Broken intermediate shaft pulley

[peterman]

I cannot find a single instance of this happening to anyone else online. Which makes me think think theres some funky things going on inside.
Relavent specs:
Fresh rebuild
M50b28 crank
Je pistons
325i rods
Replaced the intermediate shaft bearings - spun freely but no excessive play.
Pulley is the sintered version
I havent pulled it apart yet but looking for thoughts. Has anyone seen this happen before? The only thing I can think of is that the pump locked up. I had a good 40 psi at idle.

[hasa]

Does the shaft pulley turn by hand?

[digger]

Maybe bolt came loose (insufficient preload or it wasn't seated correctly) and rattled around pounding the pin and bolt. I have seen the pins broken off before.

[peterman]

The broken bolt has a consistent 45 degree helix break associated with a torsional failure. My thought is that the pin broke, then the bolt continued to tighten because of some crazy drag or hard stop in the oil pump or pump shafts. Essentially, timing belt constantly tightening the bolt until failure.
I have not tried to spin it yet.
I am glad to hear that you have seen that failure though, digger

[digger]

The broken bolt has a consistent 45 degree helix break associated with a torsional failure. My thought is that the pin broke, then the bolt continued to tighten because of some crazy drag or hard stop in the oil pump or pump shafts. Essentially, timing belt constantly tightening the bolt until failure.
I have not tried to spin it yet.
I am glad to hear that you have seen that failure though, digger

i have seen the pin break off but as it wasnt my engine idk what the bolt looked like or the outcome

The witness marks suggest the gear spun on the end of the shaft so it was atleast loose for quite some period of time so that will likely be why the bolt failed. If the bolt came loose then this could explain why the pin sheared off.

the belt tension and resulting torque probably isnt enough to over tighten the M10 gr10.9 bolt as it acts over a fairly small radius so the gear just slips and can oscillate

[varg]

Most of the time a key or pin fails in something like this, it's because the bolt wasn't torqued properly and the pin saw loads it shouldn't. The interface between shaft and sprocket does all of the heavy lifting in torque transmission, not the pin itself.

[litu]

Can someone tell the torque for that bolt?, thanks

[digger]

60Nm for the intermediate shaft sprocket and 65Nm for the cam gear sprocket

[peterman]

Hoping to have an answer for you guys on Saturday. Bolt was tightened and loctited in place... But anything can happen.
Varg, I wholly disagree with your statement. The pin is what drives the shaft in the same way a woodruff key would. They are immensely strong in shear, which is why the failure is so uncommon. The bolt is there to ensure the pin remains engaged, and suppress any gear wobble. To rely on the clamping force and static friction of two faces would be crazy. A taper lock would work, but also could be suspect in this application.
All that being said, I am hoping it was improper torque. That would be great.

[digger]

Hoping to have an answer for you guys on Saturday. Bolt was tightened and loctited in place... But anything can happen.
Varg, I wholly disagree with your statement. The pin is what drives the shaft in the same way a woodruff key would. They are immensely strong in shear, which is why the failure is so uncommon. The bolt is there to ensure the pin remains engaged, and suppress any gear wobble. To rely on the clamping force and static friction of two faces would be crazy. A taper lock would work, but also could be suspect in this application.
All that being said, I am hoping it was improper torque. That would be great.

Varg is correct these things are usually designed to be friction driven with the pin/key for alignment only. The reason is because they usually fail due to the harmonics in the system being an internal combustion engine the pins and keys get pounded as the torque varies and they fatigue due to the hole or notch etc. Whereas with a friction drive the torque transfer isn't so concentrated so you don't get that failure mode when things happen as they should. Some rough calcs show that torquing the bolt to 65 Nm gives about 200Nm of friction drive before slip

In this instance the pin would be probably just for redundancy (and the two work together to stop bolt loosening limiting the relative rotational movement) as there is no need for phasing the intermediate shaft gear. If you wanted the pin to drive it then you would need to design it properly (the first thing is not make the pin hollow, loose and not some sintered BS)

Look up S55 crank hub slip issue it was just friction with a fancy washer with single bolt to stop the timing going out of whack it works most the time

[peterman]

Never one to not investigate, I decided to run the analysis on the bolt clamping pressure and how that correlates to contact pressure and subsequently shear stress on the intermediate shaft. Using a static friction coefficient of .8 and a bolt pretension of 48000N which pretty much equates to 65Nm on a grade 10 M10 bolt. I assumed the bolt head to be 18mm in diameter to define the clamping pinball range of the "bolt" (beam element).
I fixed the "intermediate shaft" and applied a ramped moment from 0 - 300 Nm over 6 seconds. A stock M20 will put out about 220Nm of torque.
Sure enough, shear stress is an order of magnitude below the contact pressure confirming that the surface contact provides enough pressure to keep the gear from spinning.

I stand corrected!

[peterman]

Pulled the front apart and the intermediate shaft was locked up. The oil pump shaft spins freely but the intermediate was not moving. I was able to pry it out with two pry bars on either side of the retaining groove. My clearance when I honed the bearing was clearly insufficient.

[digger]

When you took that photo in the first post wasn't whether it spun by hand the first think you checked?

[varg]

I have never seen someone fact check one of my forum posts in ANSYS

Look up S55 crank hub slip issue it was just friction with a fancy washer with single bolt to stop the timing going out of whack it works most the time

Probably the diamond microparticle impregnated friction washers/shims that are popping up in the over-optimized world of modern automotive design. I don't pay much attention to modern BMWs, so I only know that the S55 hub slip issue exists and aftermarket parts exist that may or may not fix it. Is a root cause known? Torsional vibration? Just cost cutting in a bad place?

[peterman]

When you took that photo in the first post wasn't whether it spun by hand the first think you checked?

I didn't check it. I know I torqued the bolt and added loc-tite so I was very confident that it was a torsional failure. As I said before, the bolt had the marks of a torsional failure. One thing I should note was when it sheared the pin, it screamed for about 3 seconds before shutting down the engine. The bolt itself sheared when I was cranking the engine over by hand. It was difficult to spin, but then became very easy and the cam lost timing. That's when I knew what happened.
I tried to spin the shaft today and it had no intention of turning. I took out the mating gear to the oil pump and was able to use a 6 mm socket to turn the pump with my fingers.
Intermediate shaft was stuck and was difficult to remove. There was bearing material stuck in the oiling groove.

Varg, it's the engineer in me. Had to be done. I did modify the analysis and run it again with more realistic numbers, ( 35000N clamping force and a few other modifiers to bring the results in from the fringe) even still, if the pump consumes .5hp, thats equivalent to around 10Nm at 400RPM-ish. So if it sees 20 Nm that is 15 times less than what was calculated at slip. What I will say, however, is that there is clearly a difference between analysis and real life, because the pin sheared and the bolt failed due to slip, so my coefficient of friction was clearly off or else the engine would have stalled without breaking the gear.

The S55 crank slip stuff kind of proves my point. Compressive force is not enough. Why not pin it and never worry about it? Every aftermarket solution is some type of keyed component that doesn't rely on friction.

Also, I had to use an easy-out on my lathe to get the bolt out