Water: What we're up against

greenspun.com : LUSENET : Millennium Salons : One Thread

The following is a brief excerpt from a question asked by David Hunter, on the Y2K Water List, and a very clear response from David Hall that shows the size and complexity of the y2k problem in water and sewage treatment facilities. It may help you if you want to ask your local water people questions. You may also want to make them aware of teh Y2K Water list (click link above):

Dave Hunter wrote:

Why should water or sewage treatment info be any more difficult to come by than electricity? Since many, if not most, water and sewage districts are government entities, wouldn't this information be publicly available if asked for? Would we probably have to use the Freedom of Information Act to get it?

Are there certain types of systems (software (mainframe or PC) and embedded) that water utilities use that are pretty much common to most water utilities? Have tests shown there is more of a likelihood for problems to occur in one type of system over another?

What kind of bodily harm and property damage are we talking about - flooding, lack of water to customers, lack of clean, safe water? Can you identify a list of probable outcomes of Y2K failure in relation to water?

[snip]

From : Dave Hall
Date : June 6, 1998

David,

Water plants are no different than any other industry sector in that they use "similar" type systems to control and purify ware and sewage treatment. Note, however, that for Year 2000, there is NO SUCH THING as "identical" systems using microprocessors. EVERY piece of equipment, every PC, every system using microprocessors MUST be tested to determine it's INDIVIDUAL response to Year 2000 dates. That means that there cannot be type (or generic) testing for vendor equipment. Testing has found that even within model numbers with consequential serial numbers (001, 002, 003, etc.) the response to year 2000 dates VARIES widely.

This inability to type test means that each water system must test their individual systems on-site, in operational mode, with all variables to be sure that they can continue to operate successfully in Year 2000. We have noted that embedded systems many times require up to 30 tests to exercise all logic (or at least most) paths.

This is the problem. Does each water system have engineers capable of designing such tests, ensuring that the tests do not destroy some vital part, conducting such tests while the system is needed to provide water, paying for such tests, and project management experience to ensure that everything is accomplished properly and reported? In my experience they do not. IMHO, Most water systems personnel do not even believe that a problem could exist. Just like 90% of the population of the globe. If a problem can't exist, why should they worry about testing?

Once you get the personnel and Executive Management worried, then such tests must be designed on an individual basis for each of the thousands of water systems, carried out, problems identified, and projects started to correct the problems. And we have what, about 410 working days left?

In terms of what harm, I DON'T KNOW. That's the problem. I, as a citizen of Illinois, require PROOF that MY water system has been tested and proven free of ANY Yr2K impacts because I depend upon that water to be safe and plentiful. If it has not been tested, then the water could be contaminated (if not purified), could become toxic (if all chemicals were dumped in at one time, could become not available (if pumps shut down), could become scarce (if some pumps shut down), etc. You pick the scenario.

Microprocessors run everything in many systems. What could happen if a few or many shut down or provided bad control signals? You name the impact, because if we do not test, we do not know and must GUESS. I for one do not want to GUESS OR ASSUME that my water will be pure and available. As to the status of the chemical suppliers and vendors of microprocessor-based equipment, what is their status? For your information NO plant or facility that has been tested for Yr2K impacts has passed with NO failures. Some have been minor, some serious, some catastrophic. Your guess is as good as mine until someone asks the manufacturers and vendors the question. I would assume that local water systems would be interested in that answer, but I don't know of many who have asked the question.

This is not a complete answer to your questions, but that would take a full six hour training course just to touch the surface of all possible interactions, problems and Best Practices for projects. If you are interested in that, contact me. I do that kind of course for state/local governments and companies and have conducted one for Illinois State Government personnel.

Dave Hall My opinions only, of course
Year 2000 Infrastructure and Embedded Systems Engineering
dhall@enteract.com



-- Bill (billdale@lakesnet.net), June 07, 1998

Answers

Government's Role in Reducing "Year 2000" Risks Leon A. Kappelman, Jerry L. Johnson, and Kathy Rosmond

http://www.year2000.unt.edu/kappelma/Gov-y2k.htm

Water and Waste Treatment Systems Most modern systems that control the treatment and distribution of drinking water and waste water use computers and "smart valves" with embedded microprocessors. Problems with the control systems, hardware or software, or the smart valves could result in total system failures, contamination of ground water, and/or contamination of drinking water. Recommendation: The NRCC [Natural Resource Conservation Commission] should initiate a proceeding to assess the year 2000 readiness of all water districts, municipal water supplies, and waste water treatment facilities. The proceeding should determine if each utility has conducted a year 2000 risk assessment, developed a corrective action plan, and established a date to become year-2000 ready. NRCC also should make information available about known problems and potential solutions for specific vendor-supplied systems and equipment.

-- Harlan Smith (hwsmith@cris.com), June 11, 1998.


Moderation questions? read the FAQ