Turbine blade electron-beam repair and preheating

Used to illustrate the upper extreme of preheat (1000°C, ~2000°F) for nickel-base turbine blade alloys. The blade is glowing red inside the vacuum chamber while the EB weld is made; economics ($50 scrap vs. $7000 repaired) justify the unusual practice.

Some materials are very susceptible to residual stresses and cracking, such as some nickel-based alloys. If I had a turbine blade — I've been to shops where they take turbine blades and they preheat them to a thousand degrees centigrade, which is almost 2000 degrees Fahrenheit, in order to be able to weld them, because just the residual stresses from cooling and shrinkage will crack that turbine blade alloy. It doesn't have very good ductility to begin with, and it has excellent high-temperature strength — that's why you're using it in the turbine. So it won't deform in the turbine, and with that high strength at high temperatures it just cracks. They actually put an induction heater on it; the thing is glowing red when you do the electron beam weld to repair it, and they do it in a vacuum. It's kind of pricey, but that little blade might be worth seven thousand dollars repaired, and it's worth — actually it's worth about fifty dollars scrap metal — but it's worth seven thousand if you can repair it. So it's worth doing some pretty unusual preheats in some alloys, but not steels.

Hot-section turbine blade tip repair by electron beam, with preheat to 1000°C / 1800°F to prevent nickel-superalloy cracking under thermal stress.

Another example: fancier turbine blades — the ones in the hot section of the engine. I've seen two places where they're doing electron beam, adding the tip or melting in there, and they're preheating to a thousand degrees centigrade, 1800 degrees Fahrenheit, because of this problem of sucking out the heat and the fact that the nickel-based superalloys are very prone to cracking. If you have big thermal stresses, you'll get cracks. So they preheat to within a couple hundred degrees of the melting point. We do it, it's just tricky and expensive, and you don't do this on twenty-seven-pound parts — you do it on very high-value-added parts.