This is what happened today when we were practicing one of our new strategies. We noticed we couldn’t control our lift, but then we could, and then everything seized up. We took the gearbox apart, and a sheared off 7/14 teeth!
Events and pictures like this one create “teachable” moments, and that is a good thing. One of the best things about the FRC, I believe.
Three ways to take advantage if you have not already done so:
(1) get the lead mechanical engineering mentor on your team to explain how stresses on gear teeth are calculated, or
(2) seek help with (1) from one of the fine mechanical engineers who designed this gear [looks like VexPro to me], or
(3) seek advice from one of the many experts who follow CD.
After working through the stress analysis, try to understand the loading conditions that caused the 14T gear in your elevator drive to become over-stressed. Then (depending on the cause) you might look for design improvements to keep it from happening again.
We did something similar during testing, but managed to get the same gear down to 3 teeth.
We didn’t put much concern into stall conditions given the lightweight gamepieces. Then we stuck a sturdy locking mechanism on the output side of our gearbox. Should have probably seen that one coming.
Just curious, What strategy change caused a pinion gear to break? Your lifter at Northern Lights seemed pretty solid.
Our team actually does not have any mechanical engineers as mentors, in fact the entire robot was CADed, and manufactured by students (under the supervision of a few qualified parents.) We have addressed the issue by ordering a new gear, fashioned from steel instead of aluminum.
It seems that vex pro has already been notified of this issue, which is why they offer this gear in a steel variety.
Thanks for the wise words!
It definitely was a stall condition that caused the failure! We’ll have to be a little more careful this week in Wisconsin
What happened was our operator and driver were pushing for a faster cycle time, and I believe the operator had driven the elevator down and stalled the gearbox in this haste. It is powered by two CIM motors at a 9.52:1 reduction, with a stall torque hovering around 350lbs. Looking at the gear teeth it is apparent that they are a little wimpy for high load applications :o . We’ll have a steel replacement when we get to competition and replace it on Thursday.
The broken gear: 7075 Al ~70ksi yield strength
Common steel gear: ~40ksi yeild strength
You can’t fix this problem by replacing the gear. The pinion is overloaded and the best fix is a redesign that reduces load.
Edit: I went on a gear strength rant a while back, it could help you.
I agree they should fix the design to not load such a small gear/use on final reduction. But also I believe all the steel gears by VEX PRO are 4140 steel which should have a yield at 100ksi or more.
We also were concerned about this problem - we didn’t want to run the motors against the hard stop at the top and bottom of the elevator travel.
We use a pot to control travel of the elevator. Since it’s possible for the pot to come loose and change value we put a second pot on the other side and compare the two to make sure they agree. We also added a light sensor that checks the position of the elevator in the middle of its travel as another check. (We’ve never made a robot with an elevator before so we were a bit cautious.)
If the elevator comes down on top of a tote it’s possible for the elevator to tip the robot over. (This happened at our week 0 event when the HP tried to add a 7th tote and the co-driver tried to block it by dropping the elevator.) To stop the robot from tipping over, the software/electrical team added a sensor that detects when the front of the robot lifts off the ground. (I’m not sure what sensor they added.)
A quick question, what material is the gear that this gear was meshed to? You probably want to have them be the same material, that has given us issues in the past. That doesn’t mean don’t change the gear material, but if you have to then change both gears.
^This is correct, all of our steel gears are hardened 4140, a fair amount stronger than the typical 1018 steel gear.
First off, I’d like to apologize for posting false information. I clearly did not have standard gear alloys in mind when I posted that. To correct myself: a common steel gear would be around 50-60ksi.
I’m always the first one to jump up and quell the assumption that switching to steel from aluminum isn’t necessarily a strength gain. The reason being that your average steel gear, say 4140, has a lower yield than 7075 Al. This is because a lot of steel gears are sold in an annealed state.
I was not aware that vex hardens their gears. Now that I know, I’m even more in awe of the quality and value of products company’s like vex are selling.
On a side note, to the vex pro squad: under the CIM gear section on your website at the top of the page the steel gears are said to be made from 5150. However, lower down on the page it says they are 4140. Also, I don’t believe it’s mentioned anywhere that the gears are hardened. I would personally suggest that on the website you include the fact that you harden the gears. For me, the lack of that information would be the deciding factor in whether or not I buy a steel pinon. I mean, they’re listed as 5051 (yield of ~50ksi iirc) when in actuality they’re twice the strength.
tl;dr, disregard my previous post. I was incorrect that steel pinions were weaker that their 7075 counterpart. At least for vex pro products.
Edit: Wanted to share a little snipit of wisdom I recently leaned regarding cim pinions. Vex sells cim output shafts for VPs. Be mindful when using a cim pinion on a VP, you’re putting a significantly higher load on the pinion by running it through a reduction and if not accounted for it could completely ruin your performance at competition.
It sounds to me like you have no software developed to avoid stall conditions like this. In the future, I’d recommend installing sensors so that your motors don’t cause harm to your machine. Limit switches, reed switches, and Hall effect sensors are popular choices.
Also, you should be able to use the new RoboRIO to gather current draw data from the motors and to shut them down if they exceed a certain value in case your elevator is getting caught on something else.
Also, in addition to using sensors to stop your lift, be sure to have said sensors included in autonomous. Otherwise, you can have bad things happen to your robot. Case in point: http://i.imgur.com/jTXfc1l.jpg
That’s a 10:1 stage from a versaplanetary. We had to start an autonomous mode with our elevator down at the bottom of the robot. In teleop, we have a limit switch that stops the motor from applying downward force on the elevator. It was never included in autonomous because the elevator was never intended to travel anywhere near the bottom of the robot. Needless to say, bad things happened.
(Special shoutout to 3663 CPR for the assist. We were up and running for the next qualification match like nothing had happened.)