Shimano 3CC hub adjustment?

[Frank Krygowski]

I'm finishing up the build of a sweet kids bike, including a NOS Shimano
3CC coaster brake hub of unknown vintage. (Thanks again to Andrew as the
source of the appropriate but rare shifter.) The bike is a very
lightweight and pretty little antique. It also has hand brakes.

The bike is in my workstand and ready to go, but I have one worry. When
turning the cranks, there seems to be excess resistance. It's not the
bottom bracket, because I overhauled that, plus the cranks feel very
free during the forward or backward motion until the drive or the brake
engage. It doesn't affect coasting; the rear wheel seems to spin very
freely, so the coaster brake is not dragging. The friction seems to be
in the drive motion itself.

(About five and ten years ago, I resurrected two other bikes with
Shimano three speed hubs and don't remember feeling this resistance. But
that was long ago.)

I've added lubricating oil via the axle's pushrod hole, but so far I
perceive no difference. I'm (understandably?) reluctant to disassemble
the hub, not only because I'd like to deliver the bike in three days. Is
there a reasonable chance it will "wear in" and loosen up? Should I
consider a solvent flush and adding oil? Or I wonder if there's a
bearing adjustment that applies to the drive mechanism, separate from
the main wheel bearings?

I've gazed at the Shimano hub manual available at Sheldon's site
https://www.sheldonbrown.com/sutherland/CB-IGH-5-shimano.pdf
but at least so far, I don't recognize a solution there.

Any tips?

--
- Frank Krygowski

[AMuzi]

Unusual problem these are so simple and reliable generally

1. Unlikely to be the problem but check hub bearing. Mount
the wheel in a vise adjustment side up. Hold the axle in
one hand and move the rim up and down in the other. Adjust
for a trace of play at the rim.

2. Maybe a tight spot in the chain adjustment? Easy to check
for that.

3. Possible frame ends not parallel and axle is flexing.
This would have to be extreme for a symptom.

4. Perhaps dirt/crud/rust in gearbox. Wouldn't hurt to undo
the adjustment enough to peek at the bearing. If rust shows,
these are simple assemblies don't be afraid of a
rebuild/clean/lube.

Do write back I'm curious to know what you find.
--
Andrew Muzi
<www.yellowjersey.org/>
Open every day since 1 April, 1971

[slto...@gmail.com]

You should know everything to look at: spin the rear wheel to check hub bearings and see how long it takes to slow to a stop.

With the chain removed spin the crank which is the probable source of the resistance. In general the old manually adjusted cranks are over-tightened.

Third source is that the chain has been adjusted too tightly.

Then of course you should look for dragging handbrakes and misalignment that causes the tire to rub against the chain stay.

[Frank Krygowski]

Well:

1) I put the wheel in an axle vise in my bench vise and played with the
wheel bearing adjustment. I didn't see any improvement, but I'll keep it
adjusted a tad looser than before, just in case.

2) No, the chain is definitely not tight. If anything it's a bit loose.

3) Dropouts look fine. I had to do quite a lot of cold setting to get
the hub into the frame in the first place, and the two 5/16 bolts with
nuts and washers that I used to test the dropout alignment were still in
the top of the relevant bolt jar. The point precisely at each other.
Probably more important, the dropouts are stamped and only 1/8" thick,
so they seem unlikely to exert serious bending loads even if they were out.

4) I pulled the left hand brake arm and the bits that came with it. All
looked greasy and black inside the hub. The grease was higher viscosity
than what I might choose, but it wasn't dried out

I noticed that with the axle in the vise, when I turn the sprocket
forward by hand, I could feel some reaction torque in the brake arm -
IOW, the arm wanted to rotate with the wheel. I put a spring scale on a
short length of chain. I measured about 3.5 inch-pounds of torque to
drive the wheel, and a reaction torque of about half that on the brake
arm. So half the drag was internal.

So I stripped the hub, following Sheldon's PDF, but I didn't bother to
pull the palls, planet gears, etc. out of the sub-assemblies. The hub
internals were entirely slathered in heavy grease. I cleaned most of it
out, washed all the components down reasonably well with WD-40 and
inspected everything. I saw no problems at all. In fact, I'm sure the
hub has never been used.

I reassembled using lighter grease on the ball bearings, heavier
moly-disulphide grease (used for CV joints) on the brake shoes, and
automatic transmission oil everywhere else. I squirted in additional oil
when it was all together.

It's all back together, and it may be slightly improved. (I didn't
bother putting the spring scale on it.) But it's still somewhat draggy.
I'm now assuming it's just the nature of the beast. Maybe it will "run
in" after some miles, although I didn't identify anything that would
make that happen.

BTW, this hub seems more picky about shifter cable adjustment than a
Sturmey-Archer. Gear #2 skips if the adjustment is even a tiny bit loose.

Anyway, I consider the project done, and it is what it is. Thanks for
the ideas.

--
- Frank Krygowski

[Mark J.]

One last minute idea, which may not apply to Shimano. I just overhauled
a Sturmey-Archer AW three-speed hub from the 70's. Re-reading Sheldon's
site (don't remember which page, of which there are many), I was
reminded that the "Ball ring," which screws into the drive-side
hubshell, has a double-track thread - i.e. two separate spirals cut in
it - so that you can initially engage the threads in two ways, which
result in final positions 180 degrees apart.

Sheldon says it matters which of the two ways you start it. Dumb me
said "what difference can it make". Now I know. With the ball ring
spun 180 degrees, the bearings don't adjust well, the sprockets "wobble"
while coasting (no big deal there), and you get some of the friction you
describe in your Shimano. Oh, and if you true the wheel that way, then
when you put the ball ring back where it belongs, the wheel has a small
but symmetric "wobble."

Now I've put everything back the right way, it's amazing the difference
(and yes, I did re-true the wheel). What I can't understand is how the
manufacturing process causes one thread engagement (of the
"double-thread") to be nice and concentric, and the other one to be
off-axis - or what else is causing these differences. If someone
understands what is really going on there, please enlighten!

Again, Shimano may not have copied the double thread, so this may not
apply to your hub. But if so, I can give first-hand testimony that it
can really matter.

Mark J.

[Frank Krygowski]

Interesting. I've rebuilt S-A hubs, but I don't remember thinking about
that double thread.

I doubt it matters with this Shimano hub, though. For one thing, I'm
convinced the hub was factory original, never used. And the instruction
PDF I got from Sheldon's site didn't mention that.

Anyway, it's kind of moot now. The kid's got the bike and he loves it.

--
- Frank Krygowski

[Mark J.]

I didn't think about it previously; I had the hub "backwards" for years
until this recent rebuild, though it didn't get ridden much in that
time, and I was always troubled in that period by several mysterious
symptoms including a bit of drag.

> I doubt it matters with this Shimano hub, though. For one thing, I'm
> convinced the hub was factory original, never used. And the instruction
> PDF I got from Sheldon's site didn't mention that.
>
> Anyway, it's kind of moot now. The kid's got the bike and he loves it.

All that matters, really.

Mark J.