Metallurgy at the weld joint

So if such a joint consists of a Metal A | Alloy | Metal B sandwich, how is it different from a soldered joint, other than the Alloy (which is determined at weld time instead of by a solder manufacturer)? Does the heat involved create a different boundary between the metals and the alloy in the joint than if the joint were soldered -- or is the increased strength and/or other beneficial properties derived from the fact that the metals were "soldered" with the alloy as opposed to silver soldered?

wrote: (clip)is the increased strength and/or other beneficial properties derived from the fact that the metals were "soldered" with the alloy as opposed to silver soldered? ^^^^^^^^^^^^^^^^^ A "solder" joint is made by wetting the surfaces of the two pieces with molten solder, and then allowing the solder to cool and solidify. It is always a lower melting material than the pieces being joined, and can be presumed to have lower properties. The weld you describe is made by alloying the materials on both sides of the joint, so can generally be presumed to have properties in the same range as the parent materials. After the joint is cooled, there will not be a metal/solder interface. If the two steels being welded are compatible, and the weld is properly done, you have a single piece of solid steel.

Leo, I'm speaking of welds between two different metals. Assume for the sake of argument that we're making a butt weld between two 1/8 plates of metals A and B. (Assuming that it is in fact possible to weld these metals together.)

At some point, the A plate becomes a B one, because they're all molten together at the weld joint. As I understand Roy, there is a sharply deliniated boundary between metal A and the A/B alloy that is the weld bead, and then another boundary between the alloy and metal B.

If there is such a boundary, then it seems like the structure of the joint is similar to that of a soldered one, even if the physical properties are quite different. Are the improved physical properties therefore the result of the inherent quality of the joint; that the alloy is simply better at bonding to metals A and B because it is an alloy of those metals? Or is it that the alloy is simply stronger than solder?

If not, then how does the joint differ metallurgically from a soldered one? Is the transition a more gradual one, where it is not quite possible to say where metal A ends and the alloy begins, because the crystals are interlaced?

You need to get more specific on which metals you are talking about since there are a host of different possible ways this works.

Soldering or brazing works by using a filler metal that can wet the surface of the base material or the base material can take the filler into solution (between the grains of the base material) Typically this diffusion layer is only a couple thousandths of an inch thick.

The welding process has a much larger diffusion zone where the molton alloy comes in contact with a compatable base material and forms a whole series of complex alloys. Keep in mind that an alloy is really a mixture of grains of various pure and other relatively simple compounds stuctured into grains and latices. This diffusion area is much larger than what you see in a solder/brazing joint.

My original comments were more focused on normal weld situations like welding 304 stainless to A36 mild steel.

If you want to see how to weld absolutely dissimilar metals, do a google > Leo,

I might need to be corrected on this, but I think a soldered or brazed joint is a friction bond, much like a glued joint; the three pieces of metal, A, B, and the solder, remain as three distinct pieces and are held together due to the tight fit of the solidified solder on the surfaces of parts A and B. In a welded joint, both parts A and B are actually melted/mixed together so that the end product is a single piece of metal whose chemical make-up changes as you move from its "A end" to its "B end."

I think brazed and soldered joints are held together by some sort of surface effect, though I forget what and how.

Yes, the question is fun to ponder... An inch to one side you have 100% metal A, an inch to the other side you have 100% metal B. Somewhere in the middle is a band that varies from 99:1 to 1:99 with 50:50 in the middle - so ? According to the specific metals involved there is a range of alloys, A in solution B, to B in solution A - could be good, could be bad, could explain why some metals don't/won't weld right ?

To add to the several replies already:

The exact behaviour depends on the materials and how they alloy, the temperature gradient across the joint, melting points of the metals, ratio of metal A to metal B melted into the pool, and several other factors.

In the general case, there will be a region of alloyed material, that is somewhat uniform locally, from the pool, surrounded by heat affected but unmelted material, often with an intermingling of dendrites at the boundries of the melt.

I won't try to explain metalurgy in any detail here.

Note also that with braz> If you TIG fusion-weld a joint (no filler added) between two

dissimilar

discrete