Versaplanetary Output Shaft Casing Fail

Just showing pictures of our failure mode after OCR QF2 match. After 12/12 successful traversals in Quals and QF, our climber motor VP output shaft mounting points finally had enough.

This was running a climber winch on a 20:1 reduction with a Falcon 500.
Once the motor broke out of the mounting screws, it proceeded to rotate on the driveshaft and twist and then sever about a dozen nearby power and CAN wires. Clearly this made it impossible with our single 5 min timeout to fix our climber for SF1 (and the severed CAN wires took out the rest of the bus), but before SF2 we were able to hotwire CAN back to the drivetrain to play defense.

Next event we are going to add additional mounting support and utilize the side mount tapped holes with an additional bracket.

1 Like

We’ve seen some marring on ours over the years but haven’t had this failure before. How long are your mounting screws? The damage in the photos seems to go down about 1/3 of the housing. It might have more resistance to tearout if you can use longer fasteners in addition to side mounting.

1 Like

they were 1/2" screws but going through a 1/8" plate, so there was only 3/8" thread contact rather than the full 1/2".

Unfortunately we didn’t have any 5/8" length screws, but I’ll be picking some up now! 3/4" screws were too long even with a washer

A couple of things don’t add up here.

That looks like a VP2. With a 20:1 ratio in there you should have been able to put a 1.5" 10-32 screw in there with room to spare. Anything past 5/8" would have been useless, but a 3/4" should have fit. If the VP is assembled correctly, there is a lot of space after the output housing towards the input housing.

Also, 4 threads of engagement is going to provide the full holding force. With 3/8" in the output housing, you had 12 threads of engagement.

I suspect the bigger issue is that the screws were not fully tightened and they worked loose over time. Every time you climbed they worked a little more loose until failure at about the worst time.

Consider using loctite. And/or add it to your pre-match checklist of things to check.

The rule of thumb is that the engagement should be 1x the diameter when threaded into a like material. When the material the bolt is threading into is weaker/softer then it needs to be 2x diameter. So 12 threads would be the minimum engagement for a steel 10/32 into aluminum.

However that only relates to achieving the clamping force before stripping the threads. That doesn’t relate to the thing it is threading into failing. A longer bolt could have helped with that. If they weren’t tight then yes that will cause the threads to wear

1 Like

Does your winch apply side loads to the output shaft of the gearbox? It looks like the screws were being pushed sideways causing the material in the gearbox body to tear. The threads in the hole don’t look badly damaged so I would not think the force causing the damage was along the axis of the screws.

This is probably a time for some STEM education. It would be beneficial for you to draw a free-body diagram to estimate and analyze the magnitude AND the direction of the forces to determine what the appropriate solutions are. Otherwise, you just guessing at the root cause of the failure and fixing it by trial and error, which can take a long time.

Side loads? The force is specifically torsion/shear applied through the two #10-32 bolts countering the rotational torque of the staged falcon.
The VP is bolted to a 1/8” thick titanium plate which incurred no damage or bending, and the screws were fully intact (just ripped out of the VP). So axial loads here we’re not the issue.

We did remove and replace those screws several times during competition due to unrelated winch maintenance, so based on that experience using loctite here would not be beneficial. It’s possible the multiple loosening and retightening of the screw in these threads lead to additional wear.

I will have to look into why a longer screw could not be inserted into the stage. It’s possible we mix and matched ring gears in the next stage with the older version of the VP, which used #8’s here instead of #10’s

Is the winch drum on the output shaft? If it is, is the other end of the shaft supported? You have not shared sufficient information to show how the forces act on the gearbox.

Ah, I’m following your question now – the winch has a 0.75 inch diameter aluminum “sleeve” over the hex shaft and the shaft itself is supported by a flanged bearing and another 1/8" titanium plate on the opposing side. The shaft sticks out of the opposing bearing a bit and is locked in place with a washer + #10" bolt (inside of shaft is tapped).
The plate nearer the VP does not have a bearing, it’s attached in place by those 2 bolts on the outer housing only

You should still draw a diagram showing how all the forces act on the winch cable, winch drum, the two mounting plates, screws, bearings and the body of the gearbox. Start with the tension on the winch cable to see where the force is transferred to, then keep going. I suspect you will find there is a lateral force on the screws i.e. the screws are in shear between the mounting plate and the gearbox body.

From this description and the photos it sounds like there could be mounting bolts with a sheer load. The corners of this aluminum block are likely not strong enough for repeated loading like that. VP was designed in a BAG and 775 era. The Falcon 500 is crazy powerful in comparison.

In different years we’ve typically had a weak point regarding bolts into aluminum like this - either we forget to loctite and the screws loosen we don’t use a long enough screw in the first place. Either results can result in extra motion that eventually causes breakage.

This topic was automatically closed 365 days after the last reply. New replies are no longer allowed.