Rule of thumb is that the legs of a fillet weld should be equal to the thickness of the thinest member being joined. Or is it the throat?

Either way, the filler has a higher tensile strength than the parent metal. So a properly executed weld will be stronger than the members being joined. IF skip welding, I suppose you would figure the volume of the filler metal deposited and multiply by the tensile strength. Maybe???....

What are you working on anyway?

 

Hello coastalwelder, unfortunately as you already stated, there is not a simple answer. Some items that would affect your answer: is this compression, tension, shear, tortional, cyclic, or static load, are there temperature issues involved as well, heat/cold etc.? Those are a bunch of variables and I would venture to say there might be additional ones in some instances. On the other hand, E7018 is a 70,000 psi. tensile strength electrode, the actuals are slightly higher, if you figure the area of your weld deposits you could theoretically figure it's strength. Yet, you would also have to take into account the strength of the A36 base material, it is likely not anywhere near that of the E7018 electrode, I believe it ranges anywhere from the low 30k's to upper 40k's in tensile strength. I would imagine others will have much more technically correct answers and I will look forward to their responses. Best regards, Allan

 

Aevald has the correct answer to a complex question. E7018 has a tensile strength of 70k psi, that figure is very, very misleading because nothing about other modes of loading are accounted for. What is considered a failure, the joint breaking or the joint moving 0.01 inch? Failure mode must be accounted for, the type of joint, joint shape. An engineer can calculate the strength from the dimensions of the weld from the electrode, and only estimate the additional strength by the depth of penetration if there is any.

Over welding will not make the joint any stronger that the proper weld technique, it just wastes time, filler rod-wire and may weaken the joint due to heating. The rule is to match the leg to the thickness of the base metal.

 

costalwelder - and to all that aspire to weld - the first book in your library should be Lincoln Electric's, The Precedure Handbook of Arc Welding, now in its 14th. edition. You should own this book before you own a welder, it's the best $25.00 you can spend on your welding fantasies. Your question, and virtually every question on this forum, in one form or another, is answered in this book. Over the decades I have gifted no-less than 6 books to aspiring welders. Your very question is amplily illustrated; advising you to buy the book with the answer is much better advice than answering your question. Opus

 

The safest bet is to ALWAYS lay down more weld than the thickness of the parent material. Do this every time and a proper weld will always be stronger than the parent material.

Scott
http://www.welders360.com/

 

This is an example as to why you should buy the book.
The only absolute I have experienced in my 30 years of welding is that I am going to get burned.

 

Thank you all for the input. I've actually laid down thousands of pounds of filler metal successfully following someone elses procedure and weld symbols. It wasn't till recently that I've taken an interest in actually how those were created.
Maybe it's a "bad"economy thing, but I've been seeing more repair and reinforce jobs come into the shop. Most recently was a broken receiver hitch on a RV. I've seen this happen several times before. Usually, it appears to be caused from overloading. It made me ask the ? , "could I figure out what the weak link was and how much could the customer safely tow?"

 

Thinnest member is the determiner of fillet size. Weld doesn't have to be any stronger than the thinnest piece of steel joined as a rule. It makes sense, and is well covered in the first few chapters of the Bible.

In practical experience, when I have a lift eye, I have found that increasing leg size by building up a multi pass weld keeps the lift eye from bending near the base. But that's a special application.

On a failure, where did it fail? The weld itself, the steel next to the weld, or simply the steel. Reinforcing welds doesn't make up for bad design in the overall structure. Sometimes adding a simple stiffener or gusset can reduce the stress raiser at a poorly designed intersection. Most of the time the only way I can figure it out is by taking a good long time looking at the damage, and trying to figure out all the stress that the joint might be subjected to.

 

one of the most interesting failures I ever encountered was back when I only used 6013. The weld didn't fail (per se), the steel didn't fail. The entire weld simply pulled out of the parent metal on a hydraulic arm. Clearly defined shallow pockets could be seen where the weld pulled out. It was inadequate penetration. It was the last time I relied on 6013. The repair was done with 6011, and all is well to this day.

 

In referance to the "handbook" you all are refering to. I found a chart in the 12th edition table 2-8, Titled " Allowable load for various sizes of fillet welds" It seems like a pretty straight forward reference point. Can anyone tell me what "r" "f" and "w" stand for????

 

I suppose we could really go over the edge and discuss the weld ductility, and resistance to shock loads (Charpy stuff I think).

There's seismic stuff too.

But in the everyday world I think most guys have a good feeling for just about the right amount of weld to put on a joint. Or maybe not, dunno.

Excessive welding can damage the parent metal, too little can lead to failure too.

How does the particular material respond to the high heat of welding.

There's a ton of variables besides just tensile strength from what I understand.

Quality of the weld is an issue too..................

Sometimes even the engineers get it wrong. Remember the bridge collapse in Minnesota or somwhere out that way.

It's probably beyond most of us here.

I tend to look at manufactured stuff, and observe how it's designed and welded. It's not a perfect way to learn, but it's about the best I can do without further education.

My greatest leap of faith is climbing onto an airplane:laugh:

But to be real, I probably take my life into my hands every time I go over a bridge. Walk into an elevator, or sit in a plain chair in the restaurant (metal chair)

 

Not sure if any of these would be considered easy rules of thumb but here is a link to some calculations if you'd like to do the math http://www.gowelding.com/calcs/ :dizzy:

 

Thanks for this, this is what I was looking for.

I guess the short and sweet answer to my original post is NO, there is no "simple stupid " rule of thumb that I could use as a general purpose calculator.

We've all seen things in our field that should never work or should never been done, but have been working for years. Likewise, most of us have seen things fail that have left everyone on the job site scratching there heads wondering "how"

I had an old school welder tell me a while back " no matter what someone tells you to weld, you have to know that what YOU did is going to be ok" That being said, I've never walked away from something I welded thinking "I hope that holds" The source of my original question was hoping to find a quick way of confermation.

 

was working ona pipe hanger project and i had an engineer tellme that 1" long 1/4 fillet is good for a thousand pounds. no one is gonna sign off on that , but it probably wont hurt you , either, if you use yer head in applying it.

 

A lot has to do with who is welding it. And what kind of penetration you are getting.

You can see welding's weaknesses when you weld aluminum. Often steel hides the welds weakness. Where aluminum brings it out.

You are not supposed to start or stop a weld ever. Yet it is done every second of everyday. In theory one inch of fillet weld is zero holding strength, from an expert engineers standpoint. In other words when you put your one inch intermittent beads on a heavy moving trailer and the trailer comes apart and kills a Senator and his family. There will be an expert there to show you how stupid you were to do that.

I am not saying I will not do that later today on a lawn mower grass catcher. I am just saying that push comes to shove, you are the fool. That is the power of the organized crime family, of college degree holding criminals backing themselves up. You are the butcher they are the ones that would have told you if they knew you were so stupid.

Approved plans containing specific instructions to put one inch beads intermittently on a part will vanish. I will have a set but everyone will be more interested in why I have the only set.

The truth is that when you make a fillet weld. You create a new structure. You no longer have two flat plates. That can bend and flex. You now have a new geometric structure, created by the weld. Usually the weld is now protected by this geometric shape. However upon destructive testing you often see the weakness in the weld come out.
In other words it looked and worked so good for so many years. And then suddenly just let loose. The reason is that certain stresses are created by the geometry, that can cause the weld to crack at the edges or right in the middle of the part. And then the crack just runs up the weld.

Depending on which welding engineer you talk to. A ten foot weld is not stronger then a one inch weld if the start and end of the weld are not finished. Brought all the way to a place, that leaves no starting point for a tear. Since I have seen perfect welds ten feet long open up. Just crack right down the part. I agree totally.

When you weld something you have to become an engineer. You need to know what this weld is going to do. And where it is going to let loose. I destroy things all the time for that reason.

If you have ever done a fillet weld, on two plates held 90 degrees to each other, and then closed the angle between them on the weld. You know that you can just bust a fillet weld right off. Unless you totally penetrate. And even then I have seen some surprising tears.

If you weld it well, more then likely the material will give before the weld. Aluminum of certain hardness being an exception. Hardened aluminum will often tear a weld, a weld that is often softer then the hardened material being welded.

Something that catches a lot of welders the first time they do it, is when they do a fillet on solid material that does not flex or give. Sometimes without that reliving flex. A weld will just open right up, down a whole solid piece fillet welded. Because the forces are just too much for the start of the weld.

Sincerely,


William McCormick

 

non illegitimatum carborundum, willie...
(dont let the bastards wear you down)...

 

Destroy it, it is the only way I know to see how it holds. Where it lets loose, and perhaps why it lets loose.

We found that on steel on fillets that MAG welding with 98/2 Argon/Oxygen was better then straight Argon. Partially because the straight argon had a little trouble heating the part up. Either we had to put a run in tab to preheat the part or we got a cold start. The 98/2 seemed to really preheat the part well. And we got tenacious holding with the 98/2 MAG welding process.

That is why they use Argon and CO2 MAG welding today. For the heat. Pure Argon is a much better weld if you can get it to weld.

Today they would laugh at you if they saw you welding on run in tabs. Ha-ha.

Sincerely,


William McCormick

 

I agree with what William Jr says, as a welding engineer i spent a good deal of time at one job doing failure analysis on all sorts of parts and found a lot of flaws with either the design, the welding techinque or the metallurgy.

anyways the formulas start off easy but kick up the complexity to 11 because all of the sudden you're talking about 3 dimensional stress analysis and the math becomes some tricky calculus. I call in a professional engineer to double check and sign off on those calculations because frankly I don't trust myself to not make a mistake. Past that also there's all sorts of caveats and design knowledge to where something can theoretically hold 10,000 pounds but will break about with you tugging on it. or as mentioned with short welds, tearing with rotation around the axis of the weld, forces not parallel to weld axis' etc it gets tricky fast.

The best advice as mentioned is to build one up and destroy it, a lot easier to build up two frames then load one until it fails as opposed to having a professional engineer spend time building a model to theorize structure strength to a good degree.

on my own I'll do simple loading calculations, but then again I have a degree in welding engineering and some practical experience designing pressure vessels.

 

I know something that earns some respect from me for some engineers, is when you show them something they had never seen before. They will often go into great depths investigating what the ramifications are. And everyone walks away a winner. Because I have no time or resources to go into such great depths. They sum it up for me, so from then on, I have the proper lingo to go with my understanding of it.

A few will just deny it is true or claim it is all calculated in there.

The best engineers will talk about a design to death, and I love every second of that. Two guys from two different angles, talking about engineering is about as good as it gets. Both guys walk away with great stuff. Mind boggling stuff. The last words from a great engineer will be, "well the bets are in lets go see".

Some guys, welders get angry because they do not know about all the angles of engineering. And feel left out, that is bad too. I won't let welders or architects stop knowledge or conversation about something. I will tell anyone anything if I believe it is true. Sometimes I am arguing with guys that are cage fighters on the side. Ha-ha.

A friend of mine that runs big jobs for the government. Pouring runways with those on site moving concrete plants. Was telling me that they are concreting over asphalt at JFK airport. Now last time I fooled around with asphalt when I was a kid. I blew a three foot crater in the street with a tiny electrochemical charge.

He was saying that they were paving over the asphalt with 18 inches of cement. I cannot say that, this particular asphalt could explode without a sample. However it is usually a poor idea to do something like that, over something that is explosive. This was at JFK.

A good engineer already tested the asphalt or would want to know this.



Sincerely,


William McCormick

 

You guys mentioned having an engineer double check the calculations for me. I like that idea for future projects. What type of engineer should I look for? Sorry for the lame question, I've always been on the sweaty side of the project.

 

An RPE (registered professional engineer) can stamp a designed joint or structure, he can also certify that you are qualified to assemble or weld that structure.

The RPE cert for a welded joint or proceedure is specific to that joint or proceedure only.

You can shop or ask around and maybe find someone "on staff" somewhere that will do it on the cheap.

Matt

 

Thank you all for the input. I've actually laid down thousands of pounds of filler metal successfully following someone elses procedure and weld symbols. It wasn't till recently that I've taken an interest in actually how those were created.
Maybe it's a "bad"economy thing, but I've been seeing more repair and reinforce jobs come into the shop. Most recently was a broken receiver hitch on a RV. I've seen this happen several times before. Usually, it appears to be caused from overloading. It made me ask the ? , "could I figure out what the weak link was and how much could the customer safely tow?"

In referance to the "handbook" you all are refering to. I found a chart in the 12th edition table 2-8, Titled " Allowable load for various sizes of fillet welds" It seems like a pretty straight forward reference point. Can anyone tell me what "r" "f" and "w" stand for????

You guys mentioned having an engineer double check the calculations for me. I like that idea for future projects. What type of engineer should I look for? Sorry for the lame question, I've always been on the sweaty side of the project.

Every project, large or small, has a purpose. Projects can have a run of thousands of identical pieces, or be unique one-of-a-kind projects. They serve a variety of functions; and are subject to a range of loads, whether static, wear surface, pressure vessel, cyclicly loaded, shock loaded, etc. Stress comes in a number forms, be it tensile, compressive, bending, torsional or shear, or a combination. I still struggle to understand the relationship between stress and strain, but I think that one results from the other.

If you are simply asking a general question for your own interest, then a study program would be in order. Engineering studies are always couched deeply in mathematics. And mathematics is, simply academic. We have a number of engineers who frequent the board, and their contributions here are always interesting and enlightening, at least for me personally. :blush:
Rule of thumb is, as always, overbuild it.

 

simple rule of thumb for 7018 - every inch (some guys say inch and a quarter) fillet weld with 1/4 legs is "good" for 1000 pounds. This is on A36 mild structural steel.

I use this rule to determine what minimum I need to weld picking eyes for hoisting materials and anchors for chainfalls (usually to hand set 1200-1800 pound stair stringers). We usually dont weld all around on these things for easy removal and clean up...if you have that option weld all around. Because Im the guy who is "designing" the picking eye or anchor, I put alot of consideration into it, and inspect and monitor the welds and anchor material frequently. The last thing I want to do is kill myself, second mostly is killing my partners.

Thats the dumb ironworker method. I see here already that most of you are smarter than us.