Spoke Gauge Sizing

[Yellow Saddle]

It is almost as though this were a religious matter.

I also see some religious traits in this. Some people have so much faith in something that they treat their own knowledge not as something that can be altered by new evidence but as a matter of blind faith. It is true because god said it is so.

 

[Yellow Saddle]

I already gave the example when I mentioned torque loading. If the rim rotates slightly relative to the hub, the angle of the spoke must change slightly, and the position of the spoke where it bears against the flange must also change because the forces are no longer balanced. You can try this yourself with a laced wheel in which the spokes are fairly slack. Observing this on a fully tensioned wheel is rather more difficult, because it needs extreme loading such as steep uphills or hard braking. The larger the flange hole is, the greater the movement, so it's best to have the spoke as snug as possible, or use a spoke washer.

OK, I just wanted to give you a chance to elucidate your premise.

Torque loading comes with two actions: braking and pedaling. They are the same for the same acceleration but obviously the one is positive and the other negative.

Lets consider a rear wheel because that receives pedaling and braking torque. We'll also assume that the rear traction is infinite so that the wheel will not skid when you brake.

Now, you say torque will create unbalanced forces in the wheel but you don't specify where these mis-balances are.

When you pedal, one half of the spokes (all the pulling spokes) increase tension and all the pushing spokes decrease tension. But since the tension gained and the tension lost are equal but opposite, the nett tension in the wheel remains the same. The exact opposite happens when you brake. The spokes which increased tension under pedaling, now decrease tension and the spokes that lost tension when pedaling, now gain tension. The values are the same and cancel out to zero, again, no nett imbalance of tension.

Further, you say the rim rotates slightly wrt the hub. That is called windup and can be calculated. It is easy to do because we know the mass of the bicycle and rider we want to accelerate/decelerate and we know the acceleration we want to achieve and, we know the gearing, the wheel size and spoke angle. From the spoke angle we can calculate the length of the lever at the hub and therefore the increase/decrease in tension of each spoke. This you'll find on page 42 of The Bicycle Wheel, a book you say you have access to, I think.

Brandt has done the homework for us and you can go and work through it but basically, the strongest rider doing a high-torque standing start on a track only manages to change spoke tension by 5% during that acceleration. I tried to get you earlier to guess what change in tension you expect but you didn't bite. The answer is 5%. With 1000N in a spoke, 5% will raise it to 1050N or lower it to 950N at most and never thus does the spoke ever depart from the contact edge of the hole.

You also allude to some sort of swiveling that happens as the spoke flexes. We'll, we can calculate the windup angle (again, done for you on page 42) and you'll see that it is in the order of 1 degree. We know none of this windup happens between spoke and spoke hole because a used hub and spoke does not display the rouge so typical of aluminium fretting on steel.The windup is simply a non-event for the hub.

But you insist that there is lift-off between spoke and spoke hole. This can only happen if the spoke tension on the pullnig spokes double and the tension in the pushing spokes go to zero. I've demonstrated the fallacy of this with some figures above. If you want to dispute my figures, come up with alternatives and I'll plug them into the model for you. Finally, a washer on the side of the spoke cannot prevent movement inside the hole. Even if the head is head steady in the hole by the taper of the flange chamfer and head chamfer fitting perfectly, the other side will still flay about in your model.

 

[Smurfy]

OK, I just wanted to give you a chance to elucidate your premise.

Torque loading comes with two actions: braking and pedaling. They are the same for the same acceleration but obviously the one is positive and the other negative.

Lets consider a rear wheel because that receives pedaling and braking torque. We'll also assume that the rear traction is infinite so that the wheel will not skid when you brake.

Now, you say torque will create unbalanced forces in the wheel but you don't specify where these mis-balances are.

When you pedal, one half of the spokes (all the pulling spokes) increase tension and all the pushing spokes decrease tension. But since the tension gained and the tension lost are equal but opposite, the nett tension in the wheel remains the same. The exact opposite happens when you brake. The spokes which increased tension under pedaling, now decrease tension and the spokes that lost tension when pedaling, now gain tension. The values are the same and cancel out to zero, again, no nett imbalance of tension.

Further, you say the rim rotates slightly wrt the hub. That is called windup and can be calculated. It is easy to do because we know the mass of the bicycle and rider we want to accelerate/decelerate and we know the acceleration we want to achieve and, we know the gearing, the wheel size and spoke angle. From the spoke angle we can calculate the length of the lever at the hub and therefore the increase/decrease in tension of each spoke. This you'll find on page 42 of The Bicycle Wheel, a book you say you have access to, I think.

Brandt has done the homework for us and you can go and work through it but basically, the strongest rider doing a high-torque standing start on a track only manages to change spoke tension by 5% during that acceleration. I tried to get you earlier to guess what change in tension you expect but you didn't bite. The answer is 5%. With 1000N in a spoke, 5% will raise it to 1050N or lower it to 950N at most and never thus does the spoke ever depart from the contact edge of the hole.

You also allude to some sort of swiveling that happens as the spoke flexes. We'll, we can calculate the windup angle (again, done for you on page 42) and you'll see that it is in the order of 1 degree. We know none of this windup happens between spoke and spoke hole because a used hub and spoke does not display the rouge so typical of aluminium fretting on steel.The windup is simply a non-event for the hub.

But you insist that there is lift-off between spoke and spoke hole. This can only happen if the spoke tension on the pullnig spokes double and the tension in the pushing spokes go to zero. I've demonstrated the fallacy of this with some figures above. If you want to dispute my figures, come up with alternatives and I'll plug them into the model for you. Finally, a washer on the side of the spoke cannot prevent movement inside the hole. Even if the head is head steady in the hole by the taper of the flange chamfer and head chamfer fitting perfectly, the other side will still flay about in your model.

Proving that the tension won't change much isn't the same as proving that the spoke won't move much due to the change in angle during windup.

You've already said that most flange holes fit a standard 2mm spoke, so you wouldn't expect to see fretting in most situations.

 

[raleighnut]

But how does the lacing pattern affect longevity i.e Radial, single cross, 2 cross, 3 cross, (standard) 4 cross, (my favourite pattern for rears) or 5 cross, (used on tandem 40 spoke hubs) or the one I got John at Bob Warners to build for me (4 cross with plain guage on the drive side and double butted on the non-drive side) BTW he said you've been on that internet haven't you but built it to my idea anyway (I blame Sheldon Brown, his idea )

 

[Yellow Saddle]

But how does the lacing pattern affect longevity i.e Radial, single cross, 2 cross, 3 cross, (standard) 4 cross, (my favourite pattern for rears) or 5 cross, (used on tandem 40 spoke hubs) or the one I got John at Bob Warners to build for me (4 cross with plain guage on the drive side and double butted on the non-drive side) BTW he said you've been on that internet haven't you but built it to my idea anyway (I blame Sheldon Brown, his idea )

The worse pattern for longevity is radial, thereafter all things being more or less equal because once the spokes have been seated properly and stress relieved, spoke life is just about infinite. As I said, change in tension from torque is only about 5%, way within steel's maximum fatigue range. To put things in perspective, change in tension from loading could be 20%, depending on the number of spokes used.
The number of crosses affect the spoke departure angle from the hub, the ideal being perfectly tangential (90 degrees). Anything other than tangengial requires a bit of windup before transmitting torque, which for most patterns and hub sizes is negligible.
I would say five cross is too much and may even cause outbound spokes to cross over the heads of inbound spokes if the hub flange is too small. On a hub like Shimano FH-08 (tandem hub) a four-cross gives you the ideal tangential departure angle.
Further, putting plain gauge spokes in any quality wheel is a bad idea. It offers no benefits and compromises spoke life and thus wheel reliability.

 

[PpPete]

No - somewhat premature. Someone wants to carry on talking technical.
Let's talk about grain size and orientation. Anyone done any photomicrography of different brands of spoke at the elbow - or at the transition between 2.0 mm and 1.8 mm ?

Not my work but photographs that used to float around on deja-vu, now Google Groups' Rec.bicycle.tech in the 1990s. This work has been done and is very well understood.

Which only goes to demonstrate the veracity of the long-held adage about USAnians and Irony.

 

[PpPete]

I would say five cross is too much and may even cause outbound spokes to cross over the heads of inbound spokes if the hub flange is too small. On a hub like Shimano FH-08 (tandem hub) a four-cross gives you the ideal tangential departure angle.

Spokes crossing over heads is more likely, for a given number of crosses, as the flange diameter gets larger.
I tend to build dyno hubs with two cross for this reason.
But yes, always with DB spokes.

 

[Spoked Wheels]

But how does the lacing pattern affect longevity i.e Radial, single cross, 2 cross, 3 cross, (standard) 4 cross, (my favourite pattern for rears) or 5 cross, (used on tandem 40 spoke hubs) or the one I got John at Bob Warners to build for me (4 cross with plain guage on the drive side and double butted on the non-drive side) BTW he said you've been on that internet haven't you but built it to my idea anyway (I blame Sheldon Brown, his idea )

Radial, as far as I understand offers no noticeable advantage over tangential spoke lacing, Some front low spoke count wheels can't have X spokes (too few spokes) so they have radial. There is a small increase in radial stiffness and also a small gain in lateral stiffness. Somebody did some testing and found that radial lacing with spoke's heads in increases wheell stiffness by 13% over spokes with heads out. Maybe such increase is not desirable on very wide flanges, I don't know.
There is a school of thoughts that believe a radial pattern on the NDS and a X pattern on the DS is a good combination to increase the spoke tension on the NDS. This is based again, on the brace angle, but this time the spokes heads on the outside of the flange. I haven't tried this myself and I was going to put it to the test on my next set of wheels but I changed my mind I can see, in my mind, a small gain in tension of the NDS spokes but I can also see a small lost of lateral stiffness. Some factory wheels are built this way and some factory wheels even put the radial lacing on the DS..... where one expects to see pulling spokes for torque transfer. (A topic for me to investigate one day )

The disadvantages are too much strain on the hub flanges. Some hub manufacturers specifically say no for radial lacing.

As for plain gauge spoke, I don't use them. I think the advantages of using double butted spokes are there for everybody to see, it's not a secret, I do remember reading the same thing but a long time ago, before I started building wheels, not sure if the advice is still there. Even to this day some people still believe in using 2.0 - 1.8 - 2.0 on the drive side and 2.0 - 1.5 - 2.0 on the NDS perhaps is the same school of thoughts but my guess is that is driven by weight winnies eager to save a single gram

5 cross? Wow.... is there any advantage over 4?
There is a paper written by Henri P. Gavin that might be of interest to you.

The conclusions are:

"CONCLUSIONS
Requirements of strength, stiffness, and low weight are satisfied in bicycle wheels by combining
a light-weight rigid rim with pre-tensioned wire spokes. In many rear wheels very high pre-tension
stresses in half the spokes are required to maintain an asymmetrical shape. The radial, lateral, and tan-
gential stiffness of wheels with various spoke sizes, spoke geometries and rim stiffnesses is presented.
The behavior of bicycle wheels subjected to static radial loads can be accurately modeled by idealizing
the system of interlacing spokes as a linear elastic foundation of uniform stiffness per length of circum-
ference. The spoke pattern affects the over-all radial stiffness of the wheel more than it affects the spoke
strains. From a theoretical analysis, a numerical analysis, static experimental analysis, and in-service
measurements, the spoke strains appear to be insensitive to the pattern of the spoke lacing. From a
numerical analysis, the spoking pattern has the greatest impact on the spoke strains when the wheel is
subjected to large lateral loads, such as during cornering. In this case, wheels with longer spokes have
lower strains than do wheels with shorter spokes. Small variations in measured spoke strains between
the wheel types under actual riding conditions, are attributed to variations in un-measured loads. The
extreme spoke stress cycles in the road test experiments were on the order of 150 MPa. Each cycle at
this stress level shortens the remaining fatigue life by one one-millionth of the total fatigue life. Larger
stress cycles, due to large lateral loads, for instance, could shorten the fatigue life considerably. Future
work could therefore be directed toward:
• Measurements of spoke strains due to high lateral loads, and,
• Non-linear modeling of spoked wheels subjected to spoke-slackening loads.
The fatigue resistance of the spokes, the spoke diameter, the arrangement of the spokes, and the
stiffness of the rim influence wheel stiffness and fatigue life. Wheels with 2 X, 3 X, and 4 X spoke
patterns all have similar spoke strains when subjected to radial loads. The fatigue resistance of spoked
wheels to steady cycling loads is very high for most typical service conditions."

The whole paper:

 

[raleighnut]

The rear wheel on my 653 TT bike ( A Roval Classique Pave) is laced that way radial on the NDS and 2 cross on the drive side with the front simply radial but that has very large flange hubs (and a weight limit of 108 Kg so I only just creep under the limit now but I won't be riding that til the summer)