This is my third frame and its a full squish. For background this bike will be a 160/140 travel, and being a very tall person, I have a longer than usual headtube (160mm). I got a bit paranoid about strength and rigidity and came up with the pictured monstrosity. In looking at other steel full suspension bikes I saw this was a commonly reinforced area and so added these plates to mine.
My question for others with experience building/riding full-suspension steel bikes; did I go overboard? How far overboard?
Hmm…I would say way overboard to be honest. The big issue here is you have an incredibly stiff section on the down tube stepping of into the thinner walls. I am assuming you are using straight gauge here or have you used a butted down tube. In any case you have concentrated and stress being put into the down tube right at the end of the gussets and any bending will start right on the toe of each. Having them offset will help a little but it will crack at some point in its life at the top gusset.
With a longer head tube and having the TT and DT spaced further apart you already have an inherently stronger setup and would suggest that underside Bontrager gusset is all that you’d need at most.
I’m building 170mm travel enduro race bikes and am bronze brazing and have never put a gusset on any of my frames. I had one of my riders huck to flat a massive drop and slightly buckled a butted top tube at the toe of the internal taper. It had a bi-lam style joint. I ditched the bi-lam and went up one wall thickness size in both TT and DT and it has been no issues since.
thanks for the detailed response. I had thought that a longer headtube would be weaker because as the TT and DT get further apart, you have less of a triangle and more of a rectangle. I’ll keep an eye out for cracks on it, but ideally this bike won’t be ridden long before I’m moving on to a version 2.
You’ll probably get all the use you need out if it so not a waste of time.
The wide spaceing of the two tubes creates a more stable base. I’ll try an analogy. Imagine your legs are a TT and a DT. Stand with your feet together and you are easily pushed sideways. If your feet are stuck to the ground your ankles bend etc. Move your feet apart and straight away you have a more stable base and way harder to push sideways.
When the tubes are close together its easier for the force from the forks try to rotate the head tube which in turn puts more bending in the TT & DT tubes and causes stress cracking over time. With wider spacing you get much less bending and less stress. It also helps spread the forces along the tubes further. If the tubes are butted it helps even more as the ends are strong for the joint but the thinner section bends more absorbing more of the stress force.
Under normal riding, you are safely in the elastic zone. If you case a jump or slam into a stump, that is a different story. But you could argue that is pilot error not builder error.
I feel the need to add a few corrections to @Daniel_Y’s post, hope that’s ok, as that is not really how fatigue failure works. I think I was unclear in my messy old post that you link to , as that was ended up as an attempt to simulate fatigue failures with tensile failure (as the graph you show here shows), which didn’t really work. Some extrapolations from the hardness tests were used to hypothesize about why fatigue occurs at the toe of the weld (which is a region within the heat affected zone). Not any clear answers there, was hoping to build upon that with more work but that has not happened.
Fatigue happens at nominal stress levels below yield .There is a lower limit as well, shown below as endurance limit, but the take-away is that higher stress/harder riding (or decreasing material thickness) leads to fatigue failure at decreasing cyles/quicker. The actual stress level required for any particular frame and joint is very difficult to predict since welding is both imprecise in geometry and also imparts its own stresses that vary with every little detail of welding (they add to the applied riding stress decreasing the expected cycles to failure). Some qualitative analysis with FEA or just looking at what experience (like how BMX frames are gusseted, or not) would help avoid geometries that are especially susceptible. The tube gusset like you’ve made here is a geometry they no longer use. “Doubling plate” type tube gussets seem to be the most common, though some have custom shaped tubing as well.
What this means is that if you ride hard, you will likely over enough time, crack any frame. Through my 8 years of WC DH racing production/normal bikes (15+ frames from different brands in alu and carbon, as well as a few of my own steel ones), I cracked or snapped all frames but one before each season was through. Top riders swap equipment often to avoid these issues. The recent one with Bernard Kerr in Rotorua was apparently a frame that had been used for more than one season.
So in relation to @danB, how hard do you ride? IMO this frame will work just fine, for quite a while. Keep an eye on the welds at the ends of the gusset, and at the top tube and the tube gusset. In my experience the rear ends face a much harder time on full-sus bikes (though that depends on the type you choose), so any care you can take there to avoid weld defects in crucial areas of pivoting are also well worth it. Fatigue cracks occur without any plastic deformation, so bending and buckling will not warn you of a crack. Those are, however, also failures that can occur, but those will happen from large, single occurance forces.
Thanks to everyone for the great info. I had a couple follow up questions after taking a closer look at REEB and Starling. I noticed that on both bikes, the reinforcement is slightly different than a full gusset.
Starling welds on a small plate that connects the TT and DT, but doesn’t connect to the headtube. The REEB Steezl and SST have plates that do connect all three tubes, but leaves the corners chopped off and unwelded. Do you think this is done to relieve stress? I’m still slightly curious as to what the goal of reinforcing like this is, as based on the feedback here and in another thread about gussets (can’t find it again, or I would link it) it seems these things either cause more issues, or at best, do nothing. I assume its simply for style, or to make the bike look “beefy”?
Also, a friend who built his first full-sus along side me asked me to post his gusset solution for feedback as well. Other than him clearly being a better welder than me, does this single, centered gusset have any drawbacks or advantages that differ significantly from my design?
I think there is a lot of eyeball engineering going on with smaller frame companies. Unless you test both styles multiple times, you will not get an answer about which style is better. Even then, it might be inconclusive. Fatigue is
In my opinion, I don’t like this design. If you see how the stress is distributed from a front end collision, the areas of high stress are where they placed the gusset:
This is why people weld gussets to the sides of the tube (blue areas = low stress) and avoid welding to the bottom of the tubes.
But even then, that gusset would probably be fine. People often expect engineering to have all the solutions, but often, it is more intuition and educated guessing than science.
I think that a common failure that manufacturers are trying to avoid is the top tube cracking at the HAZ along the top. My theory is that the down tube is acting like a fulcrum and the fork is a big lever. If you are smashing your bike through rock gardens or rough terrain I think the top tube ends up doing a skipping rope thing on a very small scale, which causes fatigue over time. On my full suspension bikes I have been tying the top tube to the down tube to relieve the stress on the toe of the weld and move it further down to the top tube to a section that has no weld to cause a stress riser.
Not sure if it has helped or hindered yet but I haven’t had any failures so far. I’m probably just not that extreme haha. Next bike is going to be a trail bike and the plan is to skip the gussets and go up a tube size instead on both DT and TT.
Been listening back to the SUABB podcast lately and Burf from BTR discusses their gusset design in depth in this episode.
Can’t remember the exact wording but their X gusset was designed to allow for torsional movement to help dissipate the stress throughout the HT/TT/DT assembly. They did some early FEA on their designs and it seemed like the biggest issue with gussets is actually making the front end too stiff.
That centerline gussett looks cool but doesn’t really do much and likely to cause more issues. Like Daniel said without doing a lot of testing etc. it’s mostly a guess to what to use.
‘Opinion’. My take is gussets are a band aid and that most of the time they just move the point of failure because they are too stiff. Avoiding cracking is a fatigue issue and not a strength issue. Having a sharp drop off in stiffness or cross section of material means the fatigue will happen where the next availble bending point is for the given loads. A well welded or brazed joint with the correct wall thickness in the tubes is strong enough to cope with the loads. Things like cold welds or undercutting from too much heat or filing the parent tube in fillet barzes is the main cuplprit to joint failures. If you were pushing the limit with really thin walls then some carfeully considered local increase of wall thickness could help to ensure a structure that lasts. Doubler plates work well for this sort of thing because they add material that has a chance to flex with the parent tube add not cause a dramatic drop off in section.
Something that popped up in my socials today reminded me of this conversation. This is a great reminder of how reinforcements with stress risers will ultimately create a weaker part. (Maybe, we have no idea how the frame would have faired without the gusset )
I see pictures of fancy gussets cracked and only recall seeing one surly (ugly gusset ) crack.
I wouldn’t have thought poorly about this gusset until seeing the failure and it seems obvious the shearing stress would cause the failure.
Like danB mentioned it’s got a headtube gusset that doesn’t have corners and doesn’t attach to the neutral axis of the tubes. These are both decisions that reflect the engineering and branding capability and compromise of a smaller outfit:
Manufacturing engineering: The relieved corners are there for ease of manufacturing. REEB is trying to make these things cheap enough to produce that customers will actually pay for them, and labor is the most expensive part of the frame so every minute counts. If you can design the part such that it is self-fixturing and so that you don’t have to chamfer the edges of two gussets for each frame, then you should. This is an example of the well-founded, high quality manufacturing engineering that small framebuilding outfits in the US are known for.
Mechanics of materials engineering: The gussets do not attach to the neutral axes of the tubes, i.e. the sides of the tubes. Calling these the neutral axes assumes the loads the frame is subjected to are mostly in the plane of the frame. An assumption for sure, but one that most would consider reasonable. The location of these gussets goes against the conventional wisdom Daniel illustrated with his FEA figure—they redistribute the load to a certain extent, but not in a very principled way when viewed from a mechanics of materials perspective. This is an example of eyeball engineering that is likely influenced by point 3.
Branding/appealing to customers: The gussets are almost certainly completely unnecessary. REEB is staffed with skilled and talented professional welders, and they have spec’d reasonable tubes to be joined for the SST and its intended use case. To quote Devlin Cycles: “A well welded or brazed joint with the correct wall thickness in the tubes is strong enough to cope with the loads.” It’s really unlikely the frame would fall apart without the gussets, even with Adam smashing it into everything he does. But not having gussets isn’t monster-truck enough for REEB’s brand identity and consumer base, and wouldn’t communicate well with the rest of the bike’s design aesthetic. The self-described tough guys and amateur professionals who buy (and work for) REEB want their bikes to look and be unreasonably strong (read: strong beyond what is reasonable). So the gusset stays. And who am I kidding, I bought one and I think it looks cool too.
I kinda picked on REEB a bit there, but mostly as an example. I think much of the same could be said of BTR, Starling, BTCHN, the list goes on.
I think it’s also the case that people assume that engineering around the mechanics of materials is the only factor that plays a role in these types of decisions. There’s usually a lot going on all at once.
Originally posted this in another thread accidentally. Should have been posted here…
Another idea short of any full on gussets is to double ovalize the down tube, i.e. vertical ovalization (tube higher than wider) at the head tube end and horizontal ovalization (tube wider than higher) at the bottom bracket end. This could have a similar stress distributing effect as a butted tube: like a butted tube the double ovalized tube is less stiff vertically away from the joint because it is flattened (horizontally ovalized) towards the bottom bracket end, especially if the horizontal ovalization already starts relatively early (say 30% from the HT end).
Besides that the vertically ovalized tube at the head tube should also be better at resisting the frontal impact than a round tube. I’m not sure about the latter though: the vertically ovalized downtube near the head tube will definitely create more stiffness in the impact direction but maybe the weld area that is stressed is also smaller
, as that was ended up as an attempt to simulate fatigue failures with tensile failure (as the graph you show here shows), which didn’t really work. Some extrapolations from the hardness tests were used to hypothesize about why fatigue occurs at the toe of the weld (which is a region within the heat affected zone). Not any clear answers there, was hoping to build upon that with more work but that has not happened.
)
) crack.
