Rule Of Thumb For Fillet Weld Size

[Veer007]

Hey guys
This may sound silly,
If forces are unknown, is this a good practice to determine fillet weld size by the below table? Also, fillet leg and fillet size are the same?

Thanks in advance!!

 

[engineering_patrol]

Regarding fillet width, you can estimate its size as 70% of the steel plate thickness.

 

[desertfox]

Hi Veer007
My rule of thumb for fillet welds is the leg length of a fillet weld should never exceed the thickness of the smallest section being welded.

“Do not worry about your problems with mathematics, I assure you mine are far greater.” Albert Einstein

 

[phamENG]

The "if forces are unknown" bothers me. Why are you designing a connection for which you don't know the forces? Seems like a risky proposition...

 

[dauwerda]

I agree with phamENG, if forces are unknown it would not be good practice to use a "rule of thumb" weld. You may be able to use a weld that develops the full strength of the connected part, but this may result in excessively large welds which can introduce other issues.

 

[desertfox]

Well you can’t have a fillet weld leg length that exceeds the plate thickness in practice, so if you choose a leg length that is the same as the minimum plate thickness, then that’s the strongest weld you can have unless you change the weld type to say a butt weld.

“Do not worry about your problems with mathematics, I assure you mine are far greater.” Albert Einstein

 

[dauwerda]

desertfox, that is not true. The only time a fillet weld leg is limited by thickness is when it is welded on the edge of the plate. A T joint, for example does not have this limitation. See this thread for more information on that:

https://www.eng-tips.com/viewthread.cfm?qid=471129

 

[phamENG]

dauwerda - I think desertfox is getting at the development of connected part strength strategy. Once the leg gets larger than the thickness of the thinner part in a T-joint, you start getting into base material strength controlling the connection. It depends on strength of material, strength of filler, and the actual loads applied, but for a rule of thumb it's a decent practical approach.

I still go back to the not knowing the forces. In a building structure, you should always know the design load on your welds if it needs to be specified. A seal weld on a cap plate is one thing, but if you need to define a fillet weld you should know what loads to expect.

 

[desertfox]

A quote from the site I use:-

The maximum size of fillet weld is generally that of the thickness of the thinner of the two items being joined but very large fillet welds may cause unacceptable distortion and/or extremely high residual stresses. In addition, above a certain size it may be more economical to make a T-butt, rather than a fillet weld.

https://www.twi-global.com/technical-knowledge/job-knowledge/design-part-2-091

see the link above

“Do not worry about your problems with mathematics, I assure you mine are far greater.” Albert Einstein

 

[dauwerda]

If you don't know the loads that a weld needs to be designed for, the only safe rule of thumb is going to be to provide a weld that will develop the strength of the base metal. For a fillet weld this will result in a leg size that is larger than the thickness of the base metal if it is a one-sided weld (such as is the case with a tube welded to base plate). As noted above, excessively large fillet welds can lead to other issues. Hence the reason I say there is no good rule of thumb for fillet weld size if the loads are unknown.

 

[desertfox]

dauwerda if the table shown in the op's original post shows all the fillet weld sizes to be less than the thickness of the material being welded why are you assuming he has a TEE joint?
Further to develop the full strength of of the base material using a fillet weld the leg size of the weld only needs to be 3/4 of the material thickness and this applies to TEE joints.
see link below:-

https://www.linkedin.com/pulse/rule-thumb-fillet-weld-size-thomas-lakas

“Do not worry about your problems with mathematics, I assure you mine are far greater.” Albert Einstein

 

[Veer007]

so the only considerable thing is weld size should not be lesser than 1/4Xt, Right?

Thanks in advance!!

 

[desertfox]

Hi veer

Well yes you are right but it depends on what you are doing, to develop full strength in the welds compatible with the plate you need 3/4 x plate thickness on both sides of the joint, if you are just wanting rigidity then weld leg length is 1/4 x plate thickness, but if you don’t know what the loads on the joint are then surely the best you can do is develop full plate strength in the welds

“Do not worry about your problems with mathematics, I assure you mine are far greater.” Albert Einstein

 

[dauwerda]

dauwerda if the table shown in the op's original post shows all the fillet weld sizes to be less than the thickness of the material being welded why are you assuming he has a TEE joint?

The OP didn't tell anything other than that it is a fillet weld, so my assumption is that it can be any type of joint that can be welded with a fillet weld.

Further to develop the full strength of of the base material using a fillet weld the leg size of the weld only needs to be 3/4 of the material thickness and this applies to TEE joints.
see link below:-

Did you see all the assumptions listed in your link that need to be true for this to be true, mainly the requirement that the fillet weld be on both sides of the plate. A fillet weld with a leg size 3/4 of the material thickness does not develop the strength of the material - two fillet welds with 3/4 of the material thickness do.

The OP did not give any indication of joint type, just asked if that table (which also doesn't give any more detail) is a good table to use as a rule of thumb for fillet weld sizes when forces are unknown. I have responded with reasons why it is not a good idea to use it as a rule of thumb for certain types of fillet welds. It certainly may work for some situations, but it is not a catch all (isn't that what a rule of thumb is supposed to be? or if not, at least include the situations it is and isn't applicable to) and it gets even more concerning if someone is trying to apply it to joints where they have no idea what the loads they need to resist are. Perhaps that table has footnotes to make sure it is not misapplied, but the information given in the OP does not include anything about where it should or shouldn't be applied or that it is providing information for two-sided fillet welds. I believe it is prudent to point out where it doesn't work as none of this was included in the OP.

 

[phamENG]

For what it's worth, I agree completely with dauwerda. In my last post, I was thinking of a two sided T - sorry I wasn't more clear.

If I need to rely on a weld for any kind of strength, 1/4t is too small. Take a look at Table J2.4 in AISC. Minimum fillet weld sizes are all larger than 1/4t. And, of course, that's the bare minimum. You really need to know what your forces are to determine your weld size. Just going for full capacity of the connected parts will, in most cases, result in a grossly over designed, inefficient, and expensive connection. A consulting or design engineer may get away with that occasionally or on small jobs, but you work for a fabricator where cost of each connection matters on every job. Do that very often and they'll probably show you the door.

 

[desertfox]

Dauwerds

You clearly stated in your post that I was wrong and in your next post that to develop full plate strength the weld leg length would need to be larger than the plate thickness. Which is also incorrect.

I didn’t make the assumption about double sided welds but looked at the table provided which clearly shows fillet weld leg lengths less than the material thickness which was the point I was making.
I agree that without knowing the forces on the joint the best anyone can do is develop the weld strength to match the plate strength but he doesn’t need a leg length to exceed the plate thickness

“Do not worry about your problems with mathematics, I assure you mine are far greater.” Albert Einstein

 

[dauwerda]

Dauwerds

You clearly stated in your post that I was wrong

I believe you are referring to this post:

desertfox, that is not true. The only time a fillet weld leg is limited by thickness is when it is welded on the edge of the plate.

which was in response to your post:

Well you can’t have a fillet weld leg length that exceeds the plate thickness in practice, so if you choose a leg length that is the same as the minimum plate thickness, then that’s the strongest weld you can have unless you change the weld type to say a butt weld.

Which I will again state, is incorrect.

in your next post that to develop full plate strength the weld leg length would need to be larger than the plate thickness. Which is also incorrect.

I agree that this statement is not true for a two-sided fillet weld, but it certainly is true for a one sided fillet weld which is why I specifically stated that. Again, the OP didn't give any information as to what he/she was planning on applying this to.

I didn’t make the assumption about double sided welds but looked at the table provided which clearly shows fillet weld leg lengths less than the material thickness which was the point I was making.

That table does not state anywhere that it is for double sided fillet welds. You can look at it and infer that that is what it is for because of your experience. I don't believe it is a good thing to assume that the OP (or anyone else that finds this thread in a google search looking for a rule of thumb on fillet welds) can also recognize that without pointing it out. I can very easily imagine someone seeing the information provided and misapplying it.

 

[cliff234]

If the forces are unknown, you cannot safely determine the economical weld size.

In general, a fillet weld made with E70 electrodes will provide a service level load capacity of 0.928 kips per inch per 1/16" of weld size. (example: A 1/4" fillet weld would be good for 3.7 kips per inch) There are many other issues related to weld size/strength that I did not go into, but this is a start. I would suggest talking to a structural engineer to get a comprehensive answer to your questions. Ask them to review your proposed detail. Most structural failures are connection failures. Don't try to design connections unless you know what you are doing.

By the way... That's a frightening table you posted. "Rules of Thumb Fillet Weld Sizes"?! That's like a list "Rules of thumb for disarming land mines". The fact that this table appears to have been published in a book is especially disturbing.

 

[desertfox]

Dauwerds

If you have two plate edges at right angles and can only weld on one side then the leg length Of the weld cannot exceed the thickness of Of the thinnest section.In fact in the above situation the best you can do As make the leg length the thickness of the plate to be welded as I previously stated.
The people that frequent the site are supposedly professional engineers and Therefore should have some knowledge of engineering so I don’t make assumptions about what they know and don’t know, bear in mind the OP has been a member for four years. If the OP was misapplying it he wouldn’t be asking would he.

Explain to me how with two plate edges say 10mm thick at ninety degrees to each other you can produce a leg length of weld larger than 10mm.

“Do not worry about your problems with mathematics, I assure you mine are far greater.” Albert Einstein

 

[jayrod12]

I think what dauwerda is trying to say is that, for example, if you had two 1/4" thick plates at 90 degrees to each other, there's nothing stopping you from laying a 3/8" thick fillet weld connecting them. But that's uneconomical and wasteful in most instances since the weld strength will be far in excess of the base metal thickness. If you are trying to weld too close to the end of the plate, or to the edge of the plate, then you are obviously bound by the plate thickness (or the distance to the edge of your plate) when determining the weld leg length.