The Miracle of TIG Welding!

[VictorBravo]

Just had an interesting experience this evening...

As part of my 85 HP upgrade, it was finally time to weld the little steel pieces across the strut fittings.

You may remember that last month I had washers welded onto the damaged wing attach fittings. Using TIG welding and moving slowly, the washers were able to be welded without burning the paint one inch away from the weld point. We used lots of wet rags and reflective heat shielding, and the heat transfer from the weld to the surrounding metal was a small fraction of what it would have been using gas welding.

This evening, we started to weld the strut fittings while they were still installed on the wood spar! We used wet rags and aluminum heat sinks clamped to the fittings, and woven asbestos hot-rod header wrap soaked in water on the spar surface. The welder worked very slowly, meaning that he quickly welded about 1/4 inch at a time, then backed off and I put wet rags on the joint to cool it immediately.

By being really careful, we welded one fitting in place on the wing, and the varnish on the spar was not even discolored, the Stits silver dope overspray on the spar was not bubbled, and there was almost no heat transfer. The spars never got much over "warm" to the touch, We did a close inspection to see if there was any discoloration or bubbling of the varnish. The secret was doign a little at a time and my being real quick pouring cold water on the rags as soon as he stopped welding.

Needless to say, this saved me a lot of work disassembling the compression strut and fittings from the spar, and moreover it removed any chance of the bolts not wanting to go back in. Most people would agree that is is OK to "tack weld" the fittings on the corners and then remove the fitting to finish weld. All we're doing is tack welding, then cooling it off, then tack welding some more.

Of course this method is more risky than removing the fitting, but if your welding technician is good it can save a bunch of time. This is not for brand new welders of course, but in this case I am convinced it saved me a week of time.

Bill Berle

[mhorowit]

Re: The Miracle of TIG Welding!

Just had an interesting experience this evening...
... used lots of wet rags and reflective heat shielding, and the heat transfer and I put wet rags on the joint to cool it immediately.

Bill Berle

Very timely post Bill - As you may have guessed from my posts, I'm about to become a beginning OA welder. In fact yesterday I finished chaining the tanks to the outside of the shed, put the regulators on, set for #2 each and made a neutral flame - then it began to rain, so this morning I'm going to try and make some puddles (joke: it was raining - puddles-welding - puddles... never mind, you have to be a beginner, making molten 'puddles')

Any way, I may learn to make a good cluster weld, but what you've described is 'practical welding'; that only comes from experience around real welders.

Continuing - I though you let the metal cool on it's own, otherwise you ran a risk of fractures. Wrong? - Mike

[VictorBravo]

Re: The Miracle of TIG Welding!

Yes... this definitely falls under the category of "don't try this at home, kids"
until you are an expert.

The weld was only along the very edge of two thin pieces of metal. If it had been a large area weld, or welding something onto a sheet surface, then I am guessing that fractures would be a possibility. In this case there was not a large enough weld area to fracture from cooling. This is my opinion only, and verified only by visual inspection after the weld.

A small fracture in the edge of a weld was far less of a concern to me than overheating a wood spar

This type of thing could NEVER be done with a gas torch. The welding arc was hot for only five seconds at a time.

Mike, what is it that you are trying to weld? I thought your T-craft was going to be on ebay, and presumably you would be buying another that was airworthy.

[mhorowit]

Re: The Miracle of TIG Welding!

Mike, what is it that you are trying to weld? I thought your T-craft was going to be on ebay, and presumably you would be buying another that was airworthy.

VB - I am a man of many interests. I'm learning to weld because someone gave me a rig and it looks like fun. No, I don't ever expect to get near a fuselage, but my box of 4130 scraps did arrive earlier this week

Today was spent making puddles in 1/8". Tonight will be spent discovering that I'm not getting the penetration I need (or maybe I am, won't know until after supper) - Mike

[Robert Lees]

Re: The Miracle of TIG Welding!

Mike: Bill is talking about TiG...Mike, you are talking OA...two different methods. OA needs gradual cooling, due to the HUGE heat-affected-zone close to the weld. For some reason, as yet not fully understood in the aeronautical circles I frequent, but thought to be related to the small HAZ TiG does not even need normalising or other form of post heat-treatment...but it works. Even MiG/MAG does not offer these benefits.

Most modern steel aeronautical fabrications (steel tube fuselages, etc), are TiG welded these days. And what lovely welds they do!

[paulnuss]

Re: The Miracle of TIG Welding!

Indeed TIG welding (In the welding industry, it is now called GTA welding, for Gas-shielded Tungsten-electrode Arc welding) is a useful process, offering weld-puddle control and possibly superior results. The flame is an electric current, conducted through a high-melting-temperature Tungsten alloy electrode, shielded by an inert gas, most commonly pure argon, which covers and protects both the electrode and the weld area while they are in a high-temperature and reactive state. However, the issue of "heat-affected zone" still exists. I am aware that in some instances, operators are not normalizing the vicinity of the welded zone after GTA welding. That doesn't necessarily mean that normalizing shouldn't be done; strong arguments can be made that is is still a necessary step.

"Normalizing" means after welding, raising the area surrounding welded pieces to a dull cherry-red temperature, in 4130 Chrome-moly steel (and other heat-treatable steels) and allowing to cool as slowly as possible. The purpose is to reduce the size and number of crystalline grain structures in the steel, grain structures that are large and far less ductile than the original (parent) metal, or the welded area (parent metal plus any filler metal introduced) after normalizing. This normalizing process reduces internal stresses that arise from the rapid heating and cooling of a heat-treatable alloy such as aircraft 4130 steels.

The Gas-shielded Tungsten-Arc flame at around 10,000 degrees Farenheit is approximately 4,000 degrees Farenheit hotter than the oxy-acetylene flame, though it may be, and usually is, far more concentrated into a much smaller area than an oxy-acetylene flame of approximately 6,000 degress Farenheit. The melting process of the parent metal and filler (if used) is faster with GTA, with less spread to adjacent areas, than the oxy-acetylene process, with less total heat input and faster cooling. However, the heat stresses are not spread over as wide an area, and thus can cause a sharper, narrower transition of the heat-affected zone, and perhaps cause greater loss of ductility in the welded area. Microscopic grain structures of the steel could be adversely affected in bending strength and crack formation.

Heat-treatable steel alloys have complex reactions to temperatures, localized heating, fast surface cooling and other factors they may be exposed to in the welding process. Please don't lightly enter this arena when welding aircraft components that your life, or someone else's life, may be depending upon.

[VictorBravo]

Re: The Miracle of TIG Welding!

Thanks for the (obviously more educated than my own) opinion. We did not take this lightly.

IMHO The potential failure mode in this case is that the weld between the outboard edge of the existing strut fittings and the new reinforcement plate could be more brittle than the rest because of rapid cooling (wet rags to prevent spar damage). If so, it could crack at that point, which would reduce the strength of the assembly back to the level of the original fitting.

In my opinion the reinfiorcement plates are there to prevent one or more of the following:
1. a parallelogram type movement between the trut fitting straps,
2. a twisting movement as if the stuts were twisting torsionally,
3. just keeping the ends of the strut fittings straps a fixed distance apart

Basically it appears that this reinforcement plate is a stabilizer, not a primary load path. I believe that the geometry of these type of movements is such that there is not a large load on the welds. There are three linear inches of edge weld on each side of the plate.

If we were welding a landing gear fitting subject to high impact and high stress in a small area, or if we were welding something that was taking the entire direct loads for the strut attach, then I would certainly have taken a different approach.

But because we cooled it rapidly and the heat was not allowed to get into the majority of the metal, then by definition the rest of the metal would not be affected. Because of that, I am confident that the original tensile strength of the strut attach fittings has not been compromised.

Any engineers who want to weigh in on this and confirm/deny this thought process are more than welcome. If my assumptions about load paths and the purpose of this plate are incorrect, please feel free to let me know.

[VictorBravo]

Re: The Miracle of TIG Welding!

We finished the in-place TIG welding project the other night. A close visual inspection revealed that we managed to join the strut fittings and the reinforcement plate with no damage to the spar.

There was no bubbling of the epoxy varnish at the intersection of the metal fitting bend radius and the spar, which I think is where the heat would have been worst on the spar. It didn't even harm the little bit of Stits silver dope overspray, which should have been the first to melt because it is lacquer based.

I was fortunate enough to be able to see underneath the reinforcement plate with an inspection mirror, and noted no damage on the bottom edge of the spar. Close inspection revealed that the area of bubbled paint on the steel fittings under the plate was limited to less than half an inch from the weld itself.

All in all I believe this method gave me a successful result, although if I had it to do over again I might choose to remove the fittings. The emotional stress and worry was significant, and the time it took to go through the weld process very slowly was perhaps more than the time it would have taken to remove the fittings.

It CAN be done this way, without disassembling the wing, but there was not a huge savings of effort after all the beans were counted.

If someone had previously put in the spar root bushings and the other STC requirements, and the airplane was sitting in a hangar assembled ready to fly, and all that was left to do to hang the 85 engine was to weld on these little plates, then it is possible to remove a small amount of fabric, use wet rags and heat sinks, and weld the reinforcement plates on an airplane in a few hours of time without disassembling it. This would be a very rare situation of course!

After it is all said and done, the short version is that I have to agree with the previous consensus... The STC mods should be done during restoration although it is possible to accomplish it on an airplane that is still together. I believe that the spar mods should be done any time a Taylorcraft is disassembled, so that the rest of the mods can be done more easily at a later time if 85 horsepower is desired.

[Robert Lees]

Re: The Miracle of TIG Welding!

I know little about the pros/cons of TIG versus other methods...but I do know that all modern steel-tube etc aeroplane structures (Legend Cub, Aviat Husky, Pitts etc) structures are TIG welded....and I seriously doubt that they heat-treat after welding.

Certainly here in the UK it is accepted to build a homebuilt steel structure by TIG without normalising, and all our standards are generally derived from FAA-approved standards, methods & practices.

All my structural steel modifications to my Taylorcraft were performed by a CAA approved welder, using TIG. There was no normalising after welding. He professionally builds CAA- and FAA-approved aeroplanes, under the direct guidance of the CAA and FAA (respectively). That's good enough for me.

I am sure that in the US, approvals of certified welders are required, the same as here.

As I say, I know little; my maxim is that if I do not have the skill,and am not willing to learn, I pay for the skilled to come to me.

[VictorBravo]

Re: The Miracle of TIG Welding!

Here are some photos of the process. Note the short distance from the weld where the existing paint was not burned or bubbled. We used lots of wet rags as heat sinks and direct application of water to cool the steel between the weld and the wood.

[paulnuss]

Re: The Miracle of TIG Welding!

Re: Robert Lees reply about "normalizing" weldments:

I am very much aware that many factories, including Aviat and American Champion are GTA (TIG) welding 4130 tubing structures without normalizing, and their process is FAA approved. I am also aware that in previous times Bellanca GMA (MIG) welded some wing-strut attachment fittings (!), and their process was also FAA approved (!!). Those particular wing-strut fittings became the subject of an AD after some failures which led to fatal crashes. Cracks emanated from welded corners of those attachment fittings, and the first "fix" was to replace those fittings with stronger ones that had no weldments. The wings and attachments have since been totally redesigned. The FAA never, to my knowledge, revisited their welding process approvals. I am also aware that the Bellanca/American Champion 8-series wings have continuing strength and design problems, torsional loadings, etc. that may have exacerbated the wing-strut attachment problems, and those problems continue to this day.

For the most part, aircraft tubing structures and weldment designs are excessively strong, and will support the load demands even if full 100% available strength is not acheived in each and every weld. However, much study and involvement in more than 35 years in the welding business, including setting up welding processes for commercial companies in all kinds of critical applications, is the basis for my statements previously that the normalizing process can be very helpful in acheiving full-strength weldments that are also ductile, that don't break when bent.

Perhaps it is icing on the cake only. The old-timers, when original oxy-acetylene welding of steel tubing aircraft structures was used and approved, understood the benefits of normalizing weldments in steels that are "heat-treatable", as is AISI Type 4130. Those benefits would also apply to weldments done with the GTA (TIG) process. Those benefits are still available for minimum cost and effort.

[VictorBravo]

Miracle of TIG Welding IMPORTANT!!!

Anyone welding on the strut fittings needs to read this:

Far be it for me to not admit when I made a mistake. Especially when it might prevent others from same. This mistake might result in an immediate structural failure, and then I wouldn't be around to harrass all of you.

After hearing from more than one qualified person that my method (of welding the fittings in place and then cooling them off before the wood got fried) may well have created some structural problems, I am removing the fittings and having a second look at the situation with some outside help from some overly qualified rocket-science engineers.

I'd like to thank the people who contacted me on and off the forum for their opinions and expertise. Your effort in raising the questions and offering a warning may have prevented an accident.

Because this is primary structure and involves flight safety directly, I have to advise everyone to EITHER completely disregard my previous posts about being able to weld the fitting this way, OR wait until I have proven it out one way or another. Lean towards the "disregard" until further notice.

Again, if you are doing the STC mods on your airplane, either remove the fittings from the spar and gas weld them, or TIG weld and then have the fitting torch-normalized.

Bill Berle

[VictorBravo]

Re: The Miracle of TIG Welding!

The results of an inspection by a VERY qualified facility have shown that the TIG weld we used to put on the reinforcement plates was faulty. There was "porosity" in the welds.

This may have been caused by my over-zealousness in cooling the weld, it might have been poor welding technique because I was pestering the welder to do it much more quickly.

But the bottom line is that it was a serious MISTAKE for me to have tried this, it made bad welds, and it was found to be NOT SAFE. The parts were disassembled and re-welded off the aircraft, and will be normalized, eddy current inspected, and epoxy primed.

Once again, my thanks to the people who took the time and effort to send me a warning that my big idea may not be safe. You all definitely were thinking in the right direction, and I'm glad I was not too stupid, arrogant, or pig-headed to listen.

Your strut fittings NEED to be removed in order to weld the reinforcing plate.

Bill Berle

[Carl Ellis]

Re: The Miracle of TIG Welding!

Re: Robert Lees reply about "normalizing" weldments:

I am very much aware that many factories, including Aviat and American Champion are GTA (TIG) welding 4130 tubing structures without normalizing, and their process is FAA approved. .

I seem to recall something about Maule airframes being MIG welded. Is this true?

- Carl -

[VictorBravo]

Re: The Miracle of TIG Welding!

Got two of the strut fittings back in last night after they had been inspected, found defective, disassembled, and re-welded. It's not for the faint of heart to do this removal/replacement an assembled wing that is still covered! What a pain in the empennage.

The distance between the compression strut ends, spars, and installing the strut fittings left virtually zero room to move anything around. The metal was sliding on the metal and not happily either. It would have been next to impossible to install the additional 1/32" plywood doublers (drawing 2569, thanks Don).

An interesting tidbit is that one of the little tube spacers was missing, where that steel tube is held between the fittings with a long threaded brace wire. I had to make another one, then flatten it halfway so I could get it installed past the welded doublers inside the front strut fitting. It was a real bear to get that all re-assembled, finding the blind hole to push the brace wire through, etc. etc.

HOWEVER, once that brace tube and wire were tightened down, it added a significant amount of rigidity to the fittings. I couldn't twist them at all with even hard hand pressure.

Anyone disassembling the fittings (for the STC or other reasons) should remember two things: First, try to put the short tube spacers in the strut fittings before you put the fittings back on the spar.... this will save you from having to flatten the tube. Second, IF YOU CAN, make sure you have all the spacers and the brace rod loosely assembled, THEN slide the fittings over the spars and bolt them back in. This would avoid a very frustrating game of trying to fish that brace rod through all the holes from one side.

As part of this ridiculous exercise, I had a structural engineer measure the fittings and the distances between them. As an exercise to keep his slide rule moving freely, he will calculate the stresses on each part. Although not officially FAA approved, the numbers I get back will allow all of us to know how much extra "beef" is in the structure at certain locations. I'll make this information available in a future forum post for anyone intrerested.