Not logged inAmerican Welding Society Forum
Forum AWS Website Help Search Login
Up Topic Welding Industry / Technical Discussions / How would you weld this?
- - By Pat Dief Date 11-08-2007 18:17
Hello,

My name is Pat.  I'm a senior mechanical engineering student at Washington State University. 

I'm in a senior design class, and for our project we have to design a structure capable of supporting 2,000,000 lbs that also has a factor of safety of 2.

We've decided to weld the structure together rather than bolting, but because of the large size of the support members necessary to support the loads welding will be a challenge.  We're expecting the members to be around 6" x 6" based on computer modeling we've done so far.  I'll try and throw up a couple links to some pictures so you can see better what we're dealing with.

My question would be, do you know of any resources that specify weld sizes for complex weld designs like this?  Our library has some really old books on welding and only show simple welding cases with simple loading and no mention of MIG or TIG welding which we were thinking of using.
And my second question would be since you guys are experts at welding and I know very little, do you have any thoughts or concerns with welding something of this size that we should be aware of?

Some of the specifications for the project,
Material:  High strength steel with a yield strength greater than or equal to 51 ksi.  We're looking at 1020 steel for ease of weldability.
Dimensions:  4 legs with foot spacing of 6' on one side, 6'8" on the other
Maximum height:  20" from bottom of the feet to top of the structure.

sorry for the bad pic, and the image hoster throws up some adds with the picture, could not help that.

an idea of the overall setup of our design.
Parent - - By js55 (*****) Date 11-08-2007 18:47
Omer Blodgett's 'Design of Weldments' is one of the best reousrces. Available through  AWS.
Parent - By Mwccwi (***) Date 11-24-2007 21:40
I got my Omer Blodgett's 'Design of Weldments' from my Lincoln welding rep.
Parent - By aevald (*****) Date 11-08-2007 18:58
Hello Pat Dief, sounds like you've got an interesting project ahead of you. Being a design engineer on mechanical structures and other things obviously requires a bit of knowledge into the different modes of building, assembly, and such(in this case welding). The comments I would have for you are not from an engineer, they are more from one who has had experience in both the fabrication and erection of a structure or two. A572 gr 50 material comes to mind off the top of my head as a readily weldable and strength matching material to meet the requirements of your project. You may want to take a look at this material for your proposed purpose. As far as the actual joint design and related information there are many resources available through the internet that could give you some detail and guidance here. AISC has a number of publications and presentations that are available. Go to www.aisc.org and peruse their available information, they may also have a forum that you could pose some questions to.
     I would also suggest that you consider looking into taking some classes in welding to give you a more personal understanding of the various processes and thus make you a better informed engineer. Even though it can be drawn and engineered, it isn't always so simple to fabricate or weld it. If there are any large fabrication shops close to the college see if you can take a tour of their facilities and ask questions and see for yourself how some of this sort of thing comes to fruition. Spokane isn't too far away from you so you may consider taking a trip up there to do the same. Good luck on your project and be sure to share the results on here. You will likely receive other great recommendations for approaches and methods for helping you out. Consider and apply all that is suggested as you see fit. Best regards, aevald
Parent - - By DaveBoyer (*****) Date 11-09-2007 06:01
Are those 6x6 members to be solid?
Parent - - By Pat Dief Date 11-09-2007 09:31
Yes.  It's the only way we've been able to get enough rigidity in our design.  It's a lot of fun to see our designs bend under the loads and have to go back to the drawing board...well, not really fun so much as frustrating.
Parent - By Stickboy (*) Date 11-10-2007 04:41 Edited 11-10-2007 04:46
What type of surface will this project be setting on? Might want to look at some angle gussetts for that? And, how thick is the plate on top?
Parent - - By DaveBoyer (*****) Date 11-10-2007 05:51
Since YOU are the engineering student, how much strength do You need in the welds? I think Your choice of 6x6 solid material may make the weld joints dificult. Generally speaking, a large section solid beam is an ineficient way to achive rigidity. Just My opinion, as I am not an engineer.
Parent - By Stickboy (*) Date 11-10-2007 07:39
I bet there is a mighty fine line between enough ridgity and flexibilty with that much weight. To much ridgity might break.
Parent - - By HgTX (***) Date 11-20-2007 18:02
(1) If you can design to 50 ksi yield rather than 51 ksi (where did you get a number like that?), you're in a much better world of weldability.  ASTM A 572 is a friendly steel.

(2) I agree with those who say that 6x6 solid material is a questionable choice.  Don't just keep adding area and diameter till you get what you need.  Look at hollow sections and other shapes.

(3)  You already know how to design the weld.  The Blodgett book or your steel manual will give you the formulas you need.  You're the one who knows how much shear or tension that connection needs to take.

(4)  If you want to make a huge weld, or even a fairly large weld, TIG is not your process.

(5)  You don't want to make a *really* huge weld.  There are ways to increase the length rather than the depth of your weld if you need more weld.
Parent - By DaveBoyer (*****) Date 11-21-2007 06:33 Edited 11-21-2007 06:41
That 51 KSI yield or greater is a bit odd, then He mentions using 1020. What is the yield on 1020? probably LESS than 51 KSI. If deflection is the problem, I don't see higher yield as the solution, as the modulus of elastisicity won't be any higher. Myself, I would  burn a plate for the bottom large enough to cover the support structures, from thick enough material to keep within deflection requirements, and then top it with a riser block burnt from plate. Join with a fillet weld. Heavy? That is a relative term. These supports will be child's play compaired to setting the thing they are supposed to support. Too much material & cost? Way cheaper than all that cutting, weld preps and welding to make what He pictured. Just My opinion, it isn't My project.
Parent - - By Pat Dief Date 11-10-2007 19:26
Thanks for the ideas, lets see if I can respond to some of  your suggestions.

Stickboy:  The structure will have 4 legs, and each leg has a foot that is 12" x 20", the thickness of the foot material has not yet been set, but will be of the same material as the rest of the structure.  The structure (feet included) sits on 4 concrete blocks, with one foot sitting on each concrete block.  I don't know what an angle gussett is, but will look it up.

DaveBoyer:  I want as much strength out of the welds as I can get, equal to the base material if possible.  I was looking at A514 and its weldability, and one resource said for base materials greater than ~2 inches thick the weld will be as strong as the base material, provided appropriate weld rod and technique is used.  I need a more credible source to prove that to our sponsor, but it looks intriguing so far.  I agree that 6x6 solid material will not be fun to weld, but based on the loads we are carrying, and the size restrictions we must meet, we don't have many options.  We are looking at using steel plate and removing excess material to create the support structure which will reduce some welding, but there is no way to make this a solid piece (too bad it can't be cast).  Our sponsor has very good welders, so we'll do our best to limit the number of welds, and their complexity, but they're still going to have to do some welding.  I still respect your opinion engineer or not, I'm guessing you have more experience in your little finger etc.  I think I understand what you mean about solid section beams hence the idea of I-beams and such.  They just aren't providing the rigidity we need, and have stresses exceeding their yield strength when the structure is loaded.  Honestly we would love to just give them a solid block of steel, it just weighs too much :)

Stickboy:  Probably true, I'm not sure where we are on that line.

Thanks for the continued comments.  I really appreciate the ideas you have since you have so much more real world experience in these areas than I do.
Parent - - By hogan (****) Date 11-10-2007 20:10
have you considered something like 2" dewydag bars pretensioned running leg to leg
Parent - - By Pat Dief Date 11-10-2007 20:37
Dewydag?  Don't know what that is, Google came up with nothing on that one.  So I guess that's a no.

Is there another name for it?
Parent - By hogan (****) Date 11-10-2007 21:03
it's like rebar, but threaded
Parent - - By Stephan (***) Date 11-11-2007 11:33 Edited 11-11-2007 13:31
Pat Dief,

please allow me to give an additional hint on hogan's* advice.

When I have read dewydag, I recalled a German company named "DYWIDAG", please see also:

http://www.dywidag.de/en/dywidag.html

and actually I found something of what might going in the direction hogan has pointed in, please see also:

http://www.dsiamerica.com/products/post-tensioning/bar-post-tensioning-system/system-description.html

Best regards to you and all,
Stephan

* hogan, I hope you have permitted this as well
Parent - By hogan (****) Date 11-11-2007 14:37
thanks for the help with my poor spelling
Parent - - By 803056 (*****) Date 11-10-2007 21:55
Consider heavy wide flange sections. While 2 million pounds sounds like a lot, it isn't really.

Wide flange sections, be they hot rolled jumbo sections or fabricated sections (made from plate), would offer more efficient use of the material. Better section modulus for a given mass of material. If torsional strength was needed I would suggest tubular (closed) compact sections. You can even weld plates to the jumbo sections if more cross section is required.

Stay with ASTM hot rolled materials where possible for economy. AISI 1020 plate will not be as economical because of limited availability. 

From a welding stand point, the last process you would want to use would be GTAW. SAW or FCAW would be my first choices, followed by GMAW spray transfer and then SMAW in that order of preference.

Remember the mechanical properties of hot rolled materials are anisotropic. They are easily 50% weaker in the through thickness direction. So, in those cases where the loads are transmitted in the through thickness direction, special consideration must be given to prevent lamellar tearing. Welded jumbo sections should not be subject to tensile loads in the axial direction if they are joined by butt joints. Butt joints loaded in the axial direction should be bolted.

High strength steels are typically more difficult to weld. It may be beneficial to stay with materials that do not exceed 75 ksi tensile strength and keep the carbon equivalency to 0.45 or 0.5 to ensure good weldability.

Good luck - Al
Parent - - By Stickboy (*) Date 11-11-2007 04:55
By safety factor of 2, do you mean it actually has to hold twice as much?
Parent - By Pat Dief Date 11-11-2007 21:51
Thanks for the new info, I'll look at it in detail tomorrow (mon.) writing a history paper today.

Yes, the factor of safety means we have to be capable of holding 4 million pounds without exceeding the yield strength of the material.
Parent - - By darren (***) Date 11-24-2007 07:51
i have resisted this far, but
i wouldn't. it is a "unique" design.
i will give more info if you are interested.
darren
Parent - - By Pat Dief Date 11-27-2007 01:04
I'm interested in all ideas. 

We've come up with a couple models that look better than before, we've gotten away from 6" thick plate and are down to 3" (ooooohhhhhhhhh).

I read the Omer Blodgett book and have a question about weld strength, I'll post it tomorrow to provide the correct quote as I don't have the book with me at school today.  They way they refer to strength and size requires some clarification so I don't make an incorrect assumption based on his wording.  It was a very good book, bery practical info mixed with very technical.
Parent - By darren (***) Date 11-27-2007 11:09
well i think building a parabolic webbed shape (like i beams only shaped)would be less time consuming and a lot less material and welding. think of the flying buttress in med evil architecture with some supplementary cross bracing. two intersecting parabolas with some stiffeners and some cross bracing would be way cheaper and it would reduce the reliance on a welder to do all that welding as you could use the s.a.w process for most of it. and if you want to tension load the members then the aforementioned dewy dag system would work great just a thought because the thought of being the poor welder that would have to weld your skeletal support sent shivers down my spine thats alot of welding and tough to get 100% weld without alot of grinding.
darren
good for you for going to school and coming here to be proactive about your education
Parent - By Pat Dief Date 11-27-2007 18:35 Edited 11-27-2007 18:52
Ok, here is the weld size question.  On pg. 1 of section 6.3, he says

"In order to develop the full strength of a plate by means of fillet welds, it is necessary that the leg size of the fillet be 3/4 of the plate thickness... w = 3/4 * t"

Does that mean a weld meeting the above size requirement will have the same strength as the base metal, provided the right weld material is chosen?

Is there a similar equation you know of for double-vee groove welds?

------
I kind of like the I-beam flying butress idea.  The idea we are using right now is similar in a sense.  If I have time to model it I'll play around with your idea.

http://www.sjlupload.com/image_org.php?image=17249 (This is an awful image hosting website but it's the only one I can use on school computers)

That link will show you a picture of the design we have right now.  Made from 4 plates welded at the corners, with a 5th plate welded on top to take the loads.  The feet just support the structure.  The plates are 3-4 inch thick A514 because we needed the higher yield strength.

I'm still trying to figure how to weld this, but if you have any ideas I'd love to hear them.
Parent - - By Bob Garner (***) Date 11-27-2007 20:51
Hi, Pat,

My name is Bob Garner; I'm a structural engineer that hangs out on this site for welding assistance in my structural work.  I like the flying buttress concept made up of heavy plate.  You have lots of alloy choices in heavy plate so you should be able to find high strength and weldability.  Do you have access to finite element analysis progams to determine optimum plate thickness?  Your plates are probably going to be thick enough that local and overall buckling should not occur, and allowable stress will control.  I would extend your baseplates approximately equally outwards from the "legs" of your buttress to maintain concentricity of the baseplate loading.  You can determine baseplate thickness from the AISC Manual.  And you don't want to overstress the concrete pads below.  The baseplate formulas will tell you how big your baseplates must be to accomplish this.  If you need some concrete strengths to plug into the formulas, I would use at least 5000 psi concrete.  You can go as high as 10,000 psi if you need to - we do this with bridge concrete all the time but usually only in specially loaded areas such as you have here.

Mechanical design is a little different from structures design (mechanical as in machine design; structural as in bridge/building structure design).  In structural design, the safety factor, usually 1.66, is generally built into the design formulas outlined in the AISC Manual (the bible of structures design).  If you use any AISC formulas, you will have to adust your safety factor up to 2.0 as per your assignment.

I'm dying to hear the welding recommendations for this.  If this is really heavy plate, what does everyone think of the welding procedures used to butt weld railroad track rails together?  If desired, I can find a pdf file of a rail weld detail we used on the new Carrier project we did here in Sandy Eggo.

This sounds like a fun project!!

Bob
Parent - - By aevald (*****) Date 11-27-2007 21:24
Hello Bob, you mentioned welding railroad track rail. Are you referring to thermite welding? I have only seen it applied to track sections, is it also used for joining plates? or are there section size limitations? Sorry to sidetrack, but I'm curious. I have some ex-students that work for Burlington Northern Railroad and have had demonstrations of this process but have not researched or been informed of it's additional applications. Thanks and best regards, aevald
Parent - - By Bob Garner (***) Date 11-27-2007 23:40
Aevald,

I scrounged up the detail and the drawing just details the rail bevel angles and directions.  The drawing title is "Rail End Preparation for Manual Shielded Arc Welding" so I guess that answers that question.

I, too would be interested if heavy plate can be thermite welded.  How would the crucible work?  Is there a maximum distance that thermite can cover so it may have to be done in segments?  Interesting stuff.  Certainly not a sidetrack if it's something that could be used in this project.

Best wishes,

Bob G.
Parent - - By aevald (*****) Date 11-28-2007 00:08
Hello again Bob, the thermite welding that I witnessed on railroad track sections included a somewhat interesting rail end preparation: they cut the rails so that the ends were square and there was 1" gap between the two sections, they then applied a couple of form-fitted ceramic shoes and applied clay to any gaps to prevent the loss of the thermite powder while leaving the top of the forms open, next they filled the gap and form area with the thermite powder, finally using what amounted to a fourth of July sparkler they ignited the mixture, it wasn't extremely violent, but you could tell that there was magnesium in the mix and it sparked, bubbled, boiled, and smoked for a fair length of time as it was burning. I was told that the thermite mixture is a combination of magnesium, fluxes, granular metallic fillers, and possibly a few other components. A short time after activity had ceased and everything had an opportunity to cool a bit they knocked the ceramic mold and clay pieces off with a hammer and then used a special track grinding unit to smooth the weld area and allow it to be matched so that the  two sections of rail would have a bump-free profile, they did a bit more hand grinding to take off high-spots or any other imperfections and it was done. A very interesting process. I don't know if U-tube might have any video of this, might be worth some investigation to see. Best regards, Allan
Parent - By Milton Gravitt (***) Date 11-28-2007 00:25
Yes Allan they have it on U-Tube I have seen it there and it is impressive, they have a couple of films of it.
I would like to see it in person and see how they prep. things and check it out afterwards.
          Milton Gravitt
Parent - - By Pat Dief Date 11-28-2007 00:35
We have access to the FEA program built into CATIA V5.  It's not too bad, a little touchy but helpful.  The greatest stress concentrations are where the legs attach to the feet.  There is a lot of compression at the edge of the foot.  This can be reduced with some fancy filleting.  Don't know about optimum thickness but the structure is pretty close to working with 3 or 4 inch thick plate.  The trick is how strong can I assume the welds are.

We can't change the placement of the baceplates/feet or their size.  However, I can talk to the guys doing the modeling to see if they can change where they have the legs meet the feet.  I will have to try and find the AISC Manual for baseplate thickness, might be helpful with those stress concentrations I mentioned above.  I agree about the concrete pads.  We were told to use a value of 6000psi.

Is there a handy link to the AISC manual, or is it a book?  I'll look on their website and see what I find.
Parent - - By DaveBoyer (*****) Date 11-28-2007 04:56
AW CRAP!!! I wnted to see a really big thermite weld. Thermite is usually a mixture of finely powdered iron oxide and finely powdered aluminum, altho other combinations work too. The aluminum burns and uses the oxygen from the iron oxide, reducing it to iron in the process. There have been cases of this hapening accidentally in industrial dust collectors, the fire being impossible to controll.
Parent - - By aevald (*****) Date 11-28-2007 05:01
Hi Dave, did you mean to say magnesium instead of aluminum? Just wondering, didn't think the aluminum was flamable like that. If it is, then I'm ready to learn new lessons. Regards, Allan
Parent - By 803056 (*****) Date 11-28-2007 09:52
Powdered aluminum is used in some fireworks.

Al
Parent - By aevald (*****) Date 11-28-2007 14:17
Hello darren, Al, Dave, once again my thoughts or misconceptions need to be changed. Thanks for the enlightenment gentlemen. At least when I learn something new or receive a correction a lot of others learn too! Best regards, Allan
Up Topic Welding Industry / Technical Discussions / How would you weld this?

Powered by mwForum 2.29.2 © 1999-2013 Markus Wichitill