Piet van der Horst

Piet van der Horst

In 1970 Piet made welding his trade en since then he never stopped learning about that trade. By now he is well past his retirement age, but not welding is still not an option for him. It is not just work, it is a passion.

More about Piet

More articles by Piet

Gases for welding processes

Inert and active gases

Classification according to the standard In NEN-EN-ISO 14175 standard, gases are divided into main and subgroups. In this article we discuss four main groups of

Read More

Solutions can be simple

When nobody can help you

At a national fair for welding technology I am approached by someone who says that he has a TIG welding problem. He thinks that this is caused by the TIG torch. Hence his visit to the torch supplier booth.

When asked why he thinks the torch is the problem, the following very informative answer came: “We TIG weld a 55 mm round steel tube with a robot, with a wall thickness of 2 mm, on a steel plate of 100 mm round and 5 mm thickness. Our problem is that we always have holes in the weld. We also have a very high consumption of tungsten electrodes. The tube is made of blank material and the plate is blasted. We have several times contacted the supplier of the robot and they ensured us that the robot is not the problem. Then we invited the supplier of the TIG welding equipment, but even after testing with various hypermodern equipment, this did not come to a solution. The gas supplier also tried several times to solve the problem and was not successful. In the end, we commissioned another robot supplier to do tests to solve our problem. Unfortunately without desired result. We have now come to the conclusion that the problem must be in the torch. That is the only thing we have not looked at. I might be cutting corners, but still.” We agree that I will visit the company as soon as possible after the fair.

Investigating the problem

During the visit it soon becomes clear that this is a cosmetic weld. An attempt is made to make the smallest possible angle seal without filler material. No requirements are imposed on the weld, other than that it must be as small and as smooth as possible. The problem, however, is that there are holes in the weld at various places. This is a known problem. The TIG welding of steel without filler material is in principle not possible. Steel contains a relatively large amount of carbon and other contaminants. Because the TIG process is carried out with an inert gas, a gas that is not reactive, the contaminants will cause the holes in the weld. Another problem that occurs is that the puddle occasionally “explodes”. The reason for this is that the contaminants accumulate in the puddle and then suddenly come out. That causes the holes.

You therefore need the desoxants manganese and silicon, that are in the filler material, to make a good weld. It also causes another problem and that is the contamination of the tungsten electrode. During such an explosion, metal particles adhere to the tip of the tungsten electrode, which changes its composition. As a result, the melting temperature of the tungsten electrode goes down and it will evaporate. As a result, stable arc behaviour is no longer possible.

During the check of the welding parameters, like welding current, welding speed, gas quantity, gas cup and torch position, no deviation is found. There was a problem with the tungsten electrode. They were welding with a 2.4 mm purple, with a top angle between 25 and 30 degrees. And although I know that many welders will not agree with me, the more pointed the tungsten electrode tip, the larger the weld pool becomes. A larger weld pool gives more contaminants and therefore more problems. And although the problem was not solved, there were far fewer holes with a top angle of 60 degrees.

Expertise in action

So in cases like this it is important to make the arc as small as possible, I mean that the arc burn spot should be as small as possible. There are various possibilities to reduce the arc burn spot. As said, the more blunt the top angle of the tungsten electrode, the smaller the weld pool. Pulsating welding with a frequency of 2000 Hz or higher also constricts the arc. Finally, you could opt for a different gas.

The change in the top angle of the tungsten electrode did not completely solve the problem. Testing with equipment with a high pulse frequency has been done and, according to the company, has not led to a solution. I believe that the high frequency in combination with the correct top angle could have been a solution. However, there was no possibility to invest in this expensive TIG equipment, so that was not an option.

Then it remains to look at gases. A few mixed gases are available for TIG welding. You can think of Argon with Helium, but that will not work for this application because Helium actually increases the puddle. Then there is the possibility of Argon with H2 (hydrogen). Hydrogen is a relatively cold element in the welding arc and will severely constrict it. The result will be that the puddle will become thinner (more fluid) and will therefore better degas. But there’s more. Hydrogen is also reducing. That means that it cleans up contaminants.

Now I hear you think: “Hydrogen and steel? That is not allowed at all?! “And you are absolutely right. That is, when you have high demands on the weld! In this case, the only requirement is that the weld is smooth and requires no post-processing. The tests with Argon with 2% hydrogen and a 60 degree top angle on the tungsten electrode provide the desired result. No more holes, no contamination of the tungsten electrode and a nice smooth weld appearance. Perfect!

The professional in me does still wonder what would happen if we do this test again, but then also with a high frequency pulse TIG machine. It could be that we come close to laser welding.

Recent articles and infographics

Was your visit of value?

Did you learn something from it or did it answer a question you had? Would you consider making a donation?
We will use your donation for maintaining the website and adding new content.