Lately I've noticed in the NRC daily event reports that the nuclear plants are being tripped or scrammed manually. See http://www.nrc.gov/NRR/DAILY/der.htm I had thought that the nuclear plants shut down/trip/scram automatically upon detection of a problem. For the past several days the Event Reports specify that the reactors were shut down MANUALLY following the detection of problems. DOES ANYONE KNOW WHAT OCCURS DURING A REACTOR SCRAM/TRIP UNDER NORMAL OPERATIONS? Is it possible that the nuclear reactors are being run manually to some degree rather than automatically? If so, when did this begin? Any nuclear plant engineers out there? Thanks

-- slza (slzattas@erols.com), January 25, 2000

Answers

SLZA: superb observation. I think ou may be onto somepin'.

Calling marcia and/or John H. Krempaksy (who has been mysteriously absent since the refineries started blowing up): this would be a truly worthwhile are for some quick statistical analysis: rate of manual shutdowns over the last 2-6 mos; vis--vis the rate of automatic trips/scrams/ etc. Marcia?

>"<

-- Squirrel Hunter (nuts@upina.cellrelaytower), January 25, 2000.

Not a nuclear engineer but I think the answer to your question is 'both'. Nuclear power plants must be operated within certain parameter values determined by (written) operational policy and procedures.

When something happens that it not allowed by operational procedures, there may be a requirement to shut down the reactor manually. There are also situations that would cause an 'automatic' shutdown but I'd guess that nothing is 100% automatic at most of these facilities - i.e. even an 'automatic' shutdown requires personnel to facilitate the orderly shutdown.

If Mr Cook is still lurking, I'm sure that he could give you some specifics.

-- Arnie Rimmer (Arnie_Rimmer@usa.net), January 25, 2000.

SLZA,You don't think for a Moment that any Engineer will jeopardize his or her Job,and blow his Horn about all that goes on in the Workplace,do you,particularly in a Nuke Facility??

NO NUKES is good NUKES.

-- liberator (Feeding@the Trough.com), January 25, 2000.

Ohh, Sharp eyes...
Great question.

-- Possible Impact (posim@hotmail.com), January 25, 2000.

No, they're supposed to take drugs to trip.......

-- Craig (craig@ccinet.ab.ca), January 25, 2000.

The oddest thing just happened. I thought I'd go to the NRC site and look up info on reactor trips. (I'm on WEB.TV) Three times I got to the Table of Contents and three times my system shut down and rebooted itself. Anyone else have that happen?

1) US NRC, 2) Nuclear Reactors, 3) Licensing, Operatons & Inspections, 4) NRC Inspection Manual, 5) Table of Contents.

A simple "Accessed Denied" would have been sufficient.

-- Trish (adler2@webtb.net), January 25, 2000.

As a former Navy Nuke operator, I can say that it was a source of pride for the Reactor Operator to beat the automatic scram functions by manually scramming when it was called for. I did it once myself. A good operator is always on top of what is happening in the plant. I don't know if civvie nukes are as cautious as the Navy nukes were under Rickover, but I'm sure that the civilian operators would scram manually rather than waiting for the automatic systems to kick-in.

-- John A. Shaffer (jas11@psu.edu), January 25, 2000.

Good paper; has much good info on "Nuclear Reactors".

http://www.bashar.com/GSP/houston1.htm

hope this helps some. Best,

-- steve (WhoCares@nymore.Right?.com), January 25, 2000.

Thanks John, I also was in the Navy aboard fast attack Nuclear subs.

I hated when we had any type of Nuclear incident, Admiral Rickover required enough reports to keep us busy for at least the next 72 hours.

He sure was ornery and required everyone to pay attention to the smallest detail. We did too.

-- Michael (michaelteever@buffalo.com), January 25, 2000.

To beat the automatic scram mechanisms to the punch may be a source of pride for the officer on duty exercising his diligence; not a source of pride in the equipment, if it becomes necessary to trip them manually BECAUSE they are not tripping automatically. What slza reports is an observation that the plants are being tripped or scrammed manually MORE OFTEN, which obviously in light of the possibilty of embedded failures locused on Y2K, raises the question of whether the manual trips are caused not by eagle-eyed and highly trained maintenance engineers beating the automatic systems to the punch, but by ordinary staff responding to problems that are -- perhaps -- going UNDETECTED by the automatic sxystems. BIG DIFFERENCE, needless to say.

>"<

-- Squirrel Hunter (nuts@upinaa.cellrelaytower), January 25, 2000.

Thank you all for your comments. I had noticed a lot of manual trips in the NRC Daily Event Report, however as I couldn't find the prior month's event reports on their website I couldn't compare the number of manual versus automatic. In reading the event report descriptions, it appears that a trip signal is being generated by the system detecting the problem (e.g., seems to be several events regarding feedwater circulation systems and water levels if I read the description correctly). What occurred to me was if a trip signal was generated, then why didn't the system automatically shut down? I was concerned that perhaps the system that actually did the shutdowns wasn't working and they were therefore doing it manually. However, based on the several responses above, it seems like competant nuclear plant operators purposely try to manually trip the system before the automatic shutdown kicks in.

If anyone knows where the NRC keeps the older daily event reports on their website, I'll try to compare current versus past months shutdowns.

Thanks again for your responses.

-- slza (slzattas@erols.com), January 25, 2000.

slza,

Previous event reports are found under the "previous reports" link on the front page.

http://www.nrc.gov/NRR/DAILY/drlist.htm

You'll find "Daily Report Archives" at the bottom.

If you riffle through those, you'll find a wide variety of both automatic and manual reactor trips. And many periods (check out May 1999) with a lot more reactor trips than we are having now.....

-- John H Krempasky (johnk@dmv.com), January 26, 2000.

The automatic shutdown are safety features. The operators are also safety features and their job is to indeed "scram" the reactor when any prescribed parameter is exceeded. If they fail to act and parameter continues to fall or rise then the automatic features are expected to kick in. The manual is not usually a last ditch effort to shutdown after failure of the automatic features but the reverse.

For whatever this information might be worth, there doesn't appear to be monsters under this particular bed.

-- Squid (ItsDark@down.here), January 27, 2000.

Thanks John and Squid. Slza

-- slza (slzattas@erols.com), January 27, 2000.

To continue a little bit from what Sir Squid of the Deep and other reactor operators said: the operator is present to run and control things, and of course, to operate the palnt within the boundaries (temperature, pressure, flow, operating conditions, etc.) that are required by the "rules" approved by the NRC (civilian) or Navy Reactors Division (navy) for that particular plant.

The emergency shutdowns are in place to protect the public (and the plant itself) from abnormal conditions - too high a temperature, too low a pressure, too few pumps runing, or whatever.

The "trip points" (settings) for these two conditions are different: in all cases, the operator's regular operating "limit" is less than that set for the automatic shutdown. As a plant designer, I WANT the operator to see things and monitor trends - AND TAKE ACTION - before the problem gets so high that the plant has to protect itself automatically.

For example: let's say I have a power "limit" of 100%. I teach my operators to run at full power, but not to exceed 100% power. (The plant then is designed with some safety factor to be able to actually operate at a higher limit: assume this higher limit is 135% "maximum power.") I then set several automatic controls, in case problems happen and the operator doesn't respond.

First, I'll set an alarm to sound if power reaches 102%. Then, I'll automatically start moving control rods in (called a cutback) at perhaps at 105% power to begin a slower shutdown if power exceeded the alarm point. Then I'll set a first-level automatic scram point (all rods are inserted immediately by springs and gravity) point at a little higher limit (assume 110% power), but still well underneath the design point of 135% power.

Of course, this automatic scram might not occur, so I'll add operating instructions to force my operators to also scram if power exceeds 110%. Then, I'll add a second (or third/fourth) automatic scram point - using different instruments in a different power supply and using different instrument wires - to also scram the plant at 115% power.

So, if power gets to 104% - the automatic cutback at 105% might not trip off, but the operators might use a manual override to reduce power. This becomes the incident report you see. Manual action happened before automatic action.

---...---...

Operating conditions change things too: we had one shore-based plant that required 4 fast speed pumps to remain operating (if two were lost, we had to shutdown.) But the automatic reactor trip limits for two fast rector coolant pumps were very high (because the plant was conservatively designed) and so you'd never readsonably expect to ever "automatically" scram if pumps were lost (assume shore power shutdown, and the pumps suddenly had no power.)

So, if/when shore power was lost, we would have to manually scram. Not because the trip limits were exceeded, but because the oeprations manual said we had to.

---...---...---

The feedwater and condensate controls are among the few systems in nuclear plants that were specifically affected by controllers that were actually affected by y2k-related controller problems.

-- Robert A. Cook, PE (Marietta, GA) (cook.r@csaatl.com), January 27, 2000.

do you REALLY want to force the operation of a protective feature ... relying upon its' proper operation to ensure safety? no.

-- J. Messer (coolmesser1@aol.com), August 20, 2004.