It appears that you are looking at compressive stress on the rim, but most rims these days are failing at the nipple holes. Some are much more prone to this than others, and I think the alloy used has a lot to do with it. The shape and thickness and contour have an effect too, of course.

Maybe I'm just lucky (knocks on wood and strokes rabbit's foot), but the number of my builds that have suffered from this is zero out of more than a thousand... though I hear stories all the time on the forums. Someone even managed to crack several Kinlin XR270s... though this seems like one of the least likely rims for this to happen to.

So... I think the build details matter also. Things like chamfering the nipple holes, forming the spokes above the nipple, using a marine antiseize, and thorough stress-relieving.

The 340s are very light and still pretty new, so this should give anyone cause to wonder about their durability. Stan's rims are made using 6061 aluminum, which should be a good choice... my favorite for hubs also. It isn't a very strong or exotic alloy, but it more than makes up for this in these applications by having an excellent resistance to stress corrosion cracking.

The people at Stan's seem confused about tensions and spokes and wheelbuilding in general. Even though the tubeless tires can reduce spoke tension by up to 30kg they still claim that the you cannot exceed 95kg on the DS without a tire. Well... if I don't use a tubeless tire, then I'm running 90kg DS and 45kg NDS... and if I do use a tubeless tire, it's 65kg and 32kg. Since no one rides without a tire installed, what sense does that make?

Also CX-Rays are a fine spoke to use... and they are just like any other light spoke. They do not "require or work best with high tension". Compared to 2mm straight gauge spoke, they will be stretched ~80% farther at a particular tension level, which means that much greater deflection can be tolerated before they go slack. Of course they are less stiff as well, but you can deal with that by optimizing your bracing angles or adding more spokes.

Another thing about "max tensions" for rims. The static load isn't the only consideration, because you have some control over the cyclic loading (and unloading) also. With radial forces, the spoke tension decreases *below* static levels. Lateral loads can of course result in a tension increase, but these are not high very often and also vary considerably depending on riding style. Torque seems like the most severe service, especially if you are heavy and/or ride on steep grades often. The pulling spokes can easily see a 30kg increase with every stroke (and the pushing spokes, a corresponding decrease) on typical rear wheels with 12 crossed spokes. You can get a lot of stress cycles if you climb a lot.

Which makes me wonder about alternatives like 1x DS and 2x (or 3x) NDS for rims that might be prone to cracking. There are some issues with using the NDS as the primary side for torque transfer... mostly because the spokes are lightly tensioned and can go slack fairly easily. But they will also see lower max tensions due to torque. A large NDS flange helps... and not too big of an offset. Most hubs (at least the S drivetrain models) can be laced 1x heads-in on the DS and still have derailleur clearance. The Rotax hub looks like it might be a good candidate for this lacing, but I'd have to see the offset dimensions.

I emailed Brandon at the BikeHubStore regarding the Superlight and Ultralight hubs that he sells. I will post the geometries, which are identical. The rear hub is the same for both Super- and Ultralight sets. The difference up front is the Superlight is using 2x 699 Enduro bearings and a slightly larger axle, while the Ultralight is using 4x 688 Enduro bearing and a slightly smaller axle. I assume the axles are 9mm and 8mm respectively, since those are the bearing IDs.

2 bearings will be less expensive to upgrade/replace later on down the line than 4, but which setup do you think would be best - are there any upsides of the larger bearings/axle that would justify the weight for the same cost? Would 4x 688s be better than 2x 699s? I see the extra bearings as a drag disadvantage, but loading capacity advantage; however, as I intend to never tip the scales at more than 175lb, does the extra load capacity even matter? Obviously this being my first build, I've not had any experience with either.

I'll also include the Rotax dimensions I was sent.

Thanks RRuff for your insight.

I use BHS super lights and ultralights all the time. Brandon has never let me down. I've built Stans 340s only with alchemy. The hubs are comparable weights, so for a durable build for you go 24f 1x or 2x and 28r 2x/2x. Using CX-Rays will build to 1260 grams. Nice and light but durable. I've been targeting 110kgf drive side. Drops to about 105kgf with a tubed tire at 8 bar. Any less and the nds tension will be too low, and you may start snapping heads on the nds over time.

Brad

_________________
http://www.zedwheels.com - wheels made just for you

Are you still planning on the Rotax? Need the offsets from the center of the flanges, plus the hole circle diameter would be nice too.

The other rear hubs have pretty poor flange offsets... at least for an S drivetrain. The DS offset is quite small, and the NDS large... which will result in a high tension ratio.

NDS offset from center to flange looks to be 27.4, DS looks to be 16.1. Hole circle diameter for both NDS and DS looks to be 60, right ?

Yes, the 60mm hole circle is in the diagram, and I see that we can at least estimate the offsets too. Assuming a 3mm flange width it's ~17.6mm DS and ~28.9mm NDS.

That's a very narrow NDS offset.

Is there an optimal bracing angle? It seems the larger flange hole circle diameter would make up for the narrow offset. Has anybody in the bike world done a parametric study, and if so, what is oprimal?

The larger flange diameter helps a little, but it is very slight.

For most applications, making the DS offset as great as possible (which is around 19mm for S or 17.5mm for Campy) and then making the NDS ~2x that seems ideal.

So, given those hubs, would a switch to a rim that was more tolerant of higher tension be appropriate? BHS also sells kinlin wheels: would an xr200 or xr270 be a better, albeit heavier choice?

No. Higher tension would help if the tension ratio was high... but the problem with this hub is that it is low... ie lateral stiffness is lacking. Low NDS tension won't be a problem.

If you want to use this hub, I'd lace it 1x DS and 2x NDS... heads in both sides. This is an S drivetrain, right?

So I have been riding the wheels for over a week.

They spin up faster than my Dura Ace 7850's and hold speed about the same. They are not as stiff as the 7850's but it is not so bad that they hit the brake pads. Overall ride is comfortable and I do not think about them so that is very good.

The only issue so far is I hit a pothole and even with 34mm tires I managed to pinch flat the tubeless tire and dent the rim. I know that I have done worse to my 7850's as I did an entire lap of a cross course on a completely flat tubeless tire and only had a small nick in the rim. Perhaps I am expecting too much from an ultralight rim as there are always trade offs. On the plus side the rim did stay round and true despite the dent in the bead hook.

Is that a typo?
Or do you use 34mm tires on a Alpha 340 rim (and on a road bike)?
Which 34mm tires?


/Håkan
SWEDEN

They are on my lynskey procross

Would the circus monkey hubs be a better choice in terms of hubs then? The DS offset is 19mm, the NDS offset is 40mm. Those are offsets to the outside of the flanges, with a flange thickness of approx. 3.5mm (measuring off the scale drawing). NDS and DS PCDs are 38mm and 47mm, respectively. Are those better dimensions. It blows my mind that 1.xx mm would make THAT much difference in a rim build. With a ERD of 591mm, and offsets of 19mm and 40mm as for the circus monkey hub, we are talking DS and NDS bracing angles of 3.68 and 7.71 degrees. A 1mm reduction in offset results in a .2deg reduction, or about 6% reduction. Is that really noticeable if we are not riding really close to the limit of the wheel integrity? I would love to build a few sets of wheels with identical components, just changing hubs to see what actually happens. Please point me to any threads showing either field or lab tests so I can get an idea of what kind of difference .2degrees bracing angle makes, both in terms of absolutes and in relation to the rider. I am a novice, and I love building stuff, and really want to educate myself. If the geometry of a bike wheel requires that precise of resolution, and that 1mm is that substantial, I need to rethink this whole bike and wheelbuilding thing and invest in some awesome hubs.

Edit: need to look at this more trigonometrically. I forgot to take the difference in PCDs into account when making those calcs above. Will have to come back to this later. One question though : what is the minimal acceptable tension for a proper wheel build? Not taking the different PCDS into account, with 100(units of tension) on the DS, I was coming up with about 47(units of tension) required to balance it on the drive side, or a little less than half. Obviously that is incorrect with different PCDs (will get back to the math after dinner ).

Edit #2: Did a little trig/statics homework. I figure with the BikeHubStore superlight rear hub, if I lace a 24h in a triplet arrangement (BHS does offer that 8nds/16ds drilling), I should be at ~90kgf on the NDS and 100kgf on the DS. Figure attached. Look good? I know it's not as awesome a hubset as the Elf/Orc or any other true boutique hubsets - those will have to have to be purchased when I want to build a set of sweet carbon tubs. Why would you go all out on a set of hubs, then lace them to some inexpensive alloy hoops anyway?

Ron will probably post more pointed replies to your work above..

I think you'll find that a triplet arrangement won't work for this build. Yes, it balances spoke tensions better than anything out there. However, 24 spokes is too little for these rims at 175 lbs. Also, a triplet 24 hole will be laterally weaker than a 2x/2x 24 hole, unless the NDS flange offset is huge... Like Fulcrum and Troy do.