PLC's will [did] they croak?greenspun.com : LUSENET : Grassroots Information Coordination Center (GICC) : One Thread |
http://home.att.net/~ofuzzy1/PLC.html
I created this page in nov1999, it reads correct"er" every day.
Layoffs, stock declines and other forms of corpoorate scratchings.http://content.honeywell.com/year2000/IAC/ You really should read the letter posted here. "Mid-1999 is here" and 5% of all companies had done any remediation / contingency work!
Many of the links no longer work [denoted by #].
Maybe one of the reasons why the Aluminum and Pulp plants were so quick to resell their juice was PLC problems -- Does anyone know who works ate these plants?
Will PLCs go croak?
[The orginal links and pages were revised by Honeywell, and so my comments and links had to revised, sorry this page may meanader around, but endure on you'll get the jist of the facts.]
Programmable Logic Controller (PLC), this is a fancy term for a customized / specialized PeeCee that operates valves, vents, boilers, conveyer belts and lots of other big expensive parts of factories and processing plants / refineries.
PLCs are having testing troubles in the real world.
This is a big listing of links.created 00019991101
The following is solely my personal opinion. I firmly believe that the vendors listed have performed a good faith effort to resolve these issues. And I pray that they did. Read the FACTS as they are printed. No malice is intended from my comments, as they are just that comments.
I received this short letter from a friend of mine, Mark. Mark is an Guru Engineer, with a Bachelor's in building Nuclear Power Plants -- he does kinda glow at night. He has designed and supervised installations (of non-nuke) in Asia and the USofA. (Trash burners and coal plants).
Every US refinery I have been too or heard of has very advanced Modern computers, all done within the last couple of years, Typically Honeywell TDC 3000s or Fisher Provox DCS's or better.I still firmly believe the real Y2K impacts will be in things like Furby's where the malaysian maker of the eyes can't ship or his oil shipment is impacted. The delay in shipment of Windows 2000 is more commercially significant to the US economy than y2k will be.
He still does not believe that Y2K will bite to hard. I do.You may recall Windows95 (w95) was delayed until the end of 1995 and then it was so bad that folks were on ignore (hold) for upwards of 3 or 4 hours when they installed it and it wouldn't work. Windows98 was a whole lot better, because most folks hadn't fallen for the I gotta get it first hype. Instead, they waited for a more stable version to appear [I believe they are still waiting.]
So, if you believe that Y2K is not an issue then ask anyone who bought W95 early on and tried to get help by the telephone. The issue is software does not always work as promised, especially when Bill sells it.
Starting with the legalese:(see bottom of page)# The Disclaimer
[It was this:]
Honeywell-Measurex has made a good faith effort to confirm the accuracy of this product and testing information. HOWEVER, HONEYWELL-MEASUREX MAKES NO WARRANTY, EXPRESS OR IMPLIED, OR REPRESENTATION THAT THIS INFORMATION IS ACCURATE OR COMPLETE. IN NO EVENT SHALL HONEYWELL-MEASUREX BE LIABLE FOR ANY DAMAGES ARISING FROM THE USE OF THIS INFORMATION FOR ANY PURPOSE, INCLUDING SPECIAL, INDIRECT OR CONSEQUENTIAL DAMAGES.
[You bought it, tough if your plant goes poof.][Now it's this:]
DISCLOSURE OF THIS INFORMATION DOES NOT CHANGE THE TERMS OF YOUR CONTRACT WITH HONEYWELL (IF ANY). THE INFORMATION IS PROVIDED "AS IS". HONEYWELL HEREBY DISCLAIMS ANY WARRANTY (INCLUDING ANY IMPLIED WARRANTY OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE) REGARDING THE INFORMATION POSTED ON THIS SITE.[I'd make a wise-_ss comment, but I'll only say: BEGIN WORRYING!]
What does Honeywell / Allied Signal make or do ? click here to find out.
Before we get started, observe this message:THIS IS A YEAR 2000 READINESS DISCLOSURE WITHIN THE MEANING OF s.2392, "The Year 2000 Information and Readiness Disclosure Act"
It's mentioned everywhere you turn.
[When "they" place it in every subheading on the matrix, this leads to: Me thinks I smell a RAT.][This was all over the original matrix, but the matrix was seriously updated and many comments and "issues" were removed. Now the Matrix has the title "Year 2000 Product Readiness" all over it, I think it was supposed to be a link to the disclaimer, but someone omitted the link -- I have not found a link to the actual text within any text that refers to the # disclosure @ the bottom of this page, the search function found it though. ]
# [See this Matrix, as an example]
[You know all of these companies are still going to get bitten, when lawsuits OUTSIDE the USofA start happening.]
SCADA - Supervisor Control and Data Acquisition == computers that make our current systems possible.
And we fired all the folks who know how to do it manually, too.
# See this example of what a SCADA does. Read several of these links.# Manufacturing Supervisor and the need to migrate to PlantScape SCADA
Find this about 2/3 down.
This is a big web page it will take some time to load.
From the matrix of tested products. -- and has not been updated since OCT1999
In a differnet matrix, see the following.
Mark cited the # Honeywell TDC 3000s; they are classified as
W == Withdrawn! Meaning pulled from the market -- no longer sold.
Mark, you better do some MORE checking!# Industrial Systems (ISD) [ emphasis mine] "Note: Windows NT - 1) going from SP3 to SP4 needs to be done by a knowledgeable person as it can cause severe driver issues, preventing a system from even starting. A backup is essential. 2) Some of the third party packages installed on the system might not be SP4 ready."#
# Alcont and Printa Control Systems They cite Windows NT is fixed if patched with service pack #3 (SP3). Well sp3 leads to sp4 which was promptly fixed by sp5 which had other problems. I think we're up to sp7 within less than a year, that's a lot of bug fixes for something that's supposed to be stable / finished.
# Pulp and Energy Solutions section of the matrix. [ emphasis mine]
"These are not products. They are customer specific applications. Not possible to Test Y2K."
Lime Kilns, Evaporators, Caustizings, Recovery Boilers[All not tested! Think: Paper, Diapers, Feminine Hygiene Products and Energy plants just for fun.]
# See the Third Party Vendor stipulations.
See half way down.
Product Advisory Bulletins All are from Fall 1999, humm is this just being looked into?Go read these, I've included a few. I don't get warm fuzzies from these Mark! Just like this one.
Clock And LCN Certifications Required
A customer was doing some Y2K testing on a system and encountered some problems. It turned out the problems were caused by having some clock-related hardware problems before the testing was initiated. While this in not just related to Y2K testing, all customers MUST ensure the system is operating/communicating normally and with a GOOD CLOCK system. It is particularly important on systems with mixed analog and digital clock subsystems.
[You can't test to see if they're lying to you, without worrying about your system locking up. BellSouth found OVER 50% of the vendors LIE! -- see GaryNorth.com for the link for this one.]
GUS Node Precautions (or something like this)
An LCN time change greater than 3 hours may cause Global User Station (GUS) nodes to go to a SEVERE state, if GUS Displays are running GUS trends during the time change. When a GUS node goes SEVERE due to this problem, communication between the GUS Displays and the LCN will be lost. In all known cases the LCN Native Window continues to work correctly. A reboot of all affected GUS nodes is required to recover.
[Go out on that ladder in the heavy snow and open the frozen door and press the reset button, it's just that simple.]
To prevent any possibility of this problem occurring, turn off GUS Display Trends before advancing the LCN system time.
As previously communicated, Honeywell also recommends Y2K testing be done off-line or on a development system.
[How many plants have an off line systems? or spare development systems?]
If necessary, a small (Less Than 3 Hours) LCN time advance can be made with online GUS nodes running trends from Displays. The actual length of LCN time advance that can be safely made varies widely depending on the trend data collection rate and the trend time base. However, less than 3 hours should be safe regardless of these variables.
Honeywell is planning to develop a software enhancement during 1st quarter 2000, that will eliminate the 3-hour restriction while running GUS trends. At present, this enhancement is planned for initial introduction in GUS R210 software.
[I like this one: We'll fix it next year]
Safety Manager - Triconex Y2K Testing Anomaly
[Anomaly in a safety system, UH OH!]
After changing LCN time from 4 November 1998 to 31 December 1999, SOE journal stopped reporting events. Discovered that Triconex time was 14 October 1998. System: 2 LCNs, 2 NIMs (HPM, SMM), etc. staged in Phoenix for FAT. SMM Software: Release 531.3 SMM H/W Ver. b05 F/W Ver. 4508 Tricon Ver. 8.2.3 Tristation MSW Ver 2.0.1 Procedure: Was conducting recommended Century Rollover Tests Analysis:
SMM sets the Triconex date/time only when the SMM is first loaded. Thereafter, Triconex time is synchronized only for changes not exceeding a certain magnitude, which is determined to be approximately 20 days (a change of 1 month sets the Tricon time to a wrong value). Conclusions:
1. SMM synchronizes Tricon correctly when date/time changes are not very large, i.e. not more than 20 days.
2. Need to shutdown and reload the SMM if changing the LCN time by more than roughly 20 days.
[Hmm, this sounds like fun: Increment the date by 19 days and then keep doing it that way until you get to the test dates. So something that should be easy is a long drawn out process to be repeated all over the facility -- Yup, boss I tested it works just fine -- Are you crazy? I'm not gonna crawl up there and there and there to test these stupid things. I hope they weren't put in nooks and crannies, we know that some were]
Hiway Equipment Loss of Power Precautions
[Yeah it's an old box, but it works swell. Or OOPS I forgot about this one. Or the budget is tight and I want the desk clock for my desk.....]
Over the years, some customers have reported that they try to avoid losing/dropping power to older Hiway equipment because some boxes do not come back due to various reasons. Some customers have also reported they are not always performing the recommended preventive maintenance on Hiway equipment due to lack of time or personnel to do it. Customers should be aware of this and take some necessary precautions prior to December 31st in case there are extended power failures to their plants as a result of Y2K issues from their local power company. Customers are reminded of the following:
Customers should recognize the importance of Preventive Maintenance particularly on Data Hiway equipment. Refer to the Customer Resource Manual for details. Make sure all box (Hiway) and Station databases have been saved. It is further recommended that more than one copy of the databases be available on removable media. If power has not been lost/removed for an extended period of time, particularly to Hiway boxes more than 8-10 years old, there is an increased chance that some Data Hiway devices may not restart if power is cut for any extended length of time. Customers should ensure adequate additional spares are available in case there is an increased failure rate due to power failures plus all UPS, UAC, BBU, and batteries are fully operational.
[Sounds familiar? Save your records. Oh, if the juice does not come back on before then. You can do it manually, cause this system can be used as a mini-fish condo. (Fish condos are things dropped into a lake or ocean for the little fishies to hide in.)]
There are currently significant discounts on Universal CPU (for CB/EC replacements), Universal Combo I/O, Universal Data Hiway Interface, and Universal Reserve Controller Director. Users are encouraged to take advantage of this to purchase additional spares and replacements particularly for some of the older boxes.
[We want your old boxes back, maybe we'll sell them to the third world market. I hope not!, these need to recycled into fish condos]
In summary:
WORRY, and PREPARE, PREPARE, PREPARE, and then WORRY some more.
I maybe wrong about this, I can live with myself if I am.
Will you be able to live, if you are wrong about this?For details on what to buy see Fuzzy's shopping list.
Fine print: I'll say it too:
THIS IS A YEAR 2000 READINESS DISCLOSURE WITHIN THE MEANING OF s.2392, "The Year 2000 Information and Readiness Disclosure Act" .
The above was created under the fair use quotations, as the information is for public benefit.
All names and intellectual properties belong to their respected owners, respectable or not.
Copyright 1999, all rights reserved. Permission is granted for non-profit distribution, otherwise send me a small piece of the pie.
-- (perry@ofuzzy1.com), February 02, 2001
You are right Perry, the page does meandor....What are you trying to say? In a sentence or two. ThanksPreped in PA.
-- whatareyou tryingtosay (whatsup@dot.com), February 02, 2001.
I'm saying that some problems we're reading about here could very well be caused by PLCs going duhh.
One thing is for sure, we won't find out soon.
And then there's always humans to compound those errors.
-- (perry@ofuzzy1.com), February 02, 2001.
There are an increasing number of posts on this in the last few days on GICC and on EZ Board. If you don't have the URLs, I will be glad to provide you some. I will include some information below.Regards,
Paula Gordon http://www.gwu.edu/~y2k/keypeople/gordon
http://hv.greenspun.com/bboard/q-and-a-fetch-msg.tcl?msg_id=004Vzv
http://pub5.ezboard.com/fyourdontimebomb2000.showMessage? topicID=21122.topic
______________________________________________________________________ _____________________________________________
Thread from GICC Regulators blame mechanical wear for off-line California plants http://hv.greenspun.com/bboard/q-and-a-fetch-msg.tcl?msg_id=004XWU
greenspun.com : LUSENET : Grassroots Information Coordination Center (GICC) : One Thread ---------------------------------------------------------------------- ----------
Regulators blame mechanical wear for off-line California plants HOUSTON, Feb. 2, 2001— Federal regulators who inspected out-of-service California electric plants did not find any evidence the audited companies scheduled maintenance or outages to influence prices, says a report released Friday. The staff of the Federal Energy Regulatory Commission (FERC) reviewed about 60% of December outages after questions were raised about the high number of plants taken out-of-service in the midst of continuing threats of rolling blackouts, complaints about high prices, and fear the state's investor-owned utilities would not be able to pay their bills. Rather than manipulation, the regulators said the audit determined the outages occurred at generating plants that were 30-40 years old which were being operated at much higher rates than in previous years. The report attributed most of the problems to mechanical wear, including tube leaks and casing problems, turbine seal leaks and turbine blade wear, value failure, pump, and pump motor failures. Rather than deliberately keeping the plants out of service, auditors said the owners appeared to have taken whatever steps necessary to put the units back in service as soon as possible "by accelerating maintenance and incurring additional expenses." Moreover, inspectors said the outages and price movements didn't necessarily correlate. FERC auditors found Southern Energy Inc., now Mirant Corp., continued to run plants with boiler tube leaks at reduced capacity during the period. The company told inspectors units are taken out of service, if leaks become severe enough to create an environmental or safety hazard. With the generating plants running hard, maintenance needs and costs have also risen. At Reliant Energy Inc.'s Coolwater plant, auditors said maintenance costs rose to $23.1 million in 2000, up from $1.2 million in 1999, and $1 million in 1998. "This confirms what we've been saying from the very start—that the power producers have been working around-the-clock to help keep the lights on in California," said Jan Smutny-Jones, executive director of the California Independent Energy Producers. "Hopefully, today's FERC report will quiet those who have tried to blame the producers for the crisis, when in fact we have been and continue to be an important part of the solution to California's energy woes." He noted member companies have invested billions of dollars buying, modernizing, and building power plants in California, and are preparing to spend billions more to ensure that California can keep pace with the growing demand for electricity. http://elp.pennnet.com/Articles/print_screen.cfm? PUBLICATION_ID=34&ARTICLE_ID=91427
-- Martin Thompson (mthom1927@aol.com), February 02, 2001
Answers "The report attributed most of the problems to mechanical wear, including ... casing problems, turbine seal leaks and turbine blade wear ..." I am an experienced turbomachinery design and analysis engineer, and if such problems were as rampant as the report says, I would be working on solving them, and not be unemployed. The turbomachinery manufacturers are in no panic to troubleshoot or solve any such problems, or to retrofit and uprate the turbines. This is my field of technical expertise, and demand for these skills I know, from personal knowledge in the job market daily, is weak. I believe that the ultimate report conclusion is correct, but that the true short term cause, Y2K computer bug problems in embedded control systems, is being covered up. However: It is definitely true that overdriving power plants will shorten life and increase failure rates over the longer term, which is an omen of worse (cascading) effects to come.
-- Robert Riggs (rxr.999@worldnet.att.net), February 03, 2001.
______________________________________________________________________ ____________________________________
-- Paula Gordon (pgordon@erols.com), February 03, 2001.