20 year-old Nishiki wheel hub rebuild

Hi!
This is my first post on the forum - pretty excited!
I'm actually a bit of a bike newbie: been cycling for 2 1/2 years only. Mostly mtn bike, but I also have a Nishiki Landau Tri-A road bike that I use for commuting and for triathlons. I bought it in 1988, but never really used it until I moved to a bike town (san luis obispo, ca). It has Shimano 105 group, and that it the start of my question...
The bike failed the pre-triathlon inspection last month because of a wobbly rear axle. I figured I might as well rebuild it. Of course, no one has ever seen a 20 y.o. 105 derailleur (anymore) and I've been getting scanty advice locally...
I used Zinn's books and my own tools - wish me luck!

- took the hub apart, cleaned and regreased everything
- repacked everything best I could

Couple of problems:

The reason the axle was wobbly, it turns out, is that the axle is held in place on one side by a screwed-in cone for the inner race; the other side as two nuts that lock into one another. The cone nut eventually worked itself loose enough for the setup to play. Shouldn't there be another nut to lock the hub securely? Is there a good 'bike reason' for locking only one side of the setup? I don't get it. Can't be weight: there's already so many spacers on that setup...

The drivetrain _seems_ to make more noise than it did before. I cleaned the heck out of everything and re-lubed the drivetrain with 'Rock-n-Roll Gold'... still, seems to be less smooth than before. For example, on the axle grease part (with the chain off), the rear wheel free spinning will stop much much sooner than the front wheel (untouched for 20 years). Is this 'normal'? Of course, I could be paranoid about all this.

I wasn't too careful about the exact position of all the spacers on the axle. I put them where I thought it made most sense: the thick aluminum ones are on the side opposite the cassette... Could this have something to do with the perceived noise?

Thanks for your advice and comments. Great forum!

Pat

I don't know what types of hubs your bike has, but you need one cone + one lock nut on EACH side of the hub ...

Does your rear wheel have an 7-or-8-speed cassette or a 7-speed freewheel?

The spacers are not arbitrarily placed on the axle ... if the wheel is centered in the dropouts when the wheel is aligned with the bike's central plane, then the wheel is 'okay' ... if it is off center to the left-or-right when the wheel is square in the dropouts, then you have to move the spacers around.

If there is something wrong with your rear derailleur, then you can replace it with ANY 8-/9-speed Shimano rear derailleur, otherwise you probably don't need to change it.

If the wheel is not spinning as freely as it did before it is because you either used a heavier grease OR you over-tightend the cones OR put one too many bearing in on one side of the hub ...

Thanks for the comments. I agree with you that there should be a lock-nut on both sides...
Now the 2 cones (bearing inner races) are not the same length. The long one is definitely threaded; I thought the short one wasn't... but that doesn't seem to make sense to me right now. I put the 2 lock-nuts on the short cone side. Maybe that's part of my problem. I'll have to take it apart and verify that. If the small cone's indeed not threaded, I'll buy another lock-nut like you suggest, that's all.

The wheel seems centered, so maybe I got lucky with the spacers guess (notwithstanding the lock-nut comment above, which will change the overall positioning of the wheel). I was wondering though if the only effect of the spacer was really to center the wheel... does it not also affect the function of the derailleur? Especially with index shifting?

To answer your question, the setup is a cassette-freehub (NOT a freewheel). It has to be one of the first ones that Shimano made too: it has only 6 cogs, and 1988 is about when they started these, no?

I wish there existed a catalog with exploded views of all the different iterations of 105s made. I guess that would be a pretty big catalog.

I like your breakdown of the possibilities for why the wheel wouldn't spin as well. I have to conclude that it's the grease then: I kept the cones as loose as possible, just removing the backlash with the skewer; and I made sure to only put 9 BBs on each side. Since I have to take the whole thing apart again, I'll be more minimalist on the grease.

Now, what about using CERAMIC BBs as a replacement? Are those even available in non-cartridge setups? Would it be worth my while on such an old bike? I'm assuming that they aren't that expensive either here.

Thanks a lot for your comments and expertise.

Pat

To over-simplify indexing, the derailleur's stops & indexing are set according to the location of the largest & smallest cogs on a wheel.

The "location" is ALMOST standardized, now ... the 8-speed Dura Ace hub had a slightly different left-right spacing than most other hubs, but it could be "corrected" by moving the thin (1mm?) spacer from the left side to the freehub side.

I don't know when Shimano introduced the Freehub ... I have a DA freehub (11t screw on last cog) which is from the same general era prior to Shimano standardizing the Freehub interface with the hub body, but I couldn't pinpoint a year. According to Sheldon Brown, it was in the late 70s, or early 80s-- Shimano Cassettes & Freehubs, by Sheldon Brown.

As far as I know, the '105' designation predates indexing. I have a '105' friction rear derailleur.

It isn't worth the extraordinary expense for ceramic ball bearings ...

BTW. Try some repacking the hubs with some WHITE LITHIUM grease (available in the automotive "section" of most stores) ... it's lighter than any of the "regular" automotive greases.

I'm not saying it's impossible, but the far more common solution would be to have ONE locknut that's tightened down against the cone, maybe with a couple of spacers inbetween. If you're missing a locknut on the other side it'd suggest to me that someone less skilled had the hub apart previously and simply messed up the reassembly.

Actually, no - although it does makes things easier and is by far the most common configuration.
As the axle has the same thread at both ends it means that w/o locknuts one cone would try to tighten and the other would try to come loose. The one that would try to tighten is the only one really needing the locknut, as the one trying to come loose w/o a locknut would be stopped by the tension from the Q/R or the wheel nut.

As usually, Sheldon Brown has a nice write-up about it.

Critical bit here: "On the left side, the axle nut will keep the left cone from loosening up. If the wheel is accidentally reversed, so that the adjustable cone is on the right, it can tighten itself up and ruin the hub."

Yes, I took the whole thing apart again, and there were 2 lock-nuts. So I reassembled it with one on each side. The long cone is on the side of the cassette (rhs) and the short (also threaded) is on the lhs. I rearranged the spacers according to the advice of ALFENG above, and it seems to work fine, although since there are axle position adjustment screws on the dropouts, there is more than one spacer arrangement that would 'look' good with the method he describes (but with closer inspection, the wheel would certainly be misaligned at all spacer arrangements other than the right one).

I found it hard to lock the cones without binding the axle bearing against the outer race. It's really a trial and error process. I appreciate the comment about the cone tightening on one side only as the bike rolls, since it's a right-hand thread on both sides of the axle. I hadn't thought of that... But really, there shouldn't be a lot of drag on the inner race, and the axle doesn't rotate, so I'm not sure that the concern would be a great one in practice. I felt comfortable locking the cones on both sides with the locknuts. We'll see...

Do you know of any website/source that shows detailed assembly and views of various hub designs? This all looked to me like a problem that would have been worked on for many decades, and I'm curious to see what the modern solution is to this. While I keep hearing how innovative Shimano has been in this field, my setup still looked far from an optimized solution...
Thanks for your help.
Pat

Well - no. Although tinkering with the axle position screws would let you set a wheel that isn't centered over the axle dead center between the chainstays up by the BB, this wheel would still not line up in plane with the front wheel. From a rideability standpoint it's highly unlikely to be critical(at least until you take your hands off the bar...) but it can shift your tire wear into an entirely different dimension.

There is a knack to it for sure. But having proper cone wrenches and access to a bench vise does make it easier. I crank down on one cone/locknut pair first, then I stick that end in a vise. That way I'm pretty much free only to have one cone/locknut pair to work with. With one wrench on the cone, one on the locknut and the axle not moving it's fairly easy to find the sweet spot.

I'm not too sure if it's drag causing this. There's a phenomena called precession that might be involved I suspect. At least it's a big player when it comes to pedals and BBs.

And I have had a few wheels lock up for me while JRA, mostly
coaster brake versions though.

As does most of us.

Sheldon Brown has some, and for more current designs there's always Shimano's website.

Well, some hubs have gone to cartridge bearings, but many still retain cup & cone bearings. Given that bearings have decently long service intervals, and that people tend to get the hang of fairly fast, there's been no pressing need to develop anything better. Besides, "more user-friendly during maintenance" lacks a certain something in terms of being a marketing hook.