Mechanical Reliability
 

I've searched for awhile for a thread discussing simply mechanical reliability, and I can't really seem to find one. If there is one, please feel free to correct me.
What are some of the major guidelines in designing super reliable FIRST mech. systems? Our team has been plauged with what seems like a traditional breakdown in mechanical devices(gearboxes, arms, lifts) every year at the first regional, and we've never really been able to solve it. The answer would be to run everything during build season till it breaks, then fix the problem, but we never really have time to do this.

What are team's experiences in reliability? I know many teams have the same problems. Is there a whitepaper? I'm wondering about general tips: go light on dimensions, beef up everything, beef down everything? Tight tolerances, loose tolerances, etc.

-Thanks :)

Re: Mechanical Reliability
 

in general, reliability is a matter of calculating the forces a mechanism will encounter, then choosing the materials and dimensions that can withstand those forces, including a safety factor.

The easy part is calculating the normal forces, like the acceleration from the motors on the gears as the bot speeds up or climbs a hill.

The hard part is calculating how much force will be encountered if you slam into a wall at full speed, or another bot head on. It can be calculated, but the math is more complex.

Re: Mechanical Reliability
 

A lot of it comes from knowing what works, either from your team in the past or even better, other teams in the past. I like to go with what I think will work, and a little less if I'm feeling brave and a little more if I'm feeling worried. You can build something very reliable with very few calculations of any sort involved. You just have to be able to envision what kind of forces will be applied and where, and what the possible failure points will be. Intuition is what I use most. If you don't have a lot, calculations are good I guess. I get a lot of my knowledge of how to build things and of what works well by walking around the pits and literally sticking my head into other team's robots. I don't usually ask too many questions, but I observe taking painting every detail into a mental image. If you look at any mechanism for a minute or less a walk away satisfied that you know enough about it, you are doing something wrong. Spend a lot of time observing what other teams do successfully, and eventually you'll pick up on it.

Can you point out a specific problem you had and maybe give us a picture of the mechanism?

EDIT: Also, when you have a problem, even more important than getting it fixed is figuring out why the part/assembly failed. What went wrong? If you take a good hard look into why things fail, you'll know what you need to do to prevent it from happening next time. And after you can do this, then you need to develop the "power" to prevent failures from happening the first time.

Re: Mechanical Reliability
 

In engineering that 'power' is the math of basic physics and mechanics, and the knowledge of material science.

I would not expect a high school student to be able to calculate the stress and shear forces on any mechanical part that I point to, but an engineering mentor should be able to do this. Its a big part of what you learn in mechanical engineering

and that is the trick - designing something that is strong enough for whatever will be thrown at it, but not overdesigned so that it is too big, too heavy, and too expensive.

Re: Mechanical Reliability
 

I don't know about other threads, but our '05 bot is a pretty robust design.

1. All mechanical parts were designed with big safety factors. For example, we thought our motor would lift an arm with a tetra at a speed of N feet per second. We bought a big sprocket that would only move the arm at N/3 feet per second. This marginally slowed down our lift speed, but the arm didn't break and the motor didn't overheat or burn out.

2. KISS. Keep it simple. If you don't have a big budget and the help of a high-end corporate R&D lab, don't get cute. A simple bot that doesn't break will likely be more successful than a really clever one that you don't have the materials or tools to build strong.

3. Like the others wrote, do the math. We ended up changing some provisional specs after doing the physics analysis (it's handy having your mentor be the AP Physics teacher).

I'm sure others will have more clever suggestions for you soon. Good luck.

Re: Mechanical Reliability
 

If I were to make any suggestion, these two mentioned above would be the two I would give.


I have seen a "Safety factor" mentioned a couple times. A ball park number for safety is to design for double what you think is the max. you will ever need. For example, if your arm will, at a maximum, have to support 8 pounds at full extension of 3 ft. Make sure you can support sixteen, or more. Also, don't forget to account for acceleration/deceleration forces of that mass. (Do the math!!!)By following the KISS principle, you may save a little weight. Use the extra weight to build a stronger arm.

Re: Mechanical Reliability
 

I've found that most of our failures have come not from the design, but from the manufacturing and assembly of our components. Using the wrong types of fasteners, or not using loctite, has caused many assemblies to come loose and fail. Misaligned holes, shafts, chains, etc... will also lead to early failure.

It's tough to spend the first 4 or 5 weeks designing, prototyping, and testing to find what works. Then spend the last week rushing to reassemble all your work, without much forethought to how long these assemblies need to last.

One of our stupid moments of this year came to light just a couple weeks ago at our last post-season comp. We noticed that our drive gearbox mounts had come loose, causing a misalignment with our drive chain. During the 6-week build, we had assembled all the mechanical components, then came back and installed all the electrical gear, including mounting panels and brackets. Unfortunately, the main electronical panel was installed directly over these gearbox mounting screws, and were impossible to get at without removing the entire panel. (Good times...)

Anyways... as most people have already mentioned, over-design as needed with respect to material selection, x-sectional area, and gear ratios. But, don't forget about the little details, than can (and will) come back to haunt you.

BEN

Re: Mechanical Reliability
 

Actually you are describing a STATIC safety factor. This assumes no acceleration (like being slammed by a robot). For my analysis on arms or lifting gear I assume the static load, add a 1/2g side load in the weak axis to account for getting hit. That gives a reasonable static load. I then multiply that by 4 to get a DYNAMIC safety factor. Actually the impacts we see will rarely push it to twice the load, but it doesn't hurt too much to add a little more. We have never broken a structure this way.

ChrisH

Re: Mechanical Reliability
 

Well, I didn't have anything particular in mind, just general tips now that the season's coming closer and I'm starting to think about things. One of our biggest problems I can think of is inappropriate use of set screws; they can't really hold everything, no matter how many you put in.

Re: Mechanical Reliability
 

Unless you really know what you are doing, avoid setscrews alone. (reads post above me and relizes it says the same thing)

Our team has had a traditional rule, that in some cases if somebody doesn't like it and if it can be torn off with bare hands, it leaves the robot... forcefully....

DO CALCULATIONS ON EVERYTHING!!!

Re: Mechanical Reliability
 

Actual engineering is something our team needs to start doing. I know, I know, "The point of FIRST it to promote engineering, not just building big rc cars", but with a limited engineering mentor base, our team is often between a rock and a hard place where doing the calc. might get you there better, or maybe just doing it on intuition will be faster and quicker. As we grow and begin to learn more though, I feel we will be able to spend more time on design and less on build, build, build!

Re: Mechanical Reliability
 

To quote some guy nobody here has ever heard of, Woody Flowers.

"Set screws inhale audibly!" Kickoff 1999

Never expect a set screw to take any significant load. You can use one to hold a key in a keyway, but even that can be done without if you're clever. There are extensive discussions on the best way to fix a sprocket or gear to a shaft in the archives. Some got pretty passionate about different styles of keys or D and double D shafts. Searching that stuff out would be a good place to start

Some other thoughts:

In general it is a good idea to support both ends of a shaft. A few teams get by without it and leave their wheels hanging out, but they also use extra large diameter shafts.

Keep your motors running fast. If a motor boggs down because it has too much load on it, things start to heat up real fast. This can lead to major problems, including massive escapes of magic white smoke. Design so that the torque does not exceed 25% of the motor's rated stall torque, or so that the motor runs at 75% of free speed. Which are for all practical purposes the same condition. Your motors will be cool and happy and live a long cheerful life. Yes this may mean an extra stage of gear reduction, the longer life will be worth it in the end.

Learn how to crimp wires properly. Spend the money to get a really good crimper. Backup the crimp with solder. Al and the other electron pushers may feel free to add your remarks.

Maybe I'll add some more later

ChrisH

Re: Mechanical Reliability
 

dont be shy about asking for help - even putting up your basic design and asking someone to calculate how much force, torque, shear... it will see or handle

if you dont have engineers or engineering students on your team ask another team to lend you one, either a team nearby that can stop over to your site, or someone else who can help you through emails, phone calls and posts on this forum.

Some basic calculations can go a long way to pointing out the weak link in your design. Setscrews are a good example. A quick analysis of a set screw on a drive wheel shaft shows the little screw can be stressed with thousand of pounds of pressure or torque, and there is no material on earth that can withstand that and not deform (and come loose).

Many things are not intuitive. For example, how much force do you think is exherted on the connecting rod bearings in your car, when it is accelerating at a normal rate? I did the math once, it comes out to ten TONS!

BTW, when crimping wires you need those big orange and black crimpers they use to put in the kit of parts, and you MUST make the pirate noise when you squeeze the crimp! Its required! ARRRrrrrrrrrggggggggh! otherwise the wire will fall out of the crimp when you need it the most. :^)

edit: connecting rod bearings, not tie rod bearings!

Re: Mechanical Reliability
 

Our team hasn't really ever had major problems with the actual systems on our bot breaking. We have burned plenty of motors out and all but what I have noticed us do most is keep the delicate systems on the machine inside the box and make it so the opponents hit the frame and not a shaft that could break.

When it came to set screws we made sure that if we had one in the machine it was accompanied by a hardened steel key or pin and not one of them ever sheared. I personally did the calculations for the gear reductions on the arm so it could pick up a tetra easily and still be controllable. Then in the competition we got the arm stuck in the center goal and accidentally picked up the machine so... ya.

Also, I would like to ask what a "tie rod bearing" is to the best of my knowledge the tie rods ties the steering rack to the wheel and does so using ball joints. The wheels are usually pointed a little in or out so that you get a few 10 of pounds of static force on the system and perhaps a few hundreds pounds of instantaneous force if the car hit a bump but never something like 10 tons. In fact under normal conditions no bearing in a car should have more than the weight of the car on it since cars can't accelerate that hard (of course this can change if the car hits a bump). Just wondering?

Alex

Re: Mechanical Reliability
 

Yeah KISS is golden. I'm a heavy believer of Murphy's law. I apply it to everything I do. Thats pretty much it really, if something on your robot can easily cripple you're robot make it as protected as you're comfortable with. I guess the big thing though is making it easy to repair. I know we have a habit of bringing a duplicate of all our electronics, programs, and even a bunch of raw material. At VCU the team next to us literally had their arm torn off, yet in less then an hour they had the entire thing working, thats what I call teamwork!

Anyways this is a great thing to think about now because when you build a robot you keep this foremost in mind. It's no fun to watch the robot you spent many all-nighters working on start spewing "magical electrical smoke" in the middle of a match. I speak from 100% experience here.

Since I mostly do electrical though ill give an obvious pieces advice that I think allot of teams could or do benefit from. really take the time to organize, secure, and and label wires (at both ends if possible). It makes not only tracing and diagnosing problems much easier, but it also protects the wire from being pulled or tangled.

Best of luck and I hope this helps. :)