As a civil engineer directly involved over many years in the design of submersible pumping stations, though not in Canada, I recognise and agree with some important points made by the journalist. However, I don’t agree with his comment that a dry well/wet well system is “better” than having a wet well with submersible pumps. Systems using submersible pumps in wet wells have been used successfully in Australia for many decades. There is no inherent problem with them. Their main advantage is that the installation saves capital cost because there is no need for a second well. Also, pumps are cooled by the fluid in which they are immersed and so ventilation costs are eliminated. “Dry” wells may yet be damp and humid, and corrosion of pipework is then an issue.
I would make sure to involve operators in the design of your systems, because they know what works and what does not. They would have been able to point out the stupidity of having unsealed electrical junction boxes below the possible immersion level. Or was this merely a decision forced on the engineers by the bean-counters, as happens too often?
You should carry out a risk analysis: Develop a risk register, and that includes identifying all possible modes of failure, for all phases of the project (planning, design, construction, operation, demolition), so that you address them and are prepared for them. Document them fully and have them signed off by all stakeholders.
Addressing risks does not mean gold-plating everything. It does mean knowing what they are; their likelihood; what the consequences are if they eventuate; and making a decision on what to do about them. Is it just a simple matter of cost, for example, are you prepared to accept that a part may fail after 10 years? Is it cheaper to replace that part every 10 years than pay much more for a part that lasts 20 years? A lifecycle economic analysis on the major options should determine such matters preferably at the planning stage (before design starts).
(A planning study also is useful to allow the politicians and higher level bureaucrats to see what they can get for their money; or to see what it will cost to provide what they want; before funds are committed. That can save political embarrassment later, and potentially avoids the risk to a successful project caused by underfunding. Was a planning study done for this project?)
Another example: For critical risks, as a designer, you should not be relying on the operators’ level of training. You should “Keep It Simple, Stupid” and design out the risk as far as possible.
Don’t blame the operator if you don’t train the operator. Operators should be intimately involved in the design and commissioning of the system. Commissioning should cover all failure scenarios. Think of the potential reasons why an operator might not be able to address the problem - What happens if the on-call operator is suddenly not available or cannot reach the station? For example, s/he cannot reach the station due to flooding; or has had an accident on the way to or at the station?
It is not about “over-engineering”. It is about providing a facility that works as intended, and being prepared for and knowing how to overcome problems when they occur.
In relation to the risk management, was the electrical engineering design checked and signed off by an experienced engineer? Was the electrical installation checked? The fact that three pumps were attempting to run off one generator points to:
(a) A potential design or drawing fault ( the drawings may not have been checked properly to ensure that each generator only fed two pumps);
(b) A potential installation fault (the electricians installing the wiring did the work incorrectly, and the wiring was not checked once installed).
(c) Even if each generator was assigned to only two pumps, the failure points to a potential inadequate analysis of risk (should there have been a third standby generator to kick in, in the event of one of the duty generators failing? Or should each generator have been capable of taking the full load of all four pumps?
Recent Comments
I would make sure to involve operators in the design of your systems, because they know what works and what does not. They would have been able to point out the stupidity of having unsealed electrical junction boxes below the possible immersion level. Or was this merely a decision forced on the engineers by the bean-counters, as happens too often?
You should carry out a risk analysis: Develop a risk register, and that includes identifying all possible modes of failure, for all phases of the project (planning, design, construction, operation, demolition), so that you address them and are prepared for them. Document them fully and have them signed off by all stakeholders.
Addressing risks does not mean gold-plating everything. It does mean knowing what they are; their likelihood; what the consequences are if they eventuate; and making a decision on what to do about them. Is it just a simple matter of cost, for example, are you prepared to accept that a part may fail after 10 years? Is it cheaper to replace that part every 10 years than pay much more for a part that lasts 20 years? A lifecycle economic analysis on the major options should determine such matters preferably at the planning stage (before design starts).
(A planning study also is useful to allow the politicians and higher level bureaucrats to see what they can get for their money; or to see what it will cost to provide what they want; before funds are committed. That can save political embarrassment later, and potentially avoids the risk to a successful project caused by underfunding. Was a planning study done for this project?)
Another example: For critical risks, as a designer, you should not be relying on the operators’ level of training. You should “Keep It Simple, Stupid” and design out the risk as far as possible.
Don’t blame the operator if you don’t train the operator. Operators should be intimately involved in the design and commissioning of the system. Commissioning should cover all failure scenarios. Think of the potential reasons why an operator might not be able to address the problem - What happens if the on-call operator is suddenly not available or cannot reach the station? For example, s/he cannot reach the station due to flooding; or has had an accident on the way to or at the station?
It is not about “over-engineering”. It is about providing a facility that works as intended, and being prepared for and knowing how to overcome problems when they occur.
In relation to the risk management, was the electrical engineering design checked and signed off by an experienced engineer? Was the electrical installation checked? The fact that three pumps were attempting to run off one generator points to:
(a) A potential design or drawing fault ( the drawings may not have been checked properly to ensure that each generator only fed two pumps);
(b) A potential installation fault (the electricians installing the wiring did the work incorrectly, and the wiring was not checked once installed).
(c) Even if each generator was assigned to only two pumps, the failure points to a potential inadequate analysis of risk (should there have been a third standby generator to kick in, in the event of one of the duty generators failing? Or should each generator have been capable of taking the full load of all four pumps?