Falanx

Yes, I know it's lunacy. Yes, I know it's not the done think to purists who have made some beautiful brazed, welded and silver soldered frames with and without lugs, but I'm feeling kinda investigamative. So the floor is open to bonders..

At some point in the next year or so, I feel the need to try building a steel frame, lugged, and adhesively bonded, but the awkward way. Stainless tubes or alloy steel isn't a relevant distinction - my only requirement being strength as I'm tall, strong and abusive of poor little MTB frames. I can already pick the steels that suit me, but I'm unfamiliar with the quality of lugs available.

Investment cast in an alloy steel would be ideal - I'm not a supporter of turbulent mould-filling for stainless steels, but that can be reconsidered, depending on supporting evidence.

Soooooo!


1.) Who's worked with what lugs? Stainless or normal steel, give me you recommendations. Priority is smoothness of geometry and section. Ideally, if you can confirm 4130 (or so) or the type of casting stainless used, would be astounding :-)

2.) For the carbon, bamboo, glassfibre, airfix and bathtub repairmen: Have you done any metal to metal adhesive work, and what have you discovered?


Answers on here, or a postcard if you're feeling particularly esoteric, please.

Live Wire 

I'd be interested the hear any contradicting viewpoints, but as far as I know, there aren't any 4130 lugs.
As far as quality of the finished lug, Long Shen lugs are usually pretty nice, especially the ones you get from Sacks or Pacenti....Ceeway has some nice MTb lugs.
Still ( I'm sure you know this, but just in case) wouldn't you have to make your own inner/outer BG- type lugs in order to get enough surface area for proper adhesion ?

Falanx

With the strength of single-part thermally activated epoxies available, not really :-) You'd already have the tube-to-tube jointing surface area at the mitre, and the area under the lug. The lugs used in the Dynatech for example were huge because they were aluminium, which is notoriously ****ty to bond to. A nicely phosphated steel is waaaay better for adhesion, and plastic restraint means the thinner the joining material, the stronger it is.

Sachs suggests that their lugs are either alloy steel or stainless. Seeing as it's the one area of frame materials technology I'm unfamiliar with, I was testing the veracity of this...

Andrew R Stewart 

The reference to Bruce Gordon is a good start. Andy.

unterhausen

I have no idea what alloy the investment cast steel lugs are. The stainless lugs are probably made out of one of the alloys that look good when polished, since that is the reason for using them

MassiveD

Lugs are probably made out of a material noteworthy for it's casting. There isn't any real reason to use something super strong since the joints in a lugged bike are way overplayed anyway. The same tubes can be secured with tig wire, which is basically a process of casting in place small fillets.

The thing is you will end up making the joints twice. Once when you make the lugs, and once when you glue the bike together. But that is not altogether bad news. You can control your situation better when making the lug clusters since you won't have to worry all that much about heat issues or alignment. You can't use standard lugs to epoxy a bike together.

I have made lots of golf clubs, and they glue together, and are often really difficult to separate. The heads are stainless, and the shafts are chromed. The ones that I post cured were a ***** to heat apart. But golf clubs have huge penetration, it would probably scale up to about 4.5 inches on a bike size tubing. They can be made with less, but not that much less. Epoxy adhesion to metal can be summed up in the word "crap". It always fails. However in the case where you have a socket type fitting it can be bomber. When I made my demountable trimaran, I needed to make composite ferules for the beams, which were 4.5" aluminium tubing. Even with liberal parting agent, and a release membrane, it took a comealong to shift them, and in the end I had to cut them free. Most of the loaded parts are bonded in place with studs, and these are really strong. But just surface adhering stuff... The epoxy lifts right off in a few years. Obviously in aerospace, or paint, they are doing something. I am talking about home and boatyard applications done by serious people.

Falanx

There's not that many standard alloys for casting stainless. And they're not the same alloys as tubing specs. Non stainless can be alloy, can be plain carbon manganese. I guess the only way to find out is to purloin a few and pop them under my XRF :-)

You'd be surprised what you can cast. And no, it doesn't have to be hugely strong, but the nice thing about a cast geometry, and that mechanical engineers tend to forget is that their grainflow should be perfect for their geometry. If I want them stronger, I can always heat-treat transformable ones. The same tubes can be secured with silicon bronze, too, which is a process of casting as well....

I wasn't planning on making the lugs. I wanted to see what the present understanding of the field was.

For reference, Raleigh haven't ever had a goddamned clue what they're doing with materials. Most of the Dynatech series were either spectacularly overengineered in places they didn't need to be be, precisely for that reason and the rest picked tubes purely because they could get their hands on them at the time. I only titled this thread such because it would give contributors a frame of reference. The bonding area for a head-top tubes lug junction can be of the order of a square inch. I've seen a Ford Countour Estate hung off a crane jib with an epoxy bond smaller than that, and I've used the product that did it. You'd be surprised how strong adhesives can be, especially when you remember to load them in shear.

Either a crosslinked aromatic ring is strong, or it isn't, you can't have it both ways. Bonding to aluminium is a fool's game, but that's because of the morphology and behaviour of aluminium oxide, which is why the aerospace community doesn't bond to aluminium oxide, and why they paint on either a blasted surface with a massive key or a conversion coated one. The resins used in paints and the resins used in toughened adhesives differ only in the number of functional groups. Any disparity in adherence performance is entirely due to poor preparation and not understanding the substrate's needs. But I'm not bonding to aluminium. I'm bonding to a steel, and they have sexy grip ;-).

I know I can get my dirty grubby mitts on aerospec adhesives that will do the job. I wanted to know if anyone's found equal performance from glues not exclusive to the plane crowd.

unterhausen

just to review, up until IC lugs became commonplace, we used to use thin stamped and welded lugs made out of mild steel. And the production lugged bikes were often held together by those lugs because they effectively covered up horrible workmanship. So it's fairly unlikely that they have gone out of their way to use anything particularly exotic or even all that uniform between batches.

Mark Kelly 

Since I'm going to be bonding parts of my frame together, I've looked at this recently and one recommendation which was made to me by someone in the field was to use Redux 810: https://www.imatec.it/pdf/Redux810_us.pdf

I found this useful: DSTO paper on bonding metal adherends

Canaboo

I'm confused. Your first post asks and your last seems to indicate that you have all the answers...
Are you the same guy that said glue choices make no difference with carbon lay-up.

Mark Kelly 

This image from Dazza ( Darryl McCulloch of Llewellyn) and the text in another of his posts on Framebuilder
Suggest that he uses plain old 304.




(original post replaced - I thought canaboo was referring to me not Falanx)

Falanx

Er, I don't think so.

And I didn't say I had all of them, I was just hoping one or two people could point me towards any knowledge they had of lug materials, and glues other than heinously expensive aerosepc ones...

Falanx

I thought he was referring to both of us and getting us confused...

Falanx

And, unless he's either casting them himself from remelt scrap, or welding them and fettling, he's not using 304. That's a forged composition. The nearest equivalent has different alloying limits and carbon content.

Canaboo

Turns out I'm talking about both of you... I was originally referring to falanx but I checked the other thread and it was Mark who said the matrix glue stiffness makes little difference.
I think the sensible thing to do with a glued steel fame would be a wrapped fiberglass lug, shaped well and with a few filleting coats to sculpt the exterior shape.
All epoxy seems to be fairly expensive so I'm not sure where the line between readily available types and higher grade "aerospace" types actually is.
The paste epoxies available for homebuilders through Aircraft Spruce seem reasonably priced.

tuz

Just wondering... is the circa 0.1 mm gap found between normal lugs and tubes okay for epoxies?

Falanx

With all dissimilar adhesives - filler metal for brazing included - the thinnest layer with complete wetting is the strongest. The statistical likelihood of a critical defect decreases linearly with volume and at the thinnest quantities there's near as makes no difference no volume for defects to occur in as the filler tends to two interfaces and nothing between them.

Canaboo

Epoxy needs some sort of gap to have optimum strength. Dry joints are no good.
There are plenty of Epoxy/metal joints in bikes, especially in little seen areas. Alloy fork/crown bonds are Epoxy. You can see the slight glue squeeze out where the crown race seat transitions to the steerer tube.
Also hollowtech cranks are two sections bonded together. The glueline is pretty invisible there.

tuz

And what about ductility/elongation of epoxies? Any type that match steel? Or that's not a problem...

Mark Kelly 

Falanx

How large a target does your XRF need? I just cut the points off some Long Shen stainless lugs, I could post them to you if that's any use.

AFAIK Dazza gets his lugs cast by them, probably most of the other private labels too.

Mark Kelly 

As I see it, if the bondline is thin enough and it is subjected to shear, the strain in the bond material is limited to that of the materials either side.

squirtdad

I seem to recall consistent input in other threads that a lug/tube combination designed for brazing would have issues with adhesive assemble due to 1) limited surface area and b) size of gap requiring filler.

On thought might be not to build an entire frame, but get a a lug and a couuple of tubes (or tube scraps) put it together and then see what to takes to make it fail.

A lot of refreence to aviation assembly happens in these threads. my impression is that most aviation applications involve sheets of materials and large surface areas for glue.

ksisler

1) So the ideal gap-fit for a glued up lug would be to allow each molecule of epoxy to be able to touch both the tube and the lug?
2) I know that the gap in brazed lug frame is quite a bit more that this... I have cut apart the joints in some broken bikes and seen 1/8" or more thickness of brass, especially at the tube intersections and it seems like it worked well. I had been working under the assumption for a couple of decades that this was "by design" such as to allow the equivent brass to that of a fillet brazed frame to be provided at the tube intersection?
3) Am I missing something such as perhaps a detail that the metal in the brass might actually alloy with the surface of the steel tubes gaining a lot of adhesive strength where a glue couldn't and this is why a glue joint has to have a smaller gap?