Company Standards for Welding 3

[frusso110]

I currently work for a small company. We have a new product that needs to pass mil-std-810 shock/vibration. I have designed a 6061-T6 aluminum welded test fixture for this project.

Past engineers working for this company have not actually put actual AWS welding symbols on their drawings - instead merely putting a general note such as "CONSTRUCTION TO BE WELDED - fillet radii .xx".

The vendor that builds this test fixture will certainly do a fine job if I just make a few general notes on the drawing. Since this is only a test fixture of which they will only be making one- I do not want to go crazy with this project. I've already wasted time trying to figure out how strong everything needs to be, the types of weld, the filler metal, etc.

How does your engineering company deal with welded drawings? Do you have any advice for an engineer with ~5 years of experience? Is this too much detail for a simple test fixture? How much information is correct for this particular welded assembly?

I aspire to be the best engineer I can be. I want to be a "real engineer". I'm terrified of turning out like my boss and predecessor. I want to do things the "right way".

Thank you for your time in reading this post, and for any advice you all may have.

 

[MintJulep]

So what is your question really?

AWS welding symbols are the way to communicate the required weld precisely and unambiguously on engineering drawings.

The meanings are clearly defined.

They are the way to communicate the engineering requirements for welding to the fabrication shop.

Using the note "Construction to be welded" is analogous to using a note "About as big as a microwave oven" instead of properly dimensioning a drawing.

 

[TVP]

I can only echo what MintJulep already stated. Use the proper symbols. If the fabrication shop cannot decipher them, they can read their copy of Machinerys Handbook or you can clarify it for them.

 

[dbooker630]

The drawing looks good with the proper symbols. You are missing one fillet weld dimension though.

 

[gtaw]

Fillet welds are typically measured with fillet gages that are available in 1/16 inch increments from 1/8 inch to 1/2 inch. Above 1/2 inch the gages are in 1/8 inch increments. Sizing welds to the nearest 1/100 decimal inch makes little sense in the wonderful world of welding.

AWS A2.4 is the industry standard for welding and NDT symbols. The current edition is the 2012.

The Machinist Handbook and the Mechanical Engineer’s Handbook are usually about 20 years out of sync with AWS standards, at least it was a few years ago when I looked at one in a book store.

Aluminum has no endurance limit. In other words, no matter how low the amplitude of the applied stress, the aluminum is going to fail by fatigue if there are enough cycles. The bottom line is the fixture could fail before the component being tested.

Best regards - Al

 

[dhengr]

The bigger questions might be, why is it built out of 6061-6T aluminum, and what design stresses did you use in the area of the welds? What weld process are you going to use? While I agree with MintJulep that your earlier engineers didn’t do it right, or show enough detail, it seems to me that you have gone to the other extreme in selecting your joint details. Only the military could afford to pay for that fixture, since it comes out of all of our pockets. Are your weld sizes in inches, cm or mm? Why on earth all the bevel welds, that’s a test fixture, not a museum art piece, all that weld prep. is very expensive, if it’s really not needed. Two sided bevels (a V) are twice as bad as a single sided bevel, because you have to handle two pieces for each joint detail. Then you use a fillet stitch weld at a location, btwn. the base plate and the main frame, right where there will likely be reverse bending on the welds. That is tension perpendicular to the axis of the weld, across the face in one direction, and at the root of the stitch welds in the other direction. And, all the starts and stops are about the worst fatigue detail a welder can produce.

You might consider moving the two diag. kickers in an inch or two, and moving then, and the main frame for that matter, an inch away from the edge of the base plate. Then use all fillets or variations thereof.

 

[Guest102023]

AWS A2.4 is not very expensive ...

 

[Ron]

Agree with GTAW...use fractions to denote the weld size. The missing dimension is on a V-groove weld, not a fillet weld.
Also, you do not state the material on your drawing. Further, you do not specify the welding process. The will make a difference with 6061-T6 material...two options: GTAW and GMAW with proper shielding gas and wire.

You mention that you have "wasted" time determining the member sizes and such...that's not a waste of time. It is a necessity. Also agree that aluminum might not be the better material for a test fixture, particularly if repetitive or vibratory loads are being considered.

 

[ledon]

Another thing to consider is that where aluminum is concerned your only as strong as your weld. 6061-T6 has a tensile of approx. 35ksi but as welded is reduced significantly. If your code is AWS. As welded 6061-T6 is listed in table 3.2 as having a minimum tensile of 24ksi. I have run several T6 procedures and as tested tensiles have never been much above 27ksi. While most materials can be made stronger in the weld area Aluminum generally is not.

 

[frusso110]

Hi all. OP here.. I've learned a fair bit in the past couple weeks about the design of welded joints.

Now that i'm back from a short vacation, I'm still not sure about how to size my fillet welds. I've seen rules of thumb all over the place, but at the end of the day, I really don't know what to do. I ordered a book on amazon that will be here shortly, but realistically i'll probably end up bolting everything together. Thanks!