[SURVEY]: Steering head longevity, ball vs taper roller

[dieseltech]

Since I have my own pet theory as to the taper roller bearing issues so common in bike steering heads, but relatively little experience with bikes, as opposed to cars, I wanted to gather some statistical data on this subject, in order to get past all the bias and BS surrounding the issue and get right to the hard facts, in order to further my own (and the collective) understanding of the whole issue.
And since I'm building my own bike, this is also of some practical significance to me.

So to put it simply, I'd like to hear about the steering head bearings on your bike(s). Specifically, I'd like to know at least the following things:
1. What type? 2x angular contact ball, 2x taper roller, combination 1x ball / 1x roller, or some other arrangement?
2. Average longevity. Also please state if you replace them only when they start showing signs of failure, or just preventatively while they still work OK.
3. What lubricant, and what quantity (ie. packed full, just a small dab, etc.) - don't bother giving the exact name of the lubricant, just the type, for example: wheel bearing grease, CV joint moly grease, white lithium grease, etc.
4. Contamination and sealing - when you change (or relubricate) the bearings, does the old grease seem visibly deteriorated (discolored, dirty)? What sealing/shielding arrangement is used to protect the bearings and grease from external contamination?
5. Failure mode - if the bearings did fail, what happened to them? Rolling element or raceway flaking /spalling / cracking? False brinelling (regularly spaced worn indentations on raceways)? Perhaps something else altogether? Also please describe the damage to the upper and lower bearings separately if possible.

Looking at just the raw numbers from the bearing catalogs, one would think that similarly-sized taper roller bearings would be much more durable than ball bearings under similarly heavy loads. And in the typical rotating-under-load applications that's actually very true. But what about heavily loaded non-rotating applications (ie. steering)? That's what I'd like to find out.

And before you ask - no, I can't just consult the industrial publications on this subject, because there really aren't any, at least none that are freely available. This is strictly a minority application of bearings, and outside of bikes such cicrumstances don't normally happen in the industry - and when they do, it's usually due to vibrations while in storage (stationary), and the focus is on preventing those. On bikes that would mean simply not driving at all. I have nothing more to say to this.

Finally - please, no bias and fanaticism here. I'm trying to get some actual hard data, not any speculation or hearsay. Please leave drawing the conclusions to me, at least for the time being.
Thank you for your cooperation.

[coachgeo]

If you don't get the technical answers your searching for here try advrider DOT com or hit up both now. Their the best forum I've come across that has a wide variety of knowledge about bikes, bike designs, part swapping, fabrication etc etc EXCEPT diesel powering them that is .

[Mouse]

1. What type?
I've not used the simple thrust type bearings for years as they rapidly get replaced with taper roller bearings which are my first choice.
You'll probably scream but I've used deep grove ball bearings on several bikes as have several friends as well as some of the European sidecar builders with home made front ends.

2. Average longevity.
Replaced for MOT or because steering becomes noticeably notchy and the bike starts weaving by its self.
Usually 2-3 years never any longer except the deep grove ball bearings with rubber lips. This goes for all bearings I've used including the 1200 Bandit I bought new out of the packet and essentially scrapped at about 120000+ miles. That bike was standard for most of its life and I had to replace the factory standard head stock bearings before its first MOT.


3. What lubricant,
Wheel bearing grease. Packed with as much as I can get in the bearing then a bit more.
I will only re-lubricate when I have the front end apart for other reasons or fitting new. Taking the front end off every season to repack with grease is way to much work compared to replacement when they start failing.

4. Contamination and sealing -
None at all right through to sealed lip taper bearings with a shroud - Now you bring the subject up it never makes much difference.
The grease is usually a bit manky that's all.

5. Failure mode -
I wouldn't know how to differentiate the modes of failure you describe.I also have never taken notice of upper lower differences so cant comment.
They always fail with indentations where the balls / rollers are in the straight ahead position, hence a notch that makes the bike increasingly uncontrollable and difficult to hold a straight line. Sensitivity to this varies wildly from rider to rider. The MOT test tests for slack in the bearing and notches which fits in with failure. It is also sadly hilarious that some people refuse to acknowledge that a barley perceptible notch in a steering bearing will seriously affect handling but will moan when their bike handles 'weird' or it fails a MOT and start blaming tyres, road surfaces, unleaded petrol etc for the poor handling when what it needs is £25 worth of bearing and a solid day in the shed. (From experience of being in a bike club for years and saying 'oi mate its yr bearings' and getting ignored until the MOT man forces them)
Some old bikes I have worked on have rust all over the bearings which is just long term water damage.



Also: Have you checked out Tony Foale and his book of motorcycle chassis design. He has a lot to say about geometry and handling? It is very worth while having a read if you are putting this much effort into a bike build as he covers rake, trail for various front suspension configurations.

[mark_in_manchester]

I have no experimental data in the kind of detail you're looking for - but in the ural / dnepr scene people routinely upgrade standard loose-ball races for taper-roller replacements. These are a fairly heavy bike - failure is normally associated with slackness which can't be taken up without a resulting 'notchy' feel. MZs (the other bike with which I have a lot of experience) are much lighter, and perhaps oddly use standard deep-groove ball races. One would expect them not to like the end-load, but they never seem to fail.

cheers
Mark

[tappy]

Taper rollers top and bottom, bottom one's about 55mm OD, top one's more like 45mm

Last maximum of 2 years / 15 thousand miles. Used to wear out much faster before I modified the suspension internals.
I inspect their movement when the front end gets nervous, or behaves oddly when flicking left-right or right-left, and usually find they're worn (notchy, false brinelling) and replacing them makes the bike happier. They're very sensitive to initial torque tightening on this bike compared to others I've had & I use a very delicate spring balance that I made on the bar end to set them up.

Have typically used Lithium grease, packed about 50% full. After Dieseltech's recent comment I intend to use Lith-Moly grease next time.

The old grease is usually quite dark and a bit emulsified, but never gritty or curdled. The lower seal is a thin sheet disc with a slight lip, the upper seal is just a plastic lipped affair. Neither are exactly a "seal", more of a discourager.

I understand that some manufacturers favour ball bearings as standard as these create less friction so give better handling. Personally I've always used taper-rollers in everything.