Science: Physics of a Crane Collapse

David Brown
Washington Post Staff Writer
Monday, June 16, 2008; 2:00 PM

Washington Post staff writer David Brown and John Cumings, assistant professor of materials science and engineering at the University of Maryland College Park, were online Monday, June 16 at 2 p.m. ET to discuss Monday's Science Page story about the physics behind the recent crane collapses.

Read the story: Tower Cranes: Efficient, Versatile -- but How Safe? (Post, June 16).

A transcript follows.


David Brown: Good afternoon chatters! We're talking about tower cranes and their physics this afternoon. We are lucky to have with us Professor John Cumings of the Department of Materials Science and Engineering at the University of Maryland at College Park. He will help answer the questions--and for this I am personally thankful! So let's begin.


Vienna, Va.: You wrote that jacking a crane is done by inserting a new section below the turntable. If so, how do they get it up to that level in the first place? I would have assumed that the crane is jacked up from the bottom, not near the top.

Professor John Cumings: Good question. The crane itself will the raise the new section almost into place, where it held until the hydraulic lifting is completed. After this, the new section is moved into place and bolted down. A nice time-lapse video of the process is shown at


David Brown: There is some sort of frame that is brought up from the ground and put around the tower at the point where the extension is going to occur. It is used to create a platform on which a hydraulic jack is placed to lift up the jib/cab/counter-jib/tower-top assembly. Exactly how it works I don't know.


University Park, Md.: Before March of this year, I can't recall seeing or hearing a single crane-accident story in the press. Now there seem to be several per month, and sometimes several in the same week. Is the frequency of crane accidents increasing? Is the ratio of crane accidents to cranes deployed increasing? Is the press now more likely to report crane accidents because they are seen as part of a pattern after the first high-profile accident in NYC in March? Thanks!

Professor John Cumings: This is an interesting point that you bring up. Certainly there are a lot of crane construction sites and there appears to be a boom in crane utilization. It is only natural to expect that there would be a corresponding increase in accidents. Furthermore, we all expect that high-profile accidents will attract additional press to the area. However, it is also true that the construction business is under great pressure recently. With falling housing prices in many markets, time is quite literally money, and accelerated construction schedules are expectable. I wonder if the recent trend in accidents isn't simply just a collision of the laws of economics with the laws of physics.

David Brown: I did not have time to do any reporting on the frequency of crane accidents or whether there is a trend in one direction or other. I am certain that there is increased press reporting of the accidents that do occur since the NYC accident last winter that killed 7 people. It is possible the apparent increase in accidents may be like some "cancer clusters"---a coincidental bunching of accidents that, when observed over a longer time duration do not add up to an actual increase in their rate.


Washington, D.C.: Is there a practical way of inspecting for or measuring fatigue degradation in metal structures before a crack has started?

David Brown: There are numerous ways to test and inspect for metal fatigue, the most common and oldest being visual inspection for cracks or surface irregularities that might suggest a place where there has been repeated bending. The metal can be sampled and examined by electron microscopy, but that is clearly a destructive method. There are non-descructive methods. They include using small, portable X-ray machines to look for invisible defects; ultrasonic testing to find non-surface cracks; something called eddy-current testing, which determines the electrical conductivity of a section of metal, which in turn tell you something about its crystal structure; and so-called dye-penetrant tests, in which dye is put on the surface of a metal and a developer solution is added and you can see if there are any tiny cracks where it has seeped in. Unfortunately, I don't know which or how many of these are used in the current tower crane inspection.


Alexandria, Va. - Fed. Construction Engineer: The story on tower cranes was terrific. I wonder, though, to what degree we've examined the "low-level" human-factors risks relating to cranes. As a practical matter, cranes are placed, anchored, and assembled on site by people who barely speak English and cannot (and do not) follow the ASTM and Code standards; they are under tremendous pressure to place cranes with great haste; and the on-the-ground work is typically extremely sloppy. We have simply terrible construction labor in this part of the country, and the only reason that the DC area has not experienced the kind of tragedies which have befallen NYC is that the building height limit here is sufficiently low that tower cranes in the DC area typically aren't tall enough to demonstrate fulcrum-weight-point imbalance - YET.

David Brown: You obviously are familiar with conditions on the ground, and what you say is very interesting and a good tip for further reporting. All I can say is that the construction/rigging/tower crane industry is extremely averse to press inquiries (and I would think site visits, too) and the moment because of the recent accidents. I spoke with two people in industry and both were willing to give a little information on background but neither would be quoted by name. It is a business feeling besieged, and I would bet they feel there could be all kinds of new regulations that might come out these events and they are consequently lying low.


Tampa, Fla.: I've been a crane operator for 38 years, and have seen many accidents, that were mainly due to the lack of the experience of the operator not knowing how to read a load chart,and to know at what radius the crane can safely handle the load. It would be similar to a teen getting his driver's license at the age of sixteen, and the next day someone shows him how to start a tractor trailer semi, and let's him take it for a cross country ride by himself.

I've seen this happen a lot on tower cranes ,and mobile hydraulic cranes where the contractor on the job is trying to save from having to pay for a more qualified operator. I have worked the Big Dig in Boston,and quit at least five times over me not making crane picks that I knew were unsafe picks. Many times I was told if I didn't make the pick they would find someone that whould. Both of the accidents in New York could of been prevented if the contractors were more saftey minded. One big problem also on most big jobs is that the safety man dosn't know anything about the ansi B30.5/regulations that applies to cranes on construction sites.

Steve Roberts

David Brown: Like one of the other chatters, you give a very interesting view from the field, and I would have to say quite a worrisome one. All I know is that there are undoubtedly many economic forces at play that could--I am not saying they do--affect the quality of operators, supervisors and inspectors. One is certain: using these devices is not cheap. They generally rent for a four-month minimum at rates of $14,000 to $30,000 a month. That is just for the crane, not the operator too. It also costs $25,000 to 50,000 to erect and dismantle a tower crane. So one can infer there is probably great pressure to keep them in operation and in service as much as possible when they are on a job.


Salt Lake City, Utah:"The usefulness of tower cranes is not in doubt. Several recent accidents, however, have led some to wonder about their safety."

Is the rate of failure also going up, or is the number of incidents increasing due to a growing number of construction projects?

Is there a trend to the recent failures (i.e. operator error, crane design, material supplier, crane manufacturer, material issues, etc.)?

Professor John Cumings: For what it's worth, I can see no trend in the types of accidents that have occurred recently. The March accident in New York involves an attachment unit that fell during installation when jumping the crane. In the more recent May accident, the turntable attaching the crane to the tower broke in half at a repair weld. The only trend I can see is that they both happened in New York, and that the inspection office involved has been recently tainted with scandal. Fortunately, there doesn't seem to be a verifiable trend from an engineering perspective.


Falls Church, Va.: In the event of a crane collapse, are there any safety features designed into the crane that will improve the operator's chances of survival? Something akin to an eject button with parachute?

Professor John Cumings: Most cranes are not tall enough to allow for a parachute recovery. Base jumping is an extreme sport where people attempt to parachute from cliffs, bridges, buildings etc., and it is generally illegal due to the not-infrequent fatalities. The maximum height for a freestanding tower crane is about 250 feet, which is considered an "extreme" height even for base jumping. However, there are rapid-descent mechanisms for an operator to escape a crane quickly. Usually they involve ropes and harnesses (think rappelling).


Mt. Lebanon, Pa.: Don't blame cranes. It's all a matter of torque = r cross F where r and F are both vectors and cross is the cross product.

Mr. Petroski (whose books I've read) is absolutely right - it's pretty simple, really. And if you don't understand this simple equation, get out of the crane design, erection, inspection, or maintenance business.

For the life and safety of the rest of us.

Thanks much,

Registered professional engineer

Professor John Cumings: Good point. In treating the quantities involved, one should use vectors. In our sidebar analysis, we only show the simple quantities of distance and weight (both mathematically referred to as scalars). A full treatment would take into account other forces from the side, due to a swinging load or extraneous wind forces. The total resulting forces would then be vectors.

As you point out, the mathematics is even then still relatively easy, even for a first or second-year student. This makes the design of cranes relatively straightforward. However, I can imagine that on a job site there might be a tendency to pushing the limit, even a little bit, in order to move a job along faster. No level of mathematics ahead of time can solve this problem down the road.


David Brown: I think we've come to the end of this chat. I apologize for not knowing more about the business of tower cranes, which would have let me answer some of these questions with greater detail. Thanks especially to Professor John Cumings of the University of Maryland for lending us his time and expertise.


Editor's Note: moderators retain editorial control over Discussions and choose the most relevant questions for guests and hosts; guests and hosts can decline to answer questions. is not responsible for any content posted by third parties.

© 2008 The Washington Post Company