Heat and tempering

[Willie1]

Out of curiosity, I have been wondering about the numerous discussions here regarding the downside to putting metal through additional heat cycles for repairs, or tube replacement. I have been building some custom knives and other tools recently and have been learniong a lot about heat treatments, tempering, and annealing. If a tube was overcooked, could it not be annealed back to a more plastic state to remove the brittleness? To get to the brittle state, the tube would have to get to over approx 1500f. It would have to get to the bright orange, heating to the point of losing its magnetism, then cooling quickly, such as quenching in oil or water. Brazing is several hundred degrees less with brass/bronze, and even lower with silver. The metal cools slowly when brazing in contrast to quenching as well. If a brittle tube was annealed for 2h at 200-400f, it would become much more plastic, wouldn't it? Or am I missing something?

[cfrisia]

I think that would be hard to do on a certain spot. You could do that to a frame as a whole.

also, it depends greatly on the type of alloy. Some have to go through liquid state to form a desired fine-grain structure.

Last but not least, tempering and healing is a time and energy intense process, it might simply be too expensive.

[Alistair]

Out of curiosity, I have been wondering about the numerous discussions here regarding the downside to putting metal through additional heat cycles for repairs, or tube replacement. I have been building some custom knives and other tools recently and have been learniong a lot about heat treatments, tempering, and annealing. If a tube was overcooked, could it not be annealed back to a more plastic state to remove the brittleness? To get to the brittle state, the tube would have to get to over approx 1500f. It would have to get to the bright orange, heating to the point of losing its magnetism, then cooling quickly, such as quenching in oil or water. Brazing is several hundred degrees less with brass/bronze, and even lower with silver. The metal cools slowly when brazing in contrast to quenching as well. If a brittle tube was annealed for 2h at 200-400f, it would become much more plastic, wouldn't it? Or am I missing something?

Fyi, bronze brazing takes place in the mid to upper 1600 deg. F range. Above the 1500F temp. limit you mentioned. 1500 deg. F is well above the temperature that transforms the steel to the Austenitic phase. This isn't a problem as long as the cooling rate is slow enough to allow the steel to transform back to the Pearlite phase, which is what the tube started out as.

If the cooling rate is too fast there is the possibility that instead of transforming back to Pearlite, the steel will transform into Martensite. From a bike frame perspective that's not good (little toughness and elongation, brittle).
If that was to happen you could anneal the joint, as you suggest, as long as you had access to the proper facillities of course. I don't think you could do it by hand with a torch.

I think that because of the amount of heat that it takes to bronze braze a joint it would be hard to accidentally cool it quickly enough to cause problems. Maybe if you had a breeze blowing into the shop and it was really cold outside?

From what I've read, it's actually more likely that you'd get this problem with a Tig welded joint, since the overall heat input to the tubes is lower and therefore the cooling rate faster.

Alistair.

[Willie1]

Fyi, bronze brazing takes place in the mid to upper 1600 deg. F range. Above the 1500F temp. limit you mentioned. 1500 deg. F is well above the temperature that transforms the steel to the Austenitic phase. This isn't a problem as long as the cooling rate is slow enough to allow the steel to transform back to the Pearlite phase, which is what the tube started out as.

If the cooling rate is too fast there is the possibility that instead of transforming back to Pearlite, the steel will transform into Martensite. From a bike frame perspective that's not good (little toughness and elongation, brittle).
If that was to happen you could anneal the joint, as you suggest, as long as you had access to the proper facillities of course. I don't think you could do it by hand with a torch.

I think that because of the amount of heat that it takes to bronze braze a joint it would be hard to accidentally cool it quickly enough to cause problems. Maybe if you had a breeze blowing into the shop and it was really cold outside?

From what I've read, it's actually more likely that you'd get this problem with a Tig welded joint, since the overall heat input to the tubes is lower and therefore the cooling rate faster.

Alistair.

It was my understanding that tig would be more likely to cause problems with brittle joints due to the high localized heat. From what I read it would be difficult to accidentally harden the steel without quenching it. The thin tubes might be cooling with localized spots due to the high surface area to metal ratio, but I wouldn't think it would be much more extreme than a knife blade. So theoretically it would be possible to anneal the joint with the right equipment.

[Willie1]

In case anyone is curious, I found the data on the 4130. The heat treatment point is 1600f, and requires an oil quench. I don't see how this happens with brazing a joint. The braze filler would slow the cooling of the joint making hardening unlikely. Heating the joint again isn't cumulative in terms of metallurgy. Specific temperature points need to be met with specific cooling rates to change the material properties. Does anyone have an explanation that causes the brittleness that is reported from multiple heat cycles?

http://www.suppliersonline.com/propertypages/4130.asp

[Willie1]

Fyi, bronze brazing takes place in the mid to upper 1600 deg. F range.

Thank you for the clarification. I must have misread somewhere that it happens at 1100-1400f. I may have been looking at flux ranges or something like that and misunderstood what I was reading.

[steve garro]

You will see beginner bikes get spider cracks in the main tubes within the HAZ within a few years, if over cooked.
Maybe 50% of my 1st 10 bikes did this.
I did total replacements on all of them.
The originals cost me $500 to make, and I sold them for $500, and then had to fully replace them..........
You see the bad math here.
I did not know what I did not know...........
Rody may be able to give some actual real world beta on heat treating, I think Grove used to do it, the head scratcher for me is how to avoid the bowing you see in heat treated tubing.
- Garro.

[Alistair]

Does anyone have an explanation that causes the brittleness that is reported from multiple heat cycles?

My understanding is that the extra heat cycles aren't causing brittleness exactly, but are effecting a change in the metals microstructure, causing grain growth. The larger size grains act as stress risers, more effectively so than the finer sized grains that the metal in good quality tubing starts out with.

These larger stress risers, in the presence of the cyclic stresses that riding a bicycle generates within a frame, are what ends up doing it in. I don't think it's an issue of embrittlement, in the sense that you are thinking of it (ie. generation of martensite).

Alistair.

[steve garro]

My understanding is that the extra heat cycles aren't causing brittleness exactly, but are effecting a change in the metals microstructure, causing grain growth. The larger size grains act as stress risers, more effectively so than the finer sized grains that the metal in good quality tubing starts out with.


Alistair.

That's how I see it, as in the Diamond>>>>Graphite
One has excellent crystaline structure, very tight, the other, not so much.
- Garro.

[Willie1]

I have just started reading on the grain growth science. That makes a lot more sense to me. IIRC, the annealing process will help this but the tempering process won't. (Don't quote me or take this as correct. I will keep this in mind as I continue reading.) The annealing, if that is what would be needed, would require getting the metal past the 1600f point, and having the timed reduction to 900f, then air cooled. That would be much more difficult to do without affecting the braze itself than just tempering. Tempering would be quite simple with a large enough oven.

[Mike Mcdermid]

Start with annealed then heat treat
fully treated 4130 doesn't like hard spots cracks go from there

[Willie1]

I did a lot more reading, and the way to fix the grain growth problem from overheating would be to normalize the steel. This is a multi stage heat and controlled cooling of the steel. It would be way beyond the scope of a hobby shop. Here's an interesting post on the process with O-1 steel (I just bought some O-1 for my next knife.)

Grain Size and Normalizing WITH UPDATED GRAIN SHOTS

[Mike Mcdermid]

I did a lot more reading, and the way to fix the grain growth problem from overheating would be to normalize the steel. This is a multi stage heat and controlled cooling of the steel. It would be way beyond the scope of a hobby shop. Here's an interesting post on the process with O-1 steel (I just bought some O-1 for my next knife.)

Grain Size and Normalizing WITH UPDATED GRAIN SHOTS

Is 01 not a tool steel, over here in the UK it was supplied as ground flat stock or gauge plate for toolmakers and folks making punch tools
iirc it remains dimensionally and carburization free even if it was oil quenched ,i think theres lots of info on it if you google sheffield (as in where good steel used to come from) and gauge plate or ground flat stock

[Willie1]

Yes, O-1 is a tool steel. I was looking up the heat processes for it yesterday as I just bought some. The same processes apply to all alloy steel to varying degrees. It is really pronounced with O-1. Cromo that we use for bicycles experiences the grain growth problem that is shown above, but to a lesser degree. If we had a big enough oven, with precise temperature control, a bicycle frame could be normalized, but I would suspect it wouldn't maintain its alignment. The simple answer to the problem is minimize the heat the tubes experience, and let them cool as slowly as possible.

[Mark Kelly]

That's how I see it, as in the Diamond>>>>Graphite
One has excellent crystaline structure, very tight, the other, not so much.
- Garro.

I follow what you mean but that isn't a good example; diamond does not have a small grain crystal structure - a diamond is by definition a single, sometimes very large, crystal.

The carbon in diamond has a different electron structure from that in graphite, SP-3 vs SP-2. There is no analogy to this in metallurgy because the elements involved are transition elements so they aren't capable of the same type of electron hybridisation, they have too many electrons in the shells. In turn that's what makes them metallic.

A better example might be wrought iron vs cast.

[steve garro]

I follow what you mean but that isn't a good example; diamond does not have a small grain crystal structure - a diamond is by definition a single, sometimes very large, crystal.

The carbon in diamond has a different electron structure from that in graphite, SP-3 vs SP-2. There is no analogy to this in metallurgy because the elements involved are transition elements so they aren't capable of the same type of electron hybridisation, they have too many electrons in the shells. In turn that's what makes them metallic.

A better example might be wrought iron vs cast.

Not the crystalline structure per se but the molecular.
- Garro.

[JimFrain]

My father-in-law and I were having this discussion last week. (He was a frame builder in the 70's then shop owner for about 25 years). Anyway, his point was that, often when the brazed joint is overheated it causes the tube to be annealed and softened. Since the rest of the tube has already been heat treated it is much stiffer than the annealed portion of the joint/lug. Through the process of riding, the stresses to the tube being bent back and forth much like a lever causes the annealed portion to tear or brake. Its why it often happens to the seat tube right above the bottom bracket in lugged bikes (its a spot that's often overheated). He said he had seen a few bikes come into his shop that way.

The general discussion was whether or not lugs act as stress risers.

[Mark Kelly]

Since the rest of the tube has already been heat treated it is much stiffer than the annealed portion of the joint/lug.

What you meant to say was "Since the rest of the tube has already been heat treated it has a higher yield strength than the annealed portion of the joint/lug". They have exactly the same stiffness.

[JimFrain]

"What you meant to say was "Since the rest of the tube has already been heat treated it has a higher yield strength than the annealed portion of the joint/lug". They have exactly the same stiffness."

no I mean the annealed portion before the lug, if the tube has been overheated at the joint. so the portion of the tube that has not been overheated is stiffer than the portion that has been overheated.

[Todd Mosley]

Wouldn't annealing the steal make it softer and less brittle, and the point most likely to deform?