Manual GTAW Welding Aged Inconel 718 with 625 Filler

By Lawrence (*****)

Date 01-16-2008 15:18 Edited 01-16-2008 15:24

Chris,

Your plan sounds reasonable if you are looking at a short term repair.

I would not recommend what you suggest for fabrication or refurbishment.

Long term weld repairs on Green (non-annealed) inco 718 is undesirable for a long list of reasons it appears you are well aware of.

In the Aircraft engine world; if an on wing repair on unannealed inco 718 is required on a pressure vessel (compressor, TMF etc.) there are usually strict engineering constraints and detailed welding procedures that are followed by an inspection schedule to regularly reinspect the weld until the engine comes off the wing and a proper repair can be performed.

If you insist on doing this, the low heat input precaution is wise (write it into the procedure with a real value) and it will be just as important to make sure these tubular componants (I'm assuming they have seen service) are perfectly clean inside and outside in the weld and HAZ.

For Procedure purposes... You could review an established WPS for Cast Hipped Inco 718 or Cast +Hipp Waspelloy for possible heat input guidelines and the introduction of an Argon/Hydrogen 97.5/2.5 GTAW shield gas for the same purpose.

By croy Date 01-17-2008 14:35

Lawrence,

Thanks for the assistance. This project has me quite nervous, since we shouldn't be welding on aged 718 in the first place! I noticed in another post of yours that you strongly recommend abrasive grinding on weld prep surfaces just prior to welding since the oxide layer formed on 718 is so tenacious and could affect weld integrity. Is the reason that you suggest hydrogen in the shielding gas to scavenge oxygen and/or increase penetration and arc engergy? We use an 80/20 Ar/He shielding mix for most of our TIG welding as it is a mix of automatic and manual TIG and we wanted to standardize on a mix. We were thinking of using 100% Ar shielding and backing for this weld to help minimize heat input during manual TIG, but could add hydrogen if necessary. However, even with as low as 2.5% hydrogen, I can already hear our clients (Oil/Gas, mostly in U.S.) complaining about potential porosity, embrittlement, etc., even though it can't happen with the materials we are using at 2.5% H2. We MIGHT be able to get it accepted if for only (1) specialized WPS and for very good reasons (i.e., reducing oxidation).

These are for new parts, so proper surface prep is not an issue. Do you agree with using 625 filler, since PWHT is not possible, or would you suggest an alternate filler without niobium (referring to the post regarding this) like C22 or Inco 686? Then again, the 718 has even more niobium than 625, so maybe it's a moot point. We need to have at least 60 ksi yield in the weld. We were thinking of doing the root and fill passes with .045 wire and pulsed TIG to minimize heat input, is this prudent or overkill?

By Lawrence (*****)

Date 01-17-2008 17:51

Croy,

The Hy addition for GTAW is pretty common in Europe for super alloys like Inco 718. It enhanses wetting at the edges, has better cleaning action in the arc plasma, and increases heat input per amp. The Hy additions are more rare in the U.S. but as mentioned earlier, some of the majors (Pratt, GE, Allison, etc) have used this mixture with good results in procedure quals and production. There shoud be plenty of data available on Inco alloys and various GTAW shield gas mixtures. Don't know if it would be an improvement over your Helium mix.

Filler metal selection for an application like yours is above my paygrade.... Sorry.
See ASM Metals Handbook Vol 6
Here is an interesting link
http://www.stainless-steel-world.net/pdf/11012.pdf
I'm sure inco alloys has data for you as well.

Pulsed GTAW is an excellent idea for reducing heat input on semi automatic and rotated projects, Sciaky Accu-pulse is a good example. Pulsed GTAW loses effectiveness when manually operated simply because an operator can hesitate or travel more slowly than the plan and more heat just keeps pouring in.

Who ever came up with the fabrication sequence that has been put in your lap should be kicked---- hard. :)

By SWP

Date 01-17-2008 00:03

I have not worked with 718, but had a good deal of problems with liquation cracking in autogenous welds on 625 components due to the Niobium in 625. Similar component mad of Inconel 600 without Nb had no problems. I think I recall that 718 has similar cracking problems due to Nb and Laves phase formation.

By Ringo

Date 01-18-2008 16:46 Edited 01-18-2008 18:06

I work in a FAA repair station(Gas Turbine Engines),and we routinely weld 2nd and 3rd stage Low Pressure Turbine Vane Rings (Inco 718).It it basically build-up work on the vane tangs(W/L-605 AMS5796filler metal ) that get fretted from torque from the engine.Our success is welder dependent.We have one guy here who welds it too hot and too slow in my opinion,but he has the best success.We do not PWHT for fear of warping,and therfore scrapping the vane ring,Although our welding engineers say a resolutionize PWHT would work.The other welders are better than the hot guy,in that they use very little heat input,but they get cracks alot in the HAZ.In conclusion,you could try more amps with slower travel speed.

By croy Date 01-19-2008 19:39

Ringo,

That's interesting that you use Haynes 25 filler, which is mostly cobalt and tungsten, whereas 718 has no tungsten and 1% max cobalt. Do you use eddy current, fluorescent LP, etc. for HAZ crack detection? Do you know what levels of heat input on the high and low end (J/in) are being used? Are you talking normal TIG here (with pulsing?) or micro-welding of some sort? Any more of your valuable input, would be greatly appreciated! Thanks.

Chris

By Lawrence (*****)

Date 01-19-2008 21:46

The High Tungsten Haynes 25 type alloys are employed strictly for wear resistance properties on blades and surfaces that are subject to vibration, friction or even air flow.

Haynes is generally not a choice for structural connections.

By Ringo

Date 01-21-2008 15:29 Edited 01-21-2008 16:07

Your right,and the wear resistance is good (as witnessed when they come back thru on subsequent shop visits),There is a honeycomb seal that rivets onto the Vane Ring,and it does have a great deal of torsional force,but there is alot of play axially.

By Ringo

Date 01-21-2008 15:20

We use Manual tig without pulse,and use FPI for crack detection.As far as heat input we have a 130 amp max in our WPS.We are trapped on the filler metal,as we have to use whatever the engine manual has listed.The weld metal looks good when its polished and etched in the met lab.