Author Topic: Quick release vs track nuts  (Read 3728 times)

Tim Hall

  • Victoria is my queen
Quick release vs track nuts
« on: 19 December, 2019, 06:50:24 pm »
My Pretty Bike, which has gears, is fitted with quick release wheels. They work just fine and stay where they're meant to.  My Dave Yates is a fixed wheel machine and comes with track nuts front and rear. I carry a cut down 15mm spanner in my saddle pack to remove the wheels for puncture repair.

But why? Surely the forces applied to either bike are the same? They both have semi horizontal dropouts too. Is it because it's always been done like that? Or to minimise niggling chain tension issues caused by minuscule creep of the rear wheel by doing the track nuts up Bastard Tight?
There are two ways you can get exercise out of a bicycle: you can
"overhaul" it, or you can ride it.  (Jerome K Jerome)

zigzag

  • unfuckwithable
Re: Quick release vs track nuts
« Reply #1 on: 19 December, 2019, 07:09:51 pm »
i've never had axle creep with qr skewers on alloy track ends, but had on a few occasions on steel dropouts before i had to install chain tensioner. chromed steel is the most slippery finish. axle retention will also depend on how sharp the serrations are on locknuts and skewer clamps.

Re: Quick release vs track nuts
« Reply #2 on: 19 December, 2019, 10:21:49 pm »
when you ride fixed and take most of the slack out of the chain, just going over a bump can cause the tension in the chain to be momentarily  incredibly high.  [There is no such thing as a perfectly flat catenary, and the tension becomes asymptotic as you approach flatness; a modest vertical load on a nearly-flat chain run can cause a tension increase in the chain that is a hundred times higher.]  By contrast going over a bump on a geared biike just causes the derailleur to deflect slightly; no real increase in chain tension occurs.

In addition when you ride fixed you may create tensions in the chain that are several times higher than normal, because if you don't, you won't get up steep hills.  By contrast on a geared bike you are unlikely to have to push that hard on the pedals; most folk just change gear instead. 

The peak loads when riding fixed come from going over a bump at the same time as giving it what for.

The friction between two surfaces that are clamped varies from about 0.1 x the clamping load to about 1.0 x the clamping load, depending on the nature of the surfaces.

Note that to start a QR wheel moving (with the lever on the left and the QR nut on the right), all you need is for the RH locknut to start sliding; the threaded part of the QR skewer may be clear of the hollow axle and the axle may be able to move slightly without the QR nut also sliding. If the QR nut is impeding movement, it won't be able to exert a very high force to stop the wheel, because it isn't that difficult to bend a length of M5 x 0.8mm studding.

By contrast when you tighten a (good) track nut properly, both the track nut and the hub locknut have to slide past the dropout before the wheel will start to move. This means that even at the same clamping load, a nutted axle is less likely to move.

The clamping loads also vary;  roughly (estimated in skewers by the clearance change in cup and cone bearings)

- external cam QR = up to ~ 500kgf
- internal cam QR = up to ~ 500- 750kgf
- screw type 'security skewer' =  up to ~1000kgf
- breaking load of QR skewer between ~750kgf and ~1250kgf depending on quality of skewer and loading.

- axle with track nuts (3/8" or 10mm thread) up to ~ 2500 kgf.

FWIW if you are a strong rider, horizontal aluminium dropouts are likely to be a problem; unless they have steel facings they will very likely just get chewed up, even with gears. Without gears you are well advised to use a pretty solid chain tug.  I used to break QR skewers on geared bikes (with horizontal steel dropouts) trying to get them tight enough that I definitely wasn't going to pull the wheel over. In the end I had to use the best quality QR skewers and had to be sure to use new, textured/ridged locknuts on the RHS of the hub. When riding fixed the built-in washer on many track nuts just wasn't strong enough and would deform. If they were less tight than that I'd risk pulling the wheel over.

Note that the wheel adjusters in horizontal dropouts allow a geared rear axle to return to the same spot each time and that is usually where the dropout will best accommodate the locknut positively; quite a lot of QR hubs had a ridge on the locknuts which made a corresponding groove in the face of the dropout. However when running fixed or SS, the wheel never goes back in the same place twice, so the surfaces never get to mate in the same way.

I've never been able to reliably  ride SS or fixed using a wheel with a QR skewer, and nor has anyone else I've known who has been a strong/quick rider.  Riders with 'explosive power' can (briefly) be pushing four or five times harder on the pedals even than folk who are really quick over a distance.

By all means try using fixed gear with a QR skewer if you must, but IME it is bad practice and is very likely to lead to problems.

cheers

rogerzilla

  • When n+1 gets out of hand
Re: Quick release vs track nuts
« Reply #3 on: 20 December, 2019, 09:02:29 am »
QR works fine with a fixie or singlespeed if you fit a hollow axle to a cup-and-cone track hub but (as with any wheel NOT in vertical dropouts) choose a good steel QR with an enclosed cam.  My singlespeed runs happily with old straight-lever Campag QRs and has also run as a fixie in the past.  No axle creep.

There is one issue to be aware of with cartridge bearing hubs, and it's often a deal-breaker.  Some fixie hubs, like Goldtec, already have hollow axles to accommodate a QR but the sealed bearings (which presumably are a press fit to the axle AND the shell) can't cope with the axle compression and the wheel becomes stiff to turn.  Ride it like this, and the bearings will be ruined very quickly.  I wanted to use QRs on my track bike for the Dun Run, to avoid packing long allen keys, but this thwarted me.

EDIT: I agree with Brucey that aluminium dropouts can be a problem.  I had a (bolt-on) ENO hub on a Cannondale and the thing crept insufferably, being really slack after 100 miles even in dry conditions.  The nuts/bearing caps bite into alu really well but the alu itself deforms.  It's like nailing jelly to a wall.
Hard work sometimes pays off in the end, but laziness ALWAYS pays off NOW.

Re: Quick release vs track nuts
« Reply #4 on: 20 December, 2019, 09:19:08 am »
What I have read is that the nuts come from the track where you cannot have any unnecessary bits which would cause injury in a crash and that there is no reason not to use QR on a road fixed-wheel bike.  However, whoever said that (and I think it was someone with some knowledge such as Sheldon) may not have known as much about it as Brucey!

rogerzilla

  • When n+1 gets out of hand
Re: Quick release vs track nuts
« Reply #5 on: 20 December, 2019, 10:37:35 am »
The acid test is whether the chain goes slack any more quickly than with a nutted/bolted axle.  I've had nutted axles that crept worse than a QR, usually because the dropouts were chromed (which is a challenge even for Campag track nuts, which are £21 each* and pretty deeply serrated).

With nutted axles I experiment to find the minimum torque that prevents slippage and then use that.  25Nm is usually enough for a plain steel dropout, but chrome dropouts may need 35Nm or higher.  This assumes clean, lubricated threads that have been tightened a couple of times, so are smooth.  Use 25Nm on virgin components without any oil or grease, and you won't get much clamping force - torque settings are not a panacea!

*captive audience, 10mm x 26tpi thread
Hard work sometimes pays off in the end, but laziness ALWAYS pays off NOW.

Re: Quick release vs track nuts
« Reply #6 on: 20 December, 2019, 11:02:44 pm »
The acid test is whether the chain goes slack any more quickly than with a nutted/bolted axle.  I've had nutted axles that crept worse than a QR, usually because the dropouts were chromed (which is a challenge even for Campag track nuts, which are £21 each* and pretty deeply serrated).

With nutted axles I experiment to find the minimum torque that prevents slippage and then use that.  25Nm is usually enough for a plain steel dropout, but chrome dropouts may need 35Nm or higher.  This assumes clean, lubricated threads that have been tightened a couple of times, so are smooth.  Use 25Nm on virgin components without any oil or grease, and you won't get much clamping force - torque settings are not a panacea!

*captive audience, 10mm x 26tpi thread

Sorry to be pedantic but 10mmm x 26tpi??? I am sure that Mr Campagnolo wouldn't have mixed metric and imperial like that!

bludger

  • Randonneur and bargain hunter
Re: Quick release vs track nuts
« Reply #7 on: 21 December, 2019, 02:13:04 am »
So QR on track hangers is a no no. But is QR on a fork supplied with nutted wheels ok? Especially when using rim-type brakes.
YACF touring/audax bargain basement:
https://bit.ly/2Xg8pRD



Ban cars.

LittleWheelsandBig

  • Whimsy Rider
Re: Quick release vs track nuts
« Reply #8 on: 21 December, 2019, 06:25:01 am »
You might think that, mzjo, but you’d be wrong.

QR forks are effectively vertical dropouts. QRs are fine, provided the dropouts themselves are thick enough to provide clearance for the volute springs and the protruding axle ends.
Wheel meet again, don't know where, don't know when...

rogerzilla

  • When n+1 gets out of hand
Re: Quick release vs track nuts
« Reply #9 on: 21 December, 2019, 07:56:20 am »
Sorry to be pedantic but 10mmm x 26tpi??? I am sure that Mr Campagnolo wouldn't have mixed metric and imperial like that!
He did.  It's quite annoying.

Mind you, the Germans also did it with their FG threads, common on older Sachs hubs.  However, there is no FG10,0 which would be Campagnolo thread; only FG9,5 and FG10,5.  FG10,5 is almost Sturmey-Archer thread (13/32" x 26tpi) but not quite!
Hard work sometimes pays off in the end, but laziness ALWAYS pays off NOW.

Re: Quick release vs track nuts
« Reply #10 on: 21 December, 2019, 10:31:18 pm »
The acid test is whether the chain goes slack any more quickly than with a nutted/bolted axle.  I've had nutted axles that crept worse than a QR, usually because the dropouts were chromed (which is a challenge even for Campag track nuts, which are £21 each* and pretty deeply serrated).

With nutted axles I experiment to find the minimum torque that prevents slippage and then use that.  25Nm is usually enough for a plain steel dropout, but chrome dropouts may need 35Nm or higher.  This assumes clean, lubricated threads that have been tightened a couple of times, so are smooth.  Use 25Nm on virgin components without any oil or grease, and you won't get much clamping force - torque settings are not a panacea!

*captive audience, 10mm x 26tpi thread

Sorry to be pedantic but 10mmm x 26tpi??? I am sure that Mr Campagnolo wouldn't have mixed metric and imperial like that!

10mm x 26tpi is correct.

cheers

Torslanda

  • Professional Gobshite
  • Just a tart for retro kit . . .
    • John's Bikes
Re: Quick release vs track nuts
« Reply #11 on: 26 December, 2019, 05:39:25 pm »
There is a way to use a QR and keep the tension down to acceptable levels. Chain tugs.

Cheap BMXSOs don't suffer from wheel creep as such but what does happen is the nut bites the dropout and when you try to refit the wheel and retension the chain it's alost impossible to do so without fitting a chain tug.

The only downside is you either need a 10mm spanner or suitable hex key instead of a 15mm spanner, so there's no saving on tools. The plus side is that you're not going to bite into the dropout, just the chain tug...
VELOMANCER

Well that's the more blunt way of putting it but as usual he's dead right.

bludger

  • Randonneur and bargain hunter
Re: Quick release vs track nuts
« Reply #12 on: 26 December, 2019, 05:51:33 pm »
Is it possible to convert a wheelset that uses through-axle hubs to use solid axles and track nuts? Some kind of bolt you insert inside the hub and fix in with bolts? Would be a lot more cost and time effective for e.g. making a 650b fixed gear wheelset than building a set from scratch.
YACF touring/audax bargain basement:
https://bit.ly/2Xg8pRD



Ban cars.

Re: Quick release vs track nuts
« Reply #13 on: 26 December, 2019, 06:16:10 pm »
Is it possible to convert a wheelset that uses through-axle hubs to use solid axles and track nuts? Some kind of bolt you insert inside the hub and fix in with bolts? Would be a lot more cost and time effective for e.g. making a 650b fixed gear wheelset than building a set from scratch.

Yep, buy a through axle frame;)

rogerzilla

  • When n+1 gets out of hand
Re: Quick release vs track nuts
« Reply #14 on: 27 December, 2019, 03:21:12 pm »
With horizontal dropouts!
Hard work sometimes pays off in the end, but laziness ALWAYS pays off NOW.

Re: Quick release vs track nuts
« Reply #15 on: 27 December, 2019, 03:53:48 pm »
With horizontal dropouts!

You can get sliding thru-axle "dropouts" (e.g.) for some frames, but it's not going to be cheap!

The biggest problem is going to be hub width and bypassing the freewheel on the freehub. I can't see it working out cheaper or easier than getting/building the right wheel.

Re: Quick release vs track nuts
« Reply #16 on: 27 December, 2019, 04:22:21 pm »
IME:

Setup 1
Ofmega track hub with track nuts on painted steel Camapgnolo style road dropouts

setup 2
Camapgnolo 80s QR on Camapgnolo track hub on chromed Camapgnolo road dropouts*

The QR setup does possibly creep very slightly, the chain gets looser by no more than a few mm when wiggled up and down, and that is after quite a few miles. I do up the QR reasonably tight and I could make it tighter but I prefer to be cautious rather than over tighten and break the QR, which I did once when the the cam "housing" cracked.

I haven't ridden the tracknuts setup for quite a while but I remember it did need the occasional retensioning of the chain.

I would say both are similar but that's ordinary riding, not racing or training.

OTOH my geared bike with the same QR on chromed Camapgnolo road dropouts never creeps.

*The reason for using  QR is the frame is 753 and 126mm rear end

Re: Quick release vs track nuts
« Reply #17 on: 27 December, 2019, 07:18:26 pm »
you can buy solid axles in 10mm x 26tpi threading in various lengths and you don't have to buy the most expensive track nuts either. This means you can easily  convert a 126mm OLN QR hub to a solid axle, and there ought to be no more movement.

cheers

rogerzilla

  • When n+1 gets out of hand
Re: Quick release vs track nuts
« Reply #18 on: 28 December, 2019, 11:03:04 am »
I've never found a non-Campag source of 10mm × 26tpi nuts.  However, most track hubs (Formula, System EX, Halo, Miche, Shimano) will be 10 x 1mm.

Campag axles are chromed, and the threads are consequently very tight when new.  Silly, really.
Hard work sometimes pays off in the end, but laziness ALWAYS pays off NOW.

Re: Quick release vs track nuts
« Reply #19 on: 28 December, 2019, 05:18:40 pm »
you know what, I can't find a non-campag source of track nuts in 10mm x 26tpi right now either.  I'm sure that one of the other Italian hub brands used to use the same thread and I've bought them that way in the past, but that might have been a long time ago.  However the 10mm x 26tpi thread was never that common; even Spanish Zeus (which used campag hubshell forgings I think) used fully metric threads in the axles of their hubs.

I have had track nuts modified on a lathe before now; for example a 3/8" x 26tpi track nut can easily be recut to 10mm x 26tpi. I'd probably do that, bearing in mind the absurd cost of campag track nuts. Wheels manufacturing make 10 x 26tpi axles but not track nuts....?  Weird eh?

cheers


rogerzilla

  • When n+1 gets out of hand
Re: Quick release vs track nuts
« Reply #20 on: 28 December, 2019, 08:04:32 pm »
I even tried looking for "FG10,0", thinking there might be a generic German 10mm x 26tpi alternative, but that isn't one of the DIN sizes: https://www.ring-plug-thread-gages.com/PDChart/Bicycle-thread-data.html

(Oddly, FG10,5 isn't listed either but Sachs used it on lots of hubs, notably the Torpedo 3 speed on the Brompton L3.
Hard work sometimes pays off in the end, but laziness ALWAYS pays off NOW.

Re: Quick release vs track nuts
« Reply #21 on: 28 December, 2019, 10:35:02 pm »
FWIW the Sachs 10.5mm x 26tpi thread is eventually going to turn into a problem (you can buy spares at present but this mayn't last for ever), however in many cases it is possible (just) to fit a SA 13/32" x26tpi part to such axles.  Pretty snug though.

cheers

rogerzilla

  • When n+1 gets out of hand
Re: Quick release vs track nuts
« Reply #22 on: 29 December, 2019, 11:44:21 am »
Genuine SA nuts, at least the UK-made ones, are sacrificial and will strip long before the axle is damaged.  So a 10.5mm Sachs axle should be able to enlarge the 10.3mm SA thread quite easily.  Knock-off nuts for SA hubs, and more recent chromed ones, might be less malleable.
Hard work sometimes pays off in the end, but laziness ALWAYS pays off NOW.

Re: Quick release vs track nuts
« Reply #23 on: 29 December, 2019, 11:57:48 am »
the other component to that equation is of course the axle hardness. SA axles are typically hard for their full length, but that is not true of all IGH axles.  For example shimano ones are often soft at the ends (such that the axle wears as well as  the NTW in many cases). You can file these axles easily.  I don't know if there is any pattern to Sachs/SRAM axle hardness or not, but I have had to use a modified SA cone in a  SRAM hub and the SA cone was harder than the axle, i.e. the threads on the axle were eased by passage of the cone. 

SA nuts are softer than SA axles by far, like you say. I doubt that any aftermarket nuts would be as hard as SA axles though; folk have made taps out of them.

It occurs to me that you could make a 26tpi thread chaser by slitting an old SA axle lengthwise, and then using an expanding wedge of some kind to adjust the diameter accordingly.  This might work well enough to modify a 3/8" x 26tpi track nut to 10mm x 26tpi specification...?

cheers