Search found 44422 matches

by Brucey
2 Apr 2024, 4:12pm
Forum: Bikes & Bits – Technical section
Topic: save your sprockets!
Replies: 0
Views: 23

save your sprockets!

a very long time ago (in the 1980s) I dismantled my worn freewheel and de-hooked every single tooth on every sprocket. It took what seemed like ages, even though I was using an angle grinder. I did this in good part because I could no longer buy new sprockets to fit my favourite type of freewheel. 'How hard can it be?' I must have thought. Of course the correct answer to this is 'very hard indeed'. De-hooking rendered the sprockets able to run OK under light load but skipping was still evident when I put my clog down. At this point I realised I was probably wasting my time and I might even have gone off in something of a sulk.

It didn't take me very long to work out that the real problem (for derailleur sprockets) was that each tooth was worn such that there was a ramp leading up to it, and that if I waited long enough, and did enough miles under light load, I might no longer have a few um of material in the wrong place and maybe it wouldn't skip under load any more. This requires that the wear is in an unusual place, and although it does work eventually, it is enough to try the patience of a saint. So I reluctantly concluded that I was indeed wasting my time and bought new sprockets just like everybody else.

However it rankled with me that I couldn't do any better than the sprocket manufacturers managed via pressing, which is not thought of as terribly accurate way to make anything. Unsurprisingly perhaps, 40 years of sulking gets you precisely nowhere, so it took me a very long time before I even reconsidered my earlier conclusion. So what has changed in the last 40 years? At first sight, not a lot; LBS scrap bins are still full of lightly worn,but supposedly useless sprockets and plenty of folk have concluded (much as I did) that it is not possible to reverse the effects of 'the wrong kind of wear'.However I now think I might have been wrong.

As someone once said 'there's no point in getting older if you don't also get smarter'. One of the things that has happened in the last 40 years is that dremel tools have become widespread. Also, my understanding of chains has improved somewhat. Between these two things, there has been enough alteration to the circumstances for me to have changed my mind. So I now firmly believe that it is possible to reverse the effects of 'the wrong kind of wear' in most cases, if you go about it in the right way.

Galingly perhaps, it is generally only one or two sprockets that skip under load, and if they are reprofiled accurately the cassette or freewheel can usually be saved. For the longest time I considered the work to be too difficult to do with sufficient accuracy. However, finally, the scales fell from my eyes and I realised it would not be so difficult after all. The first realisation was that the tooth wear was usually much more accurate than how the sprockets were made in the first place. The second realisation was that most sprockets exhibited little or no wear in the bottom of the 'valleys'. This meant that there were two points of reference built into every sprocket upon which a new tooth profile could be based.

The simplest method is probably the 'masking' method. In this, a 'machined all over' track sprocket or other 'known good' sprocket of the same size is offered up to the sprocket you are working on. Then it is clamped in place so that they lie flat-to-flat and the worn teeth are carefully aligned so that no part of the wear is not easily visible when looking between the teeth of the 'known good' sprocket. Then it is just a question of spraying the cogs with a little paint, so as to use the 'known good' sprocket as a mask, allowing the new profile to be clearly marked on the other sprocket. The other sprocket can then be accurately reprofiled using a dremel tool. However, there are two main problems with this; first you don't really know what degree of sprocket undersizing was used originally and second it is not clear what the exact roller size in the new chain might be.

For many years the maximum roller size for cycle chain has been 7.77mm. This meant that most aftermarket/third party chains had rollers around 7.66mm and this is usually close enough. However, in recent years SRAM have muddied the waters slightly. Some of their chains use rollers which are very slightly larger. The difference is very small, so that if there is the slightest wear in a 'standard' part, it'll probably be fine with new sram chain, big rollers or no So depending on which chain you choose, the roller size could be between 7.65 and 7.95mm. If there is clearance to spare, this may not matter. However in a reprofiled part, it might. So ideally you would need to know the roller size of the new chain, but in practice it may be sufficient to assume a given roller size and blend the freshly ground area into the tooth root profile, depending on the degree of undersizing used originally,.

As ever, it is possible to make a special three-legged gauge tool which can help you. The idea here is that the nature of the contact of the three legged tool with the sprocket will tell you if the sprocket will take a new chain or not. But in this case the tool needs to be extraordinarily accurate to be of any real use. The tool is simple enough to make but difficult to do well, because it (the three-legged tool) needs to be so precise.. The first step is to machine the heads of three M5 capheads so that they match the roller size of your new chain. The next step is to take a CD-sized piece of ~1/4" plate (steel. Alu, or brass) and to centre drill it and then to mark it up accurately for later drilling. Mark radial lines at precisely 120 deg. intervals. Then use something to mark out where the PCD of the chain rollers in a 'perfect' sprocket mesh ought to lie for various sizes of sprocket.

Some of the sprocket sizes (eg n=12,15,18, 21 etc) are multiples of three; so where the PCD line crosses the 120 degree lines can be marked for drilling. This should leave a line up each 120 degree mark, comprising centres for drilled holes at about 6mm intervals. All the other sprocket sizes are either n+1 or n-1. The first hole for each can be can be marked ~1/2" either side of one of the 120 degree lines, then the second and third hole centres for each sprocket size.can be can be marked out too. Note that the spacing for all n+1 or n-1 legs is uneven, so for n+1 sprocket sizes the second/third legs can still be set n/3 away from the first leg (just like for the neighbouring 'n' drillings) but the second/third legs are now set n/3 + 1 apart. Slightly confusingly, the n/3 spacing for n+1 sprockets is actually about 1/6" less than it is for the smaller 'n' sized sprocket. Similarly, the n/3 spacing in n-1 sprockets is about 1/6" larger than it is for the larger 'n' sized sprocket. All the marked positions should be drilled and tapped M5 and the three capheads installed on the PCD of interest. Anyway, with the gauge you should be able to tell if a new chain will skip or not.

.However, this still leaves the issue of undersizing to contend with. Now I believe that all derailleur sprockets must start out slightly undersize, else it would cause all kinds of running problems, especially once everything is a bit grimy. If your sprockets start out undersized then not only will they tend to wear faster, but they will also tolerate more wear, too. Perhaps the gauge design needs to better allow for the likely sprocket/roller undersizing, but then again, it might be sufficient to use the gauge with (say) a cigarette paper between it and the sprocket, in lieu of running clearance/undersizing.

The manufacture of a suitable gauge would be pretty straightforward using a half-decent milling machine equipped with a dividing head. However, with just hand tools to use, I would regard the same task as 'extremely challenging', but not impossible. I suspect that the trick will be to first drill small pilot holes at each location, which will (almost inevitably) sometimes be in the wrong place slightly. Next the pilot holes could be filed out a bit larger where necessary and in the process recentred. Then a centre drill could be used in each modified hole, such that the pilot on the drill will pick up on the new diameter you have just filed. Having done this, any subsequent machining operations should find the same centre again automatically. This way, even if the original marking out was imperfect, it can be corrected in most cases.

The other sneaky thing you could do is to use a 'perfect' sprocket as a drilling jig; a brass bush would be made, such that it would have an ID to match the drill and an OD to match the roller size. This way it ought to be possible to make a gauge that matches your sprockets exactly, even if they are a bit undersized.
by Brucey
1 Apr 2024, 10:15am
Forum: Bikes & Bits – Technical section
Topic: Dynamo carbon fork mounting options
Replies: 24
Views: 1080

Re: Dynamo carbon fork mounting options

nomm wrote: 30 Mar 2024, 10:11pm According to SJS they get asked this constantly and no one makes a solution
-which is why I think you should make something. My 'reach-around' design should work well and be easy to make.
by Brucey
1 Apr 2024, 9:53am
Forum: Bikes & Bits – Technical section
Topic: Noisy freehub - quieten with grease?
Replies: 42
Views: 1963

Re: Noisy freehub - quieten with grease?

anywhere on the hub body will probably do. Injection of solvent will allow in-situ cleaning, Later injection of SBG (solvent-borne grease) via the same port will provide relubrication for all the bearings. I have designed all that is required to make this swift, easy and economic to do
by Brucey
31 Mar 2024, 6:03pm
Forum: Bikes & Bits – Technical section
Topic: Noisy freehub - quieten with grease?
Replies: 42
Views: 1963

Re: Noisy freehub - quieten with grease?

note that you only need to disassemble a freehub if you need to change something like the pawls, pawl springs, balls etc.; re-shimming merely requires that the dust seal is removed and one part is unscrewed, so it doesn't take too long, and nor does it require a particularly clean and tidy workshop, either. Cleaning and relubrication does not usually require any disassembly, since the unit can be flushed with solvent and relubricated in a number of different ways.

BTW if you remove one of the shims, then finger-tighten the adjusting race onto the balls, provided you know where you started from, this can tell you exactly how thick the bed of shims needs to be (in relation to whatever you just removed), or how much thinner the adjusting race needs to be. The final position of the adjusting race needs to be a little bit CW from the 'finger-tight' position, but as long as that is understood the adjusting race can be lapped.or re-shimmed in accordance with the screw thread pitch. Effectively, this arrangement contains its own built-in micrometer for use when setting up, and just for once you don't need to worry about backlash or clearance in the screw threads, because the loading is always in the same direction.
by Brucey
31 Mar 2024, 4:57pm
Forum: Bikes & Bits – Technical section
Topic: disc brake caliper rubbing against spokes
Replies: 29
Views: 5743

Re: disc brake caliper rubbing against spokes

Cyckelgalen wrote: 31 Mar 2024, 1:42pm.....Am I correct in thinking that a larger pad, with larger friction area, will provide better braking?....
only if the smaller pad routinely overheats. The larger pads ought to run cooler and last a bit longer too, but if you really want a hub brake where the friction material lasts well, get a drum brake.
by Brucey
31 Mar 2024, 2:14pm
Forum: Bikes & Bits – Technical section
Topic: How undersize are your sprockets?
Replies: 3
Views: 448

Re: How undersize are your sprockets?

thank you again for that. I don.t think that undersize rollers are part of the solution; most commonly standards for roller chains are very
prescriptive when it comes to max roller size, but rather less so for minimum roller size. Chains with slightly undersized rollers or sloppy bushings are relatively commonplace, since .they will seem to fit and run on a wider variety of sprockets. However, the penalty for this (just as with undersized sprockets) is increased noise, vibration and harshness (NVH). Not to mention faster wear.


Your recumbent sprockets may have been made with an unusually short pitch. This may be a subtle difference between MTB and 'road' stuff from the same maker since a slightly shorter pitch in the sprockets also improves tolerance to mud.. SRAM also use a slightly larger roller size in some of their equipment, so one way or another the rollers would likely appear to be undersized wrt the sprocket. Those sprockets were also plated, and on your favourite sprocket the plating was worn through. The rollers in the chain might have actually been undersized too, but I didn't really appreciate the potential significance of this at that the time, so I didn't really check this carefully. The chain did measure the correct pitch though.
by Brucey
31 Mar 2024, 1:32pm
Forum: Bikes & Bits – Technical section
Topic: 105 5700 (5703) triple shifter problem
Replies: 7
Views: 362

Re: 105 5700 (5703) triple shifter problem

PT1029 wrote: 31 Mar 2024, 7:11am
*I recall seeing instructions for 2 different routings for the same gear, depending how far offset to the left (of the gear) the BB cable guide was. I forget which exact model this was for.
IIRC more than one shimano mech was this way. The first generation 'longarm' 11s mechs were like this, plus some others eg 10s x 2 tiagra possibly
by Brucey
31 Mar 2024, 11:52am
Forum: Bikes & Bits – Technical section
Topic: 105 5700 (5703) triple shifter problem
Replies: 7
Views: 362

Re: 105 5700 (5703) triple shifter problem

there is usually a spring loaded pawl connected to the downshift lever; it sounds very much like your pawl has gummed up. The downshift lever has a weak return spring. When everything is normal it is usual that you feel this return spring when you start to move the downshift lever. After a small 'dead spot' the pawl spring throws the pawl into engagement with the downshift mechanism, which typically has a much stiffer spring in it. You can feel the stiffer spring, plus the downshift mechanism has some additional resistance to it anyway. The net result is that the downshift lever normally 'loads up' as it is moved. However, if the pawl spring cannot bring the pawl into engagement, all that you feel when you attempt a downshift is the weak return spring throughout the lever stroke.
If this is the fault, the cure is relatively simple. In mild cases spraying the whole mechanism (but not the hood) in situ with GT85 (or similar) may be effective. If the true culprit is dried grease, simply heating the mechanism (eg. by tipping hot water over it) may be enough to get things moving again
In most cases, the shifter will benefit from the following treatment;

1) remove the shifter from the bars and remove the hood from the shifter.
2) put the hoodless shifter into a ziplock bag having just spayed it liberally with GT85 (or similar).
3) leave overnight, ideally someplace warm.
by Brucey
30 Mar 2024, 8:20pm
Forum: Does anyone know … ?
Topic: Spa Cycles Tourer 540- frame alloy info for repair
Replies: 13
Views: 1142

Re: Spa Cycles Tourer 540- frame alloy info for repair

a fellow forum member kindly gave me his similarly broken Spa touring frame. I planned to weld repair it, but with a twist. The twist would be to excavate a hole between the BB shell and the DT using a die grinder.

I would aim to do the repair via MIG welding, using what I call the 'burst' welding strategy. This breaks up the usual HAZ microstructure, perrnitting much of the GCHAZ to be backtempered. The bottom bead of the weld should be slightly positive and free of undercuts, because this part of the weld is exclusively made via conduction limited melting. The weld top bead would be dressed via grinding in the usual way.

The hole would allow an internal reinforcing sleeve to be assembled in situ from ~3 pieces which would then be welded to one another. Finally the sleeve can be bonded into position using epoxy resin.

Hopefully the repair will be much stronger than it was originally, and the frame won't break there again.
by Brucey
30 Mar 2024, 5:58pm
Forum: Does anyone know … ?
Topic: Why was this 3 speed hub so cheap?
Replies: 18
Views: 1614

Re: Why was this 3 speed hub so cheap?

if the wheel is all-new you may have a NIG SRF3 internal there, in which case you can buy a much longer axle (longer than anything you can get for the older pre-NIG AW.) and make the hub fit a 135mm oln frame. 'Springing' an Al frame is a very good way of breaking it IMHO.
by Brucey
30 Mar 2024, 4:35pm
Forum: Bikes & Bits – Technical section
Topic: How undersize are your sprockets?
Replies: 3
Views: 448

How undersize are your sprockets?

a while back I found I had some nominally 'perfect' new 1/8" chain on my hands (ie. it was exactly 0.500" pitch when measured). When I idly made a loop of this chain round an old freewheel I wasn't too surprised to find that it was a baggy fit. I blamed it on sprocket wear. However, I soon found I had to think again, because it was almost equally baggy on an unworn sprocket. So what is going on there, exactly? Well, I have since come to understand that the rollers might be slightly undersized in new non-sram chain. Also, more significantly, sprockets are nearly always made with a slight built-in undersized-ness if they are to be part of a derailleur system.
This is done to ensure that a new set of sprockets nearly always works OK with a new chain and doesn't skip at the first sign of dirt, despite the relatively slack chain tension. This makes them slightly noisier and slightly more likely to wear quickly, too. When there is no danger of a real running problem being caused by a slight pitch mismatch, things are a little different. So posh track sprockets are not made undersized and can even be machined all over. By contrast, AFAIK all steel derailleur sprockets are made via pressing, which is inherently somewhat less accurate.

I do not know how much this undersizing is likely to amount to, or if it varies much between manufacturers. Maybe accurate measurements of even toothed sprockets (which are relatively easy to do) will tell all.
by Brucey
30 Mar 2024, 12:57pm
Forum: Bikes & Bits – Technical section
Topic: Vitus escarpe 2018 27.5 inch seat stay/chain stay
Replies: 5
Views: 357

Re: Vitus escarpe 2018 27.5 inch seat stay/chain stay

rareposter wrote: 29 Mar 2024, 9:46pm It's an aluminium frame so the whole thing will need heat treatment if it gets welded. Probably puts it into the "beyond economic repair" category...
most Al frames are welded using a grade of Al alloy where the recommended T6 'heat treatment' is to leave it at room temperature for a few weeks. Even if an aging treatment at elevated temperature is required a small part like this can probably be 'cooked' in a domestic oven.
by Brucey
30 Mar 2024, 12:33pm
Forum: Bikes & Bits – Technical section
Topic: SRAM 12 compatible with Shimano 11?
Replies: 3
Views: 259

Re: SRAM 12 compatible with Shimano 11?

SRAM 12s uses slightly larger rollers than normal. What you propose will work, but the chainrings will seem a bit 'preworn' if you see what I mean.
by Brucey
30 Mar 2024, 12:03pm
Forum: Bikes & Bits – Technical section
Topic: disc brake caliper rubbing against spokes
Replies: 29
Views: 5743

Re: disc brake caliper rubbing against spokes

FWIW if you use 'aero' bladed spokes rather than PG you can get another ~1mm clearance. You only need to change 1/4 of them in a built wheel.
by Brucey
30 Mar 2024, 11:06am
Forum: Bikes & Bits – Technical section
Topic: Dynamo carbon fork mounting options
Replies: 24
Views: 1080

Re: Dynamo carbon fork mounting options

Jdsk wrote: 30 Mar 2024, 8:33am If I really really had to mount this on the fork crown...
it shouldn't take very long at all for someone competent to make a suitable bracket.