Lower Cam Drive Bearing – Inner to Outer Race Misalignment Cause Discovered !

I could use an opinion or two on this one.

I am deep in the process of replacing my pair of lower cam drive bearings, which started to make some unpleasant grumbling noises, after only about 1000 miles or so of use since they were last changed. And doing some investigative triage, I think that I might have found something which could explain the (to me) early failure of the previous bearings.

Reference this well-known factory cross-section drawing of the lower cam drive bearing area:

Image Unavailable, Please Login


You can see that the inner and outer races of this bearing (no. 2 in the picture) are designed to sit flush to a pair of stops. The inner race of the bearing is to rest against the shoulder step in the drive shaft, while the outer race of the bearing is to rest against a removable snap retaining ring (no. 6 in the picture, highlighted in yellow) with a beveled spring washer (no. 8 in the drawing, also highlighted in yellow) to be placed between the bearing and the (no. 6) retaining ring. This is all perfectly clear, and is exactly as I installed the last set of bearings.

But here’s the concern. See my sketch, below:


Image Unavailable, Please Login

I have measured the distance from the edge of the shaft to the face of the step in the shaft (recall that the inner race butts up against this shaft). This distance measures 47.8 mm as is shown in the sketch. I then measured the distance from the edge of the shaft to the face of the snap retaining ring (which locates the outer bearing race) and found this dimension also to be 47.8 mm as shown. So, based upon the shaft step and the snap ring face, these are at precisely the same depth, which thus would ensure that the inner and outer bearing races could not be misaligned when the two races bear directly up against these surfaces. So far, so good ...

HOWEVER … the above does not include the presence of the beveled spring washer which fits between the outer bearing face and the removable snap ring face. The unloaded thickness of this spring washer is roughly 1.4 mm.

And recall that, according to the shop manual (first picture), one is to finally install the cog belt sprocket to the shaft, and then tighten that weirdly designed outside retaining nut to a torque of 13 kg-m, or about 95 ft-lbs. Which means that you -really- tighten the cam belt pulley to the shaft, the backside of which then shoves the inner race of the bearing as tight as can be against the shoulder step in the shaft.

But the presence of that spring bevel washer means that the OUTER race of the bearing has the potential to be 1.4 mm further away from the engine than the INNER race of the bearing (Comment: the huge force on this spring bevel washer will serve to flatten the bevel washer somewhat, so the actual mis-alignment might be more in the range of 1.0 mm or thereabouts in lieu of the uncompressed 1.4 mm thickness). But no matter the final thickness of the loaded spring bevel washer, this will serve to force the outer race of the bearing to be mis-aligned with the inner race … which easily explains the short life of the earlier bearings!


---------

Based upon the above, it is my theory that the presence of these spring bevel washers serve the only practical purpose of (intentionally ???) causing inner race / outer race bearing mis-alignment, and that the potential for long and trouble-free bearing life will be much better if these bevel spring washers are deleted altogether, so that the inner and outer bearing races are true and plumb to each other.

Has anyone ever encountered this before?

--------

All thoughts, comments, and questions are very much appreciated!

 

Steve and Scott - Many kind thanks for the both of you chiming in here! As I hold your collective experience in high regard, I had hoped that one or the other of you might catch the thread, and to have both of your comments is a genuine treat (As an aside to Scott - I have been using one of your 'blue-belt-special' round tooth pulley conversion kits for quite some time now, and it's a splendid product - I could not be more pleased, and I am changing the mint-looking 10 year old Gates belts simply as a while-I-am-in-there).

Gratuitous "before" work-in-progress picture below (and further notice the Nick's water pump!):

Image Unavailable, Please Login

A couple of additional comments:

a) I am doing this bearing swap in situ, and I've never had the front timing cover off of the engine. Nor do I have any indication that the cover has -ever- been removed from the front of the engine.

b) Grabbing the end of the discussed threaded shaft, and pulling / wiggling hard, there is no 'give' whatsoever to be felt. This appears to indicate that the internal bearing behind the cover is in good condition, and I really can't visualize this shaft having the ability to move back-and-forth axially very much during operation (thermal expansion, sure, but that movement ought to be pretty minimal - certainly not on the order of 1.4 mm).

c) It is not clear to me how there could be no force placed upon the Belleville spring washer under the tightening of the external funky cog belt pulley nut, particularly since the torque to do so is to be quite high. Tightening said nut compresses the inner race hard up against the shaft's shoulder stop. But since the outer race (in my drawing and current way of thinking) is spaced out by the earlier noted 1.4 mm bevel washer, I visualize a pretty high amount of force transferred out to the outer race via the bearing's numerous internal balls, thus with the potential of placing the inner and outer races in non-similar planes.

d) If my interpretation above is flawed ( ... and I'm always willing to learn!), then how else to explain the failure of these bearings after less than 1000 miles of use? (they were the correct SKF model number as purchased thru Superformance)

e) Is there any visualized harm done if the Belleville spring washer is placed on the -front- side of the outer bearing race, rather than the rear as is shown in the factory drawing? Since the main purpose of this beveled washer is understandably to take up manufacturing slack, I see no real difference in accomplishing this goal if the washer were to be on -either- side of the outer race. And moving it to be in front of the outer race thus ensures that my 47.8 mm dimension as shown in my sketch will be maintained, irrespective of how tightly one torques the external cam belt pulley nut.

Again, thanks and Cheers - DM

 

Ha! Once an overthinking engineer, always an overthinking engineer, eh?

Here's a rephrase of my concern:

Imagine that you are installing a new bearing, working from the outside of the cover, taking care to use a proper bearing setting tool such that you apply equal and even seating pressure on both the inner and outer races. The spring Bellville washer is in place, 'behind' the bearing that you are driving, exactly per the factory drawing. As you lightly tap the new bearing into position, the first point of contact will be when the outer race edge contacts the spring washer, at a depth of about 46.4 mm. At this point, you stop since you can't easily compress the spring washer further.

There is now a 1.4 mm gap between the edge of the inner race and the stop on the shaft. This corresponds to the unloaded thickness of the spring washer. And so you lock the bearing into place via the outer snap ring.

Now you place the cam belt pulley into place, and tighten the funky external nut to the rated (high) torque value. As the funky nut tightens, it will tend to close up this 1.4 mm gap noted above by drawing the cam belt pulley tight to the inner bearing face, and then this 1.4 mm space begins to close as you continue to really tighten the funky nut.

Which means that the inner and outer races end up being in offset planes, which differ one to the other by the final thickness of the spring washer.

...... I'm still trying to see any errors in the above thought process!

 

I do sincerely thank you folks for your thoughts and comments, and I do already clearly understand the concept of "how-this-all-should-work".

But I think that you all might be missing the main point: My car does not match the factory drawing.

Taking a look at the factory drawing (first picture in my first post), you can see that the bearing inner race abuts the shoulder on the shaft. In my second sketch drawing I have measured (repeatedly, to check) this shoulder to be at a depth of 47.8 mm.

Now, back to the factory drawing. The bearing outer race abuts against the beveled spring washer, which in turn mates tight to the snap ring retainer. So the depth of the snap ring retainer -alone- should be at 47.8 mm *PLUS* the thickness of the bevel washer, which is about 1.4 mm. That means that there ought to be about a 49.2 mm measured depth to the face of the snap ring.

But in my second sketch drawing, I have measured this snap ring depth (again, repeatedly) to be at -exactly- the same 47.8 mm depth as the inner bearing's shaft shoulder. In other words, rather than my snap ring depth measuring to be the 47.8 mm which I show, it should instead be around 49.2 mm according to the factory drawing. But it obviously is not.

Think this concept through - the factory drawing (which I dutifully followed, to the letter, back 1000 miles ago, without checking) -clearly- shows the inner and outer races to end up being in the same plane, as they should. And so, once the bevel spring washer is included per the factory drawing, then one would reasonably anticipate no bearing alignment problems whatsoever. But in my case, and perhaps others, following the factory drawing and placing the snap ring in the location as shown by the factory -creates- a 1.4 mm misalignment!

I find it to be a staggering coincidence that my pair of measurements (in my second sketch drawing) agree one to the other, right down to 0.10 mm. That "coincidence" seems to imply that the bevel spring washer should NOT be located in the position as shown in the factory drawing, but instead would be far better served to be on the outside of said bearing.

Clear now?

Thanks again and Cheers - DM

 

I do wish that this was that simple. But how many people have actually taken the time to MEASURE and to CONFIRM the dimensions which I have reported, in order to check for correct planar bearing race alignment before installation?

Darn few, I would guess.

(Cue my earlier comment about my being an "overthinking engineer", who tends not to accept things at face value)

Steve, I do sincerely appreciate and thank you for your well-experienced thoughts and comments - DM

 

Is it true, then, that the gear shaft has more than, say, 1.5 mm axial "wiggle room" so that it finds the sweet spot for the outer bearing all on its own? Grabbing ahold of the end of this shaft and trying to move it, it is rock solid, with no play or movement whatsoever in or out, or left to right. Which is what I would have expected. I have never had the opportunity to examine a front cover off of the car so that I could see the internal gears, but your comment makes me wonder if indeed this shaft has, by design intention, a pretty good amount of axial travel potential? If so, and assuming that this "pretty good amount" of travel is enough so to allow for variations in the thickness of the bevel spring washer as it compresses (or not) with heat and time, then there's no harm and no foul.

You'd asked about the amount of shaft threads sticking out beyond the end of the funky nut - see picture below taken 5 minutes ago of the front bank which I have yet to touch, and which had the bearings and washers and such installed the last time -exactly- as the factory drawing shows. There's a couple of threads protruding out beyond the locking portion of the funky nut, which makes me think that it is tightened as it should be. The nuts on both banks looked the same after my work here done years ago.

Image Unavailable, Please Login

 

Thanks again to all for your continued thoughts!

I am not inclined to try to rock the crank CW / CCW at the moment as I currently have the rear bank cam belt off (pending the outcome of this discussion) and thus all necessary timing marks are set just so. But I can assure you that the lower cam drive shaft is indeed not moving in any direction whatsoever.

Much is made of the fact that the outer bearing race is free to move axially within the recess in the timing cover, being held in place via a friction fit and the force exerted by the spring bevel washer. But this friction fit is really tight - the bearing puller that I used to remove the old bearing requires a considerable amount of force to be able to extract the old bearing. It certainly is far from a loose fit. So I do have to wonder about just how much outer race axial movement potential exists if this outer race is trying to find the sweet spot by fighting the spring bevel forces. Of course, being what appears to be an aluminum timing cover, the thermal expansion will be greater than that of the steel outer race. But I still don't see how only this thermal potential expansion difference between the aluminum cover and the steel outer race could be enough to free the outer race such that it can move axially in any noteworthy amount in an attempt to self-reduce any race misalignment.

At the moment my inner and outer race locating stops are spot-on, again as shown in my sketch way back in Post #1. I cannot help but wonder what difference it makes if the spring bevel washer is installed on the outside opposite side of the outer bearing? The spring washer should allow for minor manufacturing tolerances and axial movement of the outer race irrespective of the side it is located, so playing Devil's Advocate here, what difference should placing the spring washer at the front or the back make?

Or, for that matter, why not consider the idea of just deleting the spring washer altogether, thus permitting for an unencumbered 1.4 mm travel potential for the outer race, and without the need to fight the bevel washer forces?

DM

 

Scott - Thank you yet again for taking the time to post the drawing and pictures. Very helpful!

Referencing the drawing, IF the pair of locations noted as "slip fit" are, indeed, free to move readily, then there should be no worries. My concern now shifts to wondering IF these locations as noted actually are free to move as design intended on my own engine.

For the moment, leave this one down to me as I try a couple of ideas which have come from this most beneficial discussion. I will report back here accordingly should I have any noteworthy discoveries to share.

Thanks to all and Cheers - DM