After graduating with a master's degree in June 2006, I went to Changzhou, Jiangsu to work on foundation pit design. After becoming somewhat familiar with a few projects, I was able to do foundation pit design independently, but I just couldn't understand investigation reports (it's normal for someone who studied geotechnical engineering not to understand investigation reports). Selecting unfavorable boreholes was still okay, but choosing design parameters was out of the question—every time I had to ask the company's chief engineer to decide, and I had no impression at all of silt, clay, and the like. Once, at an expert review meeting, an expert said there was a local silt interlayer, but I couldn't find it. From then on I knew that it's not just foundation pit design—if you want to do geotechnical engineering well, you can't do it without mastering geotechnical investigation. At the end of that year, since I was also dissatisfied with that company, I switched to a small company; at the time I mainly wanted to be able to do investigation work.

And so I went to Zhejiang and engaged in geotechnical investigation and foundation pit design. By now I have done investigation for more than a hundred projects, have developed a certain concept of soils, and can handle foundation pit design with ease; for some design parameters I even dare to make appropriate adjustments. You may have encountered situations where the parameters proposed in investigation reports for foundation pit design are unreasonable—sometimes overly conservative, sometimes leaning toward being unsafe. Now I can judge which are reasonable and which are not, and many times I adjust the parameters and have them recognized by review experts. Actually, in geotechnical investigation, the key is to have some intuitive understanding of soils, and this is crucial for those of us engaged in geotechnical engineering.

Old Q says: You must thoroughly understand the surrounding environment

Every time I design a foundation pit, no matter how big or small, I go to the site to inspect it personally, mainly to see whether there are buildings, roads, pipelines, etc. around it. I think the most terrifying are sewage pipes and rainwater pipes; these two types of pipes are poorly constructed and leak year after year, causing the soil quality to deteriorate. I have encountered several projects where collapses were caused by leakage from sewage or rainwater pipes. When considering schemes, we must be cautious, leave some margin, and take corresponding measures. For gas pipelines and cables, even if the burial depth is known, the soil nail wall scheme must be used cautiously; you should know that the workers doing the work now are migrant laborers—who knows where the steel pipes they install end up. Last year I participated in a project where the gas pipeline burial depth was roughly known to be 2.5 meters. A Hangzhou company used a soil nail wall scheme, while I adopted a row pile scheme and did not dare use a single soil nail. I told the client that when migrant workers construct, it is impossible for them to control according to 2.5 meters, and steel pipes could hit the gas pipeline. In the end, the construction unit still adopted my scheme. For locations close to residential buildings, use schemes with high reliability as much as possible, such as row piles or piles with internal bracing (except where the residential buildings use pile foundations).

Old Q says: Remember to grasp the major and let go of the minor

A foundation pit always has some sections with lower risk and some sections with higher risk. We must have a concept in our minds: important places cannot have any problems and must be built quite solidly; otherwise, once an accident occurs, the consequences are very serious. For other unimportant places, save a bit—if they collapse, at worst an excavator can dig out the soil. Generally speaking, construction units can accept this. If some construction units blindly want to save money and make all kinds of demands, such a foundation pit is better left undone, because in the end it is we designers who suffer.

Old Q says: Frequently go deep into the site to obtain first-hand information

For projects you design, you must track them from beginning to end. Every project is a good opportunity for us designers to make prototype observations. Going to the site often, first, increases intuitive understanding, and second, lets you check construction quality. Especially for sections where accidents occur, you must summarize and learn lessons. Successful cases are the chips that increase our strength. On construction sites, there are often cases where construction does not follow the design, and some are even rather risky, but in the end nothing may happen; this way we also have a clearer idea. For example, if your design is a 1:1 slope, but it is actually excavated to 1:0.5 and nothing happens, then we have a clearer idea, and when designing similar foundation pits in the future we will know better.

Old Q says: Those engaged in foundation pit design must not make conceptual mistakes

I remember Professor Yu Diaomei of Tongji University said that geotechnical work is 90% experience and 10% theory, concrete work is 50% experience and 50% theory, and steel structure work is 10% experience and 90% theory, which shows how important experience is in geotechnical engineering. Successful experience is built on correct concepts. Sometimes it is not a big problem if soil nails are a bit short or a few fewer rebar are used in piles. What is terrifying is using a soil nail wall where it should not be used, or using cantilever piles where deformation requirements are strict; then no matter how long the soil nails are or how much reinforcement is used, it is useless.

Regarding the foundation pit accident at the Hangzhou Xianghu subway station, I think the design made conceptual mistakes. First, the bottom of the diaphragm wall did not pass through the muddy soil, and even more unimaginable, the reinforcement of the passive zone was canceled. Second, the first support used a steel support, and the end of the support did not form a fixed connection with the diaphragm wall. As a result, when the top of the diaphragm wall tilted backward, the steel support also detached. If the first support had been a reinforced concrete support, the situation might have been somewhat better. The establishment of correct concepts comes from a large amount of engineering practice and a certain theoretical foundation; neither can be dispensed with.

Old Q says: Beginners in foundation pit design can try the "engineering analogy method"

People just entering the industry may not have much experience and may lack confidence. I suggest using the "engineering analogy method": collect schemes from some local successful cases and study them carefully; or download some foundation pit examples with similar geological conditions from the Internet. Some conference proceedings are also good; these examples all have certain reference value.