My team has noticed some very significant wear on the 50t gear in the 3 cim ball shifters that meshes with the cim pinions. My theory is that this is because it is aluminum and the cim pinions are steel, so it would wear faster. As it itself does not mesh with any aluminum gears, would it be worth it to replace it with a steel gear of the same size? We have some lying around from AndyMark. What might be some problems with this if any?
You would have no problem with the swap unless your robot was already 120.0 pounds at inspection. The primary reason for using an aluminum gear in place of steel is weight reduction. I suppose one could argue that an aluminum gear would also be easier to machine if you were making your own.
We’ve used aluminum gears in the drive train of several of our more portly machines and haven’t worn one out yet during the machine’s <10 hour operational life.
Out of curiosity, what lubricant are you using in your gearboxes? I’ve used 3 CIM Ball Shifters for about 5 years and never had any noticeable wear of the aluminum gears inside them. Also, what ratios are you using (including external reductions and wheel diameter)? This may also be a factor.
If you’re not already using it, I STRONGLY recommend using Lucas Oil Products Red “N” Tacky #2 Grease. It stays on the gears far better than regular White Lithium grease, and, when applied liberally, in my experience typically doesn’t need to be reapplied throughout the season (though it never hurts).
We have seen serious wear on the 50t first stage gear in 3 cim ball shifters. We had to replace that gear at least once during each of the two seasons we used them as drive gearboxes.
We used the steel pinions on the motors. Aluminum pinions might have been kinder, or they might have worn out too. The gearboxes were stuffed with lubricant.
I think replacing it with a steel gear (like this https://www.andymark.com/Gear-p/am-0149.htm) is a wise choice.
One other thing that just occurred to me, another way to reduce wear on a gearbox is to implement an acceleration “ramp” in the robot code to prevent backlash due to sudden direction changes. If done properly, the driver won’t notice the difference but your gearbox definitely will. 
I still can’t figure out what could be causing you guys to have issues with these gearboxes. This year we ran our gearboxes at 2 districts, district champs, worlds, and 2 off-season events with one of the most aggressive (high speed) reductions I’ve ever tried, and every time I’ve opened up the gearboxes for inspection the ceramic coating is barely even warn off the 50t gears. 
It’s possible that we didn’t grease them enough but they wore out after one season of (very) aggressive driving in steamworks through world champs and countless outreach events and lots of driving practice. I think in the future we will use the steel gears for the first stage and also regressed after at least district competitions.
Additionally, how would you go about calculating the acceleration for this ramp?
Honestly, for us it’s just been a lot of trial and error. You see what your driver can tolerate and make sure the ramp doesn’t create too much control latency; a “sloppy” ramp can create a delay between control input and robot reaction, which can cause you to run into things you didn’t intend to. That said, a properly tuned ramp can actually make your robot easier for your driver to control (not to mention it tends to make top-heavy robots slightly harder to tip).
If your programmers already use a ramp for accel/decel autonomous control (it seems like many teams do), that can be a good starting point.
It doesn’t take much to eliminate the effects of backlash, so even if the ramp is barely noticeable to your driver it’s still MUCH better than your motors slamming from +5000rpm to -5000rpm in a fraction of a second like un-ramped controls do.
Voltage ramping is a great way to reduce the current draw on your robot, and reduce the chances for browning out, but I’m not convinced it will increase the life of your gearbox. The failures we saw were not individual tooth failures caused by overload. Rather they were gradual wear, first through the anodizing and then through the aluminum until the teeth are gone. Certainly load on the gears affects the wear rate, but the ramping only affects a small portion of the duty cycle.
Having never used the 3 cim ball shifter(we do custom) take everything i say with a grain of salt.
The only time we’ve ever worn out gears in our drive train was on our practice robot in 2018 because we never greased them. other then that we have never worn out aluminum gears. id check your grease i saw someone else posted about using red n tacky id go with that or what we use is white lithium grease.
Edit:in 2017 we did see a vary small amount of wear on the first stage of our gearbox(3 cim single speed geared for 17ish? f/s) by the end of season(2 districts a dcmp worlds and an off season event) but it was negligible.
Out of curiosity, Noah and Brendan, were either of your teams using 11t pinion gears on your CIM motors? While, in theory, the radius should be the same, the slight reduction in teeth might cause more undercutting of the 50t gear than a standard 12t pinion would, which might result in additional wear.
Just a thought.
As far as ramping affecting gearbox wear goes, my thought was less in regard to “overloading”, which, as you indicated, would result in lost teeth rather than even wear; my concern was more in regard to the gear teeth slamming into each other during any sudden change of direction in the motors, the impact of any one of these actions would be minimal, but over time would likely cause more wear on the gears than a more gentle transition between directions would.
Have either of your teams contacted VEX about these issues? They might be able to provide better insight as to the possible cause of these problems. The rest of us are basically just making educated guesses, but they’ll have a LOT more experience dealing with these sort of issues than any of us will, and their customer service and technical support (in my experience at least) is great.
I learned gear design equations, specifically Tooth Capacity in Contact Stress or Wear/Surface Durability rely on the ratio of the Modulus of Elasticity of a bull gear to a pinion gear. Design of Machine Elements, by MF Spotts.
Very often in real products designed for long-term durability, you’ll want a softer tooth pinion turning a harder tooth bull gear. Why sometimes you’ll see brass gears turning steel gears.
As for use in robotics, where the life of the gears is usually less than 100 hours a season (a guess) I’m not sure this applies, but for students reading this, it’s a useful fact to know.
I’ve only ever seen significant wear on the the ball shifter gears when used with steel pinions. If aluminium gears were the issue then why don’t we see significant wear on the 2nd stage? Additionally since it’s 3 steel pinions on one aluminium gear that’s 3x the wear of a normal gear reduction with steel exasperating the effect even more. Aluminium pinions are really the preferable choice. With the drive train forces divided over 3 pinions breakage shouldn’t be an issue.
In general, I would say that using steel gears would work just fine.
But you might be chasing a problem to the next weakest link. So it would be better to try to understand the root of the problem (pun intended) and try to address that.
Make sure you have the right gear ratios as cbale noted:
It is possible to get the wrong sized pinion. It will mesh “good enough” that, to the naked eye, you may not notice that the mesh is wrong.
The other common problem that I have seen is that, with high loading, the motor mounts can deflect and cause the contact point of the gear mesh to shift. Given the separation loads, the mesh will usually shift out toward the tip of the gear tooth and also toward the edge of the tooth closer to the motor (rather than being uniform across the entire tooth width) so if you are seeing a lot of wear on the motor end of the teeth, it is likely to be motor deflection. Make sure the motor bolts are tight and wiggle the motors to make sure there is a good amount of stiffness in the mounting. Keep in mind that the motors can also move due to inertia during robot acceleration or impacts and not just the separation forces at the gear contact point.
The 3 CIM ball shifters seem to have a decent amount of stiffness and strength so I would not expect this problem with that gearbox unless the motor bolts are loose. We used one of these gearboxes for the first time this past season for our lift and I was pretty impressed with the construction. But our application was not very abusive in terms of loading, so others may have a better input on the gearbox stiffness for more abusive applications.
The other “issue” that I would expect could cause excessive wear (although I have not experienced this directly) would be if you had one dead or poor performing motor. If one motor was dragging on the gear instead of driving, then you have the other motors working harder which would cause higher loads and more wear. Plus the bad motor would be being driven which could result in load and wear on the other side of the tooth.
We’ve been using aluminum gears for a long time (2011?). You’ll experience worse results using the aluminum gears as the pinion and I would not use them. I forget what year but we made 11T and 12T gears out of 7068 Aluminum and had issues with them wearing out. I’d much rather change the large gear out than replace each cim gear.
The first stage experiences the most wear due to high speed + multiple pinions on the same stage. The second stage doesn’t wear as badly due to slower speed and single gear contact.
We’ve used aluminum gears for two years on a Vex 3 stage gearboxes. Just took the gearboxes apart - barely can see any wear. Last year we did five competitions during the season and two offseason competitions.
Not sure what is going on for you, but in years before we combined steel and aluminum and it didn’t go well.
Most of the excessive wear I’ve seen in FRC drivetrain gearboxes was always on the second stage. It sees a higher torque load after the first stage reduction is applied, but the teeth are all still the same 20DP. It may vary depending on the application and what gear sizes are used.
I’ve seen on multiple occasions that if you run a gearbox with no grease (note: all steel gears in these cases), the second stage gets eaten away first. The one time we’ve had an issue with the first stage interface was because a motor got bumped out of position after hitting a field element and bent the transmission housing, forcing the pinion into the first stage gear.
On this note, I did run into an issue with a 3CIM Ball shifter a few years back where one motor was not seated properly before it was screwed together, which caused the back of the casing to bulge out towards the motor when someone attempted to screw the motor in and force it to seat correctly. If you notice your motors aren’t sitting completely flush against the gearbox housing, the cause of your gear-wear issues are likely deflection-related.
We’ve experienced a similar condition with the first stage of the WCP Dog Shifter gearboxes over the last 3 seasons in which we used them. We would have to change out the large aluminum gear(s) that mesh with the steel CIM pinions at least once throughout the season due to failure or fear of failure.
One theory I have to explain this wear condition is that the plates or housing that the CIMs are cantilevered from is not entirely rigid. As a result, the motor has a slight rotational degree of freedom by which it can “wobble” away from it’s axis and its shaft can deflect toward/away from the gear it meshes with. The small variation in gear mesh could contribute to the steel pinions grinding away at the aluminum gear slowly over time. Other components of the gearbox are all supported on both ends by the parallel plates of the gearbox, whereas the motors (for the most part) are cantilevered off a single plate or a plastic housing. Take this with a grain of salt as this effect has in no way been measured or proven.
Add to this that the first stage gears mesh and un-mesh at 1.5-3 times the rate and with 3 times as many contact surfaces as compared to the output stage, and it could help explain why we observe nearly no wear on the output stage gear sets but a substantial amount of wear on the first stage.
We also had even worse problems in 2014 before we 3D printed custom “snowman hole” spacers to prevent the motors from vibrating away from their intended location, but that is a separate issue that does not apply to the ball shifters as far as I am aware.
We are looking into steel gears and belts for the first stage as potential alternatives to see if it would solve this problem.
Echoing the tips to look for small assembly errors (especially related to the snowman holes!) and misc deflections, as well as adding small software ramps to your controls.
I would predict reduced wear at that interface with your small (“aggressive”) reduction, due lower torque, lower gear speeds, and fewer overall cycles at the CIM. All before software. Your motors would have been hitting electrical limits before mechanical ones.