Harry Shearer interviewed Maria Garzino, the whistle-blower who first pointed out how flawed the M.W.I. pumps were. I listened to the show and took notes and just wanted to share a few random observations. You can listen to the interview for yourself here (Sunday Sept. 13th Show).
Ms. Garzino did give a few conflicting statements over the course of the interview, but not an unreasonable amount, given it was a 30 minute interview on a technical subject.
* An error on the ACOE pump support structure drawings caused the pumps to be installed with 2' less of NPSH(a) than the pump impellers were trimmed to. Ms. Garzino thought it was not an unusual error, given the complexities of the project. It took a while to correct, but it eventually was fixed.
* The "battle test" that the ACOE is so proud of for the M.W.I. wasn't much of test at all. SCADA data shows that the M.W.I. pumps were only run once the heavy lifting was done by the direct drive pumps. For background, all major industrial equipment is controlled by Programmable Logic Controllers (PLC's). These primitive, reliable computers communicate to each other through several different protocols, one of which is SCADA. It's still a little hard to get definitive evidence, but it seems to be backed up by the runtime reports at the back of the Parsons Report (Table 3-10, etc.).
* One of the serious design flaws that kept the M.W.I. pumps from being successfully tested was air getting into the hydraulic pumps. In the M.W.I., diesel engines turn a hydraulic pump. That hydraulic pump pressurizes fluid that runs through lines to a hydraulic motor. The motor turns the impeller blades that moves the water. The reservoir that fed the hydraulic pumps was mounted incorrectly, causing air to carry into the hydraulic pump. Air in a hydraulic pump bad, on a good/bad scale. It tore the guts out of the hydraulic pumps. The reservoirs were mounted higher, to avoid air undercarry. That solved that particular problem.
* Maria Garzino then described the various stages of whistle-blowing she went through. She brought it up with various folks within the Corps. After she exhausted internal appeals, there was an Internal Technical Investigation by the Corps. I believe that was done by the Vicksburg office. She said the report was nice and all, but it didn't set its own scope and didn't really address her concerns. The Parsons Report followed on with more or less the same issues. Finally, the Office of Special Council hired their own technical advisers, went through all the reports and basically agreed with her over the Corps. Here's a letter [PDF] that describes a little more about the OSC report.
* Another issue she went through is how her actions changed some of the Corps plans. The Corps was, at one point, planning on re-using the M.W.I. pumps in the permanent pump stations. Later, the Corps decided that the M.W.I. pumps were only temporary. I don't know if there's any way of knowing whether the Corps had the original intent of using the Corps for the permanent stations, but I can say that they got lots of pushback from all sorts of sources when they discussed re-using the M.W.I. pumps. Mrs. Garzino's actions may or may not have contributed in the change of plans.
* The Corps has a developed a "How To" manual for reworking the Interim Closure Structure (ICS) into a permanent facility. The price tag on ripping out the M.W.I. pumps and replacing them with direct drives is approximately $275 million dollars. According to Ms. Garzino, the original estimate that the Corps submitted to Congress included a 50-year design life for the ICS. Spending $275 million after you told Congress that what you built had a 50-year design life is a waste of money. It's not completely an apples-to-apples comparison, but I think it's still useful to mention that large offshore oil structures have a design life of 35 years.
I listened to the radio interview to make some sort of judgment on Ms. Garzino's credibility. I listened like a hawk and took notes and while there were some contradictory statements, her interview taken as a whole leaves me to conclude her story is quite credible. All of her major assertions check out.
UPDATE- More things are coming to me in bits and pieces:
* The Corps really moved the chains for M.W.I. to pass the tests. Some of the moves were no big deal, but others were not legitimate. When competitively bidding contracts, you have to hold the winning bidders to what they said they'd do, especially with regards to testing. It violates ethical engineering practices and its unfair to the losing bidder. I understand the Corps' complaint (that they were under the gun for time), but the Corps is not the first entity in the world to face a looming deadline for a major, complicated project. If the Corps asked for testing the pumps, M.W.I. should have had a line item of a few thousand per pump for testing. If they didn't and they won the bid because of that, they should be forced to eat the difference. Remember, there's hardly ever only one vendor that can provide a piece of equipment or service.
* Verifying the pumps against their curve (making sure they work as the salesmen claim) is an important step. Hydraulic Institute standards state no worse than +/- 5% of the curve. Hydraulic Institute standards are MINIMAL; I can tell you that most clients I've dealt with have much more stringent standards on the pumps they test. If it's just a $50,000 pump or so, you're probably not going to bother testing it. Millions of dollars worth of pumps? You're going to get them tested. For critical systems, like de-ballast, firewater, etc. you're going to test the pumps to the Nth degree with a good chance of additional factory inspections.
* M.W.I.'s salesmen may have made excessive claims, but its the duty of the engineers not to just take them at their word and verify what they say. I worked on a job about a year ago where an engineering company bought some pipe for a pump discharge that wasn't pressure rated. I talked with the engineer and he said the salesmen said it was "an ideal application." I talked with the salesmen and he he told me it meets certain AASHTO, etc. specifications. I looked up those specs and the first thing they say is that these standards are meant for non-pressure rated pipe. The ratings the specs give were minimal 20 foot of head (~10 psig) ratings that should never have led an engineer to use them as force-mains. It's the engineer's job to run these things down.