Welding MC8x20 Channel to Inside of 24"x0.5" Pipe

[question_asker]

Hello. I am looking for guidance/suggestions on a tricky connection for a temporary structure I am designing.

I have attached a sketch of the proposed detail. On this structure, steel pipe pile (24" x 0.5" wall) will be supported by a HP14x73 pile located inside the pipe. The HP14x73 has a stocky angle welded to each flange as a bracket/bearing to "catch" the web of the MC8x20 channel, which is to be welded to the interior of the 24" pipe (in shop). Lateral support/blocking provided elsewhere and not shown. Only concerned about the vertical loads at this connection.

I am concerned about the welding of the MC8x20 to the inside of the 24" pipe. Using AISC SCM 14th Edition, Table 8-4, I determined the available group weld strength to be 34.5 kips. Calculations included with the attached sketch. The required load = 25 k, therefore 25k / 34.5k = 73% utilization, Weld OK. (Everything checked using service loads / ASD)

Interested to know your opinion on the following:

1. Any additional consideration required due to the acute angle of this fillet weld? The dihedral angle is approximately 67 degrees. From my review of AWS D1.1 and the AISC SCM, I did not see any reduction I needed to apply for dihedral angle between 60 and 80 degrees, it is only reduced when the acute angle drops below 60 degrees. Perhaps I need to add a clearer detail to indicate what the exact required effective throat is, although 5/16" legs with an acute angle would exceed the effective throat of a typical 90 degree tee joint and would be conservative.

2. Would you consider adding additional weld to the inside of the channel leg? My concerns about this were that it would be difficult to actually perform that weld, the weld itself would be complicated geometry (something like a flare and fillet weld), and there is a chance that in trying to strengthen the connection it is inadvertently weakened due to the heat output and difficulties mentioned.

Thank you for your input.

 

[dvd]

Can you cut slots in the pipe and weld legs from o.d.?

 

[TLHS]

I'd consider maybe beveling the toes and doing a groove weld. That's awkward in its own fun way though. If you're going to do an awkward fillet weld, I generally try to draw the geometry of it and dimension it to make it clear what I want. In theory the welding specs tell you how to communicate this stuff, but I've had people interpret things all sorts of different ways.

Another option is to do a wide flange stub or a WT stub where you cut the flange(s) to the curvature of the pipe. The web is basically perpendicular to the pipe so it's a pretty reasonable weld, and you can weld the flanges from the one accessible side

 

[JStephen]

If that channel was right at the end of the pipe, welding inside might not be that hard. If well up in the pipe, access is harder.
One problem I run into is analyzing line loads on a cylinder (typical tank issue). Bednar's Pressure Vessel Handbook has a design approximation, but resulting loads are surprisingly low.

 

[question_asker]

Thank you all for your responses.

@dvd: That is an interesting thought. I will discuss with the fabricator and see if this is feasible/preferable. My biggest concern would be how precise the cutting of the slots would be. There is a reasonable chance they would be torch cut given the application.

@TLHS: Good points. I am considering beveling the toes of the channel regardless so that the inside of the pipe is flush with the full width of the channel flanges, rather than the necked down section at the toes. I am going to include a zoomed in detail showing the weld geometry, minimum effective throat, etc.

@JStephen: The channel is near the end of the pipe (bottom of channel is approximately 12" from top of pipe, so the weld access will be OK. The geometry of the channel will not allow for much of a weld to the inside of the legs, however.

Thanks again for your feedback.

 

[azcats]

I like the idea of beveling the channel flange so you're not landing a weld at just the outside corner of the leg. Seems like a better chance for good penetration and no blow/melt thru.

 

[LittleInch]

I would bevel it some more soyou actually have a gap to weld in.

Grind a 45 degree angle face on the end so the outside is proud of the inner pipe surface.

There's no way you're going to weld on the inner face.

What's that gray outline of a wide flange beam in the middle doing?

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[question_asker]

Nice drawing @azcats. That is similar to what I was envisioning, except I am going to bevel all the way from the radius return (where the flange thickness begins changing linearly).

@LittleInch, see image below. On the left is the bevel and fillet weld I am proposing to be used on the outside of the channel (both left and right legs). However in this image I have drawn what I think you are proposing. Am I interpreting your suggestion correctly?

 

[question_asker]

@LittleInch, the grayed HP14x73 is a driven pile that will support this structure. The vertical load is transferred from the structure via the channel, which will be resting on a bracket welded to the HP14x73. This is a steel template for marine application for driving cellular sheet piles as well as interior HP14x73s - the design and sequence of the rest of this structure is a convoluted and beyond the scope of this post - just focused on the weld details for now.

 

[dvd]

A couple of options using bar

1 - cut slots and weld outside

2 - weld inside with weld prep as desired