Top Flange Analysis at Hanger Points 1

[bsmet95]

My company is a manufacturer of patented track monorail systems. Typically our track is bolted or flush-clamped directly to building steel. Occasionally the track is suspended by means of a rod with a hanger fitting and a proprietary gimbal, to prevent bending stress in the rod. See the attached.

The fitting shown has four bolts, and is sometimes used to straddle a splice connection, resulting in only two bolts on the end of one beam. If the load is great enough, we use a stiffener on the underside of the flange.

I've been asked to determine how large a load can be suspended before the stiffener is required. This seems to be similar to the prying force calc's. as per AISC Manual of Steel Construction, 9th Ed., Sect. 4, pg's. 4-89 thru 4-95.

Any help is appreciated.

 

[ishvaaag]

As a result of there being some potential inclination of the rod, some notional horizontal force needs to be assumed at the gimball.

Since dynamical forces, fatigue concerns such total number of applications and range of variation of the loadings, kind of detail, limit stress of concern etc, need be introduced.

Since close to one end, distances to the same need be normalized in order to avoid excesses from the viewpoint you are analyzing.

Since potentially webs are welded as per your detail, the actual configuration of the web-flanges union need to be cared for.

All these factors may be combined to form a set of tables for allowable or limit loading values for the production set.

 

[BAretired]

The yield line method should give a reasonably good estimate of the failure load. Then you must apply appropriate safety factors.

BA

 

[dhengr]

Bsmet95:
It seems to me you might be trying to over analyze this problem, in the wrong way. Without a side and top view of your detail, and some loads and dimensions it’s a bit difficult to comment in detail. Draw us a FBD of this detail, in several views. But, I have some thoughts and questions to get your brain in gear.
1.) If prying and/or plate bending are the critical issue, then why isn’t the top hanger fitting plate (marked HANGER FITTING) the controlling plate, it appears to be the thinnest. How do you reinf. it for these conditions?
2.) It appears you have a couple end plates on your hanger fitting to set its depth for the gimbal and the rod nut; those plates might act on the bolts in much the same way your rail beam web does, w.r.t. plate bending or bolt prying.
3.) At each of the bolts, do you have a pipe spacer, same length as the height of the end plates in 2.) abv.? You need this to clamp the bolt/nut down onto or you won’t get prying on the bolt as RCSC or AISC, sec. 4, pg 4-89 sees that issue.
4.) The yield line theory might well be applied to the top hanger fitting plate, the thinner plate, thus; supported at the two end plates, free to bend at the other two (longer) sides, clamped by your four bolts, and loaded at the center by a 2"? washer at the hanger rod. That will produce some interesting yield lines.
5.) You worry about reinf’g. the top flg. of the rail beam at this hanger, at a rail beam splice, only two bolts into each beam at that splice, and might this be a problem, with a hanger load equal to the DL and lifted load on the trolley. That being the question at hand, how do you analyze and reinf. the bot. flg. on the rail beam when a four wheel trolley rolls over this splice joint with the same loads, and a point load at the tip of the flg., under each wheel?
6.) Which is the critical condition, and who cares about the rest?

I always have more questions for you, when you come to me, than answers, call it food for thought.

 

[ToadJones]

dhengr-
CMAA no. 74 has a very detailed and thorough procedure for analyzing local flange bending do to wheel loads from trolleys.
The calculation takes into account biaxial or eve triaxial bending stresses locally to to major axis being, shear, minor axis bending and local flange bending due to the wheel loads.

On patented tracks the bottom flange is usually much narrower and made of seemingly proprietary high strength material. The bottom flange is usually of a special shape that resembles an inverted "T" and is welded longitudinally to the T shaped shown here along the web.

Another good check that maybe inverted to fit your situation is this:

http://www.modernsteel.com/steelinterchange_details.php?id=499

very practical....rather simplistic.

 

[ToadJones]

bsmet95-

You guys ever give up your material properties on the bottom flanges?

You guys design your patented tracks with ECL calcs right?

 

[bsmet95]

I am quite familar with CMAA No. 74 as well as ANSI MH27.1. I have done flange bending calc's, due to wheel loads, but was not sure if the same analysis applied in this case.

Some of our track properties are published in our catalogs, and ECL calc's. are always done (I had a hand in making a spreadsheet to calculate them for our catalog).

Any other help?

 

[bsmet95]

Additional info attached.

 

[dhengr]

Bsmet95:
Now that ToadJ mentions it, I do remember the proprietary lower flanges/rails, but it’s been a long time since I’ve dealt with one. For all my rants on rushing to computers, I would use FEA to analyze the rail flange/web detail under the trolley wheel loadings when I was producing many of these per year. And, I would probably do the same with this hanger fitting, since you are making hundred of these too. I think I would try to design at this detail, so that I didn’t have to reinforce the rail top flange. Again, I have many more questions for you than I have answers.

1.) Are the bolts 3/4", what grade?
2.) I noticed the side pls., instead of the end pls., shortly after my first post. You will have a prying type affect when you tighten the bolts down on the side pls. It causes bending moment in the bolts approx. equal to (bolt pretension)(4.375"/2 - 1.75). But, I don’t imagine you pretension the bolts to a high degree. Then you’ll have it again, likely in the opposite direction when you load the detail.
3.) I would not make the slotted holes in the beam flg. any longer than needed, you are stealing from the flg. bending strength. Punching these holes is more fatigue critical (sensitive) than drilling these holes. And, you might even consider grinding a slight chamfer around the hole edge as a burr/crack clean-up.
4.) The web/flg. weld so near these flg. bending conditions could be a fatigue problem. I would grind or nibble bevels on the rail beam web, .5 or .75" long at the upper corner, at the splice, so I could run the double sided fillet welds into a full pen. weld and off onto a run-off tab. The tab can then be ground off, eliminating the exposed fillet roots at the splice line.
5.) I would use larger, thicker, washers, and some locktite on the threads. Maybe to the point of a drilled bar above the two bolt heads on each side, at the underside of the top rail beam flg. Although, I suspect this is what you are calling a stiffener, and hoping to avoid. You might do the same thing across the hanger fitting top pl. under each pair of nuts. In each instance the idea is to rationalize a shorter bending length and/or a stiffer bending member under the bolt. The drilled bar makes sense on the rail beam flg., but a thicker hanger fitting top pl. is probably less expensive than adding the bars, but still use larger washers.
6.) Do you see any wear or beating at the tips of the rail flgs. from the trolley wheels? The real problem may be more a matter of flexibility (deflection) of the whole system at a splice, and thus instantaneous differential rail height right at the flg. tip as the trolley moves over it.
7.) Does the gimbal allow for 360° movement, or only lateral movement?
8.) I still wonder if the top pl. on the hanger fitting might not be the critical element. It’s the thinnest pl. and it is also the narrowest beam element as it spans from side pl. to side pl. Again, draw us some FBD’s of this detail, in several views.
9.) On a detail this tight and riddled with holes and load points and reaction points, it is tough to do a simple analysis, as in simple beam bending and shear stresses, etc. It can truly be said that the loads and stresses will by multi-axial within each plate, and pretty complex around so many stress raisers. If you don’t have access to FEA, and you make hundreds of these per year, maybe you could go to your local college Prof., and for a few bucks, offer this problem to one of his/her students as a term paper.

 

[bsmet95]

to dhengr

Thanks for your input. I was sure it would not be a simple analysis. We do have FEA so I'll have to use it.

 

[dhengr]

Bsmet95:
But, the first step should always the simplest rational analysis you can do. I did not mean to suggest that you should not do this simple long hand analysis, to start to get a handle on your problem. That almost always adds considerable insight into the problem, and makes the modeling and the results from the FEA all the more meaningful and on point. You know your product better than I do, but you should think about my questions in your long hand analysis, and then consider some of these same issues in your FE modeling too; the mesh selection, element shapes and dimensions (2D or 3D?), you really have to define the plate shape, the holes and loads on this problem. You have span lengths and widths which are only several times greater than the pl. thickness.

For example: the two beam elements spanning across the hanger fitting top pl., ~1.5" wide, spanning 4.375"; supported (maybe some fixity) at the side pls., but there is the bolt hole discontinuity at this point; reacted by two bolts at 3.5" o/c; free on one edge and loaded torsionally by the other edge by the gimbal washer. Maybe that gimbal washer should be 4.5" long, getting that load into the two beams better, and reducing the stress concentration at the edge of the rod hole in the top pl. I can take a first conservative shot at this long hand, to start to hone in on the problem, but if I’m making hundreds of these and might have to defend my design at some future date, and/or am trying to shave dimes off the cost of the fitting, a more detailed analysis may be quite appropriate for the file. And, don’t forget the bending in the bolts,either.