Center lock wheel nut thread direction

[penna]

I know the thread directions for center locks are meant to be left handed on the right and right handed on the left.
What I don't know is why. Initially I thought it was to do with the torque under braking or acceleration, but as they are not different front to rear this, from my logic cannot be the case. The only explanation I have heard is that it has something to do with the thread clearance between the nut and the spindle. As the wheel rotates, because the spindle is slightly smaller, with each rotation it rotates slightly more than the nut- think hula hoop. This eventually leads to the nut loosening off if it is threaded in the wrong direction.

I am still not 100% sold with this explanation so was wondering what reasoning other people have used to determine the thread direction for their center locks

[NickFavazzo]

Search nut precession, there are a few videos on it that explain it quite well on youtube, it has more to do with the wheel flexing and providing clearance progressively around the nut, slowly working it off. Our nuts have normal threads and we haven't had issues with precession (until tightening torque isn't high enough and wear comes into play..)

[Dunk Mckay]

Centre lock nuts are evil. They will get you in your sleep. Beware...


Bewaaare....

[Jon Burford]

Originally posted by Dunk Mckay:
Centre lock nuts are evil. They will get you in your sleep. Beware...


Bewaaare....

haha

[penna]

Cool thanks nick that makes a bit more sense

[Gruntguru]

I can't think why it would be different for centre lock nuts, but standard practice for normal wheel nuts is LH thread on LH side. I have experienced wheels coming loose on cars a couple of times and it was always LH side (with RH threads of course)

[exFSAE]

IMO, best approach from a practical and logistical standpoint is to just use the same threads (RH) all around the car... with a castled wheel nut and clevis pin, or some other positive lock.

You have to have the positive lock there by rule anyway so I'd say take advantage of that, let the thread direction be a non-issue, and make it easier for manufacturing and carrying spare parts.

[Claude Rouelle]

An small experience / story to share. Several years ago, I worked for a racing team who was using 2 heavy mechanics and a leverage of 2.5 meters to tight the wheels (don't even think about the torque number!) ... and they still manage to lose some of them on the race track!

The issue was simply that the thread of the hub and the wheel nut was too fine. Simply think about the amount of the torque that will be translated in a force perpendicular to the rim, holding it against the hub flange or brake disc. A simple calculation that the race car manufacturer did not do. In a panic, the team had to re-machine its own hub and central wheel nut after 2 races of the season. With a bigger pitch number they never had any problem.

[Tilman]

The issue was simply that the thread of the hub and the wheel nut was too fine. Simply think about the amount of the torque that will be translated in a force perpendicular to the rim, holding it against the hub flange or brake disc. A simple calculation that the race car manufacturer did not do. In a panic, the team had to re-machine its own hub and central wheel nut after 2 races of the season. With a bigger pitch number they never had any problem.
Is it possible that you accidentially mixed things up?

If a screw/bolt/nut loosens, you need a finer thread. This is because with a fine thread an axial force produces less torque that tries to loose the nut. In the case of an infinite fine thread, the "ridge" is perpendicular to the bolt axis which means that no loosening torque is produced at all.

In the metric system, the thread is described by outer thread diameter and pitch, in the imperial system it is outer thread diameter and threads per inch. If the thread is fine, the pitch value is low, like in M8x1,0 instead of the regular, coarser M8x1,25. With a low pitch value the value of threads per inch increases, leading to a "bigger pitch number" as you described it.

So their problem was not a thread that was too fine but a thread that was too coarse ...

[Z]

penna,

The "correct" thread direction depends on relative clearance between wheel-and-nut, and between nut-and-thread. Briefly, each of these, together with wheel-to-hub "fretting", work to loosen the nut in opposite directions. So the answer to "RH or LH?" is "It depends...".

Gruntguru is right in saying that left wheels fall off much more often when done up with RH threads. exFSAE is right in saying that logistically it makes more sense to use RH all-around. Claude is right in saying that you can do conventional centrelocks up to stupendously high torques and they still fall off...

Anyway, last year I ranted on this subject at great length on the "Center-lock hubs and wheels" thread. That thread started in 2004, but picked up again (page 4) in 2011 here (http://fsae.com/eve/forums/a/tpc/f/125607348/m/84610603711?r=62720229051#62720229051)

There are some sketches about half way down page 6 that solve most of the problems (IMO). But read the posts...

Z

[penna]

Thanks Z I have already read that thread, it just didn't have anything that explained the thread direction.
The clearance amount I have already looked into as well and believe I have it sorted.
As for using the hirth joint (see other thread) I chose not to do that for a couple of reasons. Under braking because of the angled spline the wheel would want to push itself off the spindle- hence making preload important again and if not done up tight enough possibly introducing compliance ( with a 45deg joint the preload needed is almost the same as our centerlock). This can be solved by making the angle more accute, but then you have the issue of it jamming and being hard to get off which is really not what you want on something like a wheel. The real killer was manufacturing making this joint would be a lot more involved than the traditional way of doing it.

Not saying a hirth joint would not be a good solution, just not in our case.

[Z]

Penna (or others still wondering about this),

Getting back to the issue of centrelocks unscrewing themselves...

1. The nut-to-spindle-clearance "hula-hoop" effect (aka "epicyclic precession"...) tightens RH threaded nuts on the left side of the car, and loosens RH threads on the right side.

2. If the nut has an outside conical surface (ie. male) that fits inside a female cone in the wheel, then this wheel-to-nut-clearance loosens RH threaded nuts on the left side of the car, and tightens RH threads on the right side. Since this clearance usually gets bigger than the nut-to-spindle-clearance once the nut gets a little bit loose, the left wheels usually fall off cars with all RH threaded nuts.

3. However, if the nut is shaped such that it fits on the OUTER diameter of the wheel mounting surface (ie. nut has female cone, wheel has male cone), then the tightening/loosening is as per point 1 above. This system will then tighten RH threads on the left side of the car, and tighten LH threads on the right side, as you suggested in your original post. This is the age-old "Rudge-Whitworth Centre-Lock Wire Wheel" as explained in this article. (http://www.tr-register.com.au/Files/technical/wirewheels.htm) If you really want to use centrelocks with axial spline drive (eg. pegs), then this might be the best approach.

4. A centrelock coming loose in Endurance probably means a DNF, and you end up at the bottom of the ladder ... with the other loosers. I would suggest a lot of testing with the nuts only lightly torqued, just to see what happens. http://fsae.com/groupee_common/emoticons/icon_smile.gif

Z

[onemaniac]

Originally posted by Tilman:
<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">The issue was simply that the thread of the hub and the wheel nut was too fine. Simply think about the amount of the torque that will be translated in a force perpendicular to the rim, holding it against the hub flange or brake disc. A simple calculation that the race car manufacturer did not do. In a panic, the team had to re-machine its own hub and central wheel nut after 2 races of the season. With a bigger pitch number they never had any problem.
Is it possible that you accidentially mixed things up?

If a screw/bolt/nut loosens, you need a finer thread. This is because with a fine thread an axial force produces less torque that tries to loose the nut. In the case of an infinite fine thread, the "ridge" is perpendicular to the bolt axis which means that no loosening torque is produced at all.

In the metric system, the thread is described by outer thread diameter and pitch, in the imperial system it is outer thread diameter and threads per inch. If the thread is fine, the pitch value is low, like in M8x1,0 instead of the regular, coarser M8x1,25. With a low pitch value the value of threads per inch increases, leading to a "bigger pitch number" as you described it.

So their problem was not a thread that was too fine but a thread that was too coarse ... </div></BLOCKQUOTE>

I was confused about it too. Claude, can you clarify?

[Claude Rouelle]

Onemaniac,

If you define the "angle of the pitch" as 0 degree when the thread are perpendicular to the hub axis (theoretically - but impossible - extreme fine thread) and 90 degrees when the thread is in the axis of the hub (theoretically extreme - but impossible - gross thread) what is the angle you would prefer if you goal is to avoid losing wheel without using an herculean tightening torque?