AS 5100.6 maximum fillet weld vs AS 4100 1

[Settingsun]

Hi,
In AS 5100.6 (steel bridges), the authors have sneaked in a maximum size of fillet weld under the heading Minimum Size of Fillet Welds (clause 12.6.7.2), with the maximum size equal to the size of the thinner piece of steel:
"The minimum size of a fillet weld, other than a fillet weld used to reinforce a butt weld,
shall conform to Table 12.6.7.2 except that the size of the weld shall be not greater than the thickness of the thinner part joined."

In AS 4100 (steel structures except bridges), the wording is virtually identical except 'shall be not greater than' is 'need not exceed', and therefore does not impose a maximum size:
"The minimum size of a fillet weld, other than a fillet weld used to reinforce a butt weld,
shall conform with Table 9.7.3.2, except that the size of the weld need not exceed the
thickness of the thinner part joined."

Is there method behind the bridge code, or is the intent the same as AS 4100 but poorly worded? Seems to me the effect is to force butt welds instead of large fillets. According to this webpage, the maximum fillet size is a common misconception:

https://learnwithseu.com/welding-myths-debunked-maximum-fillet-weld-thickness/

 

[Settingsun]

I asked Standards Australia and got the answer I expected: they don't respond to this type of query. They recommended I contact one of the organisations on the standard committee directly. Since it's a joint AS/NZS standard, I've half a mind to try Standards NZ as well just to see if they have the same position.

I guess I just have to hold on to this for ten years and raise it when the next draft for comment is released.

 

[IDS]

steveh49 - why not do what they suggested? The standards organisations are just administrative, they can't be expected to provide expert commentary.

Doug Jenkins
Interactive Design Services

http://newtonexcelbach.wordpress.com/

 

[rapt]

I would have said that 4100 is poorly worded and that 5100 is trying to improve that wording. "Need not exceed" sounds like stupid wording here. It is completely unnecessary in this context so why put it in.

From memory in my student days (and that was a long time ago) we were taught that a fillet weld should not be thicker than the material that was being welded, so 5100 would appear to agree with that.

 

[Ingenuity]

According to this webpage, the maximum fillet size is a common misconception:

I am 'attending' the AISC's (US steel organisation not ASI) 3-part webinar entitled: WELDED CONNECTIONS - A THREE-PART WEBINAR SERIES Link and it is presented by Duane Miller (same author referenced in your web 'welding myths' article).

Here is what AWS states for MAX fillet weld size:

Effect of edges corner 'melted' away resulting in less than t_w which is the reason why AWS has the 1/16" reduction:

And MAX fillet weld sizes do not apply to surfaces, only edges:

butt weld

Miller's 'pet peeve' is calling CJP (complete joint penetration groove welds) "butt welds". A "butt" is a joint type NOT a weld type.

 

[Settingsun]

Hi IDS, I might take it up with the Aust Steel Institute if my company is a member.

Rapt, I don't agree that AS 4100 is poorly worded. It doesn't talk about maximum weld size under the minimum weld size heading for one thing. I really think that AS 5100 should put maximum weld size under its own heading (if a maximum size is indeed the intention); and not just copy the AS 4100 commentary if the requirements are not the same.

I wanted to check the 1996 bridge code (HB77, before is became AS 5100) but I don't have a copy handy - thought I did. I do have the 1976 NAASRA code and its minimum weld size requirement is like AS 4100, with no apparent maximum weld size as in AS 5100.

 

[IDS]

Steve - the ASI should be happy to give you contact details for their representative on the committee, even if your company is not a member (although they'll probably take the opportunity to give you a sales pitch). The commentary to the new code is under preparation, and it sounds like some background on this would be a good idea.

I have an even longer history of not designing welded structures than rapt by the way, so I won't comment on whether the requirement is valid or not.

Doug Jenkins
Interactive Design Services

http://newtonexcelbach.wordpress.com/

 

[rapt]

Steve,

I thought I made it plain that my comment on 4100 wording applied to "Need not exceed".

Agreed the Heading should have been changed to something like "Weld Size Limits" when a maximum limit was included. Something you could helpfully suggest to Standards for their next Amendment.

 

[Settingsun]

Hi rapt,
What's the issue with need not exceed? That's ok/clear to me so I'm interested in how else it can be understood or found ambiguous.

(It's also the same as the 1976 bridge code and Blodgett's welded structures book from the 1960s, so I'm even less of a fan of the unexplained AS 5100 wording since it seems a sneaky change to decades-old lore.)

 

[Retrograde]

As far as I am concerned "need not exceed" means should not exceed except in exceptional circumstances.

 

[IDS]

so I'm even less of a fan of the unexplained AS 5100 wording since it seems a sneaky change to decades-old lore

I don't know about sneaky. It was in the draft for public comment in 2016.

Doug Jenkins
Interactive Design Services

http://newtonexcelbach.wordpress.com/

 

[Settingsun]

Hi Retrograde, why would you limit a large fillet to exceptional circumstances? (See below)

IDS, the wording was the same in AS 5100:2004 but I didn't mean no opportunity for comment when I said sneaky. AS 5100 doesn't exist in a vacuum: it's based on AS 4100 which is probably more widely used. I think minor wording changes could easily be overlooked especially when under the wrong heading, and especially when the wording that's been changed pre-dates AS 5100 by 40+ years so is 'familiar' to the engineering community.

More reading and I've found the document which gave me the impression (years ago) that large fillets are OK. Australian AISC's hollow sections design guide (1996). It talks about throat thickness up to 1.6 times wall thickness, so leg length of 2.2 times wall thickness. A CIDECT design guide (European, also for hollow sections) has minimum throat thickness about equal to wall thickness, so leg length 1.4 times wall thickness minimum.

I really should get in touch with ASI now. Maybe it's to do with fatigue (commentary would be good if so!) I'm still not especially convinced that the maximum weld size is intentional. It could have been handled much better if intended, and the first part of the next clause becomes redundant as currently worded.

------
Hi Ingenuity, I overlooked that your pictures come from a current webinar. I thought that they came from the webpage I linked to. Thanks for providing them. I guess I agree about the 'butt weld' nomenclature but it's the official term in the Australian standards so must be used.

 

[rapt]

steveh49

A rule on "maximum" weld thickness requiring "Need not exceed" is meaningless.

It is basically saying it does not have to exceed, but can if you want too without any limit.

So why say anything?

Standards editors would not normally accept that terminology. Obviously that one slipped through the checking process.

 

[Settingsun]

Hi rapt,
The 'need not exceed' does have a purpose. The minimum weld size is first specified in terms of the thicker part, then the 'need not exceed' provides a relaxation in the case that the thinner part is a lot thinner. In AS 5100, the relaxation becomes a mandatory requirement. One effect of the AS 5100 rule is that the full capacity hollow sections cannot be achieved with fillet welds.

 

[Ingenuity]

I have not practiced in AU for several decades - are fillet welds specified by weld 'leg size' or 'throat size'? I think it is 'throat'.

One effect of the AS 5100 rule is that the full capacity hollow sections cannot be achieved with fillet welds.

You mean in a HS to base plate connection? I know this ref is NOT AS 5100, but from the US AISC's publication entitled "[italic]Design Guide 21: Welded Connections--A Primer for Engineers[/italic]" Link it states this, re hollow sections:

 

[Settingsun]

Hi Ingenuity, it's leg length also in Australia. Your extract is exactly the issue and it also applies to other situations such as web-to-chord connections in trusses.

As background, the AS 5100 wording was brought to my attention by a bridge engineer verifying one of my designs. I'd used a 6mm fillet weld on a 5mm hollow section. The easy answer to that was the design is subject to AS 4100 and not AS 5100 but I did have to read AS 5100 a second time when he showed it to me since the relevant clause seemed at first glance to be the same as AS 4100 (hence my objection to the sneakiness of it).