[iRider]

One potential issue is that point B may need further forward to allow more shock positioning options.

Why? Just extend the core further to the front without moving point B. The attachment points of the core don't have to match with the outside dimensions of the core. If you look at it in 3d there is space for both.

 

[TrueScotsman]

I see the point C as crucial. The Core must be wide enough in that area to be able to contain 135mm wide Speedhub/Alfine. Other gearboxes are much narrower (83mm max?) which makes me think that Core side faces should be 3d in all three versions of Core. Such a structure is advantegeous, however it's more expensive to manufacture.

I think we have to differentiate between what we call the "core" and the fixing points A,B and C. For instance the seat tube assembly at point C could be defined as a certain diameter and, say, 50mm wide, but the "core" that fixes to the outside of this could be either 83mm for some gearboxes or even 135mm for a Speedhub/Alfine. The ABC points are specific to the "standard" and the Core is specific to the application. Does this make sense?

Why? Just extend the core further to the front without moving point B. The attachment points of the core don't have to match with the outside dimensions of the core. If you look at it in 3d there is space for both.

You are absolutely right- the outside dimensions of the Core could exceed the lines between the points. The problem I have with this is, whilst points A to C and points B to C are relatively direct, this is not the case with points A to B. Unless we have a strut (adding weight) between A and B there is going to be a huge stress concentration at X (see sketch). This is why in page 4 I suggested moving point A forward to reduce this stress, but no-one seemed keen. A strut between A and B (attached to the head assembly) would limit forward shock positions and if it was part of the Core it would hugely increase Core size, making manufacturing harder and more expensive.


Any ideas?

 

[Steve M]

Put A where X is. Otherwise the whole pin-jointed-truss theory kinda goes out the window anyway.

 

[TrueScotsman]

Thank you Steve- that's what I was trying to say all along!

 

[Pslide]

I'm confused, why would you move A to X? Certainly there's no point in having a core that is as big as the frame itself, or am I missing something?

Now eliminating A altogether seems like a possibility. That's how the Italians do it...

 

[TrueScotsman]

Moving A to X makes an inherently strong pin-jointed triangle. You could have a truss frame like your italian motorcycle (good example) but it would probably be heavier and also restrict the dimensions of the Core.

Now, this is where I think we need some clarity regarding terms;

I propose that we refer to the "Core" as the part joining points B and C. the part which encompasses the drivetrain and suspension components. The outer dimensions of such a Core are yet to be defined.

I propose that we refer to the points A,B and C as the "ABC concept" (for want of a better term). These are just defined points in space with specific mounting dimensions (tbc).

You are right, we can totally ignore point A if we wish. (Although it reduces the modular nature of the concept)

After consulting with others and playing with shock orientations, here is a refined drawing of the "ABC concept" points. Point A should be in such a position as to fit S,M,L,&XL frames;


Here is a possible doodle (MS Paint alert!!!) of how they could be combined together. Imagine a Alu downtube, a CF top tube/seat tower and a CNC'd alu Core;


If people could take these dimensions and see how they work (or don't) with their own design I would be grateful.
Feedback, as always, is welcome.

 

[OGRipper]

This is a pretty cool idea with a lot of interesting possibilities, and it's great to see this kind of thinking. You can look at it as an expansion of proposed gearbox "standards" that will allow other types of suspension systems. As others noted the open source and modular concepts will require design features that may be seen as compromises or negatives as compared to the top "integrated" designs out there.

And I can't see anyone contributing a protectable suspension design on an open source basis, but you could still have protected designs built into modular components. Whether that happens or not remains to be seen. Would Trek or Specialized be willing to sell a CORE/rear end so people can build frankenbikes around their protected technology? If so, other than some committed DIY'ers, who would want one over the integrated bike? Still, the whole thing is pretty neat as a concept/design exercise and I hope you guys keep it up.

 

[Steve M]

I'm confused, why would you move A to X? Certainly there's no point in having a core that is as big as the frame itself, or am I missing something?

Now eliminating A altogether seems like a possibility. That's how the Italians do it...

The Core only has to attach to B and C. You'd need some clever, highly robust interfaces at all the points.

 

[TrueScotsman]

You'd need some clever, highly robust interfaces at all the points.

Not sure I follow Steve, are you saying that even with the changes to position A we would still need highly robust interfaces? If so, what type of fixings would you recommend?

As for the ABC positions- what do most people think?

Regarding the Core "area"- I have been thinking about defining a maximum circular area encompassing points B and C. This circular area allows for more shock positions than just straight lines connecting point B and C and the BB. Obviously this area would have to defined in 3d (especially where the point connections overlap) but this is kinda where I am getting at;

 

[graham888]

As far as the "ABC" fixings go, I have always envisaged some sort of 12mm ally axle system at each point much like that found through the rear wheel, maybe with an added interface of some sort to help locate the parts and reduce any vertical flex slightly.

ABC seem fine to me as they are now, though it needs to be made clear that the Y co-ordinates are measured from rear axle, not the BB, as the BB height is surely a variable of the cores design.

Is this maximum circular area really required? I see that your trying to ensure that all toptubes etc fit all the cores, but I'm not so sure it is needed. Surely the extremities of the core are highly dependent on the suspension design and should be left to the judgment of the designer?

 

[TrueScotsman]

As far as the "ABC" fixings go, I have always envisaged some sort of 12mm ally axle system at each point much like that found through the rear wheel, maybe with an added interface of some sort to help locate the parts and reduce any vertical flex slightly.

Cheers for the feedback! I too reckon a 12mm axle system is a good idea (this way a standard M12 bolt would also get you out of a fix). The "added interface" is a cool idea (sort of like radial grooves?) but would have to be specced correctly so that the two parts meet at the right angle. Have you any thoughts/examples of such a system?

ABC seem fine to me as they are now, though it needs to be made clear that the Y co-ordinates are measured from rear axle, not the BB height, as the BB height is surely a variable of the cores design.

Very true about the Y co-ordinates- I hadn't spotted that. BB height could be adjusted by altering the swingarm height but this would have consequences for the head angle.
The X co-ordinates would have to be taken from the BB and the Y co-ordinates from the axle height.

Is this maximum circular area really required? I see that your trying to ensure that all toptubes etc fit all the cores, but I'm not so sure it is needed. Surely the extremities of the core are highly dependent on the suspension design and should be left to the judgment of the designer?

I know what you are saying, but surely we would need something defined so as not have parts conflicting?. This will be better to be defined in 3d.

Further comments anyone?..............

 

[TrueScotsman]

Right, I have amended the ABC contact points to measure from the new datum point (X co-ords from the BB centre and Y co-ords from the rear axle centre) Would someone please double check these points so we can agree and move on to the next stage?



I have also been thinking about what graham888 said about a point connection interface to resist flexing. I originally thought of a radially grooved interface (like a splined axle) but also realised that manufacture and accurate jigging may present a problem. How about if all ABC points had a double-fixture design? It may be heavier, but it would be easier to manufacture and resist rotational movement. No?



Has anyone got a better idea?

 

[Pslide]

Maxles, all around!





No, just kidding.





ISIS!


No No....kidding again.






First gen Revolt single shear hardware!


Sorry, guess I don't have the answer...

 

[fluider]

"Right, I have amended the ABC contact points to measure from the new datum point (X co-ords from the BB centre and Y co-ords from the rear axle centre)"
Sorry for jumping at it but I can't see a reason why would you want to describe a point's coordinates by two different reference points. I think the BB location within the Core is crucial so I find it to be the most logical reference point. But that's just my opinion of course.

As for the connection interface. Perhaps a female axle with eliptical outer shape could prevent rotational movement and long screw would fix it sideways. You can rotate the elipse main axis against the direction from which the highest load is expected to come.

 

[TrueScotsman]

"Right, I have amended the ABC contact points to measure from the new datum point (X co-ords from the BB centre and Y co-ords from the rear axle centre)"
Sorry for jumping at it but I can't see a reason why would you want to describe a point's coordinates by two different reference points. I think the BB location within the Core is crucial so I find it to be the most logical reference point. But that's just my opinion of course.

I originally thought that the BB was the most logical reference point, but as graham888 pointed out that this doesn't give the option of altering the height of the BB in a Core. If you used the rear axle as a reference point that doesn't give you the option of altering the length of a swingarm. But, by combining the two, you get total X and Y adjustability.

As for the connection interface. Perhaps a female axle with eliptical outer shape could prevent rotational movement and long screw would fix it sideways. You can rotate the elipse main axis against the direction from which the highest load is expected to come.

I like the idea of an eliptical interface, but I would need to see more details to fully understand it.

Cheers fluider.

 

[TrueScotsman]

Ooops, sorry -double post.

Whilst I am typing;
You've heard of the saying "square peg in a round hole"- what if the connections were like this?
What i mean is square connectors that locate into each other with a round hole for an axle through this. The square (or even hexagonal may be better) would resist rotational forces and the simple thru' axle would tie it together. I will post a sketch of what i mean.

EDIT- here we go-

 

[LMC]

I would stick to what you had drawn in the 2nd pic on post #92. The only difference i would make would be to space the holes out so that either

a) one of the bolts can serve as a shock bolt or;
b) both bolts can be used to fix a shock mount to the frame like the orabge 224 shock mount.

I would say 2 m8 bolts with a 70mm gap between the holes should do.

 

[fluider]

I'd say one tries to avoid using any square shapes in mechanical engineering, especially when they shall be used in inside corners since the sharp edges concentrate stress. Unless your aim is to concentrate stress and lead in some direction. Perhaps a head flange of the axle is safe in this (like Trek gravity bikes have their flip-chips designed).
I was meaning something like this.