Sturmey NIG 3s hub; what to look out for, what breaks.

For discussions about bikes and equipment.
mig
Posts: 2702
Joined: 19 Oct 2011, 9:39pm

Re: Sturmey NIG 3s hub; what to look out for, what breaks.

Post by mig »

i like the idea of these hubs. i've only ever used one as a child though when my bike had a flat tyre and i borrowed my sister's bike for the day to get to the football pitch in the summer.

when i look at the exploded diagrams to my mind's eye they look very complicated parts wise to shift between three gears. i'm just plain wrong with that idea i suppose? would a 'modern' take on such a hub take a similar approach?
Brucey
Posts: 44517
Joined: 4 Jan 2012, 6:25pm

Re: Sturmey NIG 3s hub; what to look out for, what breaks.

Post by Brucey »

regarding complexity, all three speed hubs are fairly simple devices, and complexity is not usually really present, only poor presentation I suppose. In the early SA brochures they were at pains to make the gear look simple, hence they often illustrated the gear broken down into subassemblies, e.g.

Image
but the exact function of these subassemblies isn't always clear.

There are exceptions ( the SA 'N' type is one such, the S3X is another; a clue with both is that there are more gears in them than is strictly necessary) but nearly all 3s hubs are trying to do the same thing which is exploit the relative difference in speed of the planet cage and ring gear (of a single gear train) to achieve the increase or reduction in gear ratio. For example the ring gear is always going 4/3 as fast as the planet cage in an AW gear, so the ratios available are 4/3, 1 or 3/4. The input to the hub is always via the 'driver' and the output to the hub is always to the hubshell, but can either can be routed via the ring gear or the planet cage. 3s hubs usually have two separate dog rings (which can be driven by pawls), either of which can be used to drive the hubshell. Below I have compiled a chart (for an AW hub in fact) but it tells you what is happening in most 3s hubs;

Gear...……..Input...……...Output...……..Ratio

Low...……...Ring Gear......Planet cage.....3/4
Middle......Ring gear...….Ring gear...…....1
High...…….Planet cage....Ring gear...…..4/3

Logically you could have a 3s gear where the middle gear uses drive to and from the planet cage instead of the ring gear, and I think there are one or two which are like that. But most work as per the chart.

The obvious moving parts in the hub to make the shift usually comprise some kind of moving clutch. The rest of the gubbins is there to make sure that the hub shifts smoothly, and/or the drive is taken from the correct part of the gear mechanism to the hubshell.

In a SA AW hub anytime the hubshell is driven from the ring gear, the hubshell is going faster than the planet cage, which means that the Low gear pawls (built into the left end of the planet cage) are overrun. Hence in SA AW gears the low gear pawls make a ticking sound in gear 2 and gear 3. In gear 1 the hubshell isn't driven by the high gear pawls (which are built into the ring gear) only because they are retracted. In the SA AW (both NIG and pre NIG types), the sliding clutch interferes with the back of the high gear pawls (which pivot about their centre, roughly) and this causes them to be retracted, so they cannot drive.

Jobst Brandt would merrily tell folk that he let his children play with SA AW hub innards, as a mechanical toy, and that they very soon were able to see how the thing worked as well as disassemble and reassemble one correctly. Out of curiosity I 'self started' myself this way by the time I was ten years old or something. But not everyone's brain is wired up the same way, so not everyone gets to grips with such things as quickly/easily. Nonetheless I recommend fiddling with the innards of an old AW to anyone who is interested.

Other hub gears are less obvious, but all are tractable given sufficient application of grey matter.

cheers
-------------------------------------------------------------------------------
~~~~~~~~~~~~~~~~~~~~~~Brucey~~~~~~~~~~~~~~~~~~~~~~~~
Tiberius
Posts: 799
Joined: 31 Dec 2014, 8:45am
Location: North East England

Re: Sturmey NIG 3s hub; what to look out for, what breaks.

Post by Tiberius »

mig wrote:when i look at the exploded diagrams to my mind's eye they look very complicated parts wise to shift between three gears. i'm just plain wrong with that idea i suppose?


I would agree that the exploded parts lists and diagrams, though accurate, do make the hubs look very complicated. But, if you are reasonably competent with mechanical things, then an hour or two with one of these things in your hands, the penny drops soon enough.

The first time that I pulled one apart I had no idea how they worked. I pulled it apart 'parrot fashion' I simply laid the components out in order and then put them all back together again and checked that it all still worked. Then I pulled it apart again and put more thought into exactly how it worked. What an ingenius thing !!

I now love these hubs. They are simple enough and so reliable needing very little maintenance. I really like the fact that they can be stripped right down (unlike most things today) and that parts are readily available and cheap enough.

They're no Rohloff, but then they aren't over a grand each either.
Brucey
Posts: 44517
Joined: 4 Jan 2012, 6:25pm

Re: Sturmey NIG 3s hub; what to look out for, what breaks.

Post by Brucey »

today I opened up a NIG AW hub; it had been used for long enough to cause a fair amount of wear on a sputnik rim, as far as I could tell, either that or it had been built on a worn rim to start with.

This is what I found inside

minced pawls, anyone?
minced pawls, anyone?


one of the two high gear pawls was completely smashed. The other was almost intact. I only found the remains of one 'R' spring, and the pieces were gathered adjacent to the near-intact pawl. It isn't possible to say for sure what happened first but the near intact pawl was probably lacking its spring at some stage and the other pawl smashed itself to pieces, presumably via multiple slippage events under load in second and third gears. Part of the broken pawl was jammed hard into a slot in the ring gear, and smaller pieces were strewn throughout the hub. All the lubricant was contaminated with bits of minced up pawl fragments; the larger pieces (as seen in the photo) were fished out with a magnet.

The following items were obviously damaged ;

1) the splines on the sliding clutch
2) the ribs inside the ring gear
3) the high gear pawls
4) the high gear pawl 'R' springs
5) the actuator plate
6) the RH cone

Possibly there were other damaged parts too, but it matters little; this internal will never be 'perfect' again because there are too many parts that have suffered some damage. However many of the damaged parts will still work, so the fate of this hub is to be rebuilt using as few new parts as possible and then probably be subjected to 'hack bike' use, or some kind of experimental modification. So the RH cone (which is marked but not very badly damaged) will be swapped with the left , where it will have an easier life, the actuator plate (which was practically falling off the driver and would have broken the hub soon enough if nothing else did) has been checked for cracks, reshaped and refitted, a bit of cleaning and deburring of parts, lockwasher for the RH cone fitted, replacement high gear pawls and R springs, and it'll go again, for a while at least.

I also recycled a couple of pre-NIG AW hubs today. One was 36 years old and the other was 40 years old. Both were sat in wheels that were rusting away around them, but I don't think either hub is faulty; if more than new R springs are required (eg because of corrosion) I shall be surprised.

[edit; I have not found any date marks on the hub in this thread; all I know is that it is a grey painted one from 2003 or later. If there is a dot matrix date mark on the hubshell (which is the usual thing for SA hubs from Taiwan) I have not found it. However I note that such marks don't last that well and can be removed with common solvents, so there might have been one which has come off. The actuator plate doesn't have a '2' on it which leads me to suppose that the hub is at least ten years old, if I'm right about the significance of the '2' on newer actuator plates. I also note that the looseness of the actuator plate I noted may have been caused by fragments of pawl in the hub.]

cheers
Last edited by Brucey on 23 Mar 2020, 10:40am, edited 1 time in total.
-------------------------------------------------------------------------------
~~~~~~~~~~~~~~~~~~~~~~Brucey~~~~~~~~~~~~~~~~~~~~~~~~
Greystoke
Posts: 482
Joined: 8 May 2018, 7:41am
Location: Lincolnshire

Re: Sturmey NIG 3s hub; what to look out for, what breaks.

Post by Greystoke »

I've read this topic with interest.
So in summary a properly adjusted non-NIG hub is a better more reliable bet.
Brucey
Posts: 44517
Joined: 4 Jan 2012, 6:25pm

Re: Sturmey NIG 3s hub; what to look out for, what breaks.

Post by Brucey »

Greystoke wrote:I've read this topic with interest.
So in summary a properly adjusted non-NIG hub is a better more reliable bet.


well sort of; a 'properly adjusted' NIG 3s hub is not a bad hub per se, just not quite as unlikely as a pre-NIG one to develop problems.

The biggest difference between them is that when (not if) the hub adjustment is poor, the pre-NIG hub signals that fact by slipping in a very obvious fashion, so it usually won't get ridden far in that state. By contrast if the hub still has some drive rather than none (which is pretty much the whole point of NIG) there are folk who will carry on trying to ride the bike, and turn a deaf ear to any horrible sounds coming from the hub.

[ IME experience when a pre-NIG hub is sufficiently far out of adjustment that when you select '1' you are part way into '2' (which is what usually smashes the high gear pawls up), it has been slipping in 2, really badly, for ages. I have seen pre-NIG hubs with smashed high gear pawls, but only very rarely (two or three hubs out of many hundreds). It usually takes real neglect and a total absence of mechanical sensitivity to break the pawls like that, just moreso with the pre NIG than the NIG type.]

The hub blow-up above is a case in point; that the hub started to fail at all is a disappointment of course and there may have been one part that was substandard in some way. However it is just as (if not more) probable that the adjustment was bad and that this precipitated the demise of the hub. What is clear is that the hub must have been ridden for some time after it started to fail; the way minced metal was distributed, the collateral damage, and the rounded edges of the pawl fragments is testimony to that.

So with the NIG hub above, the question 'death by natural causes' or 'murder'? is impossible to answer but the answer is I suspect 'murder'; given that the corpse has been given a pretty good kicking, this seems pretty likely.

I think part of the problem may be cultural as much as anything else. When I was young (and this applies even moreso for those older than me) folk were interested in engineering, often for its own sake. We were still in the throes of 'the machine age' if you like and 'the computer age' had hardly begun. But since then any appreciation for the subtlety of engineering is less likely to be part of someone's make up; the part of their mind which might be reserved for such things is more likely to be devoted to a keen understanding of computers and/or how their smartphones work. Things like gearboxes are likely to be considered as 'inexpensive old tech' and are to be neglected and/or abused, like any other part of the 'inexpensive and old tech' bicycle itself. [A similar analogy is the wristwatch; only the most expensive wristwatches (cf Rohloff hubs) get care and attention; the rest are just used/abused until they die, and are then replaced.]

The SA AW hub is 'the cheap wristwatch' of the bicycle world; it needs to be super-tough to survive.

cheers
-------------------------------------------------------------------------------
~~~~~~~~~~~~~~~~~~~~~~Brucey~~~~~~~~~~~~~~~~~~~~~~~~
Greystoke
Posts: 482
Joined: 8 May 2018, 7:41am
Location: Lincolnshire

Re: Sturmey NIG 3s hub; what to look out for, what breaks.

Post by Greystoke »

I agree with your comments Brucey. I'm told we don't need engineers like me anymore as they want digital engineers now.....that is until something breaks and I fix it
mig
Posts: 2702
Joined: 19 Oct 2011, 9:39pm

Re: Sturmey NIG 3s hub; what to look out for, what breaks.

Post by mig »

are any undue stresses put through the internals if just one gear is almost continually used?

also, rohloff hubs, how is the £1k price tag justified? or isn't it. i can remember seeing an ad for one many years ago and thinking that they looked interesting. upon seeing the price i simply couldn't believe it :D
hoogerbooger
Posts: 673
Joined: 14 Jun 2009, 11:27am
Location: In Wales

Re: Sturmey NIG 3s hub; what to look out for, what breaks.

Post by hoogerbooger »

if you want to you can fit a pre-NIG internal into a SRF3 hubshell. This also gives an easy way of mounting two sprockets if you want to do this, since the pre NIG driver has longer sprocket mounting splines than the standard NIG driver. You can probably obtain a (used) pre-NIG hub complete, overhaul it, and buy a revised axle, all for less than the cost of a BWR driver alone.


Looks like this isn't an option for a Brompton 6 speed. ( or at least with off the shelf parts/easy mods)

I have obtained a cheap AW and the driver has circa 6.5mm for sprockets and spacers. On a BWR there's circa 8mm available: there's circa 3.3mm between the two sprocket + 2x2mm for the sprockets and the rest is the plastic chain guide which sits on diver driver, turning with the sprockets.

The smallest dished sprocket I can find is a 16t. However it doesn't help make space as the dished recess is not wide enough to clear the ball ring cover or by the looks of it the plastic chain guide. So that seems to leave flat sprockets and losing width as the only option.

The plastic chain guide I presume is important to fit on a brompton. It's aperture could be enlarged to jam it on the ball ring cover instead of being seated on the driver, but that leaves 0.8-1mm mm to loose elsewhere. Could possibly grind a chamfer on the circlip side of the small sprocket, but that seems rather trial and error to get it even and I can't see it losing enough. Could narrow the sprockets on their outsides perhaps, but again not easy to do evenly, or narrow the spacers, but presumably can't lose too much clearance between sprockets without affecting function.

Does anyone know if the spacing between the sprocket can be reduced and what the minimum workable gap might be/is ?

Bright ideas welcome ( If none I shall start with reducing spacer width)
old fangled
Brucey
Posts: 44517
Joined: 4 Jan 2012, 6:25pm

Re: Sturmey NIG 3s hub; what to look out for, what breaks.

Post by Brucey »

this is a Brompton-specific problem (because of the size of the sprockets) so it probably ought to be in another thread. But the plastic dustcover which goes behind the sprockets on an SRF3 (and the last of the pre-NIG AWs) is wider than the (steel) part that goes onto an older version of a pre-NIG AW. You need to use the steel dustcover when fitting two sprockets to a pre-NIG AW driver. With the steel dustcover fitted you can fit two 1/8" thickness sprockets. The plastic guard fitted to Brompton hubs is normally meant to stop the chain from getting snarled up in the spokes, should it get unshipped (always a risk I suppose with a tensioner fitted on a folding bike). However I think you can do without it.

Since you are presumably wanting to use 3/32" sprockets, these can be modified by grinding to make them about 3/32" thickness, (because they are normally joggled so that they are nearly 1/8" thickness). This gives you another 1/16" of an inch to play with in terms of driver length. I have not tried to use a an offset 16T against a steel dustcover so I don't know if/how much of a clash there is.

Image

In terms of sprocket spacing if you are using 8s chain then you can go down to about 3.0 or even 2.8mm between the sprockets. However I don't think you can use a flat 13T (or 12T) up against a dished sprocket anyway; to fit two sprockets on an AW driver they need be large enough that one clears the dish on the other.

I have not done this but I think you ought to be able to modify some cassette sprockets to fit an AW driver. You would need to grind the splines inside each sprocket to fit the three lug format, and the length would only work if you used a #2 position sprocket from the cassette as the small sprocket; such sprockets have a built in spacer and also (crucially) have a recess in if the #1 sprocket is an 11T or 12T.

The length ought to work out at
~0.5mm for steel dustcover
1.9mm for 16T sprocket
4.85mm for #2 sprocket with integral spacer
- ~ 1.0mm for recess in #2 sprocket

= total of 6.25mm.

It is also possible to grind the driver slightly so that the snap ring sits a touch further out, but only if the rightmost sprocket uses an inbuilt (splined) spacer; you run out of driver spline otherwise. As it is there is a tolerance of about 0.5mm on the assembled thickness of the sprockets and spacers before the snap ring won't hold properly. If you need extra spacers they can be had in various thicknesses for cassette hubs, down to 0.5mm, so you should always be able to accommodate any small mismatch.

cheers
-------------------------------------------------------------------------------
~~~~~~~~~~~~~~~~~~~~~~Brucey~~~~~~~~~~~~~~~~~~~~~~~~
Brucey
Posts: 44517
Joined: 4 Jan 2012, 6:25pm

Re: Sturmey NIG 3s hub; what to look out for, what breaks.

Post by Brucey »

mig wrote:are any undue stresses put through the internals if just one gear is almost continually used?


if that gear is the middle gear then the mechanism is just coasting round and the efficiency is much like any other freewheel, less whatever parasitic drag there might be in the mechanism. If the high gear or the low gear are used then eventually the teeth on the gears will wear slightly on one side more than the other. But until many tens of thousands of miles have been done, it is more like polishing of the surfaces than 'wear' per se, provided the hub is well lubricated. Using the high or low gear imposes a torsional stress on the axle. Eventually the axle can fatigue and break, but it takes a very long time; I've broken about a dozen axles in hubs for freewheels, but (having done at least as many miles on it as on all freewheels I've owned combined) I've personally broken just one IGH axle.

also, rohloff hubs, how is the £1k price tag justified? or isn't it. i can remember seeing an ad for one many years ago and thinking that they looked interesting. upon seeing the price i simply couldn't believe it :D


well, it is made in about half the volume as many other hubs, it is twice as complicated, and it is better made. Each of which could be expected in turn to roughly double the cost. So about eight times the cost of (say) a Nexus 7 seems about right, on that basis.

cheers
-------------------------------------------------------------------------------
~~~~~~~~~~~~~~~~~~~~~~Brucey~~~~~~~~~~~~~~~~~~~~~~~~
Brucey
Posts: 44517
Joined: 4 Jan 2012, 6:25pm

Re: Sturmey NIG 3s hub; what to look out for, what breaks.

Post by Brucey »

FWIW today, I dismantled two NIG-type AW hubs, of the Taiwanese variety. Both were OK inside. I don't think either of them had done that many miles, but both had outlived the wheels they had been built into.

I did the usual checks and adjustments; adding the RH cone locking washer, and adjusting the fit of the actuator plate.

I did note one thing I've previously omitted to mention; there has been a small design change to the planet cage. This appears to have occurred between 2003 and 2010. The planet pinion pin ends are exposed at the left side in the early version of the planet cage but enclosed in the later version. This appears to have had the effect of reducing the loss of grease from the planet gears; in the AW3 hub upthread, whatever grease had been on the planet gears had seemingly escaped (via the pinion pin drillings) and formed large, useless clods inside the planet cage. In the later version of the planet cage this doesn't seem to happen in the same way.

cheers
-------------------------------------------------------------------------------
~~~~~~~~~~~~~~~~~~~~~~Brucey~~~~~~~~~~~~~~~~~~~~~~~~
Alloldiron
Posts: 3
Joined: 3 Feb 2023, 9:13am

Re: Sturmey NIG 3s hub; what to look out for, what breaks.

Post by Alloldiron »

Hope this thread is still active as I'm searching for information on repairing one of these hubs (the one in your picture on page 1, steel shell with red and yellow bands and no stampings).
I'm not good with gearboxes of any kind but I can usually get Nottingham SA hubs working again - got lots of SAs FM, AM, AW, S5 even a TF - but this type of AW is new to me and I can't find a pdf exploded view and parts list to see what's missing.
Can someone help or have photos of one stripped down for comparison?
All the internals appear to be in good condition but there's one broken pawl and the actuator plate is missing (if it should have one) - someone's been inside before me so I'm not sure I've got all the bits.
Thanks, Tony.
hoogerbooger
Posts: 673
Joined: 14 Jun 2009, 11:27am
Location: In Wales

Re: Sturmey NIG 3s hub; what to look out for, what breaks.

Post by hoogerbooger »

Think this is it. On Sturmey Archer Heritage site:

http://www.sturmey-archerheritage.com/i ... il&id=1145
old fangled
Alloldiron
Posts: 3
Joined: 3 Feb 2023, 9:13am

Re: Sturmey NIG 3s hub; what to look out for, what breaks.

Post by Alloldiron »

Brilliant! :D Thanks for that hoogerbooger, that's definitely the same internals as mine although mine has the steel shell.
Looks like all I need is pawls for the driver and an actuator plate.

Thanks again,
Tony
Post Reply