PTO Shaft to Gearbox Bolts Keep Breaking!

[airbiscuit]

It must get your attention when the PTO/Slip Clutch backs off the gearbox.

 

[ruffdog]

It must get your attention when the PTO/Slip Clutch backs off the gearbox.

That is what I was thinking too.........

 

[wirlybird]

In post 17, you say "They use a spring plate instead of individual springs. Both are properly adjusted and if anything adjusted slightly on the loose side."

First, I have no familiarity with slip clutches. There seems to be a consensus that movement is either at fault or a contributing factor. If the bolt tension is on the loose side, it may allow the rotating parts to move a little relative to each other. If the bolts are tightened to spec (or overtightened) , I'd think the only time there's a chance of that sort of movement is when the equipment encounters something that causes the slip clutch to slip.

The bolt simply holds/connects the PTO shaft onto the gearbox input shaft.
The collar on the PTO shaft that goes over the input shaft is so thick that you'd break a bolt before ever getting it to "clamp" down on the input shaft.
You can see in the picture, red arrow, how thick the collar is!
The bolt does take all the stress.
On a slip clutch type setup, such as in the picture, the bolt is meant only to connect the shafts. It should be strong enough to not shear under a large stress. The slip clutch is the safety in this case.
In a shear pin style, no slip clutch, the bolt is the safety mechanism. The use of a grade 2 or ungraded bolt connects the shafts but also is designed to "shear" under a potentially damaging load condition.

 

[wirlybird]

It must get your attention when the PTO/Slip Clutch backs off the gearbox.

That it does!

 

[chim]

Oh boy was I confused. When I saw the picture of multiple bolts, even though you did explain it I thought the broken bolts were for the plates. Big DUH for me

 

[wirlybird]

Oh boy was I confused. When I saw the picture of multiple bolts, even though you did explain it I thought the broken bolts were for the plates. Big DUH for me

It's all good and sure can get confusing!

 

[AndrewInOz]

Fracture photos add weight to non-shear failure. Too bad the fractures out of focus.
If not shear, how can a snug bolt in a boss fail in tension ?

All I can think of is out there with the aliens but was an interesting exercise.

In focus is the shank : showing the bolt has been rotating under load. Not just rattling but failing in bearing. Like the bolt walking heavily around the hole. ( surely couldn't self tighten the nut ? ).

What could make the bolt forcefully rotate......would have to be axial movement at the same time as torq applied. Hallmarks of bad alignment.

A single non CV universal joint transmitting thru an angle for a constant input speed will output a jumpy uneven speed. That uneven speed is restored to the original constant speed when transmitted back thru the second universal so long as the yolks are in the same plane AND the input (pto) is parallel with output ( gearbox input).

Off alignment will apply a series of +/- acceleration pulses around each rotation, just the thing to rotate our bolt.

Since this mower is a bitzer we should consider odd geometry.
Photos of drive arrangement and shear holes would help narrow speculation....

 

[shirleyujest]

Another possibility is the Shaft outside diameter is much smaller than the Collar inside diameter. This would allow excessive play in the joint.

 

[dicktill]

You said that the gearbox rotates smoothly. With the PTO shaft in place, the tractor not running, and the PTO disengaged, can you try rotating the PTO shaft from the tractor end to see if you can feel any anomalies?

 

[DarkBlack]

One remote scenario ,I don’t think has been mentioned, but add it to all the others…during use, your nut is coming off, allowing the bolt to start backing out. It gets hung up with the threads catching on the bolt head side holes. At that moment you at the weakest point of the bolt, and the shaft’s rotational torque is all behind applied to that one point rather than two points on a bolt, so it snaps off.

Re-insert your found remaining bolt heads from your photo. Is the remaining length exactly the distance to the nut side hole, to prove the bolt was fully in place when the end snapped off?
If so, that would likely rule out this scenario