Baltimore Bridge Collision and Collapse

[Peartree]

Not at all my field, but I would expect that a ship with a "50 ft. draft" means that it draws 50' when fully loaded and ballast tanks full. Not being fully loaded, or being in calm enough water (ie. not at sea) to use less ballast (temporarily) might allow the ship to draw less than 50'.

Am I wrong?

 

[boatgeek]

Not at all my field, but I would expect that a ship with a "50 ft. draft" means that it draws 50' when fully loaded and ballast tanks full. Not being fully loaded, or being in calm enough water (ie. not at sea) to use less ballast (temporarily) might allow the ship to draw less than 50'.

Am I wrong?

That's a good catch. Checking the registration data again, it has a 50' load line (maximum) draft. Since it was showing quite a bit of red bottom paint on departure, it was almost certainly running at least 5-10 feet lighter than that.

 

[OverTheTop]

Most strength in pillars/columns is in the wall. A hollow column is stronger than a solid column, because of how stress concentration, and load disperses around it.

I don't agree with this, regarding the hollow column being stronger than a solid. For a homogeneous construction a filled section will be stronger than the hollow, but not by much. A structural beam bends about its neutral axis. In a cylindrical beam that is the center of the circle. The material resists bending by the radius of the material CUBED. What this means is that material near the center does almost nothing for the beam stiffness, and material near the outer radius does most of the work. Making the member hollow reduces mass and cost of materials but the result is a something that is only slightly weaker than a solid beam.
https://en.wikipedia.org/wiki/Second_moment_of_area
The above is, of course, talking only about bending. If you are looking at compression then it is the overall area that will provide the load-bearing.

Happy to be told I am wrong if you have something else to add regarding the stress concentrations, but for homogeneous constructions I will stand by my answer .

 

[bjphoenix]

if you have something else to add

I'll try to do this without writing a book...

First, those are pretty massive columns. They don't look so big from a distance but if you find closeup photos, especially a photo of the large piece still hanging off the side of the ship, it is very big. In some photos showing the bases with the columns broken off you might wonder why there is no reinforcing steel sticking out, then you find a closer view and there is a whole lot of reinforcing sticking out, it just looks like hair compared to the overall size of the concrete.

Regarding pure compression strength of a column- for a short column it doesn't matter if it is hollow or not, it is the cross section area that matters. For a longer column where buckling is a factor then the section properties matter more. A larger hollow column will be stronger than a smaller solid column because it has better buckling strength. In mechanics of materials terms it higher radius of gyration, radius of gyration is a function of cross section area and moment of inertia. From a buckling standpoint it may be that a hollow column with the same diameter as a solid column could have greater buckling strength, the hollow column has smaller area but has larger radius of gyration and thus higher critical buckling stress. I could write the equations and check this.

For a member that has bending and compression it is more complicated. Bending strength is more a function of the larger dimensions so a larger column with larger void inside will be stronger in bending. The configuration of the bridge supports were creating truss action (meaning axial load only) for lateral loads in the direct along the axis of the bridge and rigid frame action (meaning bending) for lateral loads perpendicular to the bridge. Gravity loads are primarily loading the columns in compression. So the bridge supports would have been designed for various load cases involving compression and bending.

The above are generalizations based on elastic strength of materials. Reinforced concrete doesn't behave exactly that way at its strength limit, strength calculations for reinforced concrete don't involve calculating moment of inertia or section modulus. In some ways structural steel doesn't behave elastically either, residual stresses from rolling and bending behavior at the limit deviate from theoretical elastic behavior.

I've been doing this stuff for a long time, I can answer more specific questions if anybody has any.

Another tidbit that somewhat relates to the previous discussion- for a long column where buckling is a factor such that it will fail within the elastic zone- column buckling load is pi^2 x E x I / (L^2). (E is youngs modulus, I is moment of inertia, L is length) In this situation the cross section area of the column doesn't even enter into the calculation. This doesn't apply very much to reinforced concrete. Another thing I didn't mention is local buckling which could be important if someone wants to apply this discussion to rocket parts. Thin wall tubes such as large diameter Estes tubes can have local buckling failure before the overall cross section fails, limiting the strength of the system. Buildings are usually proportioned to avoid this type of failure but it can still be a limitation sometimes. I suspect that the fairly large pieces of the bridge truss itself could be limited by local buckling stresses because they have relatively thin walls compared to the width and depth of the section to keep weight down. Sometimes proportioning a beam or column from scratch is a balancing act to minimize weight while still optimizing bending strength, buckling strength and deflections.

 

[Tractionengines]

@bjphoenix THANKS, for your explanation of my overly simplistic statements. You said it a lot better than I would have, given how tired I am right now. (And it's been 20+ years since I thought about it indepth.)

 

[boatgeek]

In addition to what @bjphoenix so ably wrote above, when you're working with cost constraints (and just about everyone is!), the strength per unit of material and per unit of labor is very important. That tends to drive you toward large diameter with thinner walls, but not so thin as to run into local buckling. Depending on the situation, you might have limits on how big you can make the parts, in which case you start going to thicker walls. Labor comes into it if you have extremely dense rebar (hard to assemble the rebar and consolidate the concrete into it) or if you're welding vs. bolting parts together. My industry is a little weird among structural folks because we weld everything. Bolted connections are generally cheaper, though they'll often have some welded parts in there as well.

 

[Capt. Eric]

Not at all my field, but I would expect that a ship with a "50 ft. draft" means that it draws 50' when fully loaded and ballast tanks full. Not being fully loaded, or being in calm enough water (ie. not at sea) to use less ballast (temporarily) might allow the ship to draw less than 50'.

Am I wrong?

No offense, but yes, you're wrong (a little).

I've enjoyed this thread because it has taken some interesting turns into subjects I don't get to expound on very often. And some good questions, generally speaking.

So draft, ballast, load lines, trim and stability (where did that come from?).

The quoted draft of a ship listed in stats is typically "summer draft" or max draft allowed, as boat geek says. But what that means is that the draft marks on either side midships are as deep as you may load the ship. Every commercial vessel has them. (How they are placed is another long story, but it basically is an issue of safety, angle of roll, or incline, and required free board to avoid deck edge immersion and down flooding risks.) This allows you to determine how much weight you can load on a ship, including cargo, fuel, ballast, stores, etc. But to really get that maximum weight onboard, you have to have your ship on an even keel, fore and aft, and straight up (no list to one side). This means you have to distribute the weight so that the ship sits nice and flat. But a ship doesn't sit flat. It sags in the middle, or "hogs" with too much weight at the ends, causing the middle to arch up. So it has to be distributed just right to maximize the cargo lifting ability of the ship. That can be hard to achieve perfectly.

Loading to your marks, or maximum draft, is not usually reached. This is due to the volume of many cargos taking up all available space, before reaching the maximum weight able to be loaded. Typically, a ship has to be designed to carry heavy cargoes to reach it's marks regularly. A bulk ship that carries grain, ore, or other commodities does it often, but most other ships don't.

I don't know what the drafts of the Dali were when it left the dock, but I'm sure it was deep, but not to it's marks. Looking at pictures, you will see the red bottom sticking out of the water near the stern. That is not because it is light. That is because it has a bridge resting on it's bow. Bridges are heavy. Push the bow down, and the stern goes up. (It's like a seesaw.) My guess is that if it had a max draft of 50ft, it might have been around 45ft. While it's cargo might have been light, since we don't export as much as we import, the ship had a long voyage ahead, and probably had a large fuel load. Fuel can have a significant affect on a ship's draft. That thing probably burns about 200 tons a day at sea, and had about a 30 or 40 day trip ahead of it. A full fuel load would be significantly more than that.

Ballast - ships don't get paid to load and carry ballast. It causes a ship to have poor fuel economy since you increase the volume below water and that is harder to push through. Plus, with cargo on board, it can put you below your marks, which is not legal. A ship loads ballast to trim it in a desirable condition (flat, or slightly down by the stern). Also to provide positive stability. In other words, keep the ship from being top heavy and rolling over. Ballast can also be needed to ease stress on the ship's structure. You can easily break a ship in half if stressed too much by weights not being distributed well. Ballast levels out that distribution. But the whole "keep it from rolling over" thing is the biggest reason on a ship like the Dali. See all those containers stacked up high? That raises the center of gravity, and if it gets too high, the ship rolls over like a cheap toy in a bathtub. So they usually load some ballast down low to keep "positive stability." That cannot be removed if stability is low.

An empty ship will also load ballast to sink down lower so the rudder and propeller are submerged properly. On rare occasions, an empty ship may load additional ballast to keep it from being damaged in bad weather (bobbing like a cork is bad). But that would be temporary, and only on an empty ship.

So ballast is often necessary, but you never want to load more than you have to. As I said, a deeper ship takes more power to push through the water. It costs more fuel, which is expensive. So the ship likely didn't have too much room to remove ballast. And depending on where and how much cargo is loaded, you might not be able to load any more if you wanted to. Not without excessive stress or too much trim in the wrong way.

I hope this makes sense. I tried to keep it as simple as I could. If it's confusing, let me know how so and I'll try to do better.

 

[teepot]

You guys are amazing. So much knowledge.

 

[Capt. Eric]

You guys are amazing. So much knowledge.

Ship operations are far more complicated and diverse than most people can appreciate. The breath of subjects we have to be familiar with is amazing, even to me when I think about it. It's not just navigation and ship handling. Caring for different cargo types could be a career in itself. Ship construction and naval architecture, welding, mechanical engineering and maintenance, steel preservation are others. Add in industrial type safety, firefighting (in some ways different from shore-side firefighting), medical care, feeding and sanitation, accounting (Captains do the payroll), insurance and business, a massive host of legal compliance and regulatory matters, plus dealing with a wide range of personnel and cultural issues. You have help with some things, but you need to be familiar enough with all of it to make good decisions and head off conflicts. One of the things I liked about the job was that diversity. I could be planning a voyage one minute, then addressing a crane maintenance issue the next, back to catch up on the payroll, when someone needs first aid. All that could easily be addressed within one hour of any given day. It's not easy. 12hr days are the minimum.

 

[boatgeek]

Adding a couple of things to @Capt. Eric's descriptions above.

Calculating the load line has three parts. The first is all about the ship's geometry. You choose a reference deck, and from the hull shape you get a load line freeboard (distance from the deck to the deepest draft) after 3 tabs of Excel calculations. Then you check to make sure that doors, ventilators, etc. all have the right sill heights. Finally, you have to prove that the ship can meet stability and longitudinal strength requirements at the load line. To take a rocketry analogy, Let's say that you want to know if you can fly an I1299 at a particular field. First you have to know if you can fly I's at all based on the site dimensions. Then you have to check if you have a rocket that will take that thrust, is stable, and will stay under the waiver. Then you have to consider whether you'll land in trees or lakes if you do fly that motor.

One more issue of ballast water is that it's not necessarily easy to get rid of once you have it on board. Because of invasive species concerns, you can't generally take ballast water from one port to another without treating it. Treatment systems are kind of finicky and prone to breaking down (see discussion about automatic circuit breakers earlier in this thread, then add the complication of living things). That leaves you with the option of just carrying it around for a long time (not great) or doing a ballast water exchange >200 miles offshore. That's when you pump off the ballast you have on board and pump on new "clean" ballast. Stuff can go wrong during a ballast exchange, as happened to the Cougar Ace. A naval architect acquaintance of mine fell and died on that salvage effort. It's not easy moving around a ship with a 60-degree list. Of course, you can also ignore the rules and pump off the ballast whenever you feel like it, but that risks jail time.

You guys are amazing. So much knowledge.

We all have our areas of expertise. This is literally my day job (I currently have a project trying to get a deeper load line for a client's boat, and another project where I'm doing longitudinal strength calculations). You know infinitely more about guns and golf than I do, and others have lots of other experience.

 

[wsume99]

I don't know what the drafts of the Dali were when it left the dock, but I'm sure it was deep, but not to it's marks. Looking at pictures, you will see the red bottom sticking out of the water near the stern. That is not because it is light. That is because it has a bridge resting on it's bow. Bridges are heavy. Push the bow down, and the stern goes up. (It's like a seesaw.)

Not trying to argue with your logic but didn't the Dali run aground? I believe the ships bow is resting outside the channel as well. In this case the weight of the bridge is being supported by the river bed. Surely the additional weight would push the bow further into the mud. However if the stern is not grounded is it possible the stern is not being pushed up?

 

[Capt. Eric]

Not trying to argue with your logic but didn't the Dali run aground? I believe the ships bow is resting outside the channel as well. In this case the weight of the bridge is being supported by the river bed. Surely the additional weight would push the bow further into the mud. However if the stern is not grounded is it possible the stern is not being pushed up?

The stern is not being pushed up. The bow is being pushed down. The stern is lifted by the ship's structure. Again, the seesaw thing. But while the bow of the Dali is currently firmly aground, part of that is the huge freaking weight (layman's terms) of the bridge that fell on top of it. I mean, the bow reached the point of hitting the thing, so it couldn't have been aground very much, no? Also, mud is soft and compresses, squishes, squeezes, and gets pushed aside. If just being outside a designated channel on a chart means the ship instantly stops, it never would have hit the bridge.

 

[Capt. Eric]

Adding a couple of things to @Capt. Eric's descriptions above.

Calculating the load line has three parts. The first is all about the ship's geometry. You choose a reference deck, and from the hull shape you get a load line freeboard (distance from the deck to the deepest draft) after 3 tabs of Excel calculations. Then you check to make sure that doors, ventilators, etc. all have the right sill heights. Finally, you have to prove that the ship can meet stability and longitudinal strength requirements at the load line. To take a rocketry analogy, Let's say that you want to know if you can fly an I1299 at a particular field. First you have to know if you can fly I's at all based on the site dimensions. Then you have to check if you have a rocket that will take that thrust, is stable, and will stay under the waiver. Then you have to consider whether you'll land in trees or lakes if you do fly that motor.

One more issue of ballast water is that it's not necessarily easy to get rid of once you have it on board. Because of invasive species concerns, you can't generally take ballast water from one port to another without treating it. Treatment systems are kind of finicky and prone to breaking down (see discussion about automatic circuit breakers earlier in this thread, then add the complication of living things). That leaves you with the option of just carrying it around for a long time (not great) or doing a ballast water exchange >200 miles offshore. That's when you pump off the ballast you have on board and pump on new "clean" ballast. Stuff can go wrong during a ballast exchange, as happened to the Cougar Ace. A naval architect acquaintance of mine fell and died on that salvage effort. It's not easy moving around a ship with a 60-degree list. Of course, you can also ignore the rules and pump off the ballast whenever you feel like it, but that risks jail time.


We all have our areas of expertise. This is literally my day job (I currently have a project trying to get a deeper load line for a client's boat, and another project where I'm doing longitudinal strength calculations). You know infinitely more about guns and golf than I do, and others have lots of other experience.

Ballast treatment systems absolutely suck and aren't worth the money being wasted on them. Figure it out, then add the requirement. They did the other way around. Stupid.

 

[bjphoenix]

One of the latest youtube videos by Jeff Ostroff shows pieces of the bridge after they have been cut up and placed on shore. Some of the rectangular hollow sections appear to be relatively thin with respect to their external dimensions.

 

[cwbullet]

I hope they fix it soon

 

[Peartree]

I hope they fix it soon

Best case is still years.

 

[cwbullet]

Best case is still years.

sad but you are probably right.

 

[boatgeek]

sad but you are probably right.

Absolute best case is probably 2 years. 3-5 is more likely. 6-8 is possible if they completely rework the bridge design.

 

[teepot]

They are trying to get the federal channel open first. I heard they hope to do that by the end of May.

 

[Mick Kelly]

Is it true that this "accident" blocks some 20 coastal defense ships from leaving port? Any chance this was cyber attack? Not a conspiracy theory just asking a question

 

[Capt. Eric]

Is it true that this "accident" blocks some 20 coastal defense ships from leaving port? Any chance this was cyber attack? Not a conspiracy theory just asking a question

No chance whatsoever. Commercial ship controls are not normally hooked up to the internet. A small percentage can send performance data to the home office. A very small number of small vessels are being controlled remotely, but just as trials and for research. No ships of this size can be affected like that that I have ever heard of.

As for "coastal defense", well, that's not really how the Navy works. I don't know exactly what ships are upstream from the bridge, but I thought the Navy is out in Norfolk, not all the way in at Baltimore. So don't worry. We're safe. Of course, the USS Constellation will be stuck for some time.

 

[Banzai88]

As for "coastal defense", well, that's not really how the Navy works. I don't know exactly what ships are upstream from the bridge, but I thought the Navy is out in Norfolk, not all the way in at Baltimore. So don't worry.

Yeah, we're down here in Norfolk, and all is well.

As for 'coastal defense ships', if any are actually in port up there, if you're referring to the LCS Freedom and Independence class ships......harsh language on the radio would be more effective anyway.

 

[Mick Kelly]

No chance whatsoever. Commercial ship controls are not normally hooked up to the internet. A small percentage can send performance data to the home office. A very small number of small vessels are being controlled remotely, but just as trials and for research. No ships of this size can be affected like that that I have ever heard of.

As for "coastal defense", well, that's not really how the Navy works. I don't know exactly what ships are upstream from the bridge, but I thought the Navy is out in Norfolk, not all the way in at Baltimore. So don't worry. We're safe. Of course, the USS Constellation will be stuck for some time.

How do they send performance data "back to the home office"? Ukraine has many autonomous kamikaze small boats that have sunk at least 4 Russian warships. Were u in the Navy? I was. For starters no one in their right mind would attack Norfolk, why run into the teeth of an enemy?
Was just wondering if this was a message sent in retaliatory response to something we did related to either Iran or Russia.
It was the electrical system that was malfunctioned.
Just some thoughts outside the box is all.

 

[Capt. Eric]

How do they send performance data "back to the home office"? Ukraine has many autonomous kamikaze small boats that have sunk at least 4 Russian warships. Were u in the Navy? I was. For starters no one in their right mind would attack Norfolk, why run into the teeth of an enemy?
Was just wondering if this was a message sent in retaliatory response to something we did related to either Iran or Russia.
It was the electrical system that was malfunctioned.
Just some thoughts outside the box is all.

On ships with the capability, which isn't all ships, by any means, the engine monitoring data (temperatures, pressures, timing info, fuel and lube oil consumption) might be sent to the office where the office engineers can take a look. This can be sent periodically (once a day for example) via email data packet. It is also possible for a data connection through satellite service to provide on demand or more frequent snapshots, or allow the office to ping for info. Somewhat limited usefulness in my opinion. The ship has engineers on board who are supposed to be monitoring this stuff and taking needed action, so the office guy is just "big brother" most of the time. He doesn't get the full picture that someone working in the engine room has. But these systems that are in widespread commercial use do not allow for remote control of anything.

I know there are crazy conspiracy theories out there, but it's all crap. These kinds of breakdowns happen. It's usually something like a lube oil sensor error or some other silly thing that causes the controls to shut down an engine (main engine, or generators), which cascades into the ship going dark. (Just guessing at a possibility. The NTSB report will get it nailed down.) The difference here is that it happened to occur just before passing under a bridge. Just bad luck. At worst, some maintenance was skipped, or mechanical problems not addressed properly, but nothing more. It wasn't Iran, Chinese, Russians or any pimple faced punk hacker.

Since you asked, I was in the Navy reserves way back when (part of my obligation for attending a federal academy), and have worked with the Navy as a contractor operating ships, and carrying military cargoes. But that has nothing to do with this event. Not a military issue.

 

[boatgeek]

Yeah, we're down here in Norfolk, and all is well.

There are a couple of naval supply ship terminals on the wrong side of the bridge. Not that supplies aren't important, but a couple of them being stuck wouldn't be a dealbreaker. I suspect they could get out of the side channel being used by the bridge removal fleet.

As for 'coastal defense ships', if any are actually in port up there, if you're referring to the LCS Freedom and Independence class ships......harsh language on the radio would be more effective anyway.

Hey, you don't just need harsh words on the radio. You also need a butter knife to take one of those bad boys out of action. A slingshot might do as well.

 

[Greg Furtman]

An interesting video about the cause.

 

[jsdemar]

An interesting video about the cause.

The home circuit breaker example is quite cringeworthy. He should stick to evaluating structures.

 

[OverTheTop]

The home circuit breaker example is quite cringeworthy. He should stick to evaluating structures.

Completely agree!

 

[bjphoenix]

The home circuit breaker example is quite cringeworthy. He should stick to evaluating structures.

But the thing is he's an electrical engineer, not a structural engineer.

 

[boatgeek]

An interesting video about the cause.

Arrgh. There's a lot of misinformation (at best) in there. First of all, the discussion about the construction vehicles at around 3:15 is a red herring. He's concerned that the construction vehicles that were on the bridge (potentially with bodies inside) are going to be left down there while the temporary channel opens. The NTSB clearly shows that the debris will be removed from the bottom in way of the temporary channel, so any bodies down there will be recovered as well. All along, the response team has talked about removing debris in order to recover bodies, so it's pretty clear that this is a priority, as it should be.

At 6:20 the presenter seems surprised that they're investigating the circuit breakers when there was a systemwide blackout. Given all of the discussion here in the early days, it's weird this is a surprise. It also shouldn't be a surprise that if there's a potential failure, the equipment manufacturer would get involved. These are more or less bespoke systems (Hyundai building small volumes of products for itself), so of course you'd get the manufacturer in.

On to the circuit breakers. Oh that residential panel. I can't see how the refrigerated containers could possibly take down a [properly functioning] electrical system. Even if they were overloading the system, the circuit breakers on this size ship should be coordinated so that the breaker closest to the load trips first, minimizing systemic issues. It's possible that someone down the line started fiddling with the system to prevent breakers on subpanels from tripping, but it would take a lot of screwing around to get that all the way to blowing power to the entire ship. It also would not surprise me at all if there was a power management system on the refrigerated containers that would keep the total load from the refrigerated containers under control. If there was trouble with refrigerated containers before departure (as per news today), that was almost certainly a symptom rather than a cause.

One useful item was that I was completely wrong about the NTSB's timeline for announcing preliminary findings. That will apparently be in early May. That's later than I expected, but not really a surprise given that they're still in the field collecting data.