Why do wheel spinners tighten clockwise on the left?

Even later then that. I think the Chrysler newyorker had that into the 90's. Bentley and rolls same thing. Learned that the hard way on the Chryslers. At least rolls and Bentley has an arrow on the wheel nuts.

 

Andy,

To me it just makes sense that you should tighten the spinner in the opposite direction of the hub. From a dead start, or during acceleration, the driven hub will spin but the wheel will resist turning due to the weight of the car on the tire and the mass of the wheel and spinner. For an instant, the hub will spin and the spinner will stay in place. If the spinner is not threaded in the opposite direction, it will loosen.

It’s the same reason why when you tighten a nut on a bolt, the bolt head and nut spin in the opposite directions with respect to each other.

For the non-driven wheels the mass of the wheel, spinner and tire will tend to resist turning when the hub begins to turn but to a much lesser degree.

It’s also the same reason why there is a rim lock on a motorcycle tire. The driven rim will turn and the tire will stay stuck to the terrain. Without a rim lock, the tire will rotate on the rim and the tube stem would get pinched and you’ll get a flat. Front tires are not driven and thus have no rim lock.

Back to the car example. I’ll bet you can tighten a loose spinner at speed by just hitting the brakes. The hub should stop turning instantly and the spinner will tend to continue turning effectively tightening itself on the hub. I don’t know why on one side the spinner is female and the other male. I would think the spinner would be better off being male on both sides so it could use the taper as a wedge for a stronger fit – same as the bolts on the existing Ferrari wheels. It could be simply to remind you that you need to tighten the spinners differently from side to side.

Physics is your friend but try not to dig too deep into the theory or the world gets complicated pretty fast.

Hope this helps.

 

Thanks very much, Solid State.

I was thinking like you did initially. But if that were true, spinners would tighten in the same direction on both Rudge-Whitworth and Chapman (Lotus Elan) wheels. But they don't (Chapman spinners tighten clockwise on the right side and vice versa on the left).

Also, the front wheels aren't driven in rear-wheel-drive car.

The tapers on a R-W spinner (such as Borrani's) are always female, regardless of the side of the car they're on. The tapers on a Chapman spinner always male. It is that difference that is supposed to account for why they are threaded reverse of each other.

That is when explanations of inertia and acceleration (which should apply to both female and male tapers) fall apart.

I forgot to mention that mechanical precession operates when the nut (the spinner) is tightened fully, not when it's loose. There is always a gap, albeit microscopic, between the unloaded side of the spinner and the wheel. It is that gap that causes the spinner to tighten or loosen purely by reason of the rotation of the load. Or so I've read (probably incorrectly).

I wish I hadn't begun trying to understand this, but I have a car with R-W wheels (female taper) and a Lotus Elan with Chapman's spinners (male taper). They tighten differently from each other, as noted, and it has always bugged me that I don't understand why.

 

My two cars that started me on this. The same laws of physics apply to both cars and should explain why the spinners on one car tighten one way and the other way on the other car.

Andy
Image Unavailable, Please Login

 

Here are knock-offs (spinners) for the early Dino, the 206. Note that they have female tapers, following the Rudge-Whitworth system, and therefore tighten counter-clockwise on the right side of the car (and vice versa on the other side).
Image Unavailable, Please Login

 

Oops, yes, the D-Type is the the same as the Borrani/RW.
However the F40, 288GTO, TR and the Porsche centre lock nuts have the male taper like the Elan yet tighten the opposite way to the Elan.

They can't all be correct!

 

Well, re-read what I detailed and try and use physics (scientific method) to disprove it. I have spun many a rim on the tire and if it had a spinner with a right hand thread it would loosen and fall off!

Don't think your statement is quite correct here (or I am misunderstanding): "It is that gap that causes the spinner to tighten or loosen purely by reason of the rotation of the load" because in our example the load is the spinner and the rotation occurs on the hub upon acceleration. Unless, of course, you immediately stop the rotation of the hub which will transfer the load back to the hub as the spinner tends to continue to rotate.

Try this: Put a threaded rod in the end of a hand drill. Run a nut a couple threads up. Initiated the drill. Repeat with drill in opposite direction. Which way did the nut come off? Answer: __________

 

Andy,

If you want to read / learn about precession, more power to you -- it is an interesting phenomenon.

However, it is not relevant to the situation with wheel spinners / nuts that is being discussed. There is no precession going on in this system --- not sure where you read that it was, but that information is incorrect.

Precession only occurs for "free" bodies (though they may be simply supported as long as they still have the remaining degrees of freedom) which are acted upon solely by their own inertial mass and angular momentum, and with or without the presence of external gravitational forces.

These wheel nuts in this type of system are too constrained (in space) against completely free motion by both the threads and by the spindle / axis they are mounted on. Also, there are frictional forces and reaction (contact) forces acting upon them at all times.

Therefore, they cannot experience precession (i.e., they cannot "precess").

 

David, thanks for your note. I think what you are saying is that precession cannot occur between a knock-off and the wheel, because they are pressed tightly against each other. They are not "free bodies," able to move relative to each other.

I am not saying that is wrong. You are probably right. But what i read is that precession actually occurs between two objects tightly screwed together. A good example is the spindle of a bicycle pedal screwed as tightly as possible into the crank.

This is a quote from the link I posted earlier -

"In machinery, [epicyclic] fretting [precession] is the micro-motion of tightly fitting parts that superficially appear immobile with respect to each other."

“What may be less evident is that even tightly fitting parts have relative clearance due to their elasticity, metals not being rigid materials as is evident from steel springs. Under load, micro deformations, enough to cause motion, occur in such joints.”

There is always apparently always a gap, because of lb elasticity of metals, between the spindle and the crank and, in our case, the knock-off and the wheel.

I am not arguing for precession but that seems to be where most technical articles lead to.

I'm in the car at the moment. But in the meantime, read the link above when you have time. It may answer your point about "free bodies."

 

Solid State, many thanks. Let me state that the reasons are not clear in my mind, and you may well be right.

This is what I understand (probably wrongly):

The "load" spoken of in discussions on knock-off direction is the weight of the vehicle. In Chapman's spinner, which has a male thread, the weight of the car corner is borne by the male taper of the spinner, specifically at 6:00, as it rests on the corresponding taper on the wheel. In the R-W system, that load is borne by the spinner at 12:00 of its female taper, as it rests on the upper, corresponding male taper of the wheel. The photo below (from the book "Lotus Engineering) illustrates what I mean. The arrows point to the load.

It is this load that spins in an epicyclic motion (counterclockwise) as the wheel moves clockwise.

Precession they say (I think) moves opposite to that epicyclic motion.

Here is another link that explains where the load is on a bicycle pedal (a prime candidate apparently for precession) and also the microscopic gap that allows movement between tightly fitting parts.

Everyday Scientist » why is the left bike pedal left-hand threaded?

I am not citing these to support one position or the other, but simply passing it along.

Andy
Image Unavailable, Please Login

 

There were times in the past when I wondered why I didn't take physics or engineering. Now I know why.

 

Will, yes. That puzzles me. F1 cars also have a male taper, ditto Cobras and GT40s (notorious for losing wheels in the early days), and they tighten RW style. All of those cars seem to have locking devices or use safety wire.

Someone told me that some Porsche race cars use right-handed threads on both sides, relying on the locking system. I read there was a recall on their earlier centre nuts, which were prone to loosening in racing. I'm not sure. My GT3 has bolt-on wheels.

Chapman was no intellectual slouch and always innovative (he had after all more F1 Constructor Trophies than Ferrari up to the 70s), so I am hesitant to write his theory off. And Elans don't shed spinners. (But I safety wired mine when I was racing, just in case.)

Dayton's male-taper knock offs apparently follow Chapman's theory.

I'm still hoping someone will come forth with an explanation that makes sense of both R-W and Lotus practices. I think I understand, but F40s, etc., throw me off. I hope someone's unified theory will explain that too.

Best,

Andy

 

As mentioned earlier this visually explains it well.

http://en.wikipedia.org/wiki/Precession_(mechanical)

Basically there is a difference in friction between the bottom of the spinner where it touches the wheel and the top. That and the slight difference in diameter between the contact point on the wheel and the spinner causes a rotational force to be applied, and even though the wheel itself cannot rotate freely from the hub, there is distortion and shifting slightly out of center due to play in the spline through the range of rotation that causes a rotational force.

In the case of the elan spinner this force is applied inward via the outside radius of the spinner, the other designs apply that force outward via the inside radius of the spinner. The elan spinner is analogous to the blue part of the wikipedia diagram, however for the Borrani/RW design you have to imagine that the inner blue part is the wheel/hub and the outer part is the spinner.

What's hard to know is whether Colin Chapman really though of this ahead of time or had the wheels fall off his car a few times before he worked it out

 

The fly in the ointment for me is that the wheel/hub is subjected to both forward acceleration of the car and braking. Thus it would seem that the correct solution depends upon one being assumed to be more important in actual driving conditions than the other. I guess this a function of both time spent in each condition (probably greater in forward acceleration) and torque under each condition (probably greater under braking).

 

Andy,

You have already been given the answer. The proof is above. Did you try it? When you do, it will prove why you tighten in the opposite direction. If two manufacturers do it differently then why should that be so perplexing? Racers tighten to proper spec and then wire-tie. In engineering, you take your shot and move on. Too much to achieve to sweat such small details. Try not to get too "wrapped around the axle"!

 

IMO it is the necessary small rotational clearance in the splines which when faced with being on the wrong side of the car and first loosened allows unwinding of the spinner to the amount of the clearance each time the brakes are applied. This stays until the next braking which adds another small unwind until after a few dozen stops the wheel falls off.

I speak from personal experience having received a 1965 Corvette which had the knock offs and wheel hubs wrong sided. I had purchased the car with the proviso the knock off wheels from another dealership car be installed. My left front wheel departed the car taking a chunk out of the fender. Easy fixed but could have been very nasty.

 

You are exactly correct ---- it is the backlash between the splines of the internal wheel hub and spindle that causes the problem.

The only reason some makers went to conical / tapered interfaces between the nut and the hub was to attempt to increase the joint clamping force (more surface contact) enough to mitigate this backlash --- for some it worked, for some it did not..... depends on the loads and configuration in each application.

Most (nearly all) modern race cars & road cars (the few that actually still use them) went to keyed center nuts to eliminate any concerns. The exception being F1 cars, where minimizing pit stop times (for wheel changes) is too crucial, so they take the risk of a nut coming loose ---- which they rarely do because most F1 cars also used tapered splines to reduce backlash to essentially zero.

....... A practice they 'stole' from Aerospace, where we developed tapered splined joints for articulated gimbal transmission joints about 35 years ago.

Other than F1 and Aerospace few can justify the cost of the precision machining involved to make tapered splines and the sophistication level of design analysis required to ensure that allowable contact stresses are not exceeded.

 

Many thanks, Casey. I am still trying to understand this, but the points you mentioned from the article seem valid to me. I think the main point I drew from it (and trying to understand the formula) is that the load point on the inner circle makes a full 360 degrees ahead of the load point on the outer circle, because the load point on the inner circle has a shorter distance to travel. The two circles are not concentric, so the load points will not coincide always. That difference places the load point on the inner circle (if the wheel is rotating clockwise) ahead (e.g., now at 6:00) of the original load point on the outer circle (e.g., still at 5:00). The "removable" part - the spinner - will then either move clockwise or counterclockwise, depending on whether it is the inner (male taper) or outer (female taper) circle. Or so I think.

 

Will, if its precession that moves the spinner, and therefore rotation that moves the spinner microscopic amounts over time, then it seems time spent might be more crucial than torque. I read that the force of mechanical precession is quite powerful.

 

Thanks, Solid State. Yes, of course, we know where the nut will go. But I am not sure that is an analogous example. As I said, if that is the force that were loosening (or tightening) spinners, then both RW and Chapman (Lotus Elan) wheels would tighten in the same direction. But they don't, and both don't spin off. That means either: (1) there is no physics principle involved - R-W and Chapman were propagating hogwash - and you can thread spinners any way you like ; or (2) there is a principle that applies to both, the different effect on each being dur to their respective tapers.

One experiment I read is this - hold a round pencil at one end in one hand somewhat firmly and the other end loosely in the other hand. then move both hands in a circle before you without rotating the pencil. The pencil should rotate on its axis.

You are right we don't need to know the answer. It just bugs me that I don't.

best,

Andy

 

David, Willy,

Thanks for taking the trouble to explore this and explain. But if it were the clearance between the splines that causes the movement, why does a bicycle pedal's spindle, which is tightly screwed into the crank (no splines), come loose if it were not "handed" correctly?

I'm not arguing the point. but just trying to understand. From the explanations, including the link mentioned several times above, there is a always a microscopic clearance between two parts screwed tightly together. This clearance allows the "wobble," the epicyclic motion illustrated in the moving drawing, that allows one part to creep out of alignment with the other.

 

It seems Formula One cars would lose wheels regularly were it not for the locking devices on the hubs. The host opens by saying, "One would expect wheels to fall off road cars, but why would they fall off million-dollar Formula One cars, with all those expensive mechanics?" Perhaps the answer is that they should have listened to Colin Chapman.

https://www.youtube.com/watch?v=PeIqXgaoSNM

 

British cars are said to have a habit of shedding parts on the driveway. I was determined not to have that happen to me. When I was building and racing Lotuses (I had an Elan and a 23 sports racer), I read whole books just on fasteners, mostly by Carroll Smith. Smith is very entertaining author, by the way; he was the chief mechanic on Carroll Shelby's GT40 development program (remember they whupped Ferraris in Le Mans). But he wasn't a very good driver apparently; I asked Carroll Shelby over dinner about him, and Carroll said he kept crashing into everything in sight. This is just to explain that I am OC by nature, so please bear with me.

Charles Darwin, having spent months on board the Beagle studying penguins, said "Now I know more about penguins than I care to." I sometimes feel that way about knock-off wheels.

 

I think this may go back to some concept in physics? I vaguely recall in college physics class, principles related to conservation of rotational equilibrium or something to that point. I clearly remember the professor's demonstration: standing on a rotating platform, he held an axle with a spinning bicycle wheel, which made the platform spin. He then flipped the axle so that the bicycle wheel was spinning opposite - the net result was that the platform on which he stood, stopped and then began to rotate in reverse, without any other action on the professor's part. Have no idea if something like this is in play with a knock-off, just an idea. A friend has an Aston DB4. Finishing the restoration, he went for a test drive, and quickly the wheels came loose, repeatedly. I suggested he had the L/R handed hubs on the wrong sides. Indeed this was the case. When he swapped the hubs to the correct sides, the problem went away.

 

The problem is not caused by backlash in the rotational direction but by loading in the radial direction. Thats why it is not related to braking or acceleration. If you were to drive the car at a constant speed the very small movement of the wheel vs the hub/spinner in a vertical direction would cause the spinner to unwind if wrongly threaded.
The spinner mating surface is a slightly different diameter than the wheel mating surface owing to the spinner taper not being in uniformly perfect engagement because of loads, so if they are turning at the same RPM one will attempt to turn slightly faster than the other at the circumference and relative movement will occur hence the spinner will try to move.
On the Elan the wheel surface is the larger diameter whereas on the other system the spinner is the larger diameter so the resulting movement is opposite.

Chapmans demonstration would have easily shown this. He likely had a small wheel on an axle and a larger ring around it and ran it along a desk. The larger wheel would have turned more slowly, being driven at the bottom. That would be the same relative movement (exaggerated) of the spinner vs wheel.