[kjmccarx]

I'm a rocket scientist, not a bike engineer... that being said, it seems like disc brakes could be pretty bad for a wheel. I haven't crunched any numbers but the amount of torsion about the hub seems like it is enormous. It also seems like using a disc brake can introduce a large amount of angular acceleration in the opposite direction of the motion which means that the spokes are transferring torques in directions they are not designed to.

Has anyone damaged their wheels and/or spokes using disc brakes? They are used for downhill riders, so I'm guessing there is no real danger, but the torques involved seem like they are very large.

 

[127.0.0.1]

This subject has been examined here 10 years ago

http://forums.mtbr.com/wheels-tires/disc-brake-lacing-revisited-please-read-15776.html

lots of info here
Let me google that for you

 

[Joules]

so as a rocket scientist, if you had on one hand tens of millions of test flights as data, and on the other some numbers you haven't crunched, and yourself admit you don't have a lot of faith in, which seem to contradict your tens of millions of data points... wouldn't that imply, to you, that maybe your hunch is wrong? That maybe your understanding of the basic physics involved is lacking?

The only mountain bikes these days that have rim brakes are wal-mart types, there hasn't been a mid- or high-end bike sold in nearly a decade without discs, longer than that for downhill bikes, not to mention when is the last time you saw a motorcycle with v-brakes? The problems you fear have never happened.

 

[DethWshBkr]

They are very large, and you are correct. Stopping the hub, while the wheel attempts to turn puts tremendous load on the spokes. That said - I don't believe the stress placed on the spokes would be any worse than the loads placed on them by the wheel attempting to go out of round when riding it. The load on the spokes (when properly adjusted) is still a tension force.

 

[kjmccarx]

so as a rocket scientist, if you had on one hand tens of millions of test flights as data, and on the other some numbers you haven't crunched, and yourself admit you don't have a lot of faith in, which seem to contradict your tens of millions of data points... wouldn't that imply, to you, that maybe your hunch is wrong? That maybe your understanding of the basic physics involved is lacking?

I was just trying to ask a question to the community, I don't appreciate your sas. Obviously disc brakes don't cause damage to the wheel normally, but I was asking if it has every happened. Whether or not my before referenced "hunch" was true the amount of torque involved is very large, the hub is under a lot of torsion, and all the spokes experience strain when using a disc brake.

If you would like to correct anything I just said, please do so.

I, myself, have been using disc brakes for the last 5 years. But just because I haven't experienced any problems doesn't mean I am not curious.

 

[Saul Lumikko]

That's why they are laced with cross patterns. Radial spoked wheels would be torn apart from disc braking forces. That's why they aren't made.

 

[kjmccarx]

That's why they are laced with cross patterns. Radial spoked wheels would be torn apart from disc braking forces. That's why they aren't made.

Ahhh. That makes sense. I wasn't thinking about that. Thank you.

 

[127.0.0.1]

since well over 10 years ago

it is pretty much a non-issue

the biggest danger or risk is an improper build or assembly, or brake fade.
[improper build includes adapting a disc to a non-disc frame or fork]


forks are built tougher to handle disc brake forces
stays are built tougher to handle disc brake forces
the rules of building and lacing spoked wheels are written to cover various disc hubs and dishing (shimano has it's own rules, other have them too)
rims are built tougher to handle disc brake forces
spokes are designed to handle it
nipples are designed to handle it

 

[Max24]

That's why they are laced with cross patterns. Radial spoked wheels would be torn apart from disc braking forces. That's why they aren't made.

Yep, that's definitely the case. Oh and being a rocket scientist must be really cool, congrats OP

 

[kjmccarx]

Yep, that's definitely the case. Oh and being a rocket scientist must be really cool, congrats OP

Thank you, that means a lot. I really do enjoy it.

 

[Kanik]

OP, if you’re really a rocket scientist, I’d expect you to be able to analyse this fairly simple kinematics and solid mechanics situation. Let’s go over a typical braking situation:

A rider has a mass of 80 kg and is traveling at 10 m/s (36 km/h, 22 mph), and wants to stop in 10 m (33 ft) distance.
•By using v^2 =(v_0)^2+2a(x), we find the acceleration is 5 m/s^2 (backwards, as the rider is braking). •The braking force is found with F=ma, and it is 400 N.
•If the wheel is a 26” wheel with a MTB tire, we can assume the wheel is about 600mm diameter. This means a radius of 300 mm or 0.3 m.
•Thus the moment (torque) about the wheel hub is 400*0.3 = 120 Nm.
•The force at the brake caliper, for a 180mm rotor, is 120/.09 = 1333 N.

Now for the solid mechanics. This will be a bit less precise, there’s a lot of fudge factor thrown in here.
•If the hub flange is 40 mm diameter, and 16 spokes are tensioned by the moment (32 spokes, half get tensioned by a moment and half get de-tensioned), then the load on a single spoke is (120/.02)/16 = 375 N per spoke.
•If the spoke is 2mm diameter, then its area is π(.001)^2 = π*10^-6 m^2. (0.00000314 m^2)
•Thus the tensile stress on the spoke is 375/0.00000314 = 119.4 MPa.
•This is way below the yield stress for drawn 316 stainless steel of 400+ MPa.
Thus, the spokes don’t seem to be overloaded. Obviously, there are many more factors to consider here, such as stress concentrations and shear stresses at the spoke elbow where it meets the hub, and at the rolled threads. These two locations are where spokes actually break.

The hub itself would need a more detailed analysis, but the answer here is to just have reasonably sized flanges and you won’t have an issue. I would suspect that some impacts, both on road or off road, load the spokes and hub flanges more than disc brakes so.

The rotors would have a similar analysis, except you assume that the rotor spokes are loaded in shear and not tension. In reality, it is combination of shear and compression. Regardless, a reasonably designed rotor will not fail either.


I'm just a second year mechatronic eng student typing this on my way to uni. So I may have gotten something wrong. Hope this gives you some insight of the forces involved.

 

[kjmccarx]

Kanik, thanks for that... It would be cool to see someone attach tension sensors to every spoke and see what the real-world forces and tensions look like.

As for whether I'm "really a rocket scientist," I'll leave that up to everyone else to decide, because I don't really care. I was just trying to make a joke... I didn't think people would get upset from a seemingly simple question, but I guess that's just pretty common online - lots of angry people.

 

[Kanik]

You wouldn't be able to apply tension sensors to each spoke, but you could use a laser extensiometer on a single spoke in a static loading test to measure the tension (by calculating with the tensile modulus of the spokes).
You could also just use finite element analysis on a computer to calculate stresses on a 3D model of the wheel. This is what engineers do when designing new wheelsets and hubs, especially lightweight ones and those with novel spoke designs. FEA divides the model into a finite but very numerous number of individual elements which can be individually solved by the computer to determine stresses in complex geometries. It is similar to how a computer might numerically evaluate an integral rather than solving it using algebraic principles as a human would do.

 

[ca_rider]

You wouldn't be able to apply tension sensors to each spoke, but you could use a laser extensiometer on a single spoke in a static loading test to measure the tension (by calculating with the tensile modulus of the spokes).
You could also just use finite element analysis on a computer to calculate stresses on a 3D model of the wheel. This is what engineers do when designing new wheelsets and hubs, especially lightweight ones and those with novel spoke designs. FEA divides the model into a finite but very numerous number of individual elements which can be individually solved by the computer to determine stresses in complex geometries. It is similar to how a computer might numerically evaluate an integral rather than solving it using algebraic principles as a human would do.

Strain gauges could be used to experimentally monitor spoke extension and relaxation under load.

Jobst Brandt shows some interesting FEA spoke loading results in his book <i>The Bicycle Wheel</i> where he develops the concept that sufficiently tensioned spokes support the applied load by compressing to a less tensioned state. It is a great read with respect to wheel building, and is an excellent means to visualize spoke loading. However I believe he only analyzes the static case with no moments applied but I may be wrong.

 

[ca_rider]

re: "As for whether I'm "really a rocket scientist," I'll leave that up to everyone else to decide, because I don't really care."

To me and probably others it sounds fishy because you literally asked a 1st year physics question and allude to the fact that the spokes have directionality by indicating "the spokes are transferring torques in directions they are not designed to".

As you likely know, spokes are essentially 1-dimensional elements which do not support bending and only transfer forces axially. One convenient artifact of the typical spoke arrangement at proper tension is that the standing load is mostly supported by spokes near the ground, but torque is transferred by the remaining spokes which are not compressed by load. Even though the load is not distributed evenly about the wheel and hub, the torque nearly is. Since the torques can greatly surpass the standing load, this distribution allows flange volume to be relatively small.

Kind regards.

 

[Flamingtaco]

There isn't a "Rocket Scientist" field last time I checked. Many differenent fields come together to design, build, and launch rockets. One could be a chemical engineer, for example, and have no knowledge whatsoever of structural engineeiring. If you worked for NASA, you'd still be a "Rocket Scientist".

-Former "Air Combat Scientist"

 

[Kanik]

Strain gauges could be used to experimentally monitor spoke extension and relaxation under load.

Jobst Brandt shows some interesting FEA spoke loading results in his book <i>The Bicycle Wheel</i> where he develops the concept that sufficiently tensioned spokes support the applied load by compressing to a less tensioned state. It is a great read with respect to wheel building, and is an excellent means to visualize spoke loading. However I believe he only analyzes the static case with no moments applied but I may be wrong.

Yeah, I suppose there actually are some strain gauges that could attach to the spokes. The laser extensiometer idea I just had because all you have to attach to the spoke is a pair of reflector stickers, like in a tensile test of a specimen on a universal testing machine.

This makes me think of how you could possibly measure tension or strain in a dynamic test. All I can think of now is using a flexible sticker, on which are printed, extremely finely, some thinly spaced lines, like the graduations on a rule. This could be applied to every spoke (that has the same angle to the hub). Then, a laser diffraction pattern could be projected onto the stickers on the spokes as the wheel rotates, and a high speed camera could record the wheel during the test. If the spokes tension or de-tension, the pattern on the sticker would stretch, but the laser diffraction pattern wouldn't, and it would create the effect of a vernier scale, possibly allowing accurate reading of the strain during a dynamic test.

You might also just be able to mount a strain gauge to the spoke and mount a little wireless transmitter to the hub shell. In thinking about it, that actually seems a lot easier.

I definitely want to read that book. I haven't gotten my hands on it yet. I'm also just learning about this stuff at uni. A year ago I didn't really know the difference between stress and strain because I hadn't taken a course on it yet.

But still, I would hope that any engineer would have taken first year physics. Like, I knew the rocket scientist thing was phoney; people are just being rude about it because its definitely a weird thing to assert oneself as whilst asking a question that is a little bit silly, even if the curiosity behind it is not silly (it is in fact, very interesting and an important part of our bikes).

re: "As for whether I'm "really a rocket scientist," I'll leave that up to everyone else to decide, because I don't really care."

To me and probably others it sounds fishy because you literally asked a 1st year physics question and allude to the fact that the spokes have directionality by indicating "the spokes are transferring torques in directions they are not designed to".

As you likely know, spokes are essentially 1-dimensional elements which do not support bending and only transfer forces axially. One convenient artifact of the typical spoke arrangement at proper tension is that the standing load is mostly supported by spokes near the ground, but torque is transferred by the remaining spokes which are not compressed by load. Even though the load is not distributed evenly about the wheel and hub, the torque nearly is. Since the torques can greatly surpass the standing load, this distribution allows flange volume to be relatively small.

Kind regards.

I dunno if I would really call spokes one dimensional due to their elbow (straight pull spokes however, sure, 1D, unless you're engineering the shape of their ends). The elbow is where most failures happen, so it is reasonable to say that you need more than one dimension to analyse them properly. I mean, it's for that reason that we have so many straight pull designs now...the elbow is a major weak point for the spoke.

Also, why do you say that the standing load is supported by the spokes near the ground? I haven't seen evidence for or against that, but it's just not what I would expect to be the case. Why would moments be transferred between spokes so easily, but not radial loads? I could totally be wrong here, I have no idea. Would be cool to find out.

 

[Cotharyus]

At least you didn't imply Avid brakes somehow apply less harmful directional stress to wheels than Shimano brakes. Then you'd have really made people made.

 

[127.0.0.1]

Yeah, I suppose there actually are some strain gauges that could attach to the spokes. The laser extensiometer idea I just had because all you have to attach to the spoke is a pair of reflector stickers, like in a tensile test of a specimen on a universal testing machine.

This makes me think of how you could possibly measure tension or strain in a dynamic test. All I can think of now is using a flexible sticker, on which are printed, extremely finely, some thinly spaced lines, like the graduations on a rule. This could be applied to every spoke (that has the same angle to the hub). Then, a laser diffraction pattern could be projected onto the stickers on the spokes as the wheel rotates, and a high speed camera could record the wheel during the test. If the spokes tension or de-tension, the pattern on the sticker would stretch, but the laser diffraction pattern wouldn't, and it would create the effect of a vernier scale, possibly allowing accurate reading of the strain during a dynamic test.

You might also just be able to mount a strain gauge to the spoke and mount a little wireless transmitter to the hub shell. In thinking about it, that actually seems a lot easier.

I definitely want to read that book. I haven't gotten my hands on it yet. I'm also just learning about this stuff at uni. A year ago I didn't really know the difference between stress and strain because I hadn't taken a course on it yet.

But still, I would hope that any engineer would have taken first year physics. Like, I knew the rocket scientist thing was phoney; people are just being rude about it because its definitely a weird thing to assert oneself as whilst asking a question that is a little bit silly, even if the curiosity behind it is not silly (it is in fact, very interesting and an important part of our bikes).



I dunno if I would really call spokes one dimensional due to their elbow (straight pull spokes however, sure, 1D, unless you're engineering the shape of their ends). The elbow is where most failures happen, so it is reasonable to say that you need more than one dimension to analyse them properly. I mean, it's for that reason that we have so many straight pull designs now...the elbow is a major weak point for the spoke.

Also, why do you say that the standing load is supported by the spokes near the ground? I haven't seen evidence for or against that, but it's just not what I would expect to be the case. Why would moments be transferred between spokes so easily, but not radial loads? I could totally be wrong here, I have no idea. Would be cool to find out.

pretty simple (complicated to set up, but simple afterward)

polarized strain inspection to monitor strain on all spokes

1) give all spokes a polycarbonate or acrylic coating

2) use a rainbow viewer, take a movie(s) of the wheel in use and under force

3) calculate the forces

 

[Joules]

I dunno if I would really call spokes one dimensional due to their elbow (straight pull spokes however, sure, 1D, unless you're engineering the shape of their ends). The elbow is where most failures happen, so it is reasonable to say that you need more than one dimension to analyse them properly. I mean, it's for that reason that we have so many straight pull designs now...the elbow is a major weak point for the spoke.

well, the second dimension you'd need in this case would be time (or frequency depending on how you want to look at it). The elbow doesn't fail under static loading, rather from fatigue cycling (which is why more tension prevents those failures). If you took a single wheel and hung 10,000 pounds on it, and just let it sit, the j-bend would never fail. Ever (assuming it survived the initial loading).

I think it's helpful to back up and think of a single spoke rather than trying to contemplate the wheel as a whole; put a single j-bend spoke in a hub. Try rotating it; do you see any way it would carry any load in any direction than on it's axis? In fact, a straight-pull spoke is actually subject to more side loading than a j-bend, again simply illustrated by a single spoke not attached to anything but the hub.
Now thread that single spoke into the rim and a nipple. Push on it. Is there any way that's going to carry any load in compression? No because the nipple would just push away from the rim, the same way a threaded hole can't push on a bolt.

I see young engineers thinking of systems as systems with behaviors and properties not consistent with their component parts. Sometimes that happens, but when you see properties inconsistent with the component parts, that's a good sign that you aren't understanding something. For example the standing wheel load supported by the spokes at the bottom. Obviously that's impossible; tension in the spokes between the hub and the ground can't push the hub up, and we just saw in that thought experiment that the spokes can't be compressed.

I've heard physics 101 summarized as follows:
1: f=ma
2: you can't push or twist a body with a rope
spokes are a rope.