Hub geometry

[morganb]

I've seen some breakdowns of this before but didn't quite grasp everything. What makes a hub geometry good (for a given purpose). Flange height, offset, etc. I know a few hubs are regarded as having geometry that builds stiff wheels (Wheels MFG/Alchemy, WI) while others less so DT 240.

[ergott]

Flange diameter plays a much smaller role so for brevity I'll skip that.

Wider flange spacing gives better lateral support to a rim. The limitations are the fork clearance on the front wheel and the cassette on the rear wheel (more on the left, rear in a moment). With those limitation you can have a front hub flange spacing of about 39mm from the center of the hub and use it on any bike (radial laced with spoke heads facing out).

For the rear wheel, not on the cassette is in the way, you don't want the rear derailleur to touch the spokes when in the lowest gear. For most bikes this means up can put the flange 18mm from the center of the hub. There really isn't any reason for using less than 18mm other than greater room for error.

Now for the left, rear side. This is a compromise. If you put the flange too far from the middle of the hub the result is spoke tension that is too low to support the rim under riding loads. For example, if the right flange is 18mm and the tension is 120kgf on each spoke, you have 60kgf on each spoke going to the left flange (assuming equal number of spokes on each side). I personally think about 55kgf is the absolute minimum because spokes are likely to go slack under load (bumps in the road, accelerating/climbing out of the saddle, etc.) . When that happens, durability is greatly reduced and spokes break more often. You also end up with a wheel that won't stay true and will have poor lateral integrity.

Because of this I personally prefer the spacing of the Wheels Manufacturing hub (formerly Alchemy). These factors were all considered and I think achieve the best compromise. The resulting tension is high enough for a durable wheel and the lateral stiffness of the wheel is at it's greatest potential.

One other design aspect that can't be ignored is axle flex. In order to further reduce lateral wheel flex caused by the hub you want to support the bearings as close to the axle ends as possible. This is another aspect of the Wheels Mfg hub that was considered and makes its design better in my opinion.

People have other hubs that they prefer, but those are the reasons I think Wheels Mfg hubs are the best design out there. They are also what I consider to be competitively light. All the lighter hubs I've seen have missed out on either the flange spacing or bearing placement. Compromises in each design have to be considered when choosing what's best for you.

[sugarkane]

Amen brother

[morganb]

Thank you, this is pretty much exactly the information that I was looking for. That explains why NDS tension seems to be a problem on most wheels. One other question about front wheels with radial lacing. I have always built with the spoke head on the outside of the hub, but realized the other day the other way around should give a slightly better bracing angle. Is there any reason not to do this other than aesthetics? I've seen them laced that way on cheap OEM wheels but thats about it.

[lwk]

Interesting topic. I’m in the market for a set of road hubs and have noticed that rear hubs are also laced 2.1 . This apparently helps with the NDS spoke tension? How would the following two builds compare with the same type of hub with straight pull spokes: 28 spokes(14/14) and 24 spokes(2.1)?

[kavitator]

28 spokes will be stiffer.

2:1 lacing - it is good because nds flange can be wider. But for that spoke angle you need special drilled rims.

[ergott]

Thank you, this is pretty much exactly the information that I was looking for. That explains why NDS tension seems to be a problem on most wheels. One other question about front wheels with radial lacing. I have always built with the spoke head on the outside of the hub, but realized the other day the other way around should give a slightly better bracing angle. Is there any reason not to do this other than aesthetics? I've seen them laced that way on cheap OEM wheels but thats about it.

Some flanges can't take the load of elbows out. Another issue I have seen when I did it is increased bearing bore warp. It would make the bearing fit too loose under tension, enough that the bearing would fall out if you turn the wheel on it's side without the axle in place. This doesn't mean it will happen to every hub though.

Finally, you have to make sure there's enough clearance for the fork. With a really wide flange spacing it could interfere.

[ergott]

Interesting topic. I’m in the market for a set of road hubs and have noticed that rear hubs are also laced 2.1 . This apparently helps with the NDS spoke tension? How would the following two builds compare with the same type of hub with straight pull spokes: 28 spokes(14/14) and 24 spokes(2.1)?

Same as in same flange spacing for both?

If you are talking about triplet specific hubs then triplet 24 can be made stiffer than 14/14 28 spoke. It's just really hard to find hubs and rims that are compatible. I have a couple here. It would be better if the rims were drilled for the purpose so the spoke runs straighter into the rim. As it is it has to take a sharp bend as it exits the nipple and rim.

[Marin]

Front flange spacing also plays a big part in wheel aerodynamics.

Wider is better for stiffness, but most people don't need a front wheel to be much stiffer than a rear, and spokes make more of a difference in typical low yaw conditions than a tall rim.

Regarding aero, at low yaw a tall rim has the main advantage that it allows for shorter spokes.

[lwk]

Thank you kavitator and ergott for your answers. @ergott, yes same flange spacing

[ergott]

I spoke with Jeremy at Alchemy hubs (at the time). His math (and I trust it more than mine) revealed that using the same flange dimensions and comparing triplet 24 spoke triplet to 24 12/12 you would have a laterally stiffer with wheel with the 12/12. The reason being that halving the left spokes without compensating by moving the flange further outboard is a bigger detriment to lateral stiffness than you recover with the 16 spokes on the right side.

I haven't seen it tested, but it makes sense to me. I'd only consider triplet with hubs designed for it. Unfortunately there aren't many around. It's a comparatively tiny segment of the market.

[lwk]

I spoke with Jeremy at Alchemy hubs (at the time). His math (and I trust it more than mine) revealed that using the same flange dimensions and comparing triplet 24 spoke triplet to 24 12/12 you would have a laterally stiffer with wheel with the 12/12. The reason being that halving the left spokes without compensating by moving the flange further outboard is a bigger detriment to lateral stiffness than you recover with the 16 spokes on the right side.

I haven't seen it tested, but it makes sense to me. I'd only consider triplet with hubs designed for it. Unfortunately there aren't many around. It's a comparatively tiny segment of the market.

Once again thank you. I’ll look for normal j bend hubs then

[bm0p700f]

Some hubs also big ds flanges and disc brake hubs have big disc side flanges. The reason is not to improve bracing angle as it does not in any meaningful way. What bigger flanges do is reduces the loads placed on the spoke elbows. While this is reasonably important for disc brakes it is inconsequential for the DS as pedalling torque is not large unless you are in a very low gear.

Ergot is perfectly correct about flange spacing but riders like me can have a reliable wheel with very low nds rear tensions. Not every rider can but i dont get get spoke slacking off. The hubs I have used in these wheels are 24 spoke rear miche supertype and SWR for 1:1 lacing. Radially stiff rims are needed as nds rear tension is around 300N but because the flange is 49mm from centre. The result is though a very stuff wheel. I know wheels with thus hub have been tested in miche torture chamber which is a jug with a wheel and a 50 kg load and the rear wheel is driven and rocked severely. I saw thus jig in March and it is a severe test. Wheels built with such hubs like thus must be laced to rims which can take high tensions and spoke tensions must be very even. It is not for everyone for sure but it can be made to work. A hub like this is still better in a 2:1 pattern but there will be a loss in lateral stiffness. When talked to miche wheel designer in March they determined the old SWR hub (triplet) and the new SWR hub (and both hubs as a centre nds rear flange 49mm from centre of the hub) showed a 15% improvement in lateral stiffness for the new hub.

For a triplet hub to be stiffer than a normal hub the nds rear flange has to be 45mm from centre or greater. The triplet hub i want miche to make next year will have a nds flange 49mm from centre and i will have to have triplet drilled rims made in carbon and alloy. This is going to be very expensive but it is the only way as ergot has pointed out before to maintain nds spoke tension with tubeless tyres. A wheel with such a triplet hub would be stiffer than a 24 spoke conventislly laced hub with more normal flange spacings but will have much higher nds tension especially when tubeless tyres are installed. A asymmetric rim works well with is this kind if hubs as it increases nds tension further as a 3mm offset changes the tension balance from 65% to 83%. You also get a reduction in the bend and the nipple and there is only so much angled drilling you can do in a road rim.

[bm0p700f]

Oh and as ergot has noted another reason for a wide spaced triplet hub is I can have the nds bearing really close to to the axle ends. There are plenty of hubs like some in bring ranger wheels that have the nds flange and bearing far in board. There is no good reason for this.

[kavitator]

I like that debates - thanks for infos

BTW: RAR has his own new hub- 160g/60g
just saw it on Facebook