riveting alu to titanium 1

[christopher67]

I need to replace a rivet that is holding an alu part to a titanium tube of a 35 year old bicycle. The rivet is quite rusty.

It will see only mild weather, and will mostly be inside, with possibly an occasional exposure to rain.

What material rivet will be best for this application?

 

[LittleWheels]

Was the Titan noisy because of no anti-seize on the BB threads or something else? I'm always interested in the engineering of early materials experiments in bicycles. The Victorians did a lot (both well and unsuccessfully) but there were a lot of innovations last century.

 

[EdStainless]

The Titans had sleeves in a couple of locations to add strength and/or stiffness.
These were notorious for making noise as they shifted under load.

 

[LittleWheels]

Down tube towards the head tube and suchlike, AFAIR. I guess the bonding was not always up to scratch (the sleeves were bonded in place?). I've never seen a Titan in the flesh, only read about them since the late '80s.

 

[christopher67]

The development in titanium frame technology is interesting, I have the impression that great progress was made between ~1990 until 1995, mostly due to the two companies in Somerville, Merlin Metalworks that was instrumental in developing the procedures that led to very nice welds (double pass I believe), and Fat City Cycles that worked on butting to ensure stiffer tubes where needed with thinner tubing where possible. First with welded outer sleeve tubes, then outer buttings by machining (very cool to feel the tube variance along the tubes), and then later "ordinary" butting internally I believe. Straight gauge titanium frames could end up flexy if they had smaller diameter tubing and hence a poor ride quality, or overly heavy with larger diameter tubing.

 

[3DDave]

Titanium is about 2/3rd the density of steel and 2/3 the bulk stiffness of steel. Similarly, aluminum is 1/3 the density of steel and 1/3 the bulk stiffness of steel. The way to make up for the loss of bulk stiffness is to use sections of larger diameters/sections to handle bending loads and thicker sections to handle axial loads. In bicycle frames there are reasonable amounts the frame members can be enlarged; thicker sections aren't required as the axial compression of the members is minor compared with bending/twisting.

The bending stiffness goes up with the cube of the section while the weight goes up linearly for same thickness, so the stiffness to weight ratio ends up increasing by the square of the section.

I recall that early aluminum designs had large amounts of flex because makers tried to keep the same tube sizes; then came the fat-tube frames and that seemed to have made the market happy enough. I presume the same thing happened with titanium.

Aside from being much more expensive there isn't a lot of advantage to titanium in a bike frame. Where titanium really shines is that, though typically weaker than high strength steels, it retains that strength at higher temperatures that are above the softening temperature of steel. I also think it is non-magnetic, for what that is worth. It is also corrosion resistant, but many parts of the bike are not, so maybe if the goal is to have a 100 year old frame that is left out of doors the entire time?

Carbon fiber seems to have pushed other materials off the high-price cliff. They can be formed with continuously varying section, have very high stiffness to weight ratio, don't seem to fatigue, certainly not the same way that aluminum can.

Titanium frames are an interesting ox-bow in the development of bicycles.

 

[Stick.]

there isn't a lot of advantage to titanium in a bike frame

You forgot the most important advantage of titanium frames: They're cool as hell

 

[christopher67]

Titanium is about 2/3rd the density of steel and 2/3 the bulk stiffness of steel. Similarly, aluminum is 1/3 the density of steel and 1/3 the bulk stiffness of steel. The way to make up for the loss of bulk stiffness is to use sections of larger diameters/sections to handle bending loads and thicker sections to handle axial loads. In bicycle frames there are reasonable amounts the frame members can be enlarged; thicker sections aren't required as the axial compression of the members is minor compared with bending/twisting.

The bending stiffness goes up with the cube of the section while the weight goes up linearly for same thickness, so the stiffness to weight ratio ends up increasing by the square of the section.

I recall that early aluminum designs had large amounts of flex because makers tried to keep the same tube sizes; then came the fat-tube frames and that seemed to have made the market happy enough. I presume the same thing happened with titanium.

Aside from being much more expensive there isn't a lot of advantage to titanium in a bike frame. Where titanium really shines is that, though typically weaker than high strength steels, it retains that strength at higher temperatures that are above the softening temperature of steel. I also think it is non-magnetic, for what that is worth. It is also corrosion resistant, but many parts of the bike are not, so maybe if the goal is to have a 100 year old frame that is left out of doors the entire time?

Carbon fiber seems to have pushed other materials off the high-price cliff. They can be formed with continuously varying section, have very high stiffness to weight ratio, don't seem to fatigue, certainly not the same way that aluminum can.

Titanium frames are an interesting ox-bow in the development of bicycles.

Hmm, I take it that you are not an avid mountain biker?

 

[3DDave]

I am an engineer who has analyzed deflection of structures. Is this like people who can hear the difference in the quality of sound from oxygen-free copper wires in their stereos?

 

[christopher67]

Nope, we just ride bikes. And calm ourselves down that way. Get big grins on our faces. Feel less hostile, and less need to fuel conflicts. Lots of hippies, for sure.

 

[3DDave]

There is a section here - Engineers with Hobbies - that may be more appropriate for discussing what mountain bikers know.

 

[christopher67]

.

 

[EdStainless]

Titan started making frames in 71, with Teledyne investing in '73.
The first frames were CP Ti and this lead to a number of fatigue failures.
Eventually they went to 3-2.5 for more strength (and some fatigue resistance.
The designers were well aware of the impact of lower modulus, but the question was how much stiffness and strength does a frame need.
This is not a highly quantified set of parameters.
Ti remained a nice product as the welding is so difficult.
I own and ride a very early Al frame, an ALAN.
Small diameter tubes and screwed-and-glued construction.
It is great if you ride smoothly but flexes like mad on climbs.
Then commercial Al frames went to extreme oversized tubes resulting in frames that were so stiff that they were brutal to ride.
By the 90's they had sorted Al frames out and could deliver good ride comfort, adequate strength, and good stiffness.

 

[christopher67]

Thanks for sharing insights!

I know more about developments from the mid 80s forward and the mountain bike segment, so can mostly share insights about that. Yes, welding ti is difficult and the material cost is high. Still, objectively speaking, titanium is a fantastic material for hardtail mountain bikes. I would say it is probably the best material for non electric powered hardtail mountain bikes all things considered, unless you are 1) a "Formula 1" (world cup pro) rider with an unlimited budget, then carbon is certainly better, bc carbon frames can be made quite a bit lighter and also manipulated into more complex shapes and geometries, and because it does not matter that carbon frames generally do not last nearly as long, eg are more prone to wear and also catastrophic failure even with a relatively small crash. Or if you are 2) an everyday rider on a budget and who does not ride much, in which case a mass produced robot welded aluminium frame may be more ideal. An aluminium frame is quite stiff and harsh to ride very hard, I dont care what anyone that does not ride hard says, steel and titanium certainly does have a different feel to it. But my main concern with alu is that it does wear out over time. There were some very nice alu MTB frames made in the early 90's but they were prone to cracking, lasted max a few seasons, later mass produced ones eg TW made robot welded frames by Merida / Giant are better made objectively speaking, are relatively light weight (maybe ~1200 grams for a very light alu frame, not as light as "formula 1" carbon frames which can weigh as little as 850 grams eg Scott Scale), but will also have a very finite life, and the lighter the frame the shorter that use life will be. I prefer something that lasts longer than one or two seasons of hard riding, especially given the high cost of a quality frame ($1k-$3k). My main rider is a mid 2000s titanium frame that is relatively light weight at 1450 grams and has been ridden real hard and is still like new. 20 years old, yet no rust, no dents, no cracks, not anything but a few small chainstay scratches which of course oxidized immediately so don't matter. An alu or carbon frame would have been long gone by now with that type of riding, and a steel frame even with very high quality steel would have been a few hundred grams more and would have had scratches for sure and maybe rust here and there.

 

[EdStainless]

The only road frames that I have ever seen fail by fatigue were Ti.
Ti has terrible fatigue resistance unless you go to alloys like 6-4ELI.
I understand that the Al frames need to be built to a strength, but why don't they slim the tubes down a bit to put a bit of compliance back into them?
I have had a chance to ride some very light steel road frames (R 953).
Full Ultegra build, 8.5kg, not a spec of carbon on it, felt great and handled fantastic.
Too bad it wasn't more aero, and a less expensive.....

 

[christopher67]

Hmm.

Regarding low fatigue resistance, this does sound a bit counter intuitive. Surely, CP is considered not fatigue resistant enough for cycling frames, but 3-2.5 which can be drawn (while 6-4 cannot and hence is used as sheet or machined) is considered very strong, so much so that less material can be used, hence lower overall weights in the particular use case of a cycling frame. But no one uses CP, US makers if they tried at all must have stopped experimenting early, and Russian makers may have made some until the 90s from leftover aerospace materials, but brands using those frames never caught on. Just about all road and mtb ti frames since the mid 80s have been 3-2.5, one exception being the Litespeed Blade I believe, which was 6-4, with welded sheet "blades" instead of tubes in the main frame.

Wrt alu tubes, smaller diameter tubes need to have thicker walls, and that increases weight. Also, my understanding is that alu cannot be bent too much and too many times because of its lack of elasticity compared to ti or many steels, it will crack.

High quality steel like 853 or stainless 953 is amazing. I have a few frames with Dedacciai's 18MCDV6. 1690 grams for a mountainbike hardtail steel frame, the 1997 world championship winning frame I believe.

 

[LittleWheels]

I have cracked a few steel frames, presumably from fatigue, mostly at stress risers (e.g. dropouts, downtube shifter bosses) and tube ends (e.g. just above BB) and a Cannondale 2.8 (typical failure at the cantilevered dropout).

I've not spent enough time on a Ti or carbon frame to fatigue them. My Al Alan was driven over by a car (as was I) but the seat stays had come unglued and been rebonded well before.

Reynolds 953 frames have an intergranular corrosion problem, particularly when silver-soldered and ridden in winter slop, so they can keep coming 'unglued'. A friend of mine found the issue early on (confirmed by his company's metallurgists) and my partner has experienced this issue.

 

[EdStainless]

I tried to talk Reynolds into not allowing brazing of 953.
I wanted TIG only.
I was making the starting tubes for them.
It is the long term heat exposure that causes issues.
The only steel frame failures that I have worked on were related to localized overheating in fabrication.
These very weak locations were then susceptible to failure.
And yes some designs made it worse.
I dismantled and re-glued an ALAN, not mine but a friends.
I used a lot better adhesive than the Italians did.

 

[LittleWheels]

Colour me impressed, EdS. Reynolds kept telling frame builders and riders with broken frames that they had never heard of intergranular corrosion of 953.

Regluing an Alan is a lot harder than doing a Vitus. Didn't Alan have threaded tubes (opposite at each end) and had to be dismantled/ reassembled in a specific sequence if your debonded joint was right near the front? I stayed away from those early Al/carbon frames. Even back then, I knew about galvanic corrosion.

The scuttlebutt was that prison labour was used to assemble Alan frames but old Italian frame building had all sorts of surprises.

 

[christopher67]

Eds, sounds like 953 had some issues. Is 853 better in your view?

 

[christopher67]

Here's an interesting thing

Titanium & Wood - Nevi Titanium

nevi-titanio.com