steel grades

while we're on the subject... I'd value some input on steel for the following:

one of the things I'm doing requires a stub axle to carry a wheel. This is

20mm dia., and I currently make them (as described in more detail recently) from 20mm bright MS, drilled for lightness.

Now, 20mm bright, as I drill it (8mm hole at the maximum stress point, working out to 16mm at the extreme ends) is plenty strong enough. It's got to take half the weight of an adult trike plus rider (assuming the other wheel is lifted in a corner) and currently the same axle is used on tandems (hence twice the load) - not heard of any problems, so I assume the solo one is rather over-spec.

The axle has to be 20mm so it fits 20mm ballraces in the hubs, BTW.

The axle is currently 124mm long, it fits in a mount and a hub with a combined length of just over 124mm, and is all held together with a Q/R 5mm skewer down the middle (basically, a long 5mm bolt with an overcentre clamp on one end, if you've never seen one). The main thing this does, apart from stopping the wheel from falling off, is to pull all the spacers etc. between the ballraces and so on together solidly. I could equally well us a long, thin bolt, but then it's not possible to remove the wheel without tools, which it is now.

The main load is obviously a bending moment, at the point between the mount and the wheel inner side bearing. This has an alloy spacer about 14mm long, and it's in this area that the centre hole is 8mm.

OK, that's the job described. The current axle was tested by the unscientific but basically effective technique of holding it tight in the vice, putting about 8ft of tube over the end, and trying to bend it. It didn't; not at the sort of load that left the vice still attached to the bench, anyway.

in the early prototype stages, we made one using standard bike axles, which are 9mm solid but not plain mild steel, and these bent. Sturmey Archer supply hubs for singe-sided mounting with what appears to be a fairly ordinary 12mm bolt, and one would hope that they've tested this. I don't think they recommend those for tandems, though. So, your thoughts, please, if you've not got bored and wandered off. Can I get better steel which I can still machine? I don't really want to get into heat treatment, haven't got the gear for it.

If I stick with Bright MS, am I being over-cautious with the drilling?

Reply to
Austin Shackles
Loading thread data ...

I'm no expert on steels, but EN8 would probably do your job, machines quite easily though it's not always easy to get a fine finish, and not much dearer than ordinary mild steel.

Tim

Reply to
Tim Leech

In article , Austin Shackles writes

Austin,

Interesting - I have also been making some bicycle parts recently for a friend.

Not altogether clear what is unsatisfactory about your EN1A (?) axle, since you say you couldn't bend it (I assume you mean you couldn't give it a permanent bend, i.e. bend it past its yield point). Is it wear resistance, corrosion resistance, stiffness or yield point you are worried about?

Assuming it's just yield point, then I would consider EN8 (080M40) or (better) EN19 (709M40) or EN24T (817M40). I have made a few parts out of EN24T recently, and find it machines fairly well. If it's corrosion resistance you seek, I'm not too well up on the thousands of grades.

What I doubt you will get, much, is an increase in stiffness, since AIUI steels tend not to vary that much in Young's modulus.

It may be that the mounting, whatever that is, would give up the ghost before the axle.

David

Reply to
David Littlewood

I'd sort of agree with Tim here. All steel has roughly the same modulus (stiffness), but higher carbon and alloy steels usually have a greater yield strength.

I'm not an expert either, but what this means is that they will all bend pretty much the same up the yield point, after which a plain mild steel may bend further and easier than a higher carbon or alloy steel.

If BMS is doing the job then carry on using it. I know the stuff will dent and scratch as soon as you look at it, and an EN8 and above will be a little bit better here, but if it works why change?

Macreadys have a freecutting EN8 (ref Usacut 45 / BS 226M44) which cuts to a better finish than standard EN8, and you can flame harden it too. I have a stock of this in several bar sizes as it's possibly a bit more versatile than mild steel for not much more money.

Peter

Reply to
Peter Neill

On or around Fri, 2 Nov 2007 18:32:03 +0000, David Littlewood enlightened us thusly:

OK, maybe I need to be a bit more specific - although from the sound of it, it's not really worth changing it.

There are 2 issues with the current design:

1) it's quite heavy.

- I'd like to make it lighter by drilling it more, but I'm bothered about the bending resistance. It's good if it fails by bending rather than brittle fracture, though. It shouldn't flex in use, either as it has a hub-mounted brake and if the axle flexes then the brake gets misaligned.

2) it's fairly soft.

- It has relatively narrow ballraces on it, and although these don't rotate on the shaft, nevertheless they apply load to it on only a small area, it's possible that over time it will deform at that point and cause problems.

I guess what I need to do is make up a proper test rig and do some experimental axles, and see what effect removing more material has.

The mounting is made from 35mm long 1" bright MS, drilled 20mm. This is welded into half a hole (Imagine drilling a hole sideways through a tube and then cutting it in the middle of the hole - you get 2 bits of tube with half-a-hole in the end of each) on the end of a 34-ish mm diameter thick walled tube. The mount is mig welded along either side, and bronze welded around the ends, so the bronze runs into the gap between the sides of the tube and the mount tube.

sod it, it needs a picture, too much hassle describing it in words.

I'll take a picture of the one in progress, and stick it on the web.

Reply to
Austin Shackles

Even if you go to EN8, 16 or 24 they will still be quite soft as supplied. My concrete floor has won a number of arguments with freshly made parts that have dropped of the bench. I know you don't want to do it, but heat treatment could give you 50% more strength with any of those grades than the mild steel that you are using now. They would also be hard enough to resist marking much better than the mild steel.

If the steel that you are using now is EN1A (AKA warm butter) or similar, then changing to EN8 will give a perceptible improvement as-is, but If you are using a non-freecutting mild already, then I see a small salt bath and small domestic oven in your future :-)

Mark Rand RTFM

Reply to
Mark Rand

My beloved rule of thumb: Bore with 50% OD, you lose 25% of weight and 5% of stiffness.

Nick

Reply to
Nick Mueller

On or around Sun, 04 Nov 2007 12:35:46 +0100, Nick Mueller enlightened us thusly:

and a handy one it is too.

The key thing about this is that it's used on a safety-critical place on a product I sell, so it has only to break under excessive abuse. It's got to be as break-proof as practical in normal use, and it's better if it fails by bending than by sudden fracture. A previous axle we used for another hub was made for us by a machining company and was case-hardened by another company and cost a lot, 'cos we weren't wanting 10,000 of 'em - the axles ended up costing about 8% of the selling price, which was silly.

This one I can make myself from fairly cheap material and it takes maybe 15 minutes to produce one. At that rate I can afford to make what amounts to one-offs

I'll knock up a credible test rig and see what I can work out. Should be feasible to calculate at least the maximum static load it can get - and come to that, I could look up the bearings the hub people use and see what the quoted maximum dynamic loads are - if the bearings are deemed up to it, then the axle doesn't need to take any more load than that, plus a safety margin.

Reply to
Austin Shackles

In article , Austin Shackles writes

...But obviously best of all if it doesn't fail either way. As I and others have said, most steels and ferrous alloys have fairly similar Young's moduli, and stiffness is likely your limiting factor. You don't have to drill out to the same width all the way through, if you open out the free end of the cantilever by more than the supported end you will reduce the stiffness by less than removing the same weight uniformly along it, as long as you leave the (thinned) outer end strong enough not to collapse, of course.

Gut feel tells me that the pressure required to cause plastic flow under a bearing inner would be colossal. You should be able to work it out, but I would bet the rest of the bike would be crushed long before. Use of one of the harder steels mentioned earlier should certainly reduce even that remote chance though.

Yep. I would use something like a 10-ton hydraulic press and a travelling microscope to measure the deflection.

Fatigue at the weld may be your greatest weakness. I'm certainly no welder, but theory says ensure you leave a good fillet and grind it smooth - but I suspect you know that better than I do!

David

Reply to
David Littlewood

A pikchewer would be helpful or a drawing. But from what I gather I'd have thought the highest stress is going to be at the end of the socket into which the stub axle fits. You perhaps need to consider an internal radius to the outer end of your 35mm mounting. The mild steel isn't going to fail by brittle fracture, it's much too soft, but fatigue could eventually lead to a sudden failure.

If you're really trying to optimise it, a bit of proper stress analysis would be appropriate, but for that we'd need some accurate drawing(s) and some sensible guesses at likely loads.

A fat bloke and 4g on one corner would be a fair starting place assuming it's not motorised and has no suspension.

Richard

Reply to
Richard

On or around Sun, 4 Nov 2007 16:29:57 +0000, David Littlewood enlightened us thusly:

well, ideally, I'd test the complete thing. I had in mind something big and solid to hold the unit under test, and about 8ft of heavy gauge tube and pull on the end to supply a bending moment. But your plan is more scientific.

I might have a go at it sometime. I've got a spare one that went wrong in the making process knocking around - the wrongness was in the machining of a bearing seat, not in the structural stuff.

Reply to
Austin Shackles

You have to be aware that the bearing is an extra load to the shaft. There is a stress concentration where the shaft leaves the bearing's bore. The reduction-factor is about 2.5! You are also aware of that the yield strength is getting way down with oszillating (SP?) loads.

Nick

Reply to
Nick Mueller

On or around Sun, 04 Nov 2007 17:09:48 +0000, Richard enlightened us thusly:

aye, worth a thousand words, innit. trouble is, the drawing I have is in an obscure format. I guess I could screengrab it.

I could do that, if anyone has the ability and/or the software to analyse it. I'm working on trying to make the whole machine lighter, and this bit is one of the heavy pieces.

Reply to
Austin Shackles

I play with karts. Front wheels have a 17mm mild steel solid axle with the wheel spaced anything up to 5 cm from the stub shoulder. The do 70mph and hit things like concrete kerbs up to 2 inches hig regularly.

I've only rarely seen the axle itself bend (though I have seen th mounts bent and 8mm kingpins sheared)

As an experiment we put a 17mm tube with wall thickness of 2mm onto carbon fibre axle and deliberately tried to wreck by some ver ferocious driving and then ran it on a dyno for a few hours. It was th bearings that failed, not the axle. It was markedly lighter than th all steel model (and illegal for karting....no composites allowed). Sleeving or putting inserts into an otherwise composite part is norma practice.

Obviously a carbon part will be more expensive than its stee counterpart, but you might gain advantages* by getting rid of tha doubly welded joint.

  • Speed of manufacture
  • Less failure modes
  • Lightness

Robi

-- rss

----------------------------------------------------------------------- rsss's Profile:

formatting link
this thread:
formatting link

Reply to
rsss

On or around Wed, 7 Nov 2007 09:50:12 -0600, rsss enlightened us thusly:

carbon fibre isn't an option on this - and the nature of the thing is that it has to be welded.

Reply to
Austin Shackles

PolyTech Forum website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.