AnkleWork

Visualization much? Advance thinking + preparation + careful execution = good result.

Add a foam plug just above the top bottle mount to prevent fouling lower down. After careful measurement, abrading, cleaning, prepping, marking, dry fitting, jigging, generous masking, etc, then apply glue thickly to the entire bond areas -- inside and out -- and slowly insert the post to minimize voids. Use glue with plenty of working time and reasonable viscosity. Apply a large excess of glue that will be pushed out of the joint and have plenty of rags and solvent to clean-up the overflow (and in case a bail-out for retry is needed). Some in this forum can improve this concept. Should not be difficult to get good results.

(Also, before deciding on the desired final position consider whether you'll ever want to swap in a thicker or thinner saddle.)

sweeks

If it's filed or ground smooth, then there's less to keep the seat from rotating. I haven't seen any crack propagation on my frame, but the mating surfaces at the fracture have sort of hammered each other down so there aren't any sharp spots. In any case, there's not as much of a "lever arm" left on the seat tube, and the amount of metal there is massive.

Yeah, this is a bit of an annoyance. The folder has a sleeve shim inside the seat tube which to an extent minimizes the looseness of the seatpost. There is some rotation when I'm pedalling, but it's tolerable.

WizardOfBoz

AnkleWork, your first point illustrates my point that its going to be "a real bugger".

The point you don't address is the issue that Mad Honk points out from his decades worth of experience bonding steel shafts to golf club heads: to be effective, epoxy should be used in situations where you have a glue gap of about 0.005". At this thickness, most force in the glue is shear and the glue is transferring stress to the metal parts. When thicker, epoxy (with 2.8 ksi strength) becomes the structural element rather than transferring stresses to the bonded elements of aluminum and steel (20-44ksi and more than 100ksi strength respectively). Note too that the intact design has already broken aluminum - the stresses put on the seatpost are not trivial with respect to the seat tube.

If you mean a post sticking out of the frame, sure. If you mean something that will be durable enough to last (especially since the intact original design broke), I have my doubts. If the OP can successfully get shim tubes (say 2-3 inches wide) glued into the seat tube so that the tube-shim and shim-seatpost gaps are about 0.005, and he's done the prep and cleaning work you've (properly imho) suggested above, maybe this will work.

Your note prompts another idea. If, instead of epoxy, one filled the seat tube with an epoxy/carbon fiber mix, the strength of the carbon fibers might allow the "fill and stuff the seatpost in" approach (which is, admittedly, much easier than trying to fit and glue shim tubes). The question is: does such stuff exist? West System does offer an additive that it suggests adds strength and is used to add fillets to glue joints and to fill gaps. Be interesting to know what West suggests in this application.

If the collar had be swaged down (and possibly reamed) to get 27.2mm, and the man tube is 28mm, you have a 0.8mm gap. That's a 0.010" gap around the tube, which is exactly the range that Mad Honk suggests is weak. I trust a guy that's probably experienced the head of a driver coming off at 110mph.

AnkleWork

Don't know how any of that is applicable to a "bugger" but maybe it's time to think beyond theory into practice. Full 3-D wet-out in smaller spaces after dry assembly is achieved in industry every day; re: vacuum/pressure impregnation. And other methods are also well developed. Having only one example to perform, the OP can utilize simpler, slower methods.

Maybe read a little more and post a little less. The radial clearance question is addressed several times up thread. HTH

sweeks

Looking again at the OP's image, it looks like the crack may have originated at the stress-breaker hole at the bottom of the slot in the top of the seat tube. The seat tube flexes at the top as the seatpost clamp is tightened and loosened. Aluminum is not very good at withstanding flex, and even with the larger radius at the bottom of the slot, cracks may originate there. That's where it began on the bike I pictured above. What I *should* have done is enlarged the hole before the crack propagated too far. I've had this happen on another bike and increased the size of the hole in the seat tube with no further progression of the crack.
Here's an image of the handlepost of the bike I pictured above. It has a clamp to adjust the height. A few years ago I noticed cracks propagating out from the stress relief hole. Rather than make a hole large enough to include the cracks, I drilled "crack-stop" holes at the ends of the cracks. There has been no further progression of the cracks. It's not pretty, but it works. I don't understand why the manufacturers don't make the holes bigger to start with... it's surely not "rocket science"!

Stress-breaker holes to stop crack propagation

erileykc

Rad has already told me that they do not have spare frames to send as replacements, warranty or no and that the best they will offer is a $250 discount on a new bike from them. I declined that offer.

unterhausen

the seat tube was too big for the seat post, and so it was just rocking back and forth on the clamp. Broke at the front of the seat tube/top tube juncture as demonstrated by how dirty the fracture surface is there. It's pretty common for aluminum bikes with a lot of seatpost sticking out to have a failure at that point. I have seen fisher mountain bikes with that failure there. But those have a frame warranty.

I would have thought this frame would be made with one of the aluminum series that doesn't require post-weld heat treatment. I thought 6061 wasn't used much in bikes.

WizardOfBoz

Interesting comment on failure mode.

Regarding 6061, there's a good site describing aluminum weldability by type. The high strength alloys are 2000,5000,6000, and 7000 series. 5000 is weldable but is available in mostly sheet or plate for structural applications, and isn't heat-treatable. 2000 is high strength aerospace which are mostly unweldable. This leaves 6000 and 7000 series. Both of these when welded are affected by the heat of the weld and have a weak heat affected zone. 6000 series are generally much more hydroformable and I think that's why 6061 is often used.
https://metalpress.onlinemetals.com/...ding-aluminum/

Two specific alloys often used if welding is part of the plan are 6061 and 7005. Here's a comparison, specifically regarding bikes, from Ibex bikes. It is a bit dated (2012 or so).
https://tuulparg.eu/KB/alumiinium/AL6061vs7005frames.pdf

Pinkbikes uses 6061. Fascinating article about how they fabricate and heat treat. They also claim that alumunum bikes are repairable to full strength - but only after scrupulously removing paint, grease, and dirt, and then re-heat-treating the frame. Check it out: https://www.pinkbike.com/news/To-the...um-Frames.html. In that article they point out that they use 7075 alloy (among the highest strength Al alloys but completely unweldable and averse to hydroforming) for a lot of parts that aren't welded.

WizardOfBoz

Nicely done. I especially like the fact that you've chamfered the holes and left them with a good finish to minimize stress risers.

If may be of interest to some that in auto racing, when weight reducing holes or holes for rivets are drilled in small parts, chamfering is indicated. One technique to reduce stress risers in that hole is to put a ball bearing larger than the hole on a solid surface, then put the hole in the part on top of the bearing, and put another bearing on top of that. Then you whack the top bearing with a hammer! Hones are sometimes used to smooth the chamfer, too.

WizardOfBoz

Bummer, but thanks for sharing.
Good luck with your project. I hope my ranting was more helpful than demoralizing. To answer your original question, if the seatpost you are using "rattles around" then I think your idea of shims or shim tubes gives you the strongest solution. Let us know what you do and how it turns out.

Jeff Neese

I see neither. I see a seat tube that wasn't in far enough and would cause just about any aluminum frame to break like that. That's not the first time that exact failure has occurred.

Jeff Neese

Have you asked RAD if they will sell you just the frame, and if so are they willing to give you a discount because of what happened? That's where I would start.

Maybe I missed it above, but what exactly did RAD Customer Support tell you when you called them? Did they have a theory about why it broke?

Koyote

I know it's a long thread, but if you read through it all, you'll find that the OP explicitly ruled that out.

joel1952

In other words, it was the wrong size seat post (too small diameter)
Apparently Welding/heat treating thoughts were an illusion, based on incomplete information

Furthermore,
Shims are problematic.and Sure epoxy is used for lugs and golf clubs but they are different. The holes are blind. The forces are not sheer, they are lateral.
Maybe just weld the whole thing together. Get the proper size seat tube. Have that welded in place, including drilling a couple holes in the seat tube and weld the tube and post together that way

guy153

This is all good info and quite correct. Last year it was reported in the news that someone has invented a welding rod for 7075 which may be a game-changer

https://www.nature.com/articles/s41467-018-07989-y

And there's also Pole Bicycles in Finland who CNC mill the entire frame in two halves out of a solid piece of 7075. Then they glue them together lengthwise and also throw in a few bolts for good measure. Amazingly they only charge a few thousand Euros for one of these. I think they only make full suss MTBs so the front and rear triangles are separately milled components.

Duragrouch

OP: The problem may be not only excessive fatigue loads for the frame material and near a weld zone, it may also lay in the design of the seatpost clamping slot. EDIT: I NOW SEE THAT THE CLAMPING SLOT IS ON THE REAR, REVERSED FROM ON MY BIKE. SO MUCH OF THE BELOW MAY NOT APPLY, UNLESS, THE CRACK BEGAN AT THE SLOT AFT (DUE TO CLAMPING STRESS, NOT BENDING STRESS FROM THE SEATPOST LOADS) AND WORKED AROUND FORWARD TO THE WELD. Either way, file a letter with the Consumer Product Safety Commission. My letter below with regard to a very similar failure on a different brand of bike, after the bike maker blew me off:

I own a <my model bike> folding bicycle. There are structural cracks in the frame.

I, the sole rider of the bike, am well below the bicycle design maximums of 231 pounds weight and 6'3” height (I am 170 lbs. and 5'8” height.) The bike has been ridden solely on smooth roads, and has low mileage (I am estimating 2000-3000 miles based on tire wear.)

The frame has cracks at the top of the seat tube, already half-way around the tube on both sides, below the seatpost clamp. If the cracks continue, it will result in separation of the top portion of the tube, and seatpost, from the frame. The cracks also lead to the welds joining the top of the seat stays to the seat tube, possibly causing failure at that joint as well. Either of the above failures risk fall hazard and/or loss of bicycle control and risk of severe injury or death.

As reference, please see recall number 16-270; This recall of a different bike brand involved “The top clamp of the bicycle's seat post can crack, posing a fall hazard to the user.” While due to a different part, this failure mode is similar to the failure mode above, and thus proves precedence with regard to it being a safety issue and the necessity of a recall if the failure proven. Please see link: https://www.cpsc.gov/Recalls/2016/Ad...tional-Recalls

This frame design is used on multiple <my model bike> bicycles. A quick search on the internet shows this failure mode to be common on this frame design.

I contend that the frame cracks are the result of a design defect. (I am a retired mechanical engineer and manager, and have authored patents of, among other things, clamped structural connections loaded in fatigue.) At the top of the seat tube is a slot parallel with the tube axis, to allow the seat post clamp to close the seat tube tightly around the seat post. All bikes have a similar slot, but <my model bike> positioned the slot at the front of the tube, rather than the back of the tube as on all other bikes I have seen. The cracks initiate from the radius at the bottom of the slot, and propagate on each side circumferentially, progressing rapidly, as the crack increases the local stresses in a vicious cycle. When the rider sits on the bike, most of the force on the seat post is down, but due to the seat post angle, some is aft, which causes a bending moment (torque) into the seat tube, and this moment is especially high on <my model bike> due to the extremely long seat post, due to the folding design. That bending moment stresses the seat tube in tension in front, and compression in the back. This crack is a fatigue failure, and those only happen in tension, not compression. Thus, <my model bike> putting the clamping slot on the front of the seat tube introduced a weak point for a crack to initiate. Every other bike in the world puts this slot at the back of the seat tube, even other <my model bike> like the <other model>, which would put the slot in compression. Placing the slot on the front of the tube is a design defect. However, having the slot on the back, while definitely improving durability, would not guarantee no failure, if the stress in front still exceeded the fatigue life limits of the material. The seat tube is highly stressed; Notably, later models of this same bike are equipped with an aluminum bushing at this interface between the seat post and the seat tube, I believe to strengthen this area. The failed frame in question is equipped with the earlier plastic bushing, which is lower strength and less rigid, which may also allow excessive deflection of the stressed area under seatpost clamp loads.

I communicated the frame failure, with detailed description as noted above, and photographs, to the local <my model bike> dealer, who forwarded them to the manufacturer, <my model bike>. The manufacturer has refused to replace or repair the frame, saying the failure is due to excessive seat post extension (my seat post is 1.5” below maximum extension, it would be impossible for me to ride with the seatpost higher because I am 5'8”), and the bike is out of warranty, despite this being a safety issue.


What I am requesting from CPSC:
Ask <my model bike manufacturer>:
What was the reason for the change from plastic to aluminum for the seat post bushing, and,
Has <my model bike manufacturer> completed cyclic fatigue testing, at maximum design rider weight (231 lbs.) and maximum design seatpost extension (equal to 6'3” rider height, maximum mark on the seat post), to confirm if this is a valid solution to the seat tube failures on this frame design.

If the aluminum seat post bushing is proven to prevent frame failures, I would require the manufacturer, <my model bike>, to:
Replace any frames with cracks at the upper seatpost, and,
Replace the plastic seatpost bushing with the newer aluminum seat post bushing, including on frames with no cracks, as a preventive measure.

If, however, the aluminum seatpost bushing is not proven to prevent the frame failure as noted, I would request that CPSC require <my model bike manufacturer> to:
- Replace completely, all frames of this design, with frames proven to not fail in this manner.

Best regards,

(me)
Mechanical Engineering Manager (retired)


Attachments: photographs of failure

WizardOfBoz

Right back at ya - very much enjoyed learning about both welding 7075 and Pole bikes.

I looked Pole up. Interesting.
https://singletrackworld.com/2017/11...s-the-machine/

sweeks

They're chamfered on the inside as well!

alo

You may or may not choose to take it up with the CONSUMER PRODUCT SAFETY COMMISSION.

If you choose to repair it, whatever method you use, it is important that the seat post goes a long way into the frame. If it only goes a short way into the frame, you may get another break. You need a long seat post (or tube).

I think it is a bad idea to make the seat height fixed.

If you are planning to epoxy it in the frame in a fixed position. You would be better off, with your long seat post, drilling a small hole through the frame and seat post, and putting a thin bolt through it. You could, for example, use the inside of a quick release bolt.

There are many different ways you can repair it. Use a good one.

I mentioned one earlier with a U bolt under the bar. That is a good method.

yukiinu

Trash the frame keep the components. Go to Goodwill buy old but good Trek bike for $40. I am USAF trained to find cracks on aircraft, if your frame has a bad weld like in the photo its bound to have more dangerous discontinuities in the welds and or frame.

Mad Honk 

To the OP, Looks like you have found a solution to your problem. I trust all is well, Smiles, MH

fly135

I would agree that if you have a death wish, then substituting a real weld with JB Weld is the way to go!

tammons

Tig welding would be correct for aluminum tubing. But I would have to wonder why it broke there in the first place.

To me, and just guessing, I would say the aluminum tubing is substandard.

Duragrouch

I don't think so. Aluminum in general is not great in terms of fatigue resistance, and per unit size it flexes more because the modulus of elasticity is 1/3 compared to steel for instance. Both contribute what you see, that is a fatigue failure. Aluminum designs subjected to fatigue have to be a bit overdesigned in terms of stiffness (lack of flexing, "strain") to hold up, and can still be lighter than steel or aluminum, the problem is that the stiffness results in poor ride. However the stiffness is good for things like race bikes where there is a lot of sprinting by folks with thighs the diameter of telephone poles, or at least good sized hams. Discontinuity in geometry due to an abrupt weld bead hurts even more, that is why Cannondale (at least my 1989), et al, sands their welds into a smooth fillet. Heat affected zone hurts also, that's why quality frame makers use the correct weld rod and post-heat-treat the frames to restore full material strength to an alloy such as 6061-T6. Titanium, while lighter than steel, has excellent fatigue resistance, so that, combined with the lower density than steel allowing large diameter tubes where needed, provides frames that are stiff where needed but also flex where needed for good ride, that is why it makes an excellent material for bike frames. Carbon fiber allows even more customization for flex and stiffness, but is difficult to impossible to repair, which is why I have shied away from it for a daily or touring bike.

joel1952

As explained earlier, the seat post was too small for the seat tube, so it was able to wobble and cause fatigue crack.