Flange offset "equalizing" question

[BikeGremlin]

I got this question on my YouTube video "How the number of speeds affects rear wheel strength," and that's far from the first time it's asked, so I hope that writing the answer here will help more people (YouTube comment search is pretty poor at this time):

If i just remove the freehub of sora which is wider and replace it with 7 speed specific narrower freehub. That wont help in any way right? In terms of angle of spokes to tue wheel.

The video:

 

[BikeGremlin]

The basic idea (if I got the question correctly) is to make the wheel be stronger when facing lateral forces. You see, rear wheels need to accomodate for the cassette, and that puts right hand side spokes in an awkward position - with a pretty poor bracing angle:

As you can see in the pic. above: the TL angle is awesome, but the TR angle is almost vertical, so not much bracing against the rim moving away towards the left.

Now, if we use the same total hub width (limited by the frame's rear dropout width), but make the cassette and its freehub narrower (say by switching to a 7-speed cassette), could we make the spoke angles (TL and TR) more equal?

If we just put a narrower freehub, we still need to move the flanges a bit to the right, otherwise, the cassette would just sit closer to the bicycle's centre-line. Basically, this is what we are aiming for:

The picture exagerates a bit, but I tried to show the principle:
We are using a narrower freehub, but we are also putting some narrower spacers on the right hand side (not always possible, even with cup-and-cone hubs), and putting some wider spacers on the left hand side (usually possible, you just add a spacer from an old hub between the locknut and the cone).

EDIT
I need to correct myself:
With cup and cone hubs, the right hand side cone would rest against the new, narrower freehub (because the freehub also functions as the right hand side bearing's cup).
So, it would "automatically" move the whole hub towards the right hand side.
All you'd need to do is find a spacer to put on the left side, to accomodate for the hub's movement. So, it should be a relatively easy "hack."

I hope you can see in the picture how that would result in a more even bracing angle for the TL and TR spokes, with the rim remaining in the bicycle's centre-line.

Note:
For a more uniform spoke tension (so left spokes don't as easily unwind or break from fatigue), making the right hand side hub flange move further to the right can help to a degree, even if the left hand side flange remains in place (if we could somehow re-machine the hub to make the right hand flange stick further out). However, the way it is shown in the picture above, both flanges are moved towards the right. That simultaneously makes the TL bracing angle a bit "worse," and the TR bracing angle better. So it "doubles" the effect.

Note 2:
That is the reason why, from the factory, the left hand side flange is moved a bit towards the hub's centre, even though there is no cassette and it could have been placed all the way to the left. In that case, TL spokes would need to be very loose in order for the rim to stay centred, and without the TR spokes needing a 500 Kgf of tension.
Still, there needs to be some minimal total width between the two flanges, otherwise the wheel would be too unstable when handling lateral loads, even if the left and right hand spoke bracing angles were equal.

Hope this helps explain the problem and the solution.

Relja

P.S.
This video explans the lateral and radial stiffness, strength, and why wheels are laced and built the way they are:

 

[sprinter]

The basic idea (if I got the question correctly) is to make the wheel be stronger when facing lateral forces. You see, rear wheels need to accomodate for the cassette, and that puts right hand side spokes in an awkward position - with a pretty poor bracing angle:

View attachment 305

As you can see in the pic. above: the TL angle is awesome, but the TR angle is almost vertical, so not much bracing against the rim moving away towards the left.

Now, if we use the same total hub width (limited by the frame's rear dropout width), but make the cassette and its freehub narrower (say by switching to a 7-speed cassette), could we make the spoke angles (TL and TR) more equal?

If we just put a narrower freehub, we still need to move the flanges a bit to the right, otherwise, the cassette would just sit closer to the bicycle's centre-line. Basically, this is what we are aiming for:

View attachment 306

The picture exagerates a bit, but I tried to show the principle:
We are using a narrower freehub, but we are also putting some narrower spacers on the right hand side (not always possible, even with cup-and-cone hubs), and putting some wider spacers on the left hand side (usually possible, you just add a spacer from an old hub between the locknut and the cone).

I hope you can see in the picture how that would result in a more even bracing angle for the TL and TR spokes, with the rim remaining in the bicycle's centre-line.

Note:
For a more uniform spoke tension (so left spokes don't as easily unwind or break from fatigue), making the right hand side hub flange move further to the right can help to a degree, even if the left hand side flange remains in place (if we could somehow re-machine the hub to make the right hand flange stick further out). However, the way it is shown in the picture above, both flanges are moved towards the right. That simultaneously makes the TL bracing angle a bit "worse," and the TR bracing angle better. So it "doubles" the effect.

Note 2:
That is the reason why, from the factory, the left hand side flange is moved a bit towards the hub's centre, even though there is no cassette and it could have been placed all the way to the left. In that case, TL spokes would need to be very loose in order for the rim to stay centred, and without the TR spokes needing a 500 Kgf of tension.
Still, there needs to be some minimal total width between the two flanges, otherwise the wheel would be too unstable when handling lateral loads, even if the left and right hand spoke bracing angles were equal.

Hope this helps explain the problem and the solution.

Relja

P.S.
This video explans the lateral and radial stiffness, strength, and why wheels are laced and built the way they are:

Hi, Relja

I was the one who asked the question on YouTube. Thanks for a detailed visual explanation and I also got to know that such a great forum exist through the link you provided.

In the video you mentioned that having wider freehubs on latest sora and 105, that aggravated the problem. So I thought why not simply replace the wider freehub on sofa or 105 hub with narrower 7 speed freehub and that 7 speed freehub can only accommodate 7 speed cassette. Perhaps that could help with better bracing angle to make wheel stiffer. I see through your explanation that is not the case

Thanks again

 

[BikeGremlin]

Hi, Relja

I was the one who asked the question on YouTube. Thanks for a detailed visual explanation and I also got to know that such a great forum exist through the link you provided.

In the video you mentioned that having wider freehubs on latest sora and 105, that aggravated the problem. So I thought why not simply replace the wider freehub with narrower 7 speed freehub and that 7 speed freehub can only accommodate 7 speed cassette. Perhaps that could help with better bracing angle to make wheel stiffer. I see through your explanation that is not the case

Thanks again

Yes - the 7-speed hubs come properly designed from the factory.
With an 8-9-10 and 11-speed road hubs, changing the freehub alone would still leave the flanges in place as they were, so it won't make a difference without placing an extra spacer on the left hand side, and making the right hand side spacers thinner to accomodate.

But, I need to correct myself:
With cup and cone hubs, the right hand side cone would rest against the new, narrower freehub (since the freehub also functions as the right hand side bearing's cup).
So, it would "automatically" move the whole hub towards the right hand side.
All you'd need to do is find a spacer to put on the left side, to accomodate for the hub's movement. So, it should be a relatively easy "hack."

Relja

 

[sprinter]

O

Yes - the 7-speed hubs come properly designed from the factory.
With an 8-9-10 and 11-speed road hubs, changing the freehub alone would still leave the flanges in place as they were, so it won't make a difference without placing an extra spacer on the left hand side, and making the right hand side spacers thinner to accomodate.

But, I need to correct myself:
With cup and cone hubs, the right hand side cone would rest against the new, narrower freehub (since the freehub also functions as the right hand side bearing's cup).
So, it would "automatically" move the whole hub towards the right hand side.
All you'd need to do is find a spacer to put on the left side, to accomodate for the hub's movement. So, it should be a relatively easy "hack."

Relja

Oh, wow! Yes, I have a cup and cone hub, so what you suggested would work. I think you meant having a spacer would somehow move the whole " wheel" towards the drive side, that would in turn place the hub in center. The hub wouldn't literally move as it is attached to the spokes.

Currently 5mm is off from the dead center so having a narrower freehub would almost make it perfect bracing equal angle on both sides.

Additionally if you could please advise on which is better from the following two

A 50 mm narrower flange to flange distance hub with a perfect bracing angle, equal on both sides.

Or

A 64 mm wider flange to flange distance but with conventional problem of unequal bracing angle

 

[BikeGremlin]

O

Oh, wow! Yes, I have a cup and cone hub, so what you suggested would work. I think you meant having a spacer would somehow move the whole " wheel" towards the drive side, that would in turn place the hub in center. The hub wouldn't literally move as it is attached to the spokes.

Currently 5mm is off from the dead center so having a narrower freehub would almost make it perfect bracing equal angle on both sides.

Additionally if you could please advise on which is better from the following two

A 50 mm narrower flange to flange distance hub with a perfect bracing angle, equal on both sides.

Or

A 64 mm wider flange to flange distance but with conventional problem of unequal bracing angle

Yes, I meant the whole hub body (both flanges) would move towards the right hand side if the wider freehub is replaced with a narrower one.
Then, since the dropouts are the same width, one would need to add some spacer to the left hand side.

The question is good. Equal bracing angle should not be a goal in and of itself. The basic idea is to have the right hand side spokes (on rear hubs that are very asymmetrical) be a bit further from the hub's centre, so they have a decent bracing angle.

The left hand side spokes need to be moved towards the centre just enough so the rim can stay in place without those spokes having a very low tension.
For example:
If the right side spokes need a tension over 150 Kgf in order to keep the rim centred and the left side spokes having at least 60 Kgf tension, then the left flange should be a bit closer to the centre, so that with 60 Kgf left spoke tension, the right side spokes can stay under 150 Kgf.

Once that criteria is satisfied, it's better to have a larger total bracing angle.

So, I would rather go with 64 mm flange distance, but only if I can get the spoke tensions right with that distance.

On front hubs with disc brakes it's the left hand side spokes that are closer to the hub's middle and we have some asymmetry there too.

Not sure how well I've explained this.

Relja

 

[sprinter]

You explained perfectly

 

[sprinter]

Yes, I meant the whole hub body (both flanges) would move towards the right hand side if the wider freehub is replaced with a narrower one.
Then, since the dropouts are the same width, one would need to add some spacer to the left hand side.

The question is good. Equal bracing angle should not be a goal in and of itself. The basic idea is to have the right hand side spokes (on rear hubs that are very asymmetrical) be a bit further from the hub's centre, so they have a decent bracing angle.

The left hand side spokes need to be moved towards the centre just enough so the rim can stay in place without those spokes having a very low tension.
For example:
If the right side spokes need a tension over 150 Kgf in order to keep the rim centred and the left side spokes having at least 60 Kgf tension, then the left flange should be a bit closer to the centre, so that with 60 Kgf left spoke tension, the right side spokes can stay under 150 Kgf.

Once that criteria is satisfied, it's better to have a larger total bracing angle.

So, I would rather go with 64 mm flange distance, but only if I can get the spoke tensions right with that distance.

On front hubs with disc brakes it's the left hand side spokes that are closer to the hub's middle and we have some asymmetry there too.

Not sure how well I've explained this.

Relja

Just another question that popped in to my head People have suggested to go for a 3x on drive side and 2x on non drive side. Now that I will have equal bracing angle almost I don't need to do such pattern right ? Also I can keep the same tension on both sides?

It's a 36 holes stock durable rim with shallow aero profile with 50 mm flange to flange distance hub.

 

[BikeGremlin]

Just another question that popped in to my head People have suggested to go for a 3x on drive side and 2x on non drive side. Now that I will have equal bracing angle almost I don't need to do such pattern right ? Also I can keep the same tension on both sides?

It's a 36 holes stock durable rim with shallow aero profile with 50 mm flange to flange distance hub.

Briefly put: I see no point in going with fewer crosses than optimal, even on the left hand side.

In more detail:

1. Optimal number of crosses​

The optimal number of crosses is:
number of holes, divided by 9, then rounded down.

For 32 spokes that is 3.55, so the rounded down result is 3.

For 36 spokes, that is exactly 4.0, with no rounding down, and 4 crosses can lead to spokes overlapping the adjacent spoke heads, which is not ideal (depends on the rim and the flange diameter).

So, with 36 spokes, the optimal number of crosses is often 3.

2. Going with fewer crosses or radial?​

Generally, I see no reason to go with a fewer number of crosses than optimal. With one caveat:

Radial lacing results in spokes always gaining tension when pedalling (for the rear wheel) or when braking with a disc brake. With crossed spokes, leading spokes lose some tension when you pedal.

For highly asymmetrical hubs, where you can't get enough tension on the left hand side spokes, you could avoid the problem of left (non drive side) spokes unscrewing by lacing them radially. However, in that scenario, the better option is to just use thinner spokes on the left hand side (say 2 mm on the right and 1.8 on the left). Note that swagged ("double" or "triple butted") spokes are more durable yet elongate more because of the thinned down mid section, so that too could be a good option. I mean: you could use the 2 - 1.8 - 2 mm swagged spokes on both sides and be good. But, with extremely asymmetrical hubs, you might need to use 1.8 - 1.5 - 1.8 on the left side, paired with the 2 - 1.8 - 2 mm spokes on the right (or 2.3 - 1.8 - 2 mm "triple butted" spokes such as DT Swiss Alpine III model).

Modern hubs are pretty stiff and they transfer pedalling torque to the left flange too, so there is no reason to not use the optimal number of crosses on the left, to make the load among both sides more equal.

My experiment on torque transfer:

Also, fewer crosses, as well as radial lacing, put more stress on the hub flanges, and on the spokes.

3. What about the bracing angles?​

The difference is negligible in practice. Especially if you don't go with radial, but just go with 2x instead of 3x.

You might find this article interesting on this topic (the article, and my comments in the section 3):
SPOKE LACING: What makes Sense and what is Nonsense! Efficiency Comparison using Advanced Engineering software

Relja

 

[sprinter]

Thanks for a detailed explanation. So it's 3x on both sides then. Also thanks for pointing out bracing difference is negligeble in practice