A major part of developing a frame that strays from the normal conventions is spending the time to analyze the internal stresses to ensure that it does not fail prematurely. For our development of the Dropped Stay Concept, we ran through many iterations of Finite Element Analysis to calculate those stresses and to make sure that the deflections are all within the designed limits. To do so, we ran through three stages of the development; designing the frame geometry, determining the constraints, and analyzing the results.
Our first course of action was building out the models that we wanted to test and to ensure that they were as close to the same as possible so we could determine the impact of lowered seatstays. To do so, we built out three different models; a conventional double-diamond (Type 01), a dropped seatstay version (Type 02), and our Dropped Stay Concept (Type 03). Each of these frame models have the same front triangle geometry and chainstays while only changing the seat stays. The Type 02 frame does have an additional gusset in the front triangle but this was considered necessary as the stresses of the dropped seat stays would otherwise bend the thin seat tube. All of these models can be seen in the images below.
With these basic geometries sorted, we then applied the loads to determine how the each frame will flex while riding. After trying a number of different load cases, we settled on three cases that best demonstrate the high-stresses induced while riding; a seated impact, a standing impact, and a rear torsion test. We did run a number of additional tests that were beneficial for us to better understand exactly what is going on in the frame but the three mentioned above provide the best examples of how the frame will behave.
With the constraints set, the models built, and the test loads determined, all the remained was to run the tests and wait to see the results.
After waiting many hours for results, we finally were able to look at the deflections of the frames and determine what exactly it is that differentiates the various frame types. For the purposes of our analysis, we have converted all the deflections to spring rates (N/mm or Nm/mm) as we wish to compare stiffness and looking only at deflections is occasionally deceiving. Additionally, all comparisons of deflections in the model utilize the same scale for deflections for easier comparison. Lastly, any coloration in the frame models demonstrate the amount of deflection and not whether or not it is close to failure and all deflections are depicted as exaggerations to demonstrate what is happening and not give a realistic visualization of the actual deflection for the given load.
Vertical Seat Impact
The Vertical Seat Impact case, is designed to evaluate the amount of flex in the seat tube for the rider while seated. This case results in what we expected with the Type 03 frame exhibiting the most compliance and the Type 01 frame producing in a 70% stiffer configuration. The Type 02 frame falls between Type 01 and Type 03 configurations being biased more towards the compliance of the Type 03 frame.
Standing Impact Results
The Standing Impact Load case, imitating a 2G vertical landing with a standing rider, produces predictable results with the Type 01 frame being the stiffest again, this time by 87% over the Type 03 frame design. For these results, vertical deflections are compared at the BB center. As can be seen in the deformation image, there is also a significant deflection at the top of the seat post in the Type 03 frame caused by the deflection being spread out through the frame because of the junction of the seat stays with the down tube.
Rear Triangle Torsion
The Rear Triangle Torsion analysis isolates the flex of the rear triangle when the wheel is laterally loaded at the contact patch. While this does not directly imitate the load case of out of the saddle sprinting, it does provide a basis of comparison for out of the saddle efforts as well as torsional stiffness of the rear triangle during descending. The results show very similar torsional stiffness between the different frame types with the Type 02 being the least stiff and Type 03 being the stiffest by 50% over Type 01.
Of additional interest is again, the deflections at the seat tube, specifically of the Type 02 and Type 01 frames. Both have significant lateral deflections at the seat tube as a result of the rear triangle displacing the front triangle as well. The long seat stays of the Type 03 frame isolates the torsion loads from the front triangle.
Our results from the FE Analysis demonstrate that the Type 03 frame configuration (the Dropped Stay Concept) successfully provides vertical compliance while increasing lateral stiffness. This especially notable in the Vertical Seat Impact evaluation case were the Type 01 frame is 70% stiffer than the Type 03 configuration. This reduction in vertical stiffness was noted by our test riders in prototype frames where the seat posts were actually swapped for stiffer posts to tune the amount of flex for the rider. This tuneability is especially important as a designer as it produces a frame that through the changing of the seat post diameter and material, we are able to pair the frame stiffness with the rider weight down to lighter riders, something that is very difficult with the traditional double-diamond frame.
While the analysis does demonstrate that the Type 03 configuration has a lot of benefits for riders and designers alike, it does not speak to the utilitarian concerns of frame designs such as bottle and frame bag carrying capacity. For frames where these constraints are problematic (double-chainring drivetrains, large water bottle capacity in small frames, etc.) the Type 02 configuration presents itself as a viable alternative to the Type 01 configuration to reduce the stiffness at the seatpost and create a more supple frame.
Lastly, we also can see exactly why the traditional double-diamond frame has endured so long in the bicycle industry as it is a very efficient structure that transfers loads evenly throughout the frame and produces a very stiff structure. Exactly what was needed before modern materials and techniques necessitated a push to reduce frame stiffness to achieve a pleasant ride.