Advertise on Bikeforums.net



User Tag List

Results 1 to 21 of 21
  1. #1
    Junior Member
    Join Date
    Jun 2012
    Posts
    8
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)

    retrofitting disc brake to a track frame, design advice

    Hi everyone,

    I'm a new member here but have been reading along for a while now. I figured it'd be okay if I asked you about a problem I am trying to solve at the moment.
    I do not know if it is placed correctly here in "frame building", but I thought the audience of this subforum might be appropriate- in any case, please move this thread if it does not belong here and sorry if so.
    So let's get straight to it:
    I want to have a rear disc brake on a new school aluminum track-style frame that is rather beefy and very stiff. I would appreciate if we do not get into the whole
    "why would you wanna do such a stupid thing" discussion, but rather focus on the physics and the engineering involved in finding a solution that has the most advantages.
    I want to have disc brakes and I will do it. And I love doing it for the hell of it. haha
    So anyway, like I said, it is a track style frame, yet it has a mount for a rear road caliper brake in the short tube between the seatstays just above the tire. It also has track ends, sometimes called "horizontal dropouts" (yes, it has 120mm rear spacing, too). I have been thinking a lot about the optimal solution for the problem and I have so far come up with a couple of concepts.
    I figured I'd share the three of them with you that seem to be the most diverse in design so we have different approaches to get started.
    the first design seems to be the heaviest, the second could be lighter and the third is probably the lightest. I do not want to alter the frame, so welding or drilling is not an option.
    the dimensions and proportions in these drawings are not meant to be absolutely correct.

    here they are, I hope my explanations are not too messed up



    I'd really appreciate any advice, thoughts or ideas from you! Thanks a lot in advance,

    all the best

    meko
    Attached Images Attached Images

  2. #2
    Senior Member squirtdad's Avatar
    Join Date
    Jul 2006
    Location
    San Jose (Willow Glen) Ca
    My Bikes
    '89 Miyata 1400, '82 nishiski (current utilty/commuter project)
    Posts
    3,329
    Mentioned
    8 Post(s)
    Tagged
    0 Thread(s)
    here is a thread with some discussion of the subject. http://www.bikeforums.net/showthread...liper-location

    But before you get too far....what are you doing for hubs? AFAIK Disk brakes require hubs designed to mount the rotors. Does any one make these in a 120 mm hub? If not you will need to solve how to reset dropout size (not easy on aluminum frames) or find an adapter, which might mean adjusting wheel dish. so you need to take hubs in to consideration also

    Nice renderings by the way.
    '82 Nishiski commuter/utility
    '83 Torpado Super Strada ... cafe commuter
    '89 Miyata 1400
    Soma rush Fixie
    '78 Univega gran turismo (son's Fixie/SS)
    06 Haro x3 (son's bmx)
    Electra cruiser (wife's bike)

    looking for: De Rosa 58cm ELOS frame and fork internal cable routing

  3. #3
    Junior Member
    Join Date
    Jun 2012
    Posts
    8
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    HI squirtdad, thanks for the fast reply! I have read the thread you suggested, thank you nonetheless. It had some very valuable information. Indeed the hubs have to be disc-specific. I could either use phil wood ISO-cog hubs (expensive as hell):



    they have a 6-bolt ISO disc brake rotor pattern on both sides and I could then use a rotor on one side and an ISO 6-bolt disc cog on the other.

    or I could be really fancy and use a MTB front disc hub with 100mm spacing, rebuild it with a 10mm solid axle and use spacers to get to 120mm and then mount the rotor and the cog on the same side, maybe even left hand drive, depending on my disc brake system. a little like this:



    but the wheel could be a little too dished that way and it sure would be a very tight fit in a track frame, maybe a 140mm rotor wouldn't even fit.

    Thanks for the compliment I can't explain these things too well with words

    Oh, and yeah, it would be a fixed transmission WITH a rear disc brake. One could call that pointless but I want to be able to stop the wheel at any moment.
    Last edited by meko; 06-14-12 at 05:13 PM.

  4. #4
    Senior Member squirtdad's Avatar
    Join Date
    Jul 2006
    Location
    San Jose (Willow Glen) Ca
    My Bikes
    '89 Miyata 1400, '82 nishiski (current utilty/commuter project)
    Posts
    3,329
    Mentioned
    8 Post(s)
    Tagged
    0 Thread(s)
    Quote Originally Posted by meko View Post
    snip oh, and yeah, It would be a fixed transmission WITH a rear disc brake. One could call that pointless but I want to be able to stop the wheel at any moment.
    nothing wrong with that. I required my son's fixie we built to have a front brake. You are now at the limit of my limited help. Need to wait for some of the heavy duty guys to chime in.
    '82 Nishiski commuter/utility
    '83 Torpado Super Strada ... cafe commuter
    '89 Miyata 1400
    Soma rush Fixie
    '78 Univega gran turismo (son's Fixie/SS)
    06 Haro x3 (son's bmx)
    Electra cruiser (wife's bike)

    looking for: De Rosa 58cm ELOS frame and fork internal cable routing

  5. #5
    Junior Member
    Join Date
    Jun 2012
    Posts
    8
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Quote Originally Posted by squirtdad View Post
    nothing wrong with that. I required my son's fixie we built to have a front brake.
    I think that's sensible. I would not ride a bike without a dedicated brake system. I'll be going front and rear disc, I am required to have two brakes on my bike here in Germany anyways.

  6. #6
    Senior Member
    Join Date
    Jul 2011
    Posts
    1,543
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Something between 2 and 3 is a pretty common set-up on motorcycles, often the front wheel. This because the linkages on springers rule out the forks as a place to mount disc calipers.

    http://www.stony-point.com/BIKE-PROJ...EDITED/010.JPG

    Myself I would put two brake systems on the front wheel. The rear wheel isn`t all that effective.

  7. #7
    Junior Member
    Join Date
    Jun 2012
    Posts
    8
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Thanks for your reply!

    Quote Originally Posted by MassiveD View Post
    Myself I would put two brake systems on the front wheel. The rear wheel isn`t all that effective.
    haha what a cool idea! I haven't even thought about that. But still, on the bike I am building it would feel weird to be using two systems redundantly just to please the cops. If I have to have two brakes and I have two wheels, I think I am going to put a brake on each wheel. Plus, I like the challenge of it.


    Quote Originally Posted by MassiveD View Post
    Something between 2 and 3 is a pretty common set-up on motorcycles, often the front wheel. This because the linkages on springers rule out the forks as a place to mount disc calipers.
    Yeah, right. I think I should look into motorcycles a bit more to get some inspiration. The brake being kind of "free" to rotate around the wheel axle on its own floating arm is a good idea I think, especially with a sliding rear axle.

    Does anyone have an idea what torque would be generated when stopping / totally locking up the rear wheel? Disc brake tests I have read measured the torque of a couple of brakes and the best brake was measured at 124 N*m, but I doubt that it would be that high at the rear wheel of a non-mountain bike. Any numbers anyone?

  8. #8
    coprolite fietsbob's Avatar
    Join Date
    Jun 2010
    Location
    NW,Oregon Coast
    My Bikes
    7
    Posts
    2,564
    Mentioned
    41 Post(s)
    Tagged
    0 Thread(s)
    Seen a bike shown at the handmade bike exhibition, a fixie,
    they put the disc for the rear wheel, on the left crank arm.

  9. #9
    Junior Member
    Join Date
    Jun 2012
    Posts
    8
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Quote Originally Posted by fietsbob View Post
    Seen a bike shown at the handmade bike exhibition, a fixie,
    they put the disc for the rear wheel, on the left crank arm.
    yeah, I think I have seen that on the web. I really like the ingenuity, but I think it is rather a thing for steel frames and custom welding. If you brake at the crank arms, the torque is a lot higher because of the transmission. For a "typical" singlespeed gear ratio it would be almost three times the torque. That would equal moving the disc brake mounts on the frame further to the axle, increasing the force the frame must take. I am trying to do the exact opposite, moving the transfer point away from the axle. On top of that, such a system would put a lot of stress on the chain and the teeth.

  10. #10
    tuz
    tuz is offline
    Senior Member
    Join Date
    Dec 2006
    Location
    Toronto/Montréal
    My Bikes
    Homemade mixte, track, commuter and road, Ryffranck road
    Posts
    1,159
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Here is my estimation for the upper range of torque seen on a wheel. Say the weight on the wheel is 70kg and the wheel radius is 338 mm (700cx25). With the coefficient of fraction between rubber and asphalt = 0.9, the force is 70*9.81*0.9 = 620 N and the torque 620*0.34=210 N*m.
    homebuilt commuter, mixte, road and track (+ Ryffranck road)
    bla bla blog

  11. #11
    Junior Member
    Join Date
    Jun 2012
    Posts
    8
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Quote Originally Posted by tuz View Post
    Here is my estimation for the upper range of torque seen on a wheel. Say the weight on the wheel is 70kg and the wheel radius is 338 mm (700cx25). With the coefficient of fraction between rubber and asphalt = 0.9, the force is 70*9.81*0.9 = 620 N and the torque 620*0.34=210 N*m.
    thanks for that! I had forgotten the exact formulas. Very cool. So let's assume we take a torque of say 250 N*m to be on the safe side.
    If we then look at my third design with the rod pulling at the bottom bracket, if we attach that rod at a radius of 8cm (~3 in) from the rear axle, the rod would be pulling at the BB shell with a force of
    250 N*m / 0.08 m = 3125 N, which would equal roughly 320 Kg or 700 lbs. Since it is the rear wheel, the actual force is probably a lot less.

    Can a chainstay construction withstand such a compressing force? And, since the triangle that the rod and the chainstay form pushes the BB area down, can the frame take it?
    Would the first or second design be better, directing the force into the top area of the seatstays, almost like a rim brake would?

    I am kind of lost here, thanks for your help this far!
    Last edited by meko; 06-15-12 at 02:38 PM.

  12. #12
    Senior Member
    Join Date
    Jul 2011
    Posts
    1,543
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Quote Originally Posted by meko View Post
    Thanks for your reply!



    haha what a cool idea! I haven't even thought about that. But still, on the bike I am building it would feel weird to be using two systems redundantly just to please the cops. If I have to have two brakes and I have two wheels, I think I am going to put a brake on each wheel. Plus, I like the challenge of it..
    Cool.

    I'm not joking though. I want all the braking on the front wheel, and in my application I will have a rear brake also, but not a disc, because they are so heavy. In mud I guess the issues are different, but on the road I find regular calipers will slow the rear wheel as much as there is traction back there anyway. On tandems they have three brakes, and I like myself as much as I do any possible tandem partner.

  13. #13
    Junior Member
    Join Date
    Jan 2009
    Posts
    20
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    I have a friend who has been running a very similar setup to the one pictured here : http://gallery.mtbr.com/showphoto.ph.../63986/cat/516 for close to 10 years without any problems. He is running it on a a specailized s-works aluminum XC mountain frame and has ridden dual slalom, xc, etc. Also, just a note regarding forces- unless you have these things riding on some sort of bearing or bushing a lot of the force will be translated into your mount & axle and not along the lever arm of the mount/chainstay. I would almost venture to guess that design #3 would not rotate (much) without the tensile rod assuming you use sufficient clamping force holding the mount against the wheel. I don't see a big issue with any of the designs; I just happen to know that design 1 matches some products that are already being sold/used.

  14. #14
    Junior Member
    Join Date
    Jan 2009
    Posts
    20
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    I also found this thread: http://forums.mtbr.com/singlespeed/r...er-314985.html which has a ton of useful info- probably a few very good solutions that should be relatively easy to get a hold of.

  15. #15
    Junior Member
    Join Date
    Jun 2012
    Posts
    8
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Quote Originally Posted by MassiveD View Post
    Cool.

    I'm not joking though. I want all the braking on the front wheel, and in my application I will have a rear brake also, but not a disc, because they are so heavy. In mud I guess the issues are different, but on the road I find regular calipers will slow the rear wheel as much as there is traction back there anyway. On tandems they have three brakes, and I like myself as much as I do any possible tandem partner.
    haha I like that I know the front is the wheel to have brakes on and then some, but are you saying you will be using two brakes on a front wheel? Two discs like motorcycles have them?
    I've actually seen a front wheel on a downhill bicycle have two rotors on some bike expo... If you know of any MTB front hubs that take two discs, let me know!


    Quote Originally Posted by anhizer View Post
    I have a friend who has been running a very similar setup to the one pictured here : http://gallery.mtbr.com/showphoto.ph.../63986/cat/516 for close to 10 years without any problems. He is running it on a a specailized s-works aluminum XC mountain frame and has ridden dual slalom, xc, etc
    Yeah, that is the specialized shark fin. It is basically the same principle as in my design 1. The problem is that such a rigid torque lever must be very stiff and therefore is a little bit on the heavy side.
    The second design of mine is based on the idea of the "brake therapy" adapter: http://bikewatchnyc.files.wordpress....ke_therapy.jpg
    In my understanding, it forms that "shark fin" of a rigid torque lever with the help of the seatstay, because a rigid triangle is created and the force can be sent into the whole frame rather than just into the dropouts.

    Quote Originally Posted by anhizer View Post
    Also, just a note regarding forces- unless you have these things riding on some sort of bearing or bushing a lot of the force will be translated into your mount & axle and not along the lever arm of the mount/chainstay. I would almost venture to guess that design #3 would not rotate (much) without the tensile rod assuming you use sufficient clamping force holding the mount against the wheel.
    That is a very good point. I agree with you that the clamping force of the rear axle assembly will basically fix the brake adapter relative to the dropouts in angle and not let the brake caliper move relative to the dropouts. However, in my opinion, the whole dropout area will basically want to flex and twist as brake force is applied, thus the whole rear axle assembly including the brake system will slightly rotate/flex relative to the whole frame. So the torque arms / rods are there to couple the entire rear axis assembly to the frame as a whole.

    Quote Originally Posted by anhizer View Post
    I also found this thread: http://forums.mtbr.com/singlespeed/r...er-314985.html which has a ton of useful info- probably a few very good solutions that should be relatively easy to get a hold of.
    Thanks for digging this up, I have been reading it a few weeks back but had not bookmarked it. The thing with the adapters only engaging the dropout area is basically what I stated above, and please correct me if I am wrong, but they seem to be assuming the rear triangle is basically disc-ready and the only thing that is missing is the disc brake caliper mounts on the frame. I do not think that is a very sensible approach. It may work with a lot of frames, but at least for my application, I consider it too unsafe.

    I have so far found a few designs like #1, a lot of designs like #2, but none that works like the third one I drew. I think it could be unusable when braking the wheel while moving in reverse, but apart from that, I do not see any major flaws.

    It pulls down the chainstays at the bottom bracket rather than push down the seatstays into the seat tube like all the other designs. Is that a bad idea?
    Last edited by meko; 06-15-12 at 05:40 PM.

  16. #16
    Senior Member
    Join Date
    Nov 2010
    Location
    River City, OR
    Posts
    586
    Mentioned
    1 Post(s)
    Tagged
    0 Thread(s)
    If you consider the caliper mounting design of a motorcycle (both disc and drum), the caliper floats on the axle and rotation when brakes are applied is controlled by a torque arm, or torque link, or strut- or whatever you choose to call it. I believe this was done mostly on suspended bikes to reduce squat or rise when the brakes were applied. This wouldn't apply to a rigid frame like a bicycle.

    That being said, if I were fitting a disc brake caliper to a bicycle I'd create a mount to float on the axle. It would require some machine work but easily doable. Use a torque link to control rotation.


    Force or stress on the frame? Don't have any evidence but I'm guessing the braking forces transferred to the frame are much lower than one might think if the systerm is designed properly.


    Or... you could just do a coaster brake hub. LOL

  17. #17
    Junior Member
    Join Date
    Jan 2009
    Posts
    20
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Yeah, that is the specialized shark fin. It is basically the same principle as in my design 1. The problem is that such a rigid torque lever must be very stiff and therefore is a little bit on the heavy side.
    Depending on the geometry of the support arm you can minimize the amount of structure you include. making the bracket out of some lightweight aluminum L extrusion should be sufficient for the lever arm portion, though where the caliper is mounted you should probably add some additional material.

    However, in my opinion, the whole dropout area will basically want to flex and twist as brake force is applied, thus the whole rear axle assembly including the brake system will slightly rotate/flex relative to the whole frame.
    I agree this is true, but I question if this flexing force will be significant enough to be noticeable.

    I think it is important to think of all of the forces the rear wheel and real triangle is seeing. Yes, the disc brake will be stopping the weight of the rider, wheel, rotational forces of the wheel, etc. However on the other side of the bike you have the drivetrain which is putting quite a bit of force (especially with a heavy rider, high gearing, going up hill, etc) on the CS/SS/Dropouts without excessive flexing.

    As I mentioned before I think that all the designs will work well enough, but that some designs will be easier to manufacture than others. Design 1 can be made in a variety of fashions from a simple weld job of some off the shelf materials to casting to a full on NC made part.

    Designs 2 and 3 are more complicated, have "moving" parts (fasteners, etc) which complicate things and add weight. During my day job I subscribe heavily to the "KISS" mentality- 1 part with no fasteners vs 3+parts with fasteners is a easy decision to me. Any weight savings will likely be offset by having to fart around with threading, adjusting, & general fussing in the future.

    That being said its your project and general farting around is why I like working on bikes anyways

    Good Luck!

  18. #18
    tuz
    tuz is offline
    Senior Member
    Join Date
    Dec 2006
    Location
    Toronto/Montréal
    My Bikes
    Homemade mixte, track, commuter and road, Ryffranck road
    Posts
    1,159
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    As far as I can understand, #1 is a typical torque arm, while #2 & 3 try to triangulate the forces. I think #3 does a better job at that since you can make it act inline with the centre of the BB. If the tension is 700 lbs, a working stress of 20 kpsi (a good number for steel I think) gives a cross sectional area of 0.035 in2. That's achieved with a 1/4" rod.
    homebuilt commuter, mixte, road and track (+ Ryffranck road)
    bla bla blog

  19. #19
    Junior Member
    Join Date
    Jun 2012
    Posts
    8
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Quote Originally Posted by anhizer View Post
    Depending on the geometry of the support arm you can minimize the amount of structure you include. making the bracket out of some lightweight aluminum L extrusion should be sufficient for the lever arm portion, though where the caliper is mounted you should probably add some additional material.
    Quote Originally Posted by anhizer View Post
    Design 1 can be made in a variety of fashions from a simple weld job of some off the shelf materials to casting to a full on NC made part.
    That is true. However, I have not studied engineering or something along the lines of it and I don't have the skills (yet) to perform a stress-analysis with the help of CAD-software. Whenever I build something, I estimate, look at other similar designs that work, and then I overengineer everything to make sure it will be very unlikely to ever fail. The thing with bikes is that weight is an issue and I cannot overengineer these parts like I used to. I am afraid I am really too unexperienced to estimate the shape and material in the right scale for this application.
    Then again, I could easily calculate the forces, look at material datasheets and then build prototypes and test them with a dummy weight load. Why not...
    The thing with design 3 is that I just have to have a material that will withstand the pulling force that is easily calculated. But you are very right with:

    Quote Originally Posted by anhizer View Post
    Designs 2 and 3 are more complicated, have "moving" parts (fasteners, etc) which complicate things and add weight. During my day job I subscribe heavily to the "KISS" mentality- 1 part with no fasteners vs 3+parts with fasteners is a easy decision to me. Any weight savings will likely be offset by having to fart around with threading, adjusting, & general fussing in the future.
    really, very good point. I'll have to reconsider the one-piece design.

    Quote Originally Posted by anhizer View Post
    That being said its your project and general farting around is why I like working on bikes anyways
    Good Luck!
    haha so true! thanks. I love it as well.


    Quote Originally Posted by tuz View Post
    As far as I can understand, #1 is a typical torque arm, while #2 & 3 try to triangulate the forces. I think #3 does a better job at that since you can make it act inline with the centre of the BB. If the tension is 700 lbs, a working stress of 20 kpsi (a good number for steel I think) gives a cross sectional area of 0.035 in2. That's achieved with a 1/4" rod.
    Thanks for the reply tuz!

    I have been thinking about a 6mm tensile rod as well (roughly 1/4in) or maybe even 4mm when using high tensile strength steel (class 12.9 threaded rod for example). Do you think the frame could take that pulling force? (although I think it will be a lot less because it is the rear wheel)

    What you said about the center of the BB is quite true, I did not think of that advantage. But you mean the center of the BB seen from the side, which is the point the crank axle rotates around, right?


    Okay I have some news: I have aquired a very nice piece of machined goodness for very little money:





    SESSIONPOST6.jpgSESSIONPOST9.jpg

    it is the brake therapy floating brake system for a trek session (full suspension MTB).
    It has a very nice and long arm that the caliper mounts to and is floating around the axle via a bearing. with that piece, I can build all three designs. And the long torque transfer arm also is included!
    Thanks for all the help so far, I really appreciate that a lot!!
    Last edited by meko; 06-18-12 at 09:56 AM.

  20. #20
    tuz
    tuz is offline
    Senior Member
    Join Date
    Dec 2006
    Location
    Toronto/Montréal
    My Bikes
    Homemade mixte, track, commuter and road, Ryffranck road
    Posts
    1,159
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    I think the frame would be okay. The triangulation puts the parts in compression & tension, and materials are very strong in that direction. Looking at what is out there, generally a simple strut is used between the seatstay and chainstay. It seems to work. But that only transmits the bending load to the chainstay. A bending torque of 700x3 lbs.in will bend a 3/4x0.035 steel tube.

    Yes I meant that the forces on chainstay and link go through the same centre. In #2 the force acts slightly above the seatstay and will stress the the clamp assembly.

    However I just realized that you will loose some triangulation from the fact that the chainstays are angled with respect to the disk (seen from above). So some of the 700lbs will act in bending. Same for the link if you can't make it inline with the disk (might happen if you are constrained by the BB shell width).

    Edit. If the chainstay angle is 7 deg (kinda typical), 12% of the force will be in bending.
    Last edited by tuz; 06-18-12 at 12:09 PM.
    homebuilt commuter, mixte, road and track (+ Ryffranck road)
    bla bla blog

  21. #21
    coprolite fietsbob's Avatar
    Join Date
    Jun 2010
    Location
    NW,Oregon Coast
    My Bikes
    7
    Posts
    2,564
    Mentioned
    41 Post(s)
    Tagged
    0 Thread(s)
    My Winter Ice bike has S-A drum brakes , I didnt have to modify the frame ..
    they stop in a smooth, easily modulated manner..

    since then I got a BikeFriday with BB7 disc brakes..

    the front one, if not used carefully, can throw me out of the saddle.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •