8" Thick A36 Plate 6

[BMart006]

I have a draawing where the designer has specified 8"-thick A36 plate in their design. The plate will unfortunately be seeing combined tension, bending and torsion. Aside from the reduced yield strength, are there any other concerns I should be aware that may be information known better by experience? Any literature on the matter is appreciated.

 

[KootK]

- I might call around and see if 8" thick plate is locally available. That's pretty thick and I've seen some fabricators top out at 6" or less.

- This part stands to be mighty heavy which might warrant consideration. Will it be positioned by humans, crane, etc..

- On several occasion, I've seen thick plate replace by multiple, groove welded, thinner plats for practical reasons. For bending and tension, you could probably orient the plies to make that work. Torsion should be a non-issues as you can just use the outer plies as a box section of sorts.



I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[ukbridge]

I'm not sure what the designation/subgrade A36 means, being from the UK.

With that being said thats a very thick plate (200mm if I've done my sums right....) - I'd want to check it for brittle fracture especially if the element is in tension.

In the UK its relatively simple - you calculate a nominal Temperature based on the actual temperature of the plate, Charpy test temperature, applied stress level etc. You then use a simple look up table to determine the maximum permissible thickness of said plate. Here's a typical extract, while acknowledging that the process is likely different in other countries.

The table is quite conservative for elements not subject to any type of fatigue loading, with that being said 200mm is outside the limit everywhere (in the UK you're restricted to the 0.75fy(t) column i.e. 75% of the yield stress).

I think it warrants a closer look especially if its in tension.

gl

 

[BMart006]

KooktK, thanks for your input. I had not considered a built up section with groove welds. Unfortunately in this case I am not contractually permitted to recommend design changes, only check for "adequacy." For that reason, I'm trying to make sure there are no issues (assuming the plate can be acquired) in the common design assumptions (e.g. isotropic material). For example, I have heard that there can be issues with internal delamination on thick plates, but I have not found this written anywhere.

 

[WARose]

With a plate that thick.....keep an eye out for lamellar tearing if you are welding to the surface and there are out-of-plane loads to the plate.

 

[SwinnyGG]

I'm not sure what the designation/subgrade A36 means, being from the UK.

A36 is an extremely common low-carbon structural steel, heavily used in the US. It is standardized on mechanical properties, as opposed to chemistry. The '36' references minimum yield strength of 36,000 psi (~250 MPa).

S235J2 is the rough equivalent EU designation, if I'm remembering correctly.

As far as fabricating parts from 8" plate.. in my local area, I can easily buy A36 in thickness up to 14".

In my experience, the tricky part about fabricating sections like this isn't buying the raw material, it's finding a local fabricator who is A) capable and B) willing to work in such large sections. Finishing a part from 8" plate is much different than handling HSS/WXX sections or building weldments out of 1/2" plate or whatever.

There's also the fact that a section on this scale takes HUGE loads to flex. This can have a huge effect on implementation. For example, if I have a column base that's a 1/2" flange welded to HSS, and that flange isn't laser-perfect-flat after fabrication, well the fasteners will have enough strength to pull that flange into good bearing contact with whatever its mounted to, because the stiffness of the flange is low relative to the loads the fasteners are capable of applying, and the section is thin so the internal stress generated by a little bending are relatively low.

If my flange is 8" thick, whether or not it will be assembled with good bearing contact on its mounting surface is another matter altogether, and even if I have sufficient fastener strength to pull that flange flat, doing so may induce massive internal stress, enough to generate internal failures of the base material, in such a thick section.

I'm getting wordy here but the point is that when it comes to fabrication, moving from 'conventional' section thicknesses (<= 2" or so) to large sections (>4" or more) is a non-trivial endeavor.

 

[ukbridge]

@JGRKI Thanks for the interesting post.

In the absence of not knowing the details of the project I think its worth having a chat. 8"/200mm is very thick - not only practically from the perspective of a fabricator, but in terms of brittle fracture or loss of toughness.

For things like taper plates for bearings we rarely go significantly over 100mm which are in compression (recognising that the UK has a cooler climate relative to North America).

I'd also agree about laminar tearing being a consideration.

 

[SAIL3]

a sketch would help us understand why the 8" pl and possibility of using other options....

 

[hokie66]

Another caution about checking for lamellar defects in the thick plate.

 

[RickyTickyTavi]

Can you provide a picture of this 8" thick plate whenever its manufactured? I'm super curious to see that beast of a plate. Dont think ive ever needed anything near that level of thickness.

 

[SocklessJ]

Be very careful if welding onto plate this thick. Preheat and interpass temp will be critical, since the plate will quickly suck any heat out of the joint. You might need something like induction heaters to get the whole part hot enough, rather than just hitting the joint with a torch.

If someone else is doing the welding and you don't want to direct them how to do their job, just specify some NDE (MT or preferably UT) to make sure it didn't crack.

Also, one more vote for the lam scan.

 

[BMart006]

Attached is a rough sketch of what the situation looks like. As I said before, I unfortunately cannot recommend design changes, I can only point to issues with the contractor's proposal.

 

[BMart006]

RickyTickyTavi, I wish that I could. But security would not allow me to.

 

[jayrod12]

Geez, what's the applied load? That beam is going to weigh 4500 lbs on it's own. That seems a bit ludicrous.

 

[KootK]

Do you have force magnitudes? If that cross section were actually needed for the forces involved, there's really no way that you'd get the support welds to work. It's not like you're gonna full pen the thing. Do you what motivation lead to this being the preferred solution?

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[ukbridge]

Looks absolutely bonkers to me, without knowing anything else.

Thanks for the interesting posts.

 

[BridgeSmith]

I agree with KootK. Either this cantilever beam (it's not a plate - it's a beam) is far bigger than it needs to be, or it's going to be impossible to attach it adequately.

 

[Brad805]

I was waiting to see something unique, and now a bit let down. Seems to be poor engineering going on, and like you say earlier, that part is beyond your control.

 

[TLHS]

Be very careful if welding onto plate this thick. Preheat and interpass temp will be critical, since the plate will quickly suck any heat out of the joint. You might need something like induction heaters to get the whole part hot enough, rather than just hitting the joint with a torch.

If someone else is doing the welding and you don't want to direct them how to do their job, just specify some NDE (MT or preferably UT) to make sure it didn't crack.

I agree strongly with this. I've had guys do welding of thick plates much thinner than this, and have had problems with tearing because preheat wasn't sufficient. Technically the welding codes require them to preheat sufficiently, but this kind of thickness and the amount of thermal energy needed to evenly heat something this size is not insignificant. A lot of contractors will just follow their normal welding methods, which won't be sufficient and you'll see problems.

I wouldn't really call it poor workmanship. It's outside of the normal realm of experience of the majority of contractors. Pointing it out as a potential problem (and avoiding if at all possible) would be helpful for everyone.

 

[rb1957]

either you can do checks to convince yourself the design is adequate or you can't (convince yourself the design is adequate).

If the latter it's either because the design is shown inadequate (by calculation) or because your experience/knowledge is inadequate (no slight intended) and you don't know what to check (because the design feature is so unusual).

Handling the plate is someone else's problem, though you could note your opinion of potential issues downstream.

another day in paradise, or is paradise one day closer ?