Go Back  Bike Forums > Bike Forums > Bicycle Mechanics
Reload this Page >

Make sure that you have the correct spoke tension that we won't tell you... GRRR.

Bicycle Mechanics Broken bottom bracket? Tacoed wheel? If you're having problems with your bicycle, or just need help fixing a flat, drop in here for the latest on bicycle mechanics & bicycle maintenance.

Make sure that you have the correct spoke tension that we won't tell you... GRRR.

Old 06-07-19, 10:37 AM
  #1  
WizardOfBoz
Generally bewildered
Thread Starter
 
WizardOfBoz's Avatar
 
Join Date: Aug 2015
Location: Eastern PA, USA
Posts: 1,868

Bikes: 2014 Trek Domane 6.9, 1999 LeMond Zurich, 1978 Schwinn Superior

Mentioned: 16 Post(s)
Tagged: 0 Thread(s)
Quoted: 659 Post(s)
Liked 27 Times in 23 Posts
Make sure that you have the correct spoke tension that we won't tell you... GRRR.

Ok, the replacement spokes for the 40 yo corroded spokes from my Schwinn Superior arrive today. And I have a new Park Tool tension meter. And I've read how important spoke tension is. And I noted that Gerd Schraner, an expert, cited his skepticism about tension meters until he tried one and found out how inconsistent his tensions had been. So I want to use the right tension, and I can measure that tension, and I even may create my own calibrator*. And all the experts say "use the rim manufacturer's recommended tension.

I can find no recommendations for tension for my wheel or similar (non-boxed) rims. What kind of cruel joke is this?

Velocity Wheels kindly recommends between 110 kgf and 130 kgf. But this is for all of their rims, and they have a lot of different kinds. And they state "the mark of an excellent wheel builder is the ability to find the highest tension a rim will allow while maintaining its radial and lateral true." I don't agree. I mean, they have some experience, but this statement is at odds with people like Damon Rinard, Jobst Brandt, Musson, and Schraner. In fact, Rinard's stiffness test would seem to indicate that you get the same lateral stiffness at a range of tensions. And engineering reasoning would have you use a tension that is in that range, perhaps near the high end, but with some allowance for increased stress due to bumps, moshing the pedals, etc. Further, if you get close to the max, you stand a higher chance of potato-chipping the rim. It seems to me (PhD in engineering FWIW), that there's no need to get the rims to the point just adjacent to buckling.

Mavic has nice datasheets for their rims. For their Ksyrium SLRs They recommend 110 to 130 kg for front, and 90 to 110 kg for rear drive side. Which is weird, because it means that the non drive side is gonna be much lower. For the Aksium, they give Front: 80 to 90 kg and Rear drive side: 150 to 165 kg.

But Weinmann doesn't give values. Nor does Alex. It would be great to find an archive of historic values (as I'm using the old 27 inch rims). I can't find that table in the Sheldonopedia.

Anyway, I'm planning on using 95kgf for front and 105kgf for rear DS. Hopefully, this puts the rear NDS at about 95.

* The most accurate calibrate would use a spoke, a frame with adjustment ability, and a scale. You'd put the exact spoke type you're using into the frame, tension it to a measured tension, and then you'd note (for non-adjustable) or adjust the meter reading. Unior apparently no longer stocks their spoke tensionometer, but they had two great ideas. First, include a small bar that can be used to calibrate the meter without a frame, etc. Second, their spring adjustment was continuous and so you could "dial in" their meter. We'll see out the Park meter does. Right now, not impressed, as the edges of the meter are sharp and the parts scrape (audibly) in use.

Last edited by WizardOfBoz; 06-07-19 at 10:55 AM.
WizardOfBoz is offline  
Old 06-07-19, 10:42 AM
  #2  
fietsbob 
coprolite
 
fietsbob's Avatar
 
Join Date: Jun 2010
Location: NW,Oregon Coast
Posts: 41,928

Bikes: 8

Mentioned: 188 Post(s)
Tagged: 0 Thread(s)
Quoted: 6836 Post(s)
Likes: 0
Liked 215 Times in 179 Posts
hamster wheel ..
fietsbob is offline  
Old 06-07-19, 10:55 AM
  #3  
79pmooney
A Roadie Forever
 
79pmooney's Avatar
 
Join Date: Oct 2014
Location: Portland, OR
Posts: 6,921

Bikes: (2) ti TiCycles, 2007 w/ triple and 2011 fixed, 1979 Peter Mooney, ~1983 Trek 420 now fixed and ~1973 Raleigh Carlton Competition gravel grinder

Mentioned: 92 Post(s)
Tagged: 0 Thread(s)
Quoted: 1637 Post(s)
Likes: 0
Liked 59 Times in 50 Posts
If you shoot for the middle of the range suggested by Park for the spoke diameters you've got, you won't be far off. Right rear will have to be highish to get the left rear into acceptable. (A very good reason for going one thickness heavier for the right rear.)

These spokes are going into 40 year rims, right? No one measured spoke tension then. I doubt you will find a published recommendation for those rims anywhere.

This is not rocket science. "Feel" counts for a lot. Getting spoke tensions as even as possible for that rim and hub counts a lot. The exact number? Not nearly so much.

Ben
79pmooney is offline  
Likes For 79pmooney:
Old 06-07-19, 11:20 AM
  #4  
JohnDThompson 
Old fart
 
JohnDThompson's Avatar
 
Join Date: Nov 2004
Location: Appleton WI
Posts: 20,429

Bikes: Several, mostly not name brands.

Mentioned: 110 Post(s)
Tagged: 0 Thread(s)
Quoted: 1697 Post(s)
Liked 61 Times in 41 Posts
The use of specific spoke tensions for consumer-level wheels didn't become common until low spoke count, high tension consumer level wheels became commonly available. If your wheel has 32 or more spokes, adjusting the tension by ear and/or feel is quite adequate.
JohnDThompson is offline  
Old 06-07-19, 11:22 AM
  #5  
WizardOfBoz
Generally bewildered
Thread Starter
 
WizardOfBoz's Avatar
 
Join Date: Aug 2015
Location: Eastern PA, USA
Posts: 1,868

Bikes: 2014 Trek Domane 6.9, 1999 LeMond Zurich, 1978 Schwinn Superior

Mentioned: 16 Post(s)
Tagged: 0 Thread(s)
Quoted: 659 Post(s)
Liked 27 Times in 23 Posts
Originally Posted by 79pmooney View Post
If you shoot for the middle of the range suggested by Park for the spoke diameters you've got, you won't be far off. Right rear will have to be highish to get the left rear into acceptable. (A very good reason for going one thickness heavier for the right rear.) These spokes are going into 40 year rims, right? No one measured spoke tension then. I doubt you will find a published recommendation for those rims anywhere.
... "Feel" counts for a lot. Getting spoke tensions as even as possible for that rim and hub counts a lot. The exact number? Not nearly so much.
Ben
By that measure, we're looking at 90 kgf front, and 95kgf rear ds. I'm using 2/1.65/2 for front and rear nds, and 1.8 for rear drive side. The 1.65 matches the original spokes, the 1.8 was used for DS for the obvious reason of higher spoke tension on the dished side.

Originally Posted by 79pmooney View Post
This is not rocket science. Ben
But I want it to be.
Seriously, thanks. Glad to know that my original values are in the reasonable range.

Originally Posted by JohnDThompson View Post
The use of specific spoke tensions for consumer-level wheels didn't become common until low spoke count, high tension consumer level wheels became commonly available. If your wheel has 32 or more spokes, adjusting the tension by ear and/or feel is quite adequate.
Yeah, but where's the fun in that? And if I followed that guideline, how would I justify the purchase of another tool? Geez, John, you're getting all practical on me here.
WizardOfBoz is offline  
Old 06-07-19, 12:52 PM
  #6  
DiabloScott
It's MY mountain
 
DiabloScott's Avatar
 
Join Date: Sep 2006
Location: Mt.Diablo
Posts: 8,614

Bikes: Klein, Merckx, Trek

Mentioned: 57 Post(s)
Tagged: 0 Thread(s)
Quoted: 2195 Post(s)
Liked 31 Times in 18 Posts
Originally Posted by WizardOfBoz View Post
Velocity Wheels kindly recommends between 110 kgf and 130 kgf. But this is for all of their rims, and they have a lot of different kinds. And they state "the mark of an excellent wheel builder is the ability to find the highest tension a rim will allow while maintaining its radial and lateral true." I don't agree. I mean, they have some experience, but this statement is at odds with people like Damon Rinard, Jobst Brandt, Musson, and Schraner. In fact, Rinard's stiffness test would seem to indicate that you get the same lateral stiffness at a range of tensions. And engineering reasoning would have you use a tension that is in that range, perhaps near the high end, but with some allowance for increased stress due to bumps, moshing the pedals, etc. Further, if you get close to the max, you stand a higher chance of potato-chipping the rim. It seems to me (PhD in engineering FWIW), that there's no need to get the rims to the point just adjacent to buckling.
The idea behind this is that the higher the tension, the more of a hit you can take without a spoke going slack. How much tension you can apply is a function of the rim strength (lateral and radial and tensile - for spoke pull-out)

Basically you need enough tension that the spokes don't go slack, and not so much that the rim starts to fail or potato chip. The stronger your rims are, the more of a range in acceptable tension you'll have. I don't remember if it was Brandt's book, but somewhere in the literature I've read it said to keep laying on the tension until the wheel starts to potato chip, and then back off a half turn or so all around, and that's basically the same idea.

The real value in your tensionometer is getting all the spokes with even tension, rather than one specific tension value.

And here's a photo of my rim with the important stats printed right on the decal... I think they sell a lot of these to engineers.
I'd be really surprised if your 40-year old rims could stand this much. I'm sure you can do the conversion, but 1200N = 122kgf so that's in the middle of the range Velocity is giving.

Last edited by DiabloScott; 06-07-19 at 12:56 PM.
DiabloScott is offline  
Old 06-07-19, 01:18 PM
  #7  
79pmooney
A Roadie Forever
 
79pmooney's Avatar
 
Join Date: Oct 2014
Location: Portland, OR
Posts: 6,921

Bikes: (2) ti TiCycles, 2007 w/ triple and 2011 fixed, 1979 Peter Mooney, ~1983 Trek 420 now fixed and ~1973 Raleigh Carlton Competition gravel grinder

Mentioned: 92 Post(s)
Tagged: 0 Thread(s)
Quoted: 1637 Post(s)
Likes: 0
Liked 59 Times in 50 Posts
Originally Posted by DiabloScott View Post
...

The real value in your tensionometer is getting all the spokes with even tension, rather than one specific tension value.

And here's a photo of my rim with the important stats printed right on the decal... I think they sell a lot of these to engineers.
I'd be really surprised if your 40-year old rims could stand this much. I'm sure you can do the conversion, but 1200N = 122kgf so that's in the middle of the range Velocity is giving.

...
I actually use my Park tool for the overall tension and just listen to get the relative tensions right. So after the next round of tightening, truing and "tuning", I measure 2 or three spokes. Do this until I hit the upper portion of Park's range or until I feel it's right.

All that info on the rim? That's just completely wrong. (A weight that's around 50 grams less than actual could be there.) The ERD? You are supposed to have access to the local "Sheldon Brown" guru or have to build yourself a tool to measure it. )

Edit: the engineers screwed up on that label. A max pressure without saying the tire size tells you very little. 9 bar (130 psi in a 19c tire is very different from 9 bar in a 35c tire.

Ben

Last edited by 79pmooney; 06-07-19 at 01:24 PM.
79pmooney is offline  
Old 06-07-19, 02:20 PM
  #8  
WizardOfBoz
Generally bewildered
Thread Starter
 
WizardOfBoz's Avatar
 
Join Date: Aug 2015
Location: Eastern PA, USA
Posts: 1,868

Bikes: 2014 Trek Domane 6.9, 1999 LeMond Zurich, 1978 Schwinn Superior

Mentioned: 16 Post(s)
Tagged: 0 Thread(s)
Quoted: 659 Post(s)
Liked 27 Times in 23 Posts
Originally Posted by DiabloScott View Post
The idea behind this is that the higher the tension, the more of a hit you can take without a spoke going slack. How much tension you can apply is a function of the rim strength (lateral and radial and tensile - for spoke pull-out)

Basically you need enough tension that the spokes don't go slack, and not so much that the rim starts to fail or potato chip. The stronger your rims are, the more of a range in acceptable tension you'll have. I don't remember if it was Brandt's book, but somewhere in the literature I've read it said to keep laying on the tension until the wheel starts to potato chip, and then back off a half turn or so all around, and that's basically the same idea....
So, when we hit a bump, the rim deflects and the bottom spokes loosen up a bit. If they aren't tight enough, they go slack. Got that. But other things happen. For example, the other spokes in the wheel get MORE tension when the wheel is stressed by a bump. If the other spokes are at a limit, you have failure. This was my point about being in the higher end of the permissible range of spoke tension, but not at the very limit. Another thing would be spoke pullout. If you tight spokes so that you are just seeing the beginning of spoke pullout - you are already toast. That plastic deformation can't be recovered. So, again, I don't want to get super-near the spoke pullout limit.

I guess I'd also speculate that one hits bumps that stress you to the limit only rarely. That is, if my rims are pretty darn tight (say 80% of the start of tacos, or of spoke pullout), there will be very few bumps that will slacken the spoke even given that I'm a big heavy guy. Put another way, I'm not sure how important it is to be at 99% of max tension, vs 80%. Nor do I trust myself (or the Park tensionmeter, for that matter) to get near the 99% limit.

Originally Posted by DiabloScott View Post
I'd be really surprised if your 40-year old rims could stand this much. I'm sure you can do the conversion, but 1200N = 122kgf so that's in the middle of the range Velocity is giving.
I don't really know the max recommended, but as I pointed out I am surely going to be using less than the 122kgf. Perhaps we'll see how 85and 95kgf (front and rear DS) work, and then I may inch up to 90 and 100 or 95/105. Will advise. If this doesn't work I'll let you know how new Weinman RM-19s do.

Originally Posted by 79pmooney View Post
...
Edit: the engineers screwed up on that label. A max pressure without saying the tire size tells you very little. 9 bar (130 psi in a 19c tire is very different from 9 bar in a 35c tire.
Ben
Ben, Agree. Wonder how many people thing through this and understand that the 35mm tire while a 19mm tire has about 3800 pounds force trying to split the rim in half, while the 35mm tire has about 7000 lbs.

Last edited by WizardOfBoz; 06-07-19 at 02:31 PM.
WizardOfBoz is offline  
Old 06-07-19, 02:42 PM
  #9  
DiabloScott
It's MY mountain
 
DiabloScott's Avatar
 
Join Date: Sep 2006
Location: Mt.Diablo
Posts: 8,614

Bikes: Klein, Merckx, Trek

Mentioned: 57 Post(s)
Tagged: 0 Thread(s)
Quoted: 2195 Post(s)
Liked 31 Times in 18 Posts
Originally Posted by WizardOfBoz View Post
For example, the other spokes in the wheel get MORE tension when the wheel is stressed by a bump. If the other spokes are at a limit, you have failure. This was my point about being in the higher end of the permissible range of spoke tension, but not at the very limit.
That has been argued a LOT, and it's not initially intuitive, but it's just not true. Temporary flat spot deformation on the bottom of the rim doesn't cause a temporary bulge in the top of the rim. As far as I know, there's nothing you can do to while riding that would make any of the spokes experience an increase in tension. Brandt talks about this a little in a related but not exactly analogous example.



Originally Posted by WizardOfBoz View Post
Another thing would be spoke pullout. If you tight spokes so that you are just seeing the beginning of spoke pullout - you are already toast. That plastic deformation can't be recovered. So, again, I don't want to get super-near the spoke pullout limit.
That's certainly true, and no one is recommending that. If pull-out tension happens before taco tension, your rims are in bad shape.
DiabloScott is offline  
Old 06-07-19, 04:34 PM
  #10  
WizardOfBoz
Generally bewildered
Thread Starter
 
WizardOfBoz's Avatar
 
Join Date: Aug 2015
Location: Eastern PA, USA
Posts: 1,868

Bikes: 2014 Trek Domane 6.9, 1999 LeMond Zurich, 1978 Schwinn Superior

Mentioned: 16 Post(s)
Tagged: 0 Thread(s)
Quoted: 659 Post(s)
Liked 27 Times in 23 Posts
Ok. I'm not arguing for a "bump" in the rim on top when the rim on the bottom hits a bump. I would argue for a slight increase in circumference. This would increase in all spoke tension. I did a back of the envelope estimate below (sorry, its a large envelope).

Suppose we hit a bump, get air, come down on the wheel, and the bottom of the wheel gets deflected. At maximum deflection, our "altitude" will be a matter of millimeter lower than normal. The wheel will start to recover and will push our axle up to acclerate us (and the bike) back toward normal altitude). When the wheel is accelerating us upward, it is exerting more upward force than it does on level ground. Therefore, the net force upward of all spokes (force up minus the force down of all spokes) MUST be greater than normal. This could be, though, just the reduction of downward force of the slack spokes or an actual increase in upward force from spoke tension increase.

Back to the point about there not being a bump. Instead, I'd argue that a portion of the rim is somewhat flattened and that this increases the effective circumference and radius of the wheel.
Let's assume a 27 inch wheel, with an ERD of 630 mm.
The circumference is then 1979 mm.
Assume we hit a bump and a 3 inch chunk of rim is temporarily flattened (this is an assumption, based as much upon the drawings in Jobst Brandt's book as the SWAG method".
3 inches is 76.2mm of the circumference.
The angle this 3 inch circumferential chunk includes is 76.2/1979 * 360, or 13.86 degrees
The chord function of the circle of this is crd(theta) = 2 * sin (theta/2) = 0.24
The initial chord length is radius * crd(theta), or 0.24*630/2 = 76.014mm
But we are assuming that this is the 3 inch section that was flattened, so the chord is now the whole 3 inchs or 76.2mm.
Difference is .185mm.
Making some assumptions we assume that the wheel rim now has a circumference of the original, plus 0.185mm.
Because radius is linear with circumference, the radius increases by 315* .185/1900 , or about 0.03mm.
The pitch of spoke thread is 56 per inch, or about 0.45mm/turn.
This would make the increase in effective diameter be equivalent to about 1/15 of a turn of the spoke for each spoke not in that flattened 3 inch section.

We can back check this. The geometry assumed above would give a radius from the deflected rim to the center of the hub of about 312.7, or about 2.3 mm smaller than normal. Is this even remotely related to the normal amount of spoke stretch? If its close but less than normal spoke stretch, then perhaps we can have some confidence in our geometric assumptions. Would this amount of contraction loosen a properly tensioned spoke?

Assume a 1.8mm spoke (unbutted, for simplicity). This has an area of 2.544 square mm. At 100kg force tension, this means that there's 39.3 kgf/mm^2 of stress. Stainless steel (indeed, all steels, which is remarkable) has a Young's modulus of 29GPa, or 2960 kgf/mm^2. So divide the stress of 39.3 by the modulus of 2960 and we get a strain of 0.0133. Again, Let's assume that our spoke is about 290mm. So our total normal amount of stretch in the spoke would be about 4mm. Given that we've assumed values consistent with a well-built wheel, and that our spoke tension reduction is close, but not in excess of the normal spoke tension gives, perhaps, some confidence that our geometric assumptions are not entirely unreasonable.

We could also infer that the ratio of (0.03mm + 4mm) /4mm can be used to calcualate the increase in spoke tension elsewhere, to get about a 0.75% increase. That is, spoke tension increased from 100 to 100.75kgf.

This lines up with experience. That is, most wheels with spoke tension of 100kgf don't blow up, but some wheels - after hitting a rare bump - do fail, I'm gonna call it a day and go have a Gin and tonic.

But the point is, point rim deflection DOES affect all other spokes even if only in the 1% range. Also, if we hit bumps that flatten 3 inch sections a lot, and if our spoke tension were lower - say 70kg - then we'd expect to see failure modes deriving from slack spokes.

Last edited by WizardOfBoz; 06-07-19 at 08:43 PM.
WizardOfBoz is offline  
Old 06-08-19, 08:28 AM
  #11  
cyccommute 
Mad bike riding scientist
 
cyccommute's Avatar
 
Join Date: Nov 2004
Location: Denver, CO
Posts: 21,348

Bikes: Some silver ones, a red one, an orange one and a few titanium ones

Mentioned: 92 Post(s)
Tagged: 1 Thread(s)
Quoted: 2463 Post(s)
Likes: 0
Liked 102 Times in 69 Posts
Originally Posted by WizardOfBoz View Post
Ok, the replacement spokes for the 40 yo corroded spokes from my Schwinn Superior arrive today. And I have a new Park Tool tension meter. And I've read how important spoke tension is. And I noted that Gerd Schraner, an expert, cited his skepticism about tension meters until he tried one and found out how inconsistent his tensions had been. So I want to use the right tension, and I can measure that tension, and I even may create my own calibrator*. And all the experts say "use the rim manufacturer's recommended tension.

I can find no recommendations for tension for my wheel or similar (non-boxed) rims. What kind of cruel joke is this?

Velocity Wheels kindly recommends between 110 kgf and 130 kgf. But this is for all of their rims, and they have a lot of different kinds. And they state "the mark of an excellent wheel builder is the ability to find the highest tension a rim will allow while maintaining its radial and lateral true." I don't agree. I mean, they have some experience, but this statement is at odds with people like Damon Rinard, Jobst Brandt, Musson, and Schraner. In fact, Rinard's stiffness test would seem to indicate that you get the same lateral stiffness at a range of tensions. And engineering reasoning would have you use a tension that is in that range, perhaps near the high end, but with some allowance for increased stress due to bumps, moshing the pedals, etc. Further, if you get close to the max, you stand a higher chance of potato-chipping the rim. It seems to me (PhD in engineering FWIW), that there's no need to get the rims to the point just adjacent to buckling.

Mavic has nice datasheets for their rims. For their Ksyrium SLRs They recommend 110 to 130 kg for front, and 90 to 110 kg for rear drive side. Which is weird, because it means that the non drive side is gonna be much lower. For the Aksium, they give Front: 80 to 90 kg and Rear drive side: 150 to 165 kg.

But Weinmann doesn't give values. Nor does Alex. It would be great to find an archive of historic values (as I'm using the old 27 inch rims). I can't find that table in the Sheldonopedia.

Anyway, I'm planning on using 95kgf for front and 105kgf for rear DS. Hopefully, this puts the rear NDS at about 95.

* The most accurate calibrate would use a spoke, a frame with adjustment ability, and a scale. You'd put the exact spoke type you're using into the frame, tension it to a measured tension, and then you'd note (for non-adjustable) or adjust the meter reading. Unior apparently no longer stocks their spoke tensionometer, but they had two great ideas. First, include a small bar that can be used to calibrate the meter without a frame, etc. Second, their spring adjustment was continuous and so you could "dial in" their meter. We'll see out the Park meter does. Right now, not impressed, as the edges of the meter are sharp and the parts scrape (audibly) in use.
First, I’ve got to say that there is a perfectly adequate measure of force that could be used without resorting to the made up force unit “kilogram-force”. It’s a newton and it works. I wish that wheel manufacturers, tool makers and wheel builders would use it.

That said, look at the range of force that is recommended. 110 kgf is about 1100N. 130kgf is almost 1300N. That’s a very wide range. And everyone will tell you that the tension used is absolutely critical! Critical measurements shouldn’t have a 200N window. That’s more of an “about” than a specification.

And I don’t agree with 70Mooney about the “science” of wheelbuilding. It is akin to rocket science in that it is a highly difficult system to model and understand. There is far more art to wheelbuilding than there is science but there really should be more science. The wide range of recommended tensions (and lack of data on those tensions) says to me that the “recommendations” come more from a lawyer than an engineer.
__________________
Stuart Black
Pokin' around the Poconos A cold ride around Lake Erie
Dinosaurs in Colorado A mountain bike guide to the Purgatory Canyon dinosaur trackway
Solo Without Pie. The search for pie in the Midwest.
Picking the Scablands. Washington and Oregon, 2005. Pie and spiders on the Columbia River!
Days of Wineless Roads. Bed and Breakfasting along the KATY
Twisting Down the Alley. Misadventures in tornado alley.
An Good Ol' Fashion Appalachian Butt Whoopin'.
cyccommute is offline  
Old 06-08-19, 08:44 AM
  #12  
UniChris
Senior Member
 
UniChris's Avatar
 
Join Date: May 2017
Location: NYC
Posts: 787

Bikes: 36" Unicycle

Mentioned: 14 Post(s)
Tagged: 0 Thread(s)
Quoted: 342 Post(s)
Liked 59 Times in 51 Posts
Originally Posted by WizardOfBoz View Post
This lines up with experience. That is, most wheels with spoke tension of 100kgf don't blow up, but some wheels - after hitting a rare bump - do fail, I'm gonna call it a day and go have a Gin and tonic.
Nope, you have to earn that drink! Design an experiment where a wheel repeatedly hits a bump in the *same* place, with an indicator mark, and run it until spoke breakage or pull-out. Then see if the failure happened at the bump (slack stress), opposite it (tension increase), or if even after many trials with fresh components remains uncorrelated in position - in which case you may mix a double.

Hypothetically one could possibly do this on the road in ordinary usage with some kind of logging electronic accelerometer recording the angular position of each notable impact in the wheel's history, and study that once a failure occurs.
UniChris is offline  
Old 06-08-19, 10:04 AM
  #13  
WizardOfBoz
Generally bewildered
Thread Starter
 
WizardOfBoz's Avatar
 
Join Date: Aug 2015
Location: Eastern PA, USA
Posts: 1,868

Bikes: 2014 Trek Domane 6.9, 1999 LeMond Zurich, 1978 Schwinn Superior

Mentioned: 16 Post(s)
Tagged: 0 Thread(s)
Quoted: 659 Post(s)
Liked 27 Times in 23 Posts
Originally Posted by cyccommute View Post
First, I’ve got to say that there is a perfectly adequate measure of force that could be used without resorting to the made up force unit “kilogram-force”. It’s a newton and it works. I wish that wheel manufacturers, tool makers and wheel builders would use it.
The spoke tension meters I've seen (Park, Wheelsmith) use kgf. But it's a simple conversion, as the acceleration of gravity is 9.81 m/s/s, and so 9.81 N = 1 kgf. Easy to approximate a conversion factor by rounding to ten, impossible to confuse. I have no problem with either. I will say, though, that unit conversion has its traps. I just screwed up and misconverted the spoke gauge on the new Sapim spokes I just got. Somehow, I measured 0.080 inches and my mind registered the "8" and started thinking that it was 1.8mm. I wanted 2mm. Of course, 0.08 inches is about 2mm. Marcus, at Yojimbo's Garage, was more than patient with me about this. But we crashed a lander into Mars because of improper units conversion and so... units are important.

Originally Posted by cyccommute View Post
That said, look at the range of force that is recommended. 110 kgf is about 1100N. 130kgf is almost 1300N. That’s a very wide range. And everyone will tell you that the tension used is absolutely critical! Critical measurements shouldn’t have a 200N window. That’s more of an “about” than a specification.
With the wide variety of wheel build quality, and the rarity of wheel failure, I would challenge whether spoke tension is that critical for most wheels. I was impressed by Gerd Schraner talking about checking tension on wheels he built before using a tension meter, and how shocked he was by the wide variation in spoke tension between wheels. Damon Rinard's graph of stiffness vs spoke turns, below, shows little effect over a wide range. If we are building tandem wheels with unbutted spokes (even thought its more critical to use butted spokes for heavy loads than for light, that is, heavy riders should insist upon butted spokes with more urgency than light guys) then you need to make sure that you have the spoke stretched out enough to accomodate rim deflection because several important failure modes are due to spokes going slack. That tension IS more critical. But tension for lighter gauge butted spokes is LESS critical than for thicker or unbutted spokes (for a lower force, they stretch more and hence can accomodate more rim deflection). But given the worldwide population history of pretty good wheel survival for quality ranging from meh to sublime, and with many wheels of yesteryear having been made where there wasn't a spoke tension meter in the entire factory, I would submit that there's likely a pretty wide range of acceptable tensions. Above the limit needed to ensure that the spoke doesn't go slack, I suspect that its just not that critical.

Using the 8 turn range for which no effect on stiffness is observed in the graph below, and assuming 56 tpi nipple threads, I get nearly a 3.6mm variation in acceptable spoke stretch! Let's cut that in half to allow for some engineering safety margin. So 1.8mm. Using the numbers in my previous post, this 1.8mm window is a 50 kgf (ok, 490N) range of spoke tension.




Originally Posted by cyccommute View Post
And I don’t agree with 70Mooney about the “science” of wheelbuilding. It is akin to rocket science in that it is a highly difficult system to model and understand. There is far more art to wheelbuilding than there is science but there really should be more science. The wide range of recommended tensions (and lack of data on those tensions) says to me that the “recommendations” come more from a lawyer than an engineer.
I'm beginning to appreciate that there's a Zen sort of deal with wheel building. It's conceptually simple, and good results can be achieved with patience, and it's a craft activity that one can get completely immersed in. There's an aesthetic, and the science. And the utility and pleasure of riding on wheels you've built. All incorporated into one activity. And there's no simple analytic solution to all wheel building questions. I suspect that this is where folks thinking of wheel building as "art".

I agree that there's science in wheel building, and that we should do more to understand the wheel quantitatively. Hence my longwinded explanation, above. But the above is a "back of the envelope" approximation. The Finite Element Analysis that Jobst Brandt did was back in the day and was limited by software and hardware of the time*. Today's FEA software is much more user-friendly and you can run a program on a laptop now that once required a Cray. It would be good to get a better, more granular analysis of what happens to a rim and spokes when the tire hits a bump.

In fairness, though, what's going on in a wheel in use is surprisingly complex. You don't just come up with simple equations to get accurate answers. Bike stability is another area like this. Conceptually simple (I don't want the bike to fall down, and a 7 year old can do this while riding the bike). Analytically VERY hard (see Bicycling Science. The 1st edition has bit more math expressly stated and may be better if you are interested).

I've gotten several information sources telling me to tighten the spokes till the wheel starts to taco, then back off. I guess this works if the nipple holes don't fail at this tension. I think I'm going to start with about 85 and 90kgf and tighten to about 100kgf (DS) and 90 kgf (front and, hopefully, NDS). But if the wheel is mushy, I'll report back.

BTW, if we engineers could get a more accurate predictive model of wheel and spoke failure (ala UniChrises method), the safety margins that lawyers mandate could be shrunk.

*On edit, I note that others had suggested that software had evolved since Jobst's FEM analysis, and that Jobst had noted and rejected this criticism asking "What was deficient in the original analysis? What has changed? How would the more modern software improve the analysis?".

Last edited by WizardOfBoz; 06-11-19 at 07:53 AM.
WizardOfBoz is offline  
Old 06-08-19, 10:15 AM
  #14  
WizardOfBoz
Generally bewildered
Thread Starter
 
WizardOfBoz's Avatar
 
Join Date: Aug 2015
Location: Eastern PA, USA
Posts: 1,868

Bikes: 2014 Trek Domane 6.9, 1999 LeMond Zurich, 1978 Schwinn Superior

Mentioned: 16 Post(s)
Tagged: 0 Thread(s)
Quoted: 659 Post(s)
Liked 27 Times in 23 Posts
Originally Posted by UniChris View Post
Nope, you have to earn that drink! Design an experiment where a wheel repeatedly hits a bump in the *same* place, with an indicator mark, and run it until spoke breakage or pull-out. Then see if the failure happened at the bump (slack stress), opposite it (tension increase), or if even after many trials with fresh components remains uncorrelated in position - in which case you may mix a double.

Hypothetically one could possibly do this on the road in ordinary usage with some kind of logging electronic accelerometer recording the angular position of each notable impact in the wheel's history, and study that once a failure occurs.
Ah, to test for fatigue failure. Your experimental setup sounds impressive. That would be fun to design. I suspect that it might shorten the required experimental course, though, to add a good, modern FEA analysis of the wheel. Then calibrate the FEA model with your data, and you should be able to calculate the magnitude and maximum of stress variations at all locations in the wheel, and then find the points of highest stress in the rim, the spokes, and the hub. At that point, service life (or stress cycles per service hour / stress cycles to failure*) could be accurately predicted.

*This number is easily found in a fatigue chart for the material of interest
WizardOfBoz is offline  
Old 06-08-19, 10:19 AM
  #15  
UniChris
Senior Member
 
UniChris's Avatar
 
Join Date: May 2017
Location: NYC
Posts: 787

Bikes: 36" Unicycle

Mentioned: 14 Post(s)
Tagged: 0 Thread(s)
Quoted: 342 Post(s)
Liked 59 Times in 51 Posts
Originally Posted by WizardOfBoz View Post
add a good, modern FEA analysis of the wheel.
That would be interesting - run it at various tension levels and see when you get a flex fatigue at the spoke elbow or whatever, vs if there are places where you get a tension failure.

The analysis would have to re-load the spokes though - if I recall, conventional wisdom is that your flex fatigue cases break when whey roll off the road and get their preload tension back.
UniChris is offline  
Old 06-08-19, 12:24 PM
  #16  
nazcalines
Senior Member
 
Join Date: Jul 2013
Posts: 218
Mentioned: 10 Post(s)
Tagged: 0 Thread(s)
Quoted: 55 Post(s)
Likes: 0
Liked 0 Times in 0 Posts
I just built my first wheel (28 2.0-1.5 spokes, 11 speed hub) following Roger Musson's ebook. I took my time and tuned the spokes by tone, like he suggests. When I went to the coop to check the tension, I was spot on consistent all the the way around, but way low in tension. If I had a similar wheel at home I would've just got it to a similar feel/pitch and called it good.
nazcalines is offline  
Old 06-08-19, 01:06 PM
  #17  
WizardOfBoz
Generally bewildered
Thread Starter
 
WizardOfBoz's Avatar
 
Join Date: Aug 2015
Location: Eastern PA, USA
Posts: 1,868

Bikes: 2014 Trek Domane 6.9, 1999 LeMond Zurich, 1978 Schwinn Superior

Mentioned: 16 Post(s)
Tagged: 0 Thread(s)
Quoted: 659 Post(s)
Liked 27 Times in 23 Posts
Originally Posted by nazcalines View Post
I just built my first wheel (28 2.0-1.5 spokes, 11 speed hub) following Roger Musson's ebook. I took my time and tuned the spokes by tone, like he suggests. When I went to the coop to check the tension, I was spot on consistent all the the way around, but way low in tension. If I had a similar wheel at home I would've just got it to a similar feel/pitch and called it good.
nazcalines, I'm building my first wheel, too, even though I've been involved in bike wrenching for more than 40 years. I wanted to avoid the situation you found yourself in.

So did you go home and tighten everything? Did you buy a tension meter tool?

I have access to an older edition of Musson's book, but its good enough and so inexpensive that I'm going to buy the most recent edition.

Last edited by WizardOfBoz; 06-08-19 at 01:22 PM.
WizardOfBoz is offline  
Old 06-08-19, 01:28 PM
  #18  
trailangel
Senior Member
 
trailangel's Avatar
 
Join Date: Feb 2014
Location: Pasadena, CA
Posts: 3,635

Bikes: Schwinn Varsity

Mentioned: 16 Post(s)
Tagged: 0 Thread(s)
Quoted: 1263 Post(s)
Liked 46 Times in 31 Posts
Originally Posted by nazcalines View Post
I just built my first wheel (28 2.0-1.5 spokes, 11 speed hub) following Roger Musson's ebook. I took my time and tuned the spokes by tone, like he suggests. When I went to the coop to check the tension, I was spot on consistent all the the way around, but way low in tension. If I had a similar wheel at home I would've just got it to a similar feel/pitch and called it good.
Where did you get the 1.5 nipples?
trailangel is offline  
Old 06-08-19, 02:00 PM
  #19  
nazcalines
Senior Member
 
Join Date: Jul 2013
Posts: 218
Mentioned: 10 Post(s)
Tagged: 0 Thread(s)
Quoted: 55 Post(s)
Likes: 0
Liked 0 Times in 0 Posts
Originally Posted by WizardOfBoz View Post
nazcalines, I'm building my first wheel, too, even though I've been involved in bike wrenching for more than 40 years. I wanted to avoid the situation you found yourself in.

So did you go home and tighten everything? Did you buy a tension meter tool?

I have access to an older edition of Musson's book, but its good enough and so inexpensive that I'm going to buy the most recent edition.
No, I brought it up to the target tightness (~120kgf) at the coop using their park tension meter. At home I did the final truing. I'd like to build a tension meter at some point. I was hoping to build one at the local maker space, but they recently sold all their metal working equipment. There are plans for a 3d printed version of the Jobst Brandt meter online, so I might try to go that route if it seems doable and I can find a place to print it.

The coop's meter was pretty clearly well used, so for all I know it might have been out of calibration. I trusted it anyway, and my wheel seems solid. I guess time will tell.
nazcalines is offline  
Old 06-08-19, 02:01 PM
  #20  
nazcalines
Senior Member
 
Join Date: Jul 2013
Posts: 218
Mentioned: 10 Post(s)
Tagged: 0 Thread(s)
Quoted: 55 Post(s)
Likes: 0
Liked 0 Times in 0 Posts
Originally Posted by trailangel View Post
Where did you get the 1.5 nipples?
Sorry double butted... 2-1.5-2
nazcalines is offline  
Old 06-09-19, 04:28 PM
  #21  
63rickert
Senior Member
 
Join Date: Dec 2013
Posts: 1,067
Mentioned: 35 Post(s)
Tagged: 0 Thread(s)
Quoted: 481 Post(s)
Likes: 0
Liked 18 Times in 15 Posts
It sure would be helpful if you would tell us what rim is on the Superior. Also helpful to know if the spokes are corroded but the rim is in good shape.

See if you can find some Velocity wheels built by Velocity. Any other mfr who provides recommended tensions for their rims and also sells built wheels would be as good. Check the drive side tension in the 11 and 12 speed rears. I am not going to tell you what I have found doing just that. If you do this simple test you will worry a whole lot less. Sometimes engineering is mostly coming from legal department.

It takes a big hit to deflect a rim and loosen a spoke. It is not happening constantly. At Paris-Roubaix it will be happening often enough but not continually. There are close-up slo-mo videos posted here and there (which I have not bookmarked unfortunately) showing high speed impact and spokes flapping loose. Deflection is much more localized and much smaller in amplitude than you are imagining.
63rickert is offline  
Old 06-09-19, 09:42 PM
  #22  
WizardOfBoz
Generally bewildered
Thread Starter
 
WizardOfBoz's Avatar
 
Join Date: Aug 2015
Location: Eastern PA, USA
Posts: 1,868

Bikes: 2014 Trek Domane 6.9, 1999 LeMond Zurich, 1978 Schwinn Superior

Mentioned: 16 Post(s)
Tagged: 0 Thread(s)
Quoted: 659 Post(s)
Liked 27 Times in 23 Posts
Originally Posted by 63rickert View Post
It sure would be helpful if you would tell us what rim is on the Superior. Also helpful to know if the spokes are corroded but the rim is in good shape.

See if you can find some Velocity wheels built by Velocity. Any other mfr who provides recommended tensions for their rims and also sells built wheels would be as good. Check the drive side tension in the 11 and 12 speed rears. I am not going to tell you what I have found doing just that. If you do this simple test you will worry a whole lot less. Sometimes engineering is mostly coming from legal department.

It takes a big hit to deflect a rim and loosen a spoke. It is not happening constantly. At Paris-Roubaix it will be happening often enough but not continually. There are close-up slo-mo videos posted here and there (which I have not bookmarked unfortunately) showing high speed impact and spokes flapping loose. Deflection is much more localized and much smaller in amplitude than you are imagining.
The superior has Weinmann 27 x 1-1/4 rims. They are alloy and are oxidized and stained (normal for an old rim) but are flat with no corrosion. They are flat (they are essentially a strip of flat alloy, with no box and no extruded hollows) and are, I believe, hookless. Normandy high flange hubs. Laced 4 cross. The rims are dead flat and appear to be perfectly round. Probably not the best rims, but they are the OEM rims, and I am trying to get the bike into the same mechanical shape it was in 1978. And to learn.

For calibration, I'll probably just cob together a jig with a cheap 300kg electronic scale, and get a much more exact calibration with the exact spoke used. Sure, I can check my wheel against some bike shop bikes but I'd rather check the thing against measurable and calibrated standards. This leaves less guessing. Also, I suspect that Victory makes very few wheels in 27 inch and 4 cross. But to your point the 27 inch wheels on my World Voyaguer have similar rims (27 inch alloy, Araya though) and spokes and the spoke tension is low. But no spokes have broken in 45 years.y

The spokes were plated steel and they are pretty corroded. But they were 2.0/1.65/2.0. I've purchased the same for front and NDS, but upgraded the rear drive side o 2.0/1.8/2.0.

As a PhD in engineering, I can assure you that sometimes engineering comes from wanting stuff to work, and wanting to know why and how it might fail, and knowing how to avoid that failure. Yes, sometimes the lawyers get involved but there's probably value in understanding the problem conceptually and quantitatively and not just checking other wheels.

Jobst Brandt was an engineer. In his book, he suggests that dynamic load is spread out over about 4 spokes, which is close to but a bit wider (~8 inches) than my 3 inch assumption. So his support area is about 3 times my assumed value. But his deflection is significantly less than mine, about 0.15 mm. I'll check my units, but if Damon Rinard's graph (above) is correct and there's an 8 turn range where the wheel is stiff, then the spoke is stretching about 3.6 mm (or the rim is compressing). So how does the spoke get loose if its stretched 4mm and a bump only allows it to relax by 0.15mm? Something's fishy. I'll check my units, and also see if I can figure out the rim compression.

Spoke failure at the Paris Roubaix IS rare. Spoke failure in general is rare. I don't recall ever breaking a spoke, and I weigh 240#, and have had pretty strong legs (defensive end in a D1 football progam in college). I guess I've had good wheels, and have been careful wrt potholes and such. But that's one of my reasons for my point about sweating spoke tension. If spoke breakage is very rare, and spoke tension varies hugely between wheels, then it probably doesn't matter whether you are at 60% or 99% of TT* regarding spokes going slack when you hit a bump. Nor does it matter too much (per Rinard's graph and article) with respect to stiffness.

*TT is engineering terminology. Stands for Tension - Taco.

Last edited by WizardOfBoz; 06-09-19 at 09:46 PM.
WizardOfBoz is offline  
Old 06-10-19, 07:55 AM
  #23  
63rickert
Senior Member
 
Join Date: Dec 2013
Posts: 1,067
Mentioned: 35 Post(s)
Tagged: 0 Thread(s)
Quoted: 481 Post(s)
Likes: 0
Liked 18 Times in 15 Posts
All catalog says for OEM rims is Weinmann. In photos they look to be A-125. No concave section on top face of rim? Sidewalls are vertical, do not flare out? Probably A-125. If they are the original rims on this 41 year old bike. These are old parts. No, there is no spec for tension. I would go ahead and build to 120kgpf, with more uncertainty than usual about what happens next. One large reason to go ahead and pull tight is the load that is going on these wheels.

Jobst Brandt said all kinds of things and buried much of his content under a rhetorical style that is more than a bit offputting. He was also grounded in an earlier era. For starters no one pulled wheels as tight as we do now. Many rims would fail in the wheelstand if pulled as tight as we do now. And there were basically no customers walking into the shop who weighed 240#. If they did they were likely to be sent away without a bike. Big guys were refused because the wheels, as built then, would not hold them. Refusing to sell to large people did not put a dent in business because there were not that many people so big. Those who were that big had no aspiration to any sort of active lifestyle.

Another thing you are going to find is that the joint on the rim is nothing like as precise as with current production. I recently had a pair of dated 1977 NOS Weinmann A-124 concaves in the wheelstand. Those are big step up from the rims I believe you have. First thing noticed was the very crude grinding that had been done at factory to make the spot where the two ends of rim were joined even a little bit flat. No way were those rims going to be as true as modern rims and no way were they going to be close to uniform tension. Any who cared about precision back then were on tubulars. Any who were on clinchers were assumed to be not so fussy.

Spokes are not rubber bands. Rims are not taffy. Four millimeters of spoke elongation does not occur. The lower figure for compression is at least in ballpark.

Araya rims are way higher quality than old Weinmann clincher. Any rim or spoke alive after 45 years is in a wheel not much used.

Slightly later Superiors used Super Champion #58 rims. Those will take current spoke tension easily. Modern rims that look a lot like the old Super Champs can be had.

If this is going to work at all use widest tires available that fit and lowest tire pressure practical.
63rickert is offline  
Old 06-10-19, 11:12 AM
  #24  
Wilfred Laurier
Seńor Member
 
Wilfred Laurier's Avatar
 
Join Date: Feb 2013
Posts: 4,293
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 326 Post(s)
Liked 63 Times in 44 Posts
Discussion above about spoke tension relating to wheel stiffness is missing the point. Spokes generally suffer fatigue failures, which are caused by many excursions from high to low tension and back. The higher the tension in the wheel, the further away from zero tension, and by extension, failure, the spokes will experience as the wheel goes around.

A heavier rider generally needs wheels with more highly tensioned spokes as the excursion between high and low tension will be greater than under a lighter rider.
Wilfred Laurier is offline  
Old 06-10-19, 11:21 AM
  #25  
Nessism
Senior Member
 
Nessism's Avatar
 
Join Date: Jun 2004
Location: Torrance, CA
Posts: 2,725

Bikes: Homebuilt steel

Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 140 Post(s)
Liked 9 Times in 9 Posts
Originally Posted by WizardOfBoz View Post

Anyway, I'm planning on using 95kgf for front and 105kgf for rear DS. Hopefully, this puts the rear NDS at about 95.
This will be fine. Park has a chart (and an ap on their website) to guide you on what number to target on the tool. Don't overthink. I will say though that those rims are OLD so apply some judgement and don't overtighten.
Nessism is offline  

Thread Tools
Search this Thread

Contact Us - Archive - Advertising - Cookie Policy - Privacy Statement - Terms of Service

Copyright © 2018 MH Sub I, LLC dba Internet Brands. All rights reserved. Use of this site indicates your consent to the Terms of Use.