Reenforcing Gussets or Plates - at the HT under the DT and top of the TT - ?

[Mr_Rafi]

I'm interested in why some builders use gussets or reenforcing plates underneath the DT, and sometimes on the top of the TT as well. I understand that it essentially makes wall thicker in the area that is high stress against the HT, but I'm curious to know about the arguments for or against. Seen by and large on MTB and CX machines, I see builders such as Hunter and Soulcraft use them, while it seems the majority of builders do not.

Thank you in advance.

[sonny]

illustration:


Super Custom 29er by huntercycles, on Flickr


3 002 by frankthewelder, on Flickr

[jon_norstog]

I use a "duckbill" gusset under the DT on my adult BMX bikes, and bikes for dual slalom, 4-cross and trails/hucking/urban assault bikes like the "New Model."

The main thing, IMHO, is to set it up so the gusset transfers part of the load from the HT/DT joint to a point further down the DT, but still in the butted portion of the DT. I have had bad experiences with TT gussets and don't use them any more.

jn

[fortyfour]

My own opinion: Gussets either below the DT or above the TT basically "tell" the tube where to fail (at the end of the weld). Personally I do not use them as I feel as though the load from the HT/DT or HT/TT joint is transferred down the length of the tube rather than a point in space which is the gusset (that's just my intuition speaking and I have no hard data to support it mind you). Post #28 of this thread speaks to some math by someone who knows what they are talking about and needs no introduction. There are reasons on both sides of the aisle, but I don't see them as necessary. I increase DT / TT diameter if I am concerned. All my mountain bikes sport a 1.375" O.D. TT and 1.500" O.D. DT for the most part. That in combination with a Paragon 44mm head tube, tapered suspension steerer and 15mm through axle...

[steve garro]

Put a supertherm TT & DT on there & call it good.
- Garro.

[Mark Kelly]

Post #28 of this thread speaks to some math by someone who knows what they are talking about and needs no introduction.

Unfortunately the maths are wrong: strength and stiffness do not scale by the same factors with increase in diameter as that poster said they do.

[Jonathan]

Unfortunately the maths are wrong: strength and stiffness do not scale by the same factors with increase in diameter as that poster said they do.

You can't leave us hanging with that. What is the relationship?

[zachateseverything]

I'm interested in why some builders use gussets or reenforcing plates underneath the DT, and sometimes on the top of the TT as well. I understand that it essentially makes wall thicker in the area that is high stress against the HT, but I'm curious to know about the arguments for or against. Seen by and large on MTB and CX machines, I see builders such as Hunter and Soulcraft use them, while it seems the majority of builders do not.

Thank you in advance.

I don't like the use of gussets on bike frames. putting a gusset on a welded tube structure will strengthen the joint, but it also creates a stress concentration in the middle of the tube. I've seen a couple mass production frames fail this way.

[Mark Kelly]

Sorry, I didn't want to threadjack but maybe I can make some comment that is germane to the issue at hand.

With tubes of circular section and even wall thickness loaded in bending or torsion, stiffness is proportional to cross sectional area x square of radius (or diameter) while yield strength is proportional to cross sectional area x radius. This is generalisable to other shapes, just with more complex maths.

Since the strain produced by a given load will determine the failure mode this can cause some unintended consequences: Take a tube which has a reinforcing patch on one side and is loaded in bending. When such a patch is added to an existing structure it makes it stiffer and stronger. But let's say an ambitious maker decides to use this as an opportunity to reduce weight and use a lighter guage tube but rely on the patch to maintain the stiffness of the joint.

The axis around which the moment is applied is where the total deformation* on the stretched side matches that on the compressed side. The presence of the reinforcement moves this away from the centre line and since the stiffness increases with square but strength increases linearly, the ratio of stiffness to strength of the non - reinforced side is actually reduced by the presence of the patch.




*to be understood as the integral of areal strains x square of distance from axis

[Mark Kelly]

*to be understood as the integral of areal strains x square of distance from axis

Correction to above (Damn the 10 minute limit): Since the strains are proportional to their distance from the axis, that should be the integral of the areal strains x distance from the axis

[EricKeller]

I don't see anything in your answer that gives a reason for contradicting Don's statement. Maybe I'm misreading?

[Mark Kelly]

Strength and rigidity in tubes goes up with the cube (x³) of the diameter but only the square (X²) of the wall thickness.

Strength goes up as the cube of diameter if the ratio of wall thickness to diameter remains constant, in this case rigidity goes up as the fourth power. This rule is often cited but it's pretty impractical because the weight would also increase as the square of diameter.

Rigidity goes up as the cube of diameter if wall thickness remains constant, in this case strength increases with the square of diameter. This is a little more practical as a builder will often ponder the effect of using a larger diameter tube of the same guage for a heavier rider, fully knowing that the weight will increase linearly with diameter.

Rigidity goes up as the square of diameter if the total cross sectional area stays constant, in this case strength increases linearly with diameter. This last case is important as constant cross sectional area also makes for constant unit weight so this is the comparison made if a builder were targetting a particular weight and pondering the effects of two tubes which could achieve this.

The reasons for the different scales are easy to see: when the tube is bent or twisted, the parts further from the axis are stretched proportionally more. As long as the material stays in its linear range, the extra stretch requires proportionally extra stress. The bending moment is the area times the stress times the distance from the axis and is thus proportional to the area times the square of the distance from the axis.

The material will start to yield at the point where the strain is greatest. Since the degree of stretch is proportional to distance from the axis, this means the strength is proportional to the area times the maximal distance from the axis (half the diameter).

To the OP: sorry for the total threadjack but hopefully you've found something useful.

To everyone: ignore #10, I was right the first time as the above makes obvious.

[EricKeller]

didn't read all of your reply, but I missed the fact he said strength instead of stiffness going up as the cube. Sorry about that.

[jon_norstog]

getting back to gussets, I learned this trick from Mike Devitt at the old SE Racing operation: You don't weld/braze the gusset its full length. You Fix it as solidly as you can to the HT and tack it to the DT at its end, leaving the sides open. The idea is not to greatly stiffen the HT area, just to transfer a littl bit of the shock loading down-tube when you and the bike take a hit.

The downside is you do get some tendency to rust in there.

jn

"Thursday"

[Jamie Medeiros]

Years ago back maybe in 1999, the company I worked for sent 8 front triangles out to an independent tester to see if the tig welded gusset on the down tube was worth using. They were made with either OX gold or Reynolds 853 tubesets I don't remember. The four with gussets lasted a minimum of 40% longer in the fatigue tests.

[steve garro]

My Salsa ala Carte cracked a HT gussett
My Fat City cracked behing the "double chin" gussett
My Bontrager bent the main tubes past the gussetts
My Rock Lobster with "lobster claw" gussetts never broke.
- Garro.

[fortyfour]

My biggest concern with gussets is you're concentrating all the force on a fixed point in space somewhere along the tube. The gusset has an "abrupt" end point vs the butt of the tube has a smooth transition internally or externally along the length of the tube (ideally). In addition when using a gusset in construction, you're adding a weld down the side of the tube which increases the amount of area you have to weld, which increases the HAZ, and I was always taught / urged to minimize heat input from my peers.

What I think gets lost in translation though is that if something is going to break, not all the gussets in the world will stop it. Suspension or no suspension, if someone is riding that hard and beating something that much repeatedly, it's not a question of "if" but only "when" it will fail.

I have no doubt Steve Garro beat the living SHIT out of those frames he mentions above.

[Mike Mcdermid]

i think you should braze gussets on or glue them on or ....oh hell

edit they are brazed on nice

[Paul Sadoff]

i think you should braze gussets on or glue them on or ....oh hell

edit they are brazed on nice

Gussets can make a frame last longer or hasten it's demise. One has to know butt lengths of the tubes and make sure that the gusset does not extend beyond the butt length. If you don't know a proper way to design and/or attach a gusset you are probably better off not putting one on the frame.I put them on my frames but after all these years I think the lack of bent and/or broken frames have spoken to the success of the practice. There's a lot more to consider when adding gussets than most people realize. The practice came about because of frame failures in both brazed and welded MTB frames in the late '80's. Tubing has improved since then but I still think that a good gusset still has a place on a frame that will be ridden abusively. This isn't engineering data, just caveman/blacksmith stuff garnered from years of seeing stuff break.

[dbohemian]

They were made with either OX gold or Reynolds 853 tubesets I don't remember. The four with gussets lasted a minimum of 40% longer in the fatigue tests.

I do not doubt this but my thought is a gusset is a short term fix for the appropriate solution which is designing the tube correctly for the job. If a company had enough capacity to warrant it they could have a custom tube drawn with any number of enhancements.

For example, longer butts on one end, thicker, longer tapers and even thicker on the bottom or top. Is this not a more elegant solution than a gusset? Commonly gussets are used on tubular aircraft structures. Then again those are usually very low volume and would not justify a customized solution. When someone makes hundreds or thousands of a particular bicycle frame gussets are inelegant and only sell those customers who think it looks burly IMHO.