These inertia friction welding videos are sooooo satisfying

Originally published at:


Seems like there could be an easier way, but that shows you how much I know about welding.


Satisfying, except for the part where there is a ring around where the two pieces are stuck together. That’s the opposite of satisfying.

A few of advantages that I see right off.

  1. Perfectly even heat distribution (circumferentially) that would result in even stresses on cooling that means when the item is unclamped after cooling it will stay in perfect alignment. The radial heat distribution and resulting metallurgy would be interesting to know more about. For rotating items, the strength near the edge is most important for load-bearing, so the temperature and weld achieved at the centre may be less important. It may actually be important to not heat the centre too much (it will naturally heat more slowly due to lower radial speeds) as a hot centre with a cooled circumference could lead to some sort of fracturing.

  2. A deep weld right to the centre that is impossible to achieve on large pieces by conventional means.

  3. The weld is made of the same material as the two starting pieces, without introducing a third material.


I mean, the machine is huge and undoubtedly wildly expensive. However, with a “traditional” weld, you’d have the outer ring surfaces joined, but the strength would be low as there’d be very little welded area. You’d normally counter that by cutting the two faces at angles, and laying beads in those cavities until the area was filled. But there’s a risk of voids, there’s a lot of skilled labor involved, and alignment would be a pretty big issue.
With this rig, you get super high strength, perfect alignment, and no risk of voids. Pretty cool.
That ring that grows out could easily be machined off after, btw.


Do stir friction welding next! :slight_smile:


Dumb question time - what is the advantage in using the giant, complicated machine to weld the two smaller pieces together, when the finished item looks as if it could easily have been machined from a single billet?


Billet tends to be weak (tempering/geometry issues). Aerospace-grade alloys are difficult to machine. Kind of a laundry list of PITA. It’s why I tend to be skeptical of the promise of additive mfg, especially with metals.

The clever, expensive machines aren’t just for selling ridiculous maintence contracts, though one could be forgiven for thinking that after hearing some of the stories.


See also: Vibration-friction welding!


I am not sure which video in particular you are looking at, and in any case I couldn’t say specifically without seeing more detailed engineering drawings, but I can think of at least two reasons why this might be preferred. The first is that machining from a solid stock is expensive and can have a lot of waste. If you can cast two pieces, grind them a bit, and then weld them together that can be a lot more efficient than machining from solid billet. Second, there might be some internal machining you can’t see in these videos that can’t be performed after assembly, or would require very fancy tools.

Honestly this machine doesn’t look that complicated in terms of mass production equipment. This is the sort of process that might be cost prohibitive for custom manufacturing, where 5 axis CNC is a totally reasonable process, but not a big deal for higher volumes where you can absorb some tooling costs to use a faster process.


I see. Thank you…

1 Like

…and thank you (I was specifically referring to the first video).

1 Like

Or, you could just go with gorilla glue.


This is what I was wondering.

Thanks, @andy_hilmer, @winkybber, @nothingfuture

1 Like

Gotta say, most of the variants of solid state welding a pretty neat. Explosion welding, for instance.
And I was just reading about cold welding the other day. Pretty neat.


Take the giant valve as an example. The valve’s “disc-like” bit needs to be of a material that can withstand high temperatures without distortion or failure, as well as being durable when exposed to thermo-chemical attack from hot reactive gases surging past at high speed. The stem needs to be tough (strong and crack/fatigue-resistant) and also resistant to mechanical wear as is slides in a lubricated sleeve, but probably isn’t exposed to the same temperatures. Two different duty cycles, two different materials. Yes, you can compromise with a material that does each job OK (but not great), but I’m guessing the application of that giant valve is in a circumstance where both routine maintenance time and unplanned failure are VERY expensive indeed.


You can even weld copper to steel, or aluminium to steel.


This topic was automatically closed after 5 days. New replies are no longer allowed.