I am a mechanical engineer working on a warranty claim against a truck body manufacturer. At issue is several feet of failed weld on 3 truck bodies. The manufacturer claims the welds were as designed, I believe that the welds were not adequate for the task. The welds in question join 400 bhn plate (150 ksi tensile) to an ASTM A36 (we think) half round. The half round has been notched with the plate welded to the half round in the notch using 80 ksi electrode. The plate is 5/16". The throat of most of the weld in the failed area is only half the thickness of the plate. Am I wrong to believe that a weld designed to AWS standards would have a throat as deep as the thickness of the plate it is joining, particularly when formed using an electrode with roughly 1/2 the tensile strength of the parent metal?
I would say the weld size needs to be whatever the design, drawings, and specifications calls for. As far as I know, AWS doesn't say what size a weld needs to be, that's up to the engineers to determine according to the application. I would be careful about weld size and material thickness generalizations.
Was the welding done according to a qualified procedure? Was it required to be? Is the weld size as specified? Is the material as specified or was something substituted?
It sounds to me that you could be looking at a material/material/filler metal mismatch and you might be better off to consult a metallurgist.
CHGuilford
Thanks for the quick response. We haven't been able to obtain production drawings from the mfg with weld spec's. They did tell me during the design procedure that they use the tensile strength of the weld material for the tensile strength of the entire structure when modeling the design for FEA. That being the case, if a 5/16" plate is required, then a 5/16" thick weld is required to join that plate. In some areas the weld appears to be full thickness, which would lead me to believe that the design called for the weld to be full thickness. We have made arrangements for a CWI to come out and inspect the welds.
I agree with your point about parent metal/filler metal mismatch, we may also take samples for metallurgy if it comes to that.
Thanks for the help.
It is impossible to tell what weld would be required in a structure like this without knowing all the details of the original engineering analysis. For instance, is the A-36 half round there to increase bending resistance, and thus loaded in shear, or is it loaded in tensile or compression? If it is loaded by bending, many times undersized or intermitant welds are more than adequate. If it is loaded in compression, even less weld may be required. If tension, welds are most often "full strength", but joint design also needs to be evaluated for stress risers that can contribute to fatigue failure, as fatigue is often the culprit when falure occurs in the transportation industry.
I also agree with CHGuilford in that AWS (as in D1.1) does not specify weld sizes. If there is a governing design code for this kind of transportation equipment (I am not familiar enough with this industry to say either way), such as the ASME code for Boiler and Pressure vessels, it would specify design requirements.
What I would consider doing to tell if there is a claim here would be to do the following:
1. Detemine if workmanship conforms to the prevailing code. (You are already doing this it sounds like, as you are retaining the services of a CWI.)
2. Determine if the fabrication conformed to the design drawings. Look at the weld (material, size, welding procedure, qualification to code, etc.), and the base metal. (correct type, size, etc.)
3. Determine if the design drawings conform to any applicable design codes.
4. If no design code is applicable, you may have to hire a professional engineer to determine the adequacy of the design.
G Roberts
400 BHN plate was probably used for wear resistance and not for its structural properties. In many wear applications, the wear plate is joined with filler metal less than that of the base metal. The weld size would be dependent upon the strength needed. Since the A-36 is the weaker of the two. The UTS of the wear plate is not relevant.
I worked for a rotating equipment manufacturer that welded 300 BHN to A514 with E71T11. No failures that I was aware of due to this process. All fan components except for shroud splices were welded with 70,000 PSI tesile strength electrodes.
If the welds failed along the toe/heat affected zone, I would not be suspect of the wire, more the procedure (preheat/heat input. If the failures are in the weld metal, then the filler metal/weld size could be the culprit.
In straight tension, a weld may out perform the base metal even if the effective throat is less.
Have a nice day
Gerald Austin
MK,
I recently was involved in a major equipment failure using very high strength materials (ASTM A514). It is possible and not uncommon to undermatch filler metal with respects to the base metal, but that is determined by the intended service by the designer. In example, a statically loaded weld subjected to compressive loads (not in fatigue mode).
In our case, the manufacture would not divulge any information about the welding requirements. I had to take material specimens and send it out for analysis. My personal thoughts about our project is failure to adhere to the WPS the manufacturer said they were using.
The photos I took are really something to look at. The welds literally peeled apart in several failure modes; along the toe, cleanly through the throat, and definately between layers/beads. Some welds propogated from crater cracks.
I would not guess at what the intended design is (PJP/CJP or effective throats). AWS does provide this information in the D1.1 for prequalified joints, but I still would not make assupmtions if your the consultant. I would not say anything until the results of the failure analysis are revealed. Assumption is the mother of all ... The attornies will have you for lunch if you assume anything. Don't give them the pleasure. Wait until the manufacture provides this information, or wait until it is required by the lawyers in the usual legal format. Speculation does not mean much in the courtroom.
We are awaiting the engineering consultants to decide what they want to do for our project. If this equipment cannot be repaired, then it will go to the mat and the welding documentation (or lack of) will be evidence. Good luck.
Thanks for all the replies. The 400 bhn plate is actually being used on the outside of the body for its strength, not wear characteristics. I am not a consultant on the project, but an employee of the equipment owner. I must speculate about the design (since we haven't been able to obtain prints) to determine a course of action for us to take. Since I am an employee of the owner my opinion would mean zilch in court, but I must advise my employer on the best course of action.
Some details that may clarify: These are 200 ton mining truck dump bodies. The half round forms the top rail of the sides of the body. The trucks are loaded using large front-end loaders. The top rail of the body is thus the business end of the body, and contact with the loader arms/material, etc is inevitable. It appears that material is getting caught between the loader arms and the top rail, pulling the half round towards the inside of the body. This action creates tension in the failed weld. The mfg is claiming that we have ground away the weld by rubbing the loader arms against it, and that our loading practices are the sole cause of the failure. I am arguing that the weld was not installed as designed, or if it was then the design is insufficient. The claim that we have ground away the weld is pure bull--how much grinding would that take!?! We haven't even worn the paint off the 5/16" plate. The bodies were custom designed for our operation, so the mfg was aware of our loading practices.
Thanks again for all the replies!
From your description, it sounds as if the failure of the welds are on relatively new truck bodies. (Paint is not off yet.) If this is the case, then it is unlikely that it is a fatigue failure. (Although not impossible if you are axperiencing a great deal of vibration in this area.)
To really get down to the bottom of cracking problems, one needs to know what the cracks look like and exactly where they are occurring. Typically the info that could help would tell us if the crack is in the weld, or in the parent metal. If it is in the parent metal, which one - The 400 bhn or A36? What do the cracks look like. If they are in the weld, are they running along the centre line, or transverse to the weld. etc.
Without any further info we can only speculate widely, but I can tell you that when you are welding 400 bhn plate, it can get rather tricky. Too little pre-heat and you get cracking. Too much pre-heat and the material is weakened. Hardness tests of the weld and HAZ of the 400 bhn material is likely to give you quite a lot of information. If it is too hard, any sort of deformation of the weld in this area can result in cracking.
Let us know what the cracks look like and their location, and how long the trucks have been in service, and we can always try to speculate more narrowly.
Regards
Niekie Jooste
The truck bodies have been in operation an average of 1400 hours. The welds are not simply cracking, they have completely failed. In some areas the weld itself has separated, in other areas the weld pulled away from the 400 bhn plate, and in other areas the weld pulled away from the half-round. The weld in the failed region appears to not only be of inadequate size (in my opinion), but also to have achieved little penetration of the parent metal in some areas.
I don't believe fatigue or vibration are an issue in this case, as it would appear the failure is directly related to the loading conditions indicated in my earlier post.
I have digital photos if anyone is interested.
Thanks again.
Unless you have corrosion problems, all mechanical failures manifest as cracks. (Many corrosion problems also result in cracks.) If you state that they "completely failed", how is this taking place, and how did this start? Again, the question is where are the cracks forming, and what do they look at when they are forming? If you only see them once the whole thing is "mangled", your evidence has been eliminated.
By all means, e-mail me the pics.
Regards
Niekie
I have had a look at the pictures you e-mailed. Only in the case of one of them could I really without doubt understand what I was seeing. Without a bit of an explanation as to what the picture shows, including some idea of the magnification, it is difficult to make any conclusions.
Based on the one picture however, I can say that the failure was in the toe region of the weld where it is welded to the half round. I could not see if the failure surface exhibited a ductile or brittle nature.
Maybe you could have a look and give some feedback on this.
Based on what I saw, I would have to guess that you experienced a ductile overstess of the material. The weakest part of the joint would be the area where this failure occurred.
I assume that the half round is welded onto the top of the truck to actually serve as a means of limmiting the dammage caused by rubble hitting the top edge of the bucket. As such, its job is not a structural one, but rather to act as a "consumable" that can be removed and replaced once the wear has become excessive. I come to this conclusion because the half round appears to have a width almost the same as the top of the truck's bucket. If it was there to give structural support, it would have to be substantially wider.
Based on the above assumption, you would not actually want a large weld, as it will make it more difficult to replace this component. The question we need to ask is why there appears to be tensile stresses on this weld? If it is merely for the purpose described above, it should only see compressive stresses.
I believe you stated that the loaders tend to press against the side of the bucket while dumping the rubble into the truck. If this is what is causing these stresses, you have to ask yourself if this is a reasonable and common practice in your industry. If the answer is yes, then you may have some basis for a claim against the truck fabricator. If this is a non-structural joint, I do not think that you can argue the case on the basis of code requirements. You will rather have to argue it on the basis of "reasonable wear and tear". If on the other hand, your front end loader operators are more heavy handed than the average in the industry, you must just take the punch and either re-educate them or ensure your trucks are constructed more robust.
Let me have the explanations of the other pictures, and maybe I will be able to see something more.
Regards
Niekie Jooste
Thanks for the explanation of the pictures. Having these explanations, I understand your situation a little differently. Help me out if I still have not got the correct picture.
The half-round is the top member that joins two 5/16" plates together. (8mm in the units that I understand) The half round is about 1" in width (25mm) wich means that there is a 9mm gap between the plates at the top of the truck box.
This means that any loading from the side of the plate will result in stresses in the weld. As such, this weld can be seen as a load bearing weld.
I still could not see if the failure surfaces are brittle or ductile in nature. (Focus not good where the fracture surfaces are shown.) Not withstanding this, I saw the following:
1) The weld penetration varied between 1/3 and 2/3 (2.5 - 5mm) of the plate thickness. This suggests that the welding was possibly performed without a weld prep. They just relied on the penetration of the welding process.
2) The portion of the plate that was not welded, appears to be very uneven. It looks as though they just cut the plates thermally and welded the half round on without smoothing down the edges properly.
From the above, it is possible that the welding was performed without propper plate preperation. This would lead to possible slag entrapment in the weld from the cutting operation which could embrittle the weld as well as lead to serious porosity.
From this, I would suggest that you have ground for a claim. Just confirm that what I have said is correct.
Regards
Niekie
Here's a few points to consider...
AWS D1.1 (2000) Section 2.4.5 says-
"The MAXIMUM fillet weld size detailed along the edges of the material shall be the following:
(1) the thickness of the base metal, for less than 1/4" thick (see figure 2.1, detail A)
(2) 1/16" less than the thickness of the base metal, for metal 1/4" or more in thickness unless the weld is designated on the drawing to be built out to obtain full throat thickness"...etc.etc.
Section 5.14 says-
The MINIMUM fillet weld size, except for fillet welds used to reinforce groove welds, shall be as shown in Table 5.8. In both cases the minimum size applies if it is sufficient to satisfy design requirements.
Table 5.8 says-
Base metal thickness of greater than 1/4" and equal to or less than 1/2" minimum fillet weld size = 3/16"....etc.etc.
Section 3.3 gives you information regarding matching filler metal to base metal for prequaified procedures. The 150 KSI tensile range material is off the prequalified chart so that means the manufacturer would have had to qualify the procedure.
The CWI you hire wiil have to see the Procedure Qualification Record and the Welding Procedure Specification for the particular joint as well as the applicable drawing in order to perform a complete inspection.
There should be a pre-heat and post-welding cool down procedure for welding to the 150 KSI material to prevent cracking.
Applying more weld than necessary to a joint does not strengthen it, but actually weakens the base metal in the heat affected zone through excessive heat input causing embrittlement.
Good Luck and let us know how it turns out!
MK,
I have not seen a post saying what code the truck was welded to but I do see many post refering to AWS D1.1. The AWS D14.3 covers Earthmoving and Construction Equipment. You might be able to find some of the answers you are looking for there, if that was the code used for fabrication.
Forum
