On pages 42 and 43 of the February 09 issue of Trains is a photograph of the BNSF bridge across Lake Oahe in SD. The bridge consists of three span types…a plate deck, and a couple of through trusses (Pegram trusses by the looks of it) separated by a deck truss. My question is why are three span types required? i.e. why not a simple deck plate construction straight across?
I’m guessing that it has something to do with the piers not being equidistant, and that the type of construction is dictated by the span length. But even so…wouldn’t three deck trusses be better than two through trusses with a deck truss in the middle? It sure looks like a lot of complexity for a simple bridge across a shallow lake.
I wondered about that, too. My first guess - since other than this photo, I don’t have any actual knowledge of this situation - is that the bridge had to preserve 2 navigable channels with a certain minimum clearance above the water for sailboats and the like where the through spans are, hence the through spans vs. the deck types. Such channels - with a certain minimum width as well - may have also dictated the pier placement and hence the span distances, which would have also affected the choice of span. What’s hard to tell from the photo is the relative lengths of each span - are the through trusses long enough to be justified by their lengths alone, or are they short enough that an underclearance requirement may have driven their use ?
Other than that, the choice of type is largely affected by the span, and economics. A very rough rule of thumb is that the depth of the structure is economical when it is about 1/10 of the span’s length - although it’s not uncommon to see that ratio vary from around 1/8th to 1/12th or so, you won’t see one with a 1:20 ratio, or a 1:4 ratio. Plate girders are good for up to about 80 ft. or so, for an 8 ft. depth - I’ve seen some in the 120 ft. range/ 12 ft. depth, though. Beyond that, a truss starts to be more economical, and a deck truss is generally preferable to a through truss if there are not underclearance issues, plus the deck truss doesn’t constrain high/wide loads and is way less susceptible to being destroyed by a shifted load or derailment (see the UP RR bridge thread), plus other reasons. That’s because the deck truss’s design depth can be tailored to fit the specific span situation, whereas the through truss has to have a minimum of 22 to 25+ ft. clearance from the top of rail to the underside of the top cross-members for trains - old boxcar w/ trainman on top in the old days, double-stack now, etc. That requirement results in a practical minim
Good observations Paul…you must be an engineer. I don’t think tall sailboats are necessarily a consideration in SD though although I could be wrong… I find it interesting that the bridge is relatively recently built as I thought through truss bridge construction ended in the 30s.
If not sailboats, then maybe a cabin cruiser with antennas and masts, or a workboat of some kind.
Plus, does the lake level fluctuate at all ? It would be better if the bottom chords of a deck truss didn’t get inundated pretty often.
Through truss bridges are still constructed when circumstances say they’re best suited for the task at hand. Four or 5 years ago PennDOT erected a nice new silver-painted long 2-span through truss for the former PRR Trenton Cut-Off line where it crosses over a widened I-76 / Schuylkill Expressway, just south of the huge King of Prussia shopping center, which is also just south & east of the interchanges for U.S. Route 202 with I-76, the Pennsylvania Turnpike’s (I-76 to the west, I-276 to the east) Valley Forge/ King of Prussia interchange, and U.S. Route 422 - this is about 15 miles north of Philadelphia, and 5 miles west of Norristown. I’ll see if I can find a link for some photos of it. The existing elevations or the rail line and the road were essentially fixed - couldn’t raise them too much in either direction - so replacing the previous through 2-span plate girder with a longer 2-span through truss was the designer’s choice, and makes sense in this context.
I can’t think of any others right off-hand, but that sure doesn’t mean there aren’t any - maybe someone else can provide an another example.
Here’s a link, from the website of the bridge fabricator - High Steel Structures, of Lancaster, PA - includes an interesting set of 4 photos of rolling the bridge into place:
As you’ll see, it’s 2 spans, each 240 ft. long, 42 ft. high (doesn’t say if that includes the walkway handrails, but I’m prepared to accept that it’s the main bridge members only), 1,700+ tons of steel total, built 2002 (form another website). Depth-to-length ratio is just under 1:6, but the length - 240 ft. is right at where I said the lower limit for a through truss would be, so it does look a little “stubby” to me. But that was no doubt in response to the site conditions and constraints - had to use a through truss for sufficient clearance over the roadway, and it was too long for a plate girder, but at the lower limit for a through truss, so what else are you going to do ?
Not to hijack the thread, but I was more impressed with the high water mark on the abutments(sp). Looks like that water can get high at times. Maybe another 20-30 feet
The “UP RR bridge thread” mentioned above is titled "UP Derailment & Bridge Collapse on Overland Route" started by "MJChittick" on 12-28-2008 at 4:20 PM. Here’s the link to it - presently it’s at the top of “Page 5 of 1240” on this Forum:
Ok, I had some time to look into this a little further, and here’s how I now think it goes:
Bridge was built sometime between 1948 and 1962. So these trusses are fairly modern;
It took a long time to build this dam - like 11 years - and then 3 years for it to fill. Lake Oahe is the 4th largest man-made lake in the U.S. - though I had never heard of it before;
Corps of Engineers boating map (Sheet 17 ?) shows 2 channels as I suspected, although the eastern one is shown more towards the Mobridge (eastern) side than the easterly through truss’s location would seem to indicate;
I’m now pretty sure that this lake is part of the Upper Missouri’s navigable waterway for towboats, barges and the like - the Corps map has a lot of “River Mile” references on it in the 1,190 range, although I could be mistaken;
Accordingly, the Corps built 2 comparatively wide channels and hence spans that could get over them without obstructions. That consideration governed and hence led to a truss design of some sort being needed there;
Deck truss would not work for 2 reasons: a) Hanging down would obstruct channel and defeat the
I don’t think the Missouri River above Sioux City, Iowa is considered “navigable”, at least as to towboats and barges. As far as I know, none of the dams on this river have locks that would allow any boat to proceed upstream. But I could be wrong, too.
I think I remember seeing a few photos of this bridge in the original Journals of Lewis and Clark. They were really miffed about having to drag their boats around the spillway of the dam. The also commented on the curious construction of the Milwaukee Road bridge, albeit they had mistakenly thought that it was a C&NW bridge. Not surprising for a couple of fellas who spelled mosquito 23 different ways in their journals.
A few supplementary comments. In modern times CPR has used plate girders in the 140-150 foot range. The higher quality materials available today make these greater lengths quite feasible. They are a lot simpler to construct. Just comparing the DPGs in the pictures to the length of the SD40-2 (68 feet) indicates that those spans are quite lengthy too, perhaps around 130 feet. The depth can also be compared to the height of the locomotive, and the ratio looks to be right in the ballpark.
Location of piers is influenced by ground conditions. At this site the subsurface probably does not change greatly, but I imagine there was a free-running river prior to the dam, likely two channels with an island. It is very much easier to build foundations on dry land, and usually the pier will not need to be as tall. And if navigation was a future possibility a large unimpeded passage was required.
I think a deck truss would continue to be more economical than a through truss, but as they get longer the depth required becomes too great for most sites. In the photographs under discussion there seems to be a waterline on the piers, and a deck truss of the same length as the two through trusses might have the bottom chord underwater during a flood. Not a safe situation!
For what it’s worth: no barges or towboats go this far north. The only boat traffic is recreational, and fishing is BIG in that area. A good guess, is that the bridge was designed to get big fishing boats through there.
Is the bridge mentioned above, that originated this thread, the bridge at Mobridge, SD. Clever name for a town adjacent to a Bridge on the MO river. That didn’t click w/ me right away until my father in law pointed out that Mobridge was right on the river.
About an hr downstream of Pierre is another interesting combination bridge. This one is the Milw Rd bridge at Chamberlain. (See photos below.) The 2nd link below is kinda cool because you can pan left / right to see the whole bridge. It look like similar components to the Mobrige bridge, but different arrangement.