Are HSS members in bridge trusses a good idea? 2

[bridgeDisciple]

I'm hoping for some shared wisdom here. I'm designing a bridging system for an autonomous transit network (ATN) - light-weight modern form of APM. It is roughly in the class of pedestrian bridge with light vehicle capability, and similar in form to pony truss bridges with outboard walkways. The section is a box truss. "Cross ties" attached atop the pony cross truss extend horizontally "outboard" beyond the box, upon which the tracks are mounted. So torsional loading of the box section mid-span is a concern.

Steels, welds, and calculations for bridges are more extreme than for other structures because 1- they are exposed to weather, and 2- they are subject to extreme fatigue loads. But AASHTO and AISC guidance is a little weak on advisability of using HSS chords. The bridging is also extremely cost sensitive, so manufacturability is a key consideration.

I like the idea of HSS for aesthetics, torsion resistance, and buckling resistance. But those inner surfaces pose a long term potential hidden corrosion challenge. The ability to seal off the HSS ends to corrosion protect them seems in doubt, yet the standard docs suggest use is widespread enough that "guidance is coming".

So if any of you old bridge hands would like to steer me in the right direction, i'd appreciate it. Should i stick to angles, WT, or small beams ? Or is an HSS chorded box truss a reasonable choice ?

 

[BridgeSmith]

I think you hit on the main reason HSS sections are oftentimes avoided in bridges - hidden corrosion.

Generally, torsional resistance is not a high priority in truss bridges. If the smaller members where you could use HSS sections for your truss members need to resist alot of torsion, you've probably connected some members in way you shouldn't have, or you're missing some bracing members that need to be in the structure. The main box girder should be resisting the torsion; the other members should be braced. Torsion in members subject to repetitive loading, and therefore fatigue is not a good combination, especially if the members can't be inspected inside and out.

Rod Smith, P.E., The artist formerly known as HotRod10

 

[Settingsun]

Hidden corrosion has been a massive issue for concrete structures but for some reason it's seen as a much bigger problem for steel hollow sections.

Corrosion allowance and proper sealing will get the job done. Avoid 3mm plate with 3mm weld (only 2mm throat). 8mm plate with 6mm weld is good sealing.

 

[hokie66]

In addition to corrosion, inspection of welded connections would be my main concern with using HSS members.

 

[jorton]

HSS does not come in a steel grade commonly used in transportation projects, but I think there is now a grade that meets the same toughness criteria (Charpy V-notch testing, etc.). Also, the bridge welding code, AWS D1.5, specifically excludes HSS. Much of our fatigue and fracture control strategy is built into the material and welding specs, so I would be hesitant to operate outside them.

I worked on a similar structure carrying heavy rail transit trains. It is a straight, steel tub girder with cantilevered deck such that each track was eccentric to the tub. My project was the 3rd attempt to arrest and repair the cracks. As far as I know, the repair and retrofit have held for a decade now. I'd pay close attention to detailing to prevent distortion induced fatigue cracks.

 

[WinelandV]

I design a fair amount of walk-through truss bridges similar to your photo, but for conveyors and maintenance access. We use HSS members for the chords and side diagonals/verts, connecting them with welds (it's a gapped K connection). We use angles for the top and bottom diagonals, along with small beam sections to make the struts - but these all bolt together.

With that out of the way, some thoughts:

- I don't think you have any torsion that isn't going to be resisted via the big box truss that you have. That is, no individual member torsion, just a point load couple that your chords will have to resist in bending (which could make for a pretty beefy HSS section if you use those for your chords).​

- Yes, there is a possible hidden corrosion aspect to HSS, but the places I have worked have all dealt with that by making the tubes Hot Dipped Galvanized.

-- That in turn means your segments are limited to ~ 40 long b/c they need to fit into the tank. So if you go this route, you need to make sure that your section splices work in such a way that the splice is not interfering with your tram.​

 

[bridgeDisciple]

inspection of welded connections would be my main concern with using HSS members.

Why is HSS weld inspection necessarily a problem? All welds are on the outside surfaces. Why isn't that equivalent to the situation where non-hollow shapes are used?

 

[bridgeDisciple]

HSS does not come in a steel grade commonly used in transportation projects, but I think there is now a grade that meets the same toughness criteria (Charpy V-notch testing, etc.)

I think you are talking about ASTM A1085. And those materials are now supplied.
eg,

https://www.atlastube.com/products/structural-steel/astm-a1085/

Yes AASHTO/AWS D1.5 excludes it. But then AASHTO LRFD BDS now includes A1085. So which is it - in or out ? AISC's Steel Bridge Design Book Chapter 1, section 2.2 last paragraph encapsulates this ambiguous guidance. The last sentence of that paragraph is
"Provisions for the inclusion of tubular materials in the Code is underway and is anticipated for a future edition of the Code"

Thats why i'm here asking, and i appreciate the thoughts you shared. Your advice for hesitancy is noted and valued.

 

[hokie66]

bridgeDisciple,

I was talking about full penetration welds. You can't get to both sides. One solution is to use backing bars inside the tubes, but you then cannot remove the backing bars.

 

[jorton]

Understood. I'm actually much more open to trying new things than I sound in my post - at least from a technical perspective. After being involved in a few claims and legal disputes, my cautiousness has grown on the specs side of things. Figuring out how to get a contractor to deliver exactly what you want when working outside common specs is tough, especially seeing how things can be argued even on issues that are clearly inside the common specs.

The AISC guides are good. I think they are usually a step ahead of the code. It takes a while for it to all filter down to common practice.

 

[bridgeDisciple]

@jorton

how to get a contractor to deliver exactly what you want when working outside common specs

Excellent point! That may well be what tips me away from HSS. As i stated in my query, this design is highly cost sensitive (miles and miles of bridging).

While temped by seeing examples like that attached (which uses HSS round for the bottom chords and web), i have to recognize i won't have the internal resources and buying power of a well established player like Doppelmayr (nor are we at their same price point). Economics may drive us to bring the manufacturing in house, but its hard to see ever being completely freed from at least partial reliance upon contractors (early market and ongoing demand management).

 

[jorton]

It sounds like you have a fun project. I was picturing Oakland when reading your post. One of the advantages of that design is there is no deck, so there is less dead load to carry. I wonder if that sort of system is more cost effective than one with large cantilevered decks carrying eccentric tracks. Have you seen the Orlando (or maybe Tampa - I don't recall) airport people mover structures? Basically there is a steel beam line under each wheel. Each beam is topped with a concrete plinth. The beams are individually cambered to provide longitudinal and transverse grades, so there is also a lot less plinth dead load to carry. Not nearly as cool looking though...

 

[bridgeDisciple]

Here is another for you to picture - second half of (PRT Quick look [2 min]

https://youtu.be/lfMzERCydBk).

I wonder if that sort of system is more cost effective than one with large cantilevered decks carrying eccentric tracks

Our calculations (at least first pass) indicate cantilevered is better. First, who said anything about "decks"? Not needed, not present. Horizontal cross beams from the center box extend outboard. Track rails (angles) are mounted atop those cross beams. The pod bogies ride on the rails. So it is much as you described, "steel beam line under each wheel". The attached section drawing is the basic idea. We have narrowed in on this hybrid supported+suspended configuration to minimize wheel (and therefore guideway) loading. We're now doing a deeper pass looking at the cost impacts of fatigue design and bridge steels, etc.

I'm always looking at APMs (and roller coasters) - trying to learn. Most APMs are built like highway bridges - using steel girders and concrete. APM vehicles are soooooo heavy this is reasonable. Our ATN system has consciously made choices to minimize the bridge live loading to about 4200 lbs, to go after costs (if you are interested, Bridge Costs (31 min)

https://vimeo.com/694274246).

So roughly cost becomes proportional to the number of truss chords. If you put a truss underneath each track in a bi-directional run, that doubles the number of chords - doubles material costs. Plus putting the truss under the track adds to the vertical height of the guideway envelope. Putting it between the bi-directional tracks offers a much shorter profile (and not much wider). This has alignment advantages.

Two choices that distinguishes us from other Personal Rapid Transit (PRT) and APM providers is use of bi-directional guideways (rare for PRT), and orthogonal crossings (rare for PRT and APM). We think these are needed for a better fit to American cities. So guideway envelope height is a premium for us.

 

[bridgeDisciple]

Now that i've told you how smart our selection of this box-eccentric design is, i think i should probably open my ears and learn.

I'd pay close attention to detailing to prevent distortion induced fatigue cracks.

Where exactly were these cracks occurring, and why? How did you fix them?

 

[bridgeDisciple]

@winelandv
I assume the reason you've chosen side attachment of the web is because your web is angle and bolts, and you can tolerate the eccentric loading?

That is, no individual member torsion, just a point load couple that your chords will have to resist in bending (which could make for a pretty beefy HSS section if you use those for your chords).

I'm not sure i understand your warning. The pod loads will be applied through the bogie and wheels to the track rails. The rails transfer those loads through the cross-beams to the center box truss. These cross-beams are located only at truss panel points. So if there is no matching pod travelling in the opposite direction, this will represent an outboard load on only one side of the box, at that panel of the box. The truss members, on the box vertical side where the pod is, will resist deflection using tension and compression. The truss members on the opposite vertical side will see no such loading.

I see this differential loading trying to twist one longitudinal end of the box vs the other. I would like to avoid a diagonal member through the box (between adjacent corners) to resist this torsion via tension/compression, because this box interior would make a convenient evacuation path for riders in the case of breakdown. But currently we don't think we have that luxury, and so the box has that member.

So where do you see a "chord in bending"?

 

[WinelandV]

I wasn't sure what your setup was going to be - if your pod was going to ride on top of your tube member it would have been a concern. With the way you're going about it, you won't have to worry about that.

 

[OSUCivlEng]

Many pedestrian bridges are constructed out of HSS.

 

[bridgebuster]

Many vehicular bridges have been constructed from welded boxes over the last 50 years.