Different kinds of input or out input and output RPM, torques, horsepower, different kind of safety measures. There’s a lot that goes into these gearboxes and at the same time they cost a lot of money. We went down the path of trying to consider building our own gear relation but we realized between tolerances, our experiences, it wasn’t viable. And then when we started going down this path it took us a while to get that correct gear motor it’s a helical worm gearbox and I think once we got to the point of those controls we started to look into it, we started to see what we needed, we knew that we needed to have some sort of algorithm to set up certain timings so we don’t have any miscommunications between the motor. But as far as our expertise and abilities we didn’t get as far as we’d like with the control side which also could add a component to being a future project and stuff like that but we only got so far down that path.
Yeah there was even a point where Gus kind of mentioned that back in itself is almost a senior design project. Essentially our original design was with the hydraulic motors operated off of the forklift auxiliary port which would have been a very simple form of control and would have given all the control to the driver. Unfortunately our client didn’t want anything to do with hydraulics we had to move away from that and he more or less insisted on electric motors which wasn’t really our first intent.
Professor Gardner, what’s the typical way of handling that two motor problem? Just go to one motor?
Well yeah, I mean there isn’t a typical way. That is an unusual thing and I think one motor you know makes it easier. As I was thinking this out in my head I think the best thing to do would be to be monitoring the current that the two motors are drawing and if one of them is drawn more then you goose the command to the other one. And so you’re basically sharing the load. Yeah and without knowing the the torque profile in other words you know, you’ve got that axis of rotation kind of running close to the center of mass right is that I understand that?
That was our goal, yeah.
Yeah so you’re minimizing the load on those things so that’s the good news. The other thing you do is compliant couplings can you know help you know they would find its own equilibrium. But that’s problematic with torque that high.
The reason we strayed from a single motor is as you can see the um axle isn’t straight through axle it’s in two pieces and that avoids using chains, which chains and belts which brings in a whole another component of failures and dynamic failures that are very hard to contribute and figure out. So we ultimately decided to go with the two motors. We knew it would be challenges going forward if we needed to go forward with finding the way to coordinate them with each other. But it made more sense and reason to go with two rather than one.
Also the motors, just to mention are both non-back drivable so if the power does go out on those it’s going to hold it in place. It’s not going to come flying back down. And each of them have a manual handbrake on each of them on top of not being back drivable.
Belts and suspenders right?
Hey so what’s the margin of safety with respect to overturning the forklift? I can tell you that right now to give me one second. So it was well within the safety range. So we looked at how House of Design has two different four clip sizes for their … here it is, for their book that they have which is a Doosan model 45. Actually I’m starting the model 30. it’s only at 44 percent of its load capacity and for their larger forklift it’s only at 24 percent of their load capacity.
So we were looking at that and making sure the four those are you know one of the biggest things was that the forklift didn’t just flip over when you picked it up so we tried to keep that center of mass as close to the forklift as possible.
Is that 44 percent of the force or the moment required to overturn it or is that including a safety factor?
So like say the smaller forklift, it’s just lifting capacity in the normal range of lifting at the middle of the forks is 5,500 pounds and so our device with the biggest robot was about 1,500 pounds um but we did move the load distance out to 39 inches instead of the ideal which would be 24 inches so it looks at the the moment so normal moment would be 5,500 pounds at 2 feet where we have 500 pounds at 39 inches so doing the math it only goes to about 44 percent of the capacity so it’d be well within the range of the smaller forklift.
Very cool. I’m gonna have to run because I gotta get ready for my next thing. So nice work folks really. Yeah I appreciate it.
Yeah good luck to you all.
Thank you. Thank you.
Hey good work.
Awesome. Thank you.
Thank you. Thank you guys for coming and watching.