Making wheels stronger with a mixed spoke pattern.

[zerodish]

I use spocalc.xls available from several sources. It gives the angle of the spokes as they leave the rim. The wheel shown uses a 542mm rim a 61mm spoke hole diameter 36 holes 24mm center line to right flange and 29mm center line to left flange. If I had built the wheel with 3 cross on both sides the angle would be 5.4 degrees on the right side and 6.5 degrees on the left side. However I used a 4 cross pattern on the left side this gives a 6.2 degree angle. This increases the wheel strength two ways It allows the left side spokes to take a higher load before the spokes go to zero tension. This does happen on a hard bump. A new wheel will ping this is the spokes untwisting caused by the truing process. If you watch a high wheeler bicycle you can actually see the rim go flat on the bottom. It also allow the left side spokes to take more of the torque from acceleration and braking if you use a disk. This will also work on a front disk brake. Use 3 cross on the brake side and 4 cross on the other side. The improvement is only a few percent but it is easy to do and uses standard spokes.   https://www.flickr.com/photos/63373992@N07/13559968714/

[Pat Lamb]

Interesting concept, zerodish.  I'd never seen this suggested before.  If you don't mind, though, let me pulse you with a few questions and comments.

Before I lose 2/3 of the readers, I'll point out that the wheel deflects a bit on the bottom when ridden with a load.  If it deflects enough, the spoke tension disappears (goes to zero), which is bad in two ways.  First, it allows the spoke to unscrew, meaning the wheel goes out of true.  Second, if the spoke doesn't unscrew, the spoke cycling between tight and loose will cause the metal of the spoke to fatigue, and ultimately break like a paper clip bent back and forth.  It's actually better for the spoke to stay tight!

First, four cross is usually used for higher spoke count wheels, and two cross for lower spoke count wheels.  The idea, as I understand it, is to get the spokes coming off the hub at roughly a right angle to the radius through the center of the hub.  Are you advocating going to a 48 spoke wheel?  If you're using 4x with 36 spoke wheels, are you coming off the hub at an acute angle?

Second, the ping you note as a new wheel is ridden is caused by windup of the spoke during tensioning and truing.  This is normally fixed (by a skilled wheelbuilder) by over-correcting and then backing off during final truing.  I don't see this as something that can be corrected by changing the length and angle of a spoke.

Third, the wheel is centered by balancing the tension of the right and left side spokes.  If you're using the same number of spokes on each side, as is the case for every wheel I know of on the market now, you can lengthen or shorten the spokes on the left (non-drive) side, but the tension will have to stay the same unless you pull the rim off-center.  With the same tension on the spokes, keeping the wheel centered, the only change is going to be frictional losses as the (almost) unloaded spoke shifts.  This is unlikely to be significant, and so I doubt you'll change the load the wheel can take before a spoke goes to zero tension.

A better approach might be to replace box rims with a stiffer (V) rim.  The V rim adds some structural rigidity, meaning you share the load across more spokes.  This, in turn, means you can carry a larger load on the V wheel without the spoke losing tension.

[mathieu]

I cannot add much own insight to the question whether a mixed spoke pattern makes a wheel stronger, but I believe it does. Wheel design has advanced a bit further than Pat Lamb is probably aware of.
For at least 10 years my road bike has a set of Campagnolo Proton wheels. The front wheel, 22 spokes, has a radial (i.e. 0-cross) spoke pattern. The rear wheel has 24 spokes, 12 at each side. It has a radial pattern on the non-drive side and a 1-cross pattern on the drive side. I have used these wheels a lot on fast descents on bad mountain roads, including several via sterrata in Italy. They never needed truing.
Campagnolo Zonda wheels that are now on the market have 21 spokes on the rear, a number which is bound to violate the 'same number of spokes on each side' rule. They have 14 spokes on the drive-side and 7 on the non-drive side ; all radial. The spokes are grouped in 7 groups of 3, a pattern Campa calls Mega-G3.

The Zonda pattern may well be the result of a design where the striking looks was a prime criterion. The Proton wheel design, however, is from a time when optics were not yet a big marketing point and the finite-element calculation software for making wheels lighter and stronger was available for the industry. That's why I believe that the mixed spoke pattern for the rear wheel is a functional design.

[DaveB]

First, four cross is usually used for higher spoke count wheels, and two cross for lower spoke count wheels.  The idea, as I understand it, is to get the spokes coming off the hub at roughly a right angle to the radius through the center of the hub.  Are you advocating going to a 48 spoke wheel?  If you're using 4x with 36 spoke wheels, are you coming off the hub at an acute angle?

A 36 hole rim and hub laced 4X will produce a tangential spoke line and is the smallest number of holes that will allow 4X lacing

Second, the ping you note as a new wheel is ridden is caused by windup of the spoke during tensioning and truing.  This is normally fixed (by a skilled wheelbuilder) by over-correcting and then backing off during final truing.  I don't see this as something that can be corrected by changing the length and angle of a spoke.

Correct.  "Pinging" is prevented by proper stress relieving of the spoke line, lubing the spoke threads and preventing or correcting spoke wind-up during tensioning.  The wheel's construction geometry is not a factor.

Third, the wheel is centered by balancing the tension of the right and left side spokes.  If you're using the same number of spokes on each side, as is the case for every wheel I know of on the market now, you can lengthen or shorten the spokes on the left (non-drive) side, but the tension will have to stay the same unless you pull the rim off-center.  With the same tension on the spokes, keeping the wheel centered, the only change is going to be frictional losses as the (almost) unloaded spoke shifts.  This is unlikely to be significant, and so I doubt you'll change the load the wheel can take before a spoke goes to zero tension.

A better approach might be to replace box rims with a stiffer (V) rim.  The V rim adds some structural rigidity, meaning you share the load across more spokes.  This, in turn, means you can carry a larger load on the V wheel without the spoke losing tension.

As noted by mathieu there are asymmetrically spoked wheels these days but how much is structural and how much is a fashion statement is debatable.  One way to balance the required tension differences with a dished wheel is to use thinner spokes on the non-drive side.  Say 2.0/1.8/2.0 on the drive side and 2.0/1.6/2.0 on the non-drive side.

I agree that a deeper section, more rigid rim is also a benefit in reducing the stress load cycling as a wheel is loaded and unloaded.   

[zerodish]

Actually I did not invent the idea. Frank Berto did an article for bicycling called building a bicycle for an adult female. He did radial right and 3 cross left. He pointed out nearly all of the torque would be transferred to the left side which is under less stress. You are certain to bust a flange with radial spoking.  My wheel idea works best with 36 spokes. A 32 spoke wheel could be built with 3 cross left and 2 cross right however 2 cross will be more likely to break the flange. A 40 hole wheel could also be built with 4 cross left and 3 cross right or a 48 spoke wheel with 5 cross left and 4 cross right with no problems. If you are worried about flange breaking use a large flange hub.

[RussSeaton]

If you are worried about flange breaking use a large flange hub.

Really?  Large flange and small flange hubs have the spoke the same distance from the outside of the hub.  The metal the spoke head is bearing against is the same in both flanges.  Large flange merely moves the hole closer to the rim.  Large flange does not add more metal for the spoke head to bear against.  Flanges break with radial lacing because the spoke head is bearing against a small bit of aluminum in the hub flange.  This small bit of aluminum is the same in both large and small flange hubs.  All a large flange hub does is make the spokes shorter.

[DaveB]

You are certain to bust a flange with radial spoking.  My wheel idea works best with 36 spokes. A 32 spoke wheel could be built with 3 cross left and 2 cross right however 2 cross will be more likely to break the flange. A 40 hole wheel could also be built with 4 cross left and 3 cross right or a 48 spoke wheel with 5 cross left and 4 cross right with no problems. If you are worried about flange breaking use a large flange hub.

Very interesting.  Then all of the great number of radial spoked wheels built both individually and commercially are doomed to hub failure?  Yes, years ago both Campy and Shimano refused to warranty their hubs if laced radially but that's way in the past and both companies sell wheels with radial lacing and have removed that warning from their hubs. Your information is way out of date.

Further, as RussSeaton noted, large flange hubs do not tolerate radial or low cross lacing any better than small flange hubs. 

[mathieu]

..... All a large flange hub does is make the spokes shorter.

Yes and no. I don't think shorter spokes is a real advantage, not even in weight. What a larger hub mainly does is that it enables a larger number of spokes. You see them on tandems, where 48 spoke wheels are not unusual. But with a larger diameter flange, the spokes get more under a skewed angle at the rim, which is a recipe for spoke fracture (viz. Rohloff Speedhubs). We now enter the finesses of wheel building.

What I see over a decade or more of wheel building is that commercial wheel sets use less rather than more spokes, higher spoke tension, a tendency towards 0-cross spoke patterns rather than 3-cross or 4-cross, a reinforcement of the points of attachment of the spokes, a reduction of the rim and nave mass where there are no spokes, and a liberty to introduce arty patterns rather than maintaining traditional artisan symmetries. I guess this is all the result of finite-element design calculations that gave much better insight in wheel strength.

[Pat Lamb]

What I see over a decade or more of wheel building is that commercial wheel sets use less rather than more spokes, higher spoke tension, a tendency towards 0-cross spoke patterns rather than 3-cross or 4-cross, a reinforcement of the points of attachment of the spokes, a reduction of the rim and nave mass where there are no spokes, and a liberty to introduce arty patterns rather than maintaining traditional artisan symmetries. I guess this is all the result of finite-element design calculations that gave much better insight in wheel strength.

I suspect it's more a case of finite element analysis being employed to support marketing-driven "artistic" designs, which just happen to require proprietary rather than standard parts.  This is great for the manufacturers (don't you want to buy a new wheel?), but are difficult to repair when you break something on the road during a tour.

[DaveB]

I suspect it's more a case of finite element analysis being employed to support marketing-driven "artistic" designs, which just happen to require proprietary rather than standard parts.  This is great for the manufacturers (don't you want to buy a new wheel?), but are difficult to repair when you break something on the road during a tour.

Any cycling tourist who uses boutique designer wheels on their tour bike deserves whatever problems they run into.  As you noted, these use special spokes and other parts and most LBSs can't repair them with in-stock parts.  Standard hubs, spokes and rims in reasonable spoke count and cross patterns are the only sensible choice.

[zerodish]

I can't make this any easier. If you study first year statics you will find you will find a mixed spoke pattern will improve wheel strength by allowing the non freewheel side spokes to achieve a more equalized tension in a dished wheel.  I had the good fortune to find several books on bolted and riveted bridge structures at Powel's techinical books.  A spoke in a flange is no different than an eccentrically loaded rivet or bolt. They all recommend that rivet holes should be placed twice the distance of the diameter of the hole from the edge in this case the flange. I typically find this distance to be 1.5 times or 1 times if your count the counter sinking on available hubs. There is no excuse for this all hub makers should add the extra few grams to the flange. They also recommend holes be placed 1.5 times the diameter of the hole apart. For 36 holes and 2.6 mm diameter hole this works out to a spoke hole diameter of 74 mm. It is true Shimano and Campagnolo rates some of their hubs for radial spokes but not all of them. Campagnolo also puts a weight limit on their wheels of 80 kilograms or 176 pounds. Do you really want this on a touring bicycle?

[DaveB]

All of this is true and in theory all of these design elements should be met but it's more theory than necessity.  Are bicycle wheels collapsing left and right?  Is failure a daily occurrence?  Obviously a bridge failure can have extremely serious consequences but a broken spoke is more nuisance than disaster so the safety factors don't have to be nearly as high.  As to strength and durability, 32H 3X lacing with standard spokes have proven to give a very durable and solid wheel and more complex spoke patterns add complexity to no useful improvement

As to Campy hubs, Campy does not design for touring or MTB or any other high stress use so complaining that their hubs aren't strong enough for touring is not germane.   It's like complaining that your Miata won't hold a family of four and two weeks luggage.

[dkoloko]

Actually I did not invent the idea. Frank Berto did an article for bicycling called building a bicycle for an adult female. He did radial right and 3 cross left. He pointed out nearly all of the torque would be transferred to the left side which is under less stress. You are certain to bust a flange with radial spoking. 

There was burst in interest in weird spoking patterns at the time Berto built the wheel you mention. He told me he broke a Dura-Ace hub with a weird spoking pattern, and that was the end for him with weird spoking.

[canalligators]

There are many variables, and this discussion has a lot of drilling down to single factors.  You need to look at the bottom line, considering both design approach and construction practice.  In the end, for touring use a conservatively designed wheel built by a good wheel builder.  For most touring applications you should not use radial spoking.  Don't use low spoke count wheels, do use a good brand of spokes in a conservative pattern that doesn't require a tensiometer to build.  Use strong rims.  Make sure they're in good repair before starting your tour.

[mathieu]

In the end, for touring use a conservatively designed wheel built by a good wheel builder.  For most touring applications you should not use radial spoking.  Don't use low spoke count wheels, do use a good brand of spokes in a conservative pattern that doesn't require a tensiometer to build.  Use strong rims.  Make sure they're in good repair before starting your tour.

In June, about 120 racers will start for another edition of the Tour Divide race, 2700 miles from Banff-Canada to the US/Mex border. I guess only a few of them will have wheels that fit your recommendation. All of them will have loads that are substantially less than most tourers, probably between 15 and 30 lbs, but their pace is surely much faster and the dirt roads are much rougher. About one-third of them uses rigid forks. I guess for all of them the dynamic  impact on the wheels is much harder than for touring, whatever the load. The race rules prescribe that in case of a mechanical defect, the racer has to go back to a commercial bike shop; no private assistance or forward movement along the GD route is allowed. From the past editions, I do not remember any wheel defect. If it were a serious risk of modern MTB wheels, it would have shown.

What I want to say: there is nothing wrong with being conservative in chosing wheels, but you should be aware that your bike is probably heavier then needed!