Preheat - alternative methods

Date 03-23-2013 11:49

Hi gastonM.

I'm not an expert on this, but... I would probably just make a larger single pass fillet weld.

The whole annex consistently mentions "single pass" fillet welds, even in the figures. I would like to see what the experts say on this and why it says, "For groove welds, the hydrogen control method shall be used to determine preheat."

By gastonM (**)

Date 03-25-2013 15:06

Newinsp:

I can't do a larger single pass fillet weld because i use the minimum travel speed and maximun wire speed as permitted by WPS. My assumption is: if I let the pieces cools, to ambient temperature, in the next pass they will be in the same situation, and the cooling rate will be the same. If i do not let the pieces cools, any residual heat of previous pass will act as a preheat. But as you said, the annex I talks clearly of single pass then im not sure about my assumption. Respect of groove welds, i think that the annex, drives us, to hydrogen control because there is no sufficient laboratory test that provide consistent conclusions about the relationship of hardness, hydrogen and stress. Are you agree?

Best Regards.

Gastón

By 803056

Date 03-25-2013 20:34

Most of what we know is based on experimental results. From that, empirical equations are developed to provide "best fit" that agree with the experimental results. From that, the codes publish prescriptive criteria that fits most applications. Many codes allow the engineer to used more advanced approaches if they can show theory agree with experimental results.

Al

By gastonM (**)

Date 03-26-2013 13:07

Al, have you ever preheat a beam with 2" of web and 4" flange up to 110ºC. Can you say me which method to employ, to add heat with very high power without provide distorsion,and without waste a lot of money. Or is better, no preheat, then make hardness test?
Note that prequalified section lets determine preheat to ensure hardness below a determined value.

Regards.

By newinsp (**)

Date 03-26-2013 12:51

I see what you are saying. It sounds reasonable to let the first pass cool to the ambient temperature, then repeat.
After the first pass cooled, the question would be: "Does the first pass become part of the base metal and change the base metal thickness?" Second question would be: "Is the new thickness within the thickness range qualified?"

I agree about the groove welds.

By 803056

Date 03-26-2013 13:42

Most distortion is the result of large thermal gradients from one area to another. That is the principle employed when cambering a beam or heat straightening a member that has been damaged.

Preheat not only reduces the probability of delayed cold cracking, it can also reduce distortion by reducing the delta T from one area to another.

Depositing a single weld bead, allowing it and surrounding area to cool, and depositing the next weld bead maximizes delta T, thereby enhancing the probability of distortion.

Try this simple experiment:

Weld 3 simple T-joints consisting of two 1/2 inch plates (12-14 mm) measuring 4 x 8 inches (100 mm by 200 mm). Place the nonbutting member so it is centered on the butting member. Tack weld the plates at the very ends. The objective is to tack weld the two plates where they will offer the least "rigidity" to the assembly.

Weld one assembled T-joint at room temperature. Deposit a three pass fillet weld allowing the plate to cool to ambient temperature between beads. Measure the total angular distortion.
Preheat the second assembly to 300 degrees F (150 degrees C). Deposit a three pass fillet weld once the plates are tacked and allow for no cooling between beads. Measure the total angular distortion.
Preheat the third assembly to 300 degrees F and deposit a large single pass fillet weld that is the same size as the weld produced by the 3 bead stringer technique used for the first two assemblies. Measure the total angular distortion.

While you are at it, you might as well compare the distortion along the length of the three fillet welds.

Report your findings here on the Forum.

When you have completed the experiments and reported the results we can continue the conversation.

Best regards- Al

By gastonM (**)

Date 03-26-2013 14:54

I know what you mean, but i was no pointing on that aspect.

I'd like to be clear.

How many energy i need to add to 10 ton of steel to raise its temperature from 20ºC to 110ºC (as recommended by table 3.2)?. Really, i thought that need low energies, but when i was in front of piece with torch in hand, I noticed, that something was wrong. i made a simple calculus energies balance and heat flow out it was very high.
If i think in a oxy-fuel heater, when im heating an extreme, the other extreme (12m of distance) still its at 20ºC. If i think in a batery of heaters, i still have distorsion problems, because the other flange its cold. Then i think in a furnace or heating mantles, to develope homogeneus temperature and improve heat lost. Then my employer´s pocket says "comeback to earth". Then i read annex I and think: if laboratories test demonstrate that preheat its not necesary for single fillet welds, when hardness is less 400 HV. then think whats about multiple pass fillets welds.

For this reason, i wondered if you knew some method of preheating , inexpensive and effective for T-joints for the SAW process.

Regards.

By 803056

Date 03-26-2013 15:27 Edited 03-26-2013 15:31

It depends on the shape and geometry of the parts involved.

I've seen bridge girders preheated by attaching a long pole, about 1 1/2 meters, to the SAW tractor and the torches were attached to the pole. The fabricator was preheating the flanges measuring about 4 inches (100 mm) to about 250 degrees F (125 degrees C). If the delta T is less than 210 degrees F (105 degrees C) you should not see permanent deformation.

I've also heated a large 63 ton casting to 350 degrees F before and during welding which lasted 17 days. Very necessary, but very expensive ($84K US dollars).

I didn't convert degrees F to degrees C, you can do that. Call me lazy.

Best regards - Al

By gastonM (**)

Date 03-26-2013 20:03

I suppose, that was a hardenable steel

In my case, with A36 steel, there is no reason for do that.

The elastic deformation was a problem for me, because our machine have a fixed electrodes ,and beam, passes throught a roll system that can´t correct elastic bending provided by heating only one flange.

Regards.