Sturmey CS RK-3 stripdown - a query - now solved

[Geoff.D]

Hi,
I'm stripping a Sturmey Archer CS RK-3 hub. I've never stripped an internal hub before, let alone this particular model. I have a query and I can't find answer on the internet. I'm at the point of screwing out the planetary gear assembly from the hub body. On the internet (but applied only to other models) all I can find is that you use a punch to loosen it. There are "notches" on the circumference of the assembly, but they're crescent moon shaped, and not the rectangular, sharp edged notches that I's associate with using a punch. I'm attaching a photo of this.

My query is - Is using a punch the appropriate method in this case?

I'm assuming two things, unless anyone can correct me, (a) the assembly is a r/h thread and (b) it won't be super tight, needing only a sharp blow. So, thanks in advance for any experiences, knowledge and wisdom.

Geoff

[Airsporter1st]

There is a special tool for this, but not cheap - see eBay item #272638068093.

A punch will almost certainly upset the steel around the point of contact. If you cannot make, buy or borrow a tool, I'd suggest a brass drift might cause less damage.

[Brucey]

the steel on the ring bearing is very hard indeed; a mild steel or brass drift won't mar it appreciably.

A Brucey top tip; give it a couple of smart blows on one side, the turn the hub over and do the same thing on the other side. After a few goes at this it will unscrew. Make sure that you mark the parts so that they go back the same way; it is a RH twin start thread and every time it is assembled differently the thread in the hubshell is weakened.

IIRC Mick has the correct SA spanner for this job, purchased at some cost. I don't know of anyone else that has got one; they all use a drift of some kind, in part because there are several different patterns of spanner required to do all SA hubs.

cheers

[mjr]

Interesting. I used a round-tipped C-spanner on my hubs (an AW and a X-RD3). Am I going to hell?

[Brucey]

Interesting. I used a round-tipped C-spanner on my hubs (an AW and a X-RD3). Am I going to hell?

to hell? Hmm not sure if it qualifies as any kind of sin per se, but I would comment that you might, once the hub is well-used, have to hit the tool anyway to get the ball-ring to loosen and if the tool has only one lug there isn't that much difference then between it and a drift.

The twin start thread usually makes it easier to unscrew the ball-ring (than, say, a screw-on freewheel) but they can still become unfeasibly tight if they see high torque (typically in the middle gear; the low gear drives the other end of the hubshell and the torque in the high gear is lessened) for a long time.

cheers

[Mick F]

IIRC Mick has the correct SA spanner for this job, purchased at some cost. I don't know of anyone else that has got one; they all use a drift of some kind, in part because there are several different patterns of spanner required to do all SA hubs.

Yes, I do indeed have the correct spanner.

Yes, I know that a punch works, but it offends me, so much prefer to use the correct tool. Yes, I know that if you service different SA hubs you would need different spanners ................ but I only have one.

I mount the tool horizontally in the bench vice, and lower the rear wheel onto it. Holding it steady and giving a heave anticlockwise, the hub releases. Mine un-does easily enough even after 1000miles of Cornish hills.

This is my tool.
https://brilliantbikes.co.uk/hub-and-ax ... l-cup.html
£24 well spent IMHO.

[Geoff.D]

Thanks for the advice so far. Most informative.
I've had a preliminary attempt with a mild steel punch. It seems not to damage the hardened steel casing. But, it hasn't shifted. Maybe I don't have the confidence to give a sharp enough blow. Like Mick, I'm a little "offended" to apply this method to a precision piece. But, maybe I'm just not applying the technique skilfully enough.
I'll go back to it and give an update when successful, in one way or another.

[Mick F]

It needs some good hard blows, alternating between the two lugs.
I tried a brass drift initially, but all that did was dent the brass drift!

Then found an old steel tommy bar, and used that. I increased the weight of the hammer too, to a two pound engineering one. Keep your fingers and thumbs out of the way.
Stand the wheel upright leaning against your legs or leaning against a wall or something.
It shifts slowly but surely. Alternate the bows on the lugs. Staying with one doesn't seem to work.

The correct tool is simple and quick but you need a bench vice with soft jaws so it doesn't damage the tool's handle. You can re-fit using the tool by normal hand power as the hub tightens in use.

The ball-ring is a two-start thread and some folk would tell you to make sure you use the right one so the ring goes back rotationally where it was originally. I marked mine and the hub shell to make sure, but I'm not convinced that it would make any difference as the manufacturing is very accurate and I'm sure they don't fit them new by trial and error. They just screw 'em in and it's pot luck what rotation they are at. If it was so critical, they would mark them up when in the factory.

[Brucey]

Mine un-does easily enough even after 1000miles of Cornish hills....

ball ring only sees tightening loads from riding in the middle and high gears. Hills in low gear don't tighten the ball ring. Small wheels give the hub an easy time of it too.

The ball-ring is a two-start thread and some folk would tell you to make sure you use the right one so the ring goes back rotationally where it was originally. I marked mine and the hub shell to make sure, but I'm not convinced that it would make any difference as the manufacturing is very accurate and I'm sure they don't fit them new by trial and error. They just screw 'em in and it's pot luck what rotation they are at. If it was so critical, they would mark them up when in the factory.

The twin start thread is not (and has never been) made perfectly on either part. In the factory they assemble randomly and the result is that the hubshell (which is always softer than the ball ring, whether steel or aluminium) yields and conforms to the ball-ring threads. If you ever tighten the ball-ring in the other thread start (so it finishes 180 degrees about) then the hubshell yields again, and so forth. You don't get very many goes at this before the hubshell threads can fail. This goes double for aluminium hubshells.

If you want to prove to yourself that there is a difference, and/or you lose track of which thread start you were on, (or you are installing a replacement ball ring into a hubshell for the first time) try screwing in the ball-ring both ways (by hand). There is usually about 1/20th of a turn difference in the timing of the ball ring (hand tight) between the two thread starts. This means that there is probably about a 0.001" variation between the thread pitch and the thread spacing; so small you would never see it but large enough to definitely cause problems with repeated swapping of thread start position.

Anyway whichever thread start allows the ball ring to go on furthest (by hand) is the correct one to use.

cheers

[Mick F]

Wise words.

Yes, you must be right about small wheels giving an easier time.
What I need to do, is to experiment on a 700c wheel ..................... but there's more leverage to unscrew it with of course, so it could be swings and roundabouts.
Not as tight with 406 - but much more difficult to undo.

When my 406 wheel shifts, it goes with a bang and I need gloves on to protect my fingers from the vice. You only have nine inches of turning force available.

[Geoff.D]

The ring has shifted, using a punch. The secret was to apply a larger hammer ( thanks for the tip, Mick). A couple of strikes, even without much input from me, did the trick.

Thanks for the help everybody. Onwards I go into the depths of an internal hub, as a complete virgin (no funny moments, please ). I may be back for more advice later.

[Mick F]

First time I took mine apart, I photographed it as I went.
Hardest part - for me anyway - was getting the tiny pawl springs and pawls in the right way.
These are correct, and you can see the tiny thin springs and the shape of the pawls.

[Brucey]

The ring has shifted, using a punch. The secret was to apply a larger hammer. A couple of strikes, even without much input from me, did the trick....

It should be mentioned that the load from the hammer blow is reacted by two things;

1) the mass of the wheel and
2) through the wheel to whatever it is sitting on

In the latter case it matters (a lot) if the wheel has a tyre on it and indeed what the pressure in that tyre is. If the tyre is soft then a good portion of the hammer blow may be absorbed in the tyre. If the wheel has no tyre on it you risk damaging the rim if it is resting on a hard surface.

The weight of the hammer is significant; I think you get a more effective impulse to the hub up to head weights that about equal the hub weight, provided the spoking is stiff enough to stop the hub from twisting about its centre of mass (*). But the possibility of collateral damage is always greater with a heavier hammer too.

[(*) the worst hubs to dismantle are ones in wheels where all the spokes are slack; it is difficult to support the hub so that it doesn't just deflect and allow the energy of the hammer blow to be absorbed in the spoking.]

So I favour using a short handled hammer with a head weight of about 1-1/2 lbs, (and a force that isn't excessive, about that which might drive a 2" nail into softwood, about 1/2" at a time) with the tyre still on the wheel, inflated hard. Because (I think) the elastic energy is stored and transmitted differently in a drift of different length, up to a point a longer drift is better than a shorter one. My favourite one is about 10" long and ~3/8" diameter with the end ground to fit the hub fairly well.

[Note that if the hub is not in a wheel, you can loosen the ball ring by supporting the hub (on some softwood so that it isn't damaged) with your foot bearing on it. The hub will tend to rotate with every hammer blow (and there will never be a lot of friction this way), so the main reaction comes from the moment of inertia of the hub about its axle. This being the case a lighter hammer (3/4 to 1 lb head) and a swifter blow seems to work better.]

BTW If you have tried and failed already, simply by changing sides (accidentally or otherwise) the next blow may loosen the ball ring whether it is different or not. For this reason only by disassembling multiple hubs are you likely to reliably find out what works best or not.

cheers

[Mick F]

................ I do indeed have the correct spanner.

Yes, I know that a punch works, but it offends me, so much prefer to use the correct tool. Yes, I know that if you service different SA hubs you would need different spanners ................ but I only have one.

I mount the tool horizontally in the bench vice, and lower the rear wheel onto it. Holding it steady and giving a heave anticlockwise, the hub releases. Mine un-does easily enough even after 1000miles of Cornish hills.

Well ..............
I was doing some cleaning and maintenance, and had the sprockets off, so I thought I'd pull the hub out and have a look-see.

After struggling for ten minutes, I had to resort to the hammer and punch method.
Even then, it wasn't easy.

1,700miles since it was last out, and it was solidly in there. Looking at the marks where I'd marked the ring and the hub when tight, the ring had gone round maybe 2mm (circumferentially) further than it's been. Try as I might, the tool in the vice method wouldn't work due to me not being strong enough to turn the 406 wheel. It held well enough in the tool, but the strength to turn it eluded me.

[Brucey]

sorry to hear of that, (it must seem like a PITA given that you bothered to buy the tool) but it does tally somewhat better with my experience of these hubs.

Doubtless you could convert your spanner to a 'slogging spanner' (i.e. hit it with a hammer, ahem...) but I'm not sure it would survive that long given that treatment.

cheers