ML20042G369
| ML20042G369 | |
| Person / Time | |
|---|---|
| Site: | General Atomics |
| Issue date: | 05/04/1990 |
| From: | Asmussen K GENERAL ATOMICS (FORMERLY GA TECHNOLOGIES, INC./GENER |
| To: | Alexander Adams Office of Nuclear Reactor Regulation |
| References | |
| 67-1554, NUDOCS 9005140163 | |
| Download: ML20042G369 (45) | |
Text
-
-!L _
CENBRAL ATDhNCB May 4,1990 67-1554 Mr. Alexander Adams, Jr., Project Manager Non Power Reactor, Decommissioning and Environmental Project Directorate Division of Reactor Projects III, IV, V and Special Projects
)
Office of Nuclear Reactor Regulations i
U.S. Nuclear Regulatory Commission Washington, DC 20555
Subject:
Docket no. 50-163: Reactor License No R 67; Response to Request for AdditionalInformation
Reference:
Adams, Alexander, Jr., letter to Keith E. Asmussen, " Request for i
Additional haformation," dated April 4,1990
(
Dear Mr. Adams:
The three enclosures accompanying this letter are in response to your request (referenced letter) for additional information regarding General Atomics' 1989 annual report for GA's TRIGA Mark F Research Reactor. The three enclosures
- each address one of your three inquiries. For convenience, each of your requests are repeated singly on the cover page of the enclosure containing the corresponding _
response.
We-trust you will find the enclosures responsive to your request.
Very truly yours, Qd K W Keith E. Asmussen, Manager Licensing, Safety and Nuclear Compliance KEA/mk Enclosures cc: John B. Martin, Administrator, U.S. NRC, Region V g
9005140163 9003o4
{DR ADOCK 05000163 Ilj i
PDC 10t-55 JOHN JAY HOPKINS DFilVE SAN DIEGO. CA 92121 1194 PO. BOX 85608, SAN VIEGO. CA 92138 5600 (019)455 3000
y s
h BMIMRAL ATORNCE i
i l
e J
l ENCLOSURE 1 TO CA LEITER F1-1554 l
This enclosure responds to the following (i.e., item 1 from referenced letter):
l 1.
Your annual report discusses four FLIP fuel elements that could r.ot be removed from the core until the top grid plate was removed from the-reactor. Please provide additionalinformation concerning any measurements made of the magnitude of the s'velling.- What are the other clad deformities referred to in the report? Has a cause been determined for this problem?
Have any other fuel elements been measured for changes in diameter?
r W
T
(
10955 JOHN JAY HOPKINS DRIVE. SAN DIEGO. CA 92121 1194 PO. BOX 85608. SAN DIEGO. CA 92138 5608 (619)455 3000
4.
v r
i FLIP FUEL DAMAGE STUDIES
SUMMARY
REPORT in October-November 1984, the top grid plate in the General Atomics (GA) TRIGA Mark F reactor was changed from a circular to a hexagonal pattern.
During this conversion process, two FLIP fuel elements were found to have excessive swelling or bulging such that they would not go through the holes in the hexagonal grid plate; the holes have a diameter 30 mils greater than the outside diameter of a new fuel element.
During the following 18 months five more fuel elements had to be removed from service because of excessive swelling or bulging.
l Analysis and investigation of the reason for the excessive swelling were conducted from November 1984 to December 1987. All elements removed from service for excessive swelling up to that time had experienced significant but very diverse bumup End had been through two high power pulsing programs, one in 1975 and the other in 1980. Much of the burnup in the elements had occurred prior to any pulsing. Aside frcm the significant burnup and high power pulse operation, no other correlating factors could be found. Fuel elements with excessive swelling enconipassed the range from low burnup to high burnup. Several fuel elements have high burnup and were part of
- the high power pulsing tests in central (high power) core locations, and yet show no tendency to excessive swelling.
Neutron radiography was performed on six elements and four of the six evidenced
" cloudy" regions in the axial central half of the element. These " cloudy" regions have been seen before in the damaged FLIP fuel from the Texas A&M core. The " clouds" indicate regions where the hydrogen density has been reduced due to hydrogen migration caused by high pulsing temperatures.
Three fuel elements that had been removed from service and radiographed were destructively examined in the GA hot cell facilities during the first half of 1987. The major conclusion from this examination was that the excessive swelling or bulging of the several TRIGA-FLlP fuel elements at GA was caused by hydriding and resultant radial growth of the central zirconium rod. The radial growth was about twice the expected value, likely because of preferential radial growth of the high hydride (e phase) that was attained. The hydrogen migrated from the hydrided fuel - most of it likely rapidly and during the high power pulsing tests. The growth of the central rod cracked the fuel bodies into large radial segments. Continued thermal cycling of the elements 03 1
3 caused, in some instances, fuel segments to be progressively moved relative to each other because of particles entering the widened cracks when the fuel was cold, and causing increased stresses on the clad as the fuel was heated, in one case there is a striking example of one major fuel segment being displaced radlally relative to the adjacent segment and leaving a marked discontinuity of the outside diameter. This caused an external bump or bulge to occur on the clad.
The GA TRIGA-FLIP fuel was all manufactured as single segment extruded meats, 15 in. long. The central hole was drilled from each end, just slightly over 7.5 in. each way. The alignment of the holes was quite variable and thus the Zr rod was Inserted from each end - two segments, each about 7.5 in. long. It was not uncommon under these general circumstances that Zr rods sometimes had to be ground down (on the outside diameter) for complete insertion into the hole. Whether or not this was the case for any of the GA TRIGA-FLIP fuelis not known.
The above described mechanical Interaction effect can be progressive, as open cracks continue to have fuel segments and particles work or fall into them in either steady state or pulsing operation. In all likelihood, this is the major contribution to fuel 1
rejection by excessive swelling or distortion that has caused the rejections to be unpredictable; not correlating with such things as burnup or core location during pulsing operation, it is an unpredictable component because the clearance between the Zr rod and the fuel can be different in every fuel element, especially at the central
- axial location within the element. This causes varying crack patterns, shapes, and sizes, which in turn allows for great variation in interaction between fuel segments and particles.
The rate of severity with which the fuel damage occurs due to the hydriding of the L
central Zr rod is also very likely to be highly influenced by the frequency and magnitude of pulsing operation. Thermal stress cracking, hydrogen mobility leading to rapid hydriding of the Zr rod, and fuel segment interaction are all greatly accentuated by high temperature pulses - pulses producing temperatures much higher than the r:ormal l
cteady state operation temperatures, it is considered a reasonable conclusion that without the high power pulsing, the fuel element rejection due to " swelling" (actually mostly mechanical interaction of fuel segments) would have been greatly reduced and maybe eliminated. The radial growth of the central Zr rod alone is just not likely to be a great enough single perturbation to cause the diametral increase off 30 mils required for rejection of the fuel element from use - based on the dimensional changes calculetad to result from this effect and the l
t 03 2
4 r
.y effects of the present design clearance between the central Zr rod and the hole diameter.
Note also that there is usually some small degree of bending, seemingly constantly varying in time and magnitude, in every element. Superimposing the swelling and/or mechanical interaction effects on the bending does increase the chances of rejection due to detection in the bend gauge.
Normally, rejection of fuel elements due to bending is very infrequent and can be eliminated (or nearly so) by routine fuel.
measurements and.180 deg rotation of any fuel elements with bending greater than about 1/32 in.
All of the photos, data, and conclusions from the hot cell examination of the damaged TRIGA-FLIP fuel element from Texas A&M wer3 reexamined irt conjunction with the evaluation of.the rejected TRIGA-FLIP fuel from the GA Mark F reactor. As a result it was concluded that for the Texas A&M element the hydriding and resultant radial growth of the central Zr rod caused some cracking and exerted forces on the fuel which would increase the diameter of the fuel element. However, there was a lot more fracturing of the Texas A&M fuel compared to the GA fuel. This is most likely because of the much greater number of pulses performed in the Texas A&M core - 725 versus 40 at GA.
In the GA hot cell evaluations, the Texas A&M fuel photos were used as examples in looklog for porous regions in the fuel sections. Only two porous regions were identified, and they were not nearly as graphic as the Texas A&M photos. All of the sections from damaged areas of the Texas A&M fuel element had apparent porous regions near the outer diameter o the element. From the visual examinations of the fuel it would be concluded that there was significantly more porous fuel structure formed in the damaged, Texas A&M fuel versus that at GA. Tha porosity is caused by high hydrogen pressure within the fuel matrix during pulsing operation, and from these comparisons it would be concluded that higher temperatures were reached in the Texas A&M fuel.
Another significant difference is, that while the GA and Texas A&M radiographs 7
appear similar, the clouds are more faint and less distinct in the GA films. This coincides with the detection of far fewer regions of porous structure in the GA fuel. In fact, the cloudiness was such in the GA films that more regions containing the porous structure were expected than were actually found. One of the bases for choosing the locations for the metallographic sections was the locations of clouds in the neutron radiographs. A porous region was identified in only one of the three sections where cloudiness was evident on the neutron radiographs.
03 3
,f o
Two different damage mechanisms or TRIGA fuel have now been identified:
1.
Porous region swelling, where high hydrogen pressures within the fuel matrix have been generated during high power pulsing. The high pressures cause swelling and Iccal weakening of the matrix and some hydrogen is depleted from these regions. Continued high power pulsing apparently causes contin-ued swelling.
2.
Hydride-induced growth of the central Zr rod contained in the fuel body. This growth is likely greatly accentuated by high power pulsing operation and is radially preferential and greater than the design clearance between the Zr rod and the central hole diameter. The excessive growth of the central rod causes -
the fuel body to crack into large radial segments. Subsequent high power-pulsing operation (rapid heating and cooling) causes mechanical Interaction between the fuel segments and can result in a local increase or " swelling" of the fuel element diameter.
The damage to the GA TRIGA fuel resulted from the second mechanism. The Texas A&M fuel was apparently damaged by a combination of the two mechanisms.
The Sandia high power pulse test fuel (1966) was apparently damaged by only the first mechanism. However, no transverse sections of the Sandia fuel were taken for metallo-graphy or measurements and thus there is no documentation as to growth of the central 2r rod. The fuel was severely cracked however, and porous regions clearly identified. Thus there has been no quantification of the sweiling caused only by the first mechanism, porous region swelling.
Also, all fuel removed from service due to swelling by either of the above two
- mechanisms has been involved with very significant high power pulsing operation.
Table 1 shows a summary of all the TRIGA-FLIP fuel elements removed from service in the GA TRIGA Mark F reactor since its installation in 1973. The fuel is
- measured annually for length and bend, and also by a plate gauge for maximum diameter. The plate gauge (thickness 1.25 in.) has hole diameters from 1.485 to 1.5251n., in steps of 0.010 in. It should~ be noted that the present Mark F grid plate, installed in 1984 is significantly thicker than all other TRIGA grid plates,1.25 in. versus 0.75 to 1.00 in., and the holes are not tapered at the top and bottom. This has also contributed to the rejection of fuel, in that there is less latitude to maneuver fuel through the grid plate and elements with maybe only a 20 mil diameter growth are sometimes difficult to remove, where the value is much closer to 30 mils in the thinner grids with tapered holes.
03 4
l
._9
.f.
t, e
i TABLE 1
SUMMARY
CF FUP FUEL REMOYF0 FROM SERWCE 1
2 3 l 4 5
6 7
l 3 l 9 10 l 11 l12 l 13
-14 History Merk-F (R-67)
Puleing Removal.
Plate Gauge Mark-lli(M-100)
Poettion t
Average Axlel Last Lend Growth Height Swell Core Small Large Element Poettlon Date Why (in.)
(in.)
Poeltion (in.)
- (In.)
Burnup Position Core Core Remarks 6326 C2 7/21/01 Bond
> 1/16 0.048 OK
<0.020 H@
C C
C Hot cell
( M10 g)
PIE 6333 E18 7/21/81 Bend
> 1/16 0.029 OK
<0.020 High F/E E
E 6344 05 7/21/81 Bend
> 1/10 0.038 2
25 0.020 High D
D D
6352 015 7'/21'81 Band
> 1/16 0.031 4
16 0.040 High E
D D
6353 CIO 7/21/01 Band
> 1/16 0.051 2
15 0 020 H@.
D C
C 6377 E16 7/21/01 Bend
> 1/16 0.019 2
25 0.020 Medun F
E F
( 81010 g) 6332 F2 8/11/82 Bend
> 1/16 0.043 OK
<0.020 High C
D F/O 6330 F23 8/11/82 Bend
> 1/16 0.024 OK
<0.020 Medium 0
0 F
6394 F12 8/11/82 Bend
> 1/16
. 0.014 OK
<0.020 H@
F E
F 6415TC El 9/5/64 Bend
> 1/16 TC 4
14 0.040 Low-meoun B
B C
Hot cell.
(4 to 7 0) 6364 03 11/6/84 Swelling OK 1/16 0.021 4
18.5 0.040 High D
D D
6372 Cl2 11/6/84 Swetting OK 1/16 0.033 4
13.5 0.040 High D
C C
Hot cell
-l 6324 C3 7/16/85 Swelung OK 1/32 0.072 2
12 0.020 High C
D F
6327 E11 7/16/85 Swo6 ting OK 1/16 0.028 4
14 0.040 Hgh E
D D
6322 C1 8/12/85 Swelling OK 1/32 0.057 2
15 0.020 High B
B B
6357-E10 3/6/86 SwelHag OK 1/16 0.041 4
18-3/4 0.040 Hgh E
E E
6403 0 23 6///67 Bend
> 1/16 0.034 OK
<0.020 Low-medium ;
C/E F
F 6335 FIS 6/13/88 Bend
> 1/16 0 050 2
3 0.020 Modun F
F E
l 6340 EIS 6/13/88 Bend ;
> 1/16 0.056 2
14 0.020 High E
F F
5872 D10 6/13/88 Swemno OK 1/16 0.019 2
2 0.020 Low (< 4 g)
E ST G
5673 E23 6/13/88 Swelling OK 1/16 0.031 2
3 0.020 Low-medium o
> 1/16 0.003 3
22 0.030 High 0
C C
6370 82 6/27/80 Swelling OK 1/32 0.057 2
12 0.020 Medium o
ST o
5875 F30 6/27/80 Swelling OK 1/32 0.023 3
6.3 0.030 Medium o
ST 0
6880 F3 6/27/80 Swelin OK1/10 0.000 3
6 0.030 Low-modun
.O ST 0
5882 06 6/27/89 Sweleg CK 1/32 0.019 3
6 0.030 Low-medium o
ST o
6368 015 7/28/80 Leake+
Cid 1/32 0.046 2
14 0.020 High E
E E
1 03 5
4 g-
'g From Table 1, it is seen that 27 fuel elements have been removed from service.
between Ju y 1981 and July 1989. The first ten and four additional elements were removed because of excessive bending (> 1/16 In.). Experimental evidence shows that close observation and rotation of those elements that begin to show noticeable bending can reduce significantly the number of rejections. Thirteen elements have been rejected because of swelling and/or bulging, as discussed above and only one element was detected to leak fission products.
It is important also to point out the following unique features of the TRIGA Mark F-FLIP fuel and its operationi 1.
The fuel manufactured in 1970 consisted of single-billet,15-in. long, extruded fuel meats (versus the present 5-in long cast billets which have been man-ufactured for more than 15 years now). The long fuel billet presented greater problems with the straightness and matching of the central hole, drilled from i
each end.
2.
The reactor has been used for a great deal of development and proof-testing of various types of fuel and reactor operation, including several high power pulsing tests, mixed core operation, and power cycling from startup to full-power.
Based on evidence from destructive examinations of a select number of fuel elements, we believe that most all of the fuel rejection at the GA reactor was caused by i
results of the two unique features noted above. However, other TRIGA fuel users who perform significant high power pulsing of the core will also have some probability of experiencing the same type of fuel damage that we have described above.
l We have been monitoring, evaluating, and analyzing the problem since its incep-tion and believe it to be primarily associated with the GA core because of the combina-tion of manufacture and operating history. At the same time, we do not consider the' phenomena causing the fuel rejection to be a safety problem because it does.not lead -
l to fission product release. Rather, we consider it to be an issue associated with reactor 1
operations and economics.
03 6
CM M
'k L
t i
?
ENCLOSURE 2 'IO GA LETTER 67-1554 t
I This enclosure responds to the following (i.e., item 2 from referenced letter):
[
t 2.
Please provide the complete 10 CFR -:50.59 analysis package for the installation of the Personal Computer Console Monitoring System (PCCMS).
1 I'
(
.i l
i, l
l l
t-l'
('
t.
10955 JOHN JAY HOPKINS DRIVE SAN DIEGO. CA 921211194 RO. BOX 85608, SAN DIEGO, CA 92138 5608 (619)455-3000 r
INTERNAL CORRESPONDENCE FROM:
Junaid Razvi DATE: June 21, 1989 TO:
Reviewers
SUBJECT:
10CFR$0.59 Review Personal Computer Based Console Monitoring System for the TRIGA Mark F Reactor.
Part of 10 CFR 50.59 authorizes (1) facility changes as described in the Safety Analysis Report (SAR), (ii) changes in procedures described in the SAR, and (iii) conduct tests and experiments not described in the SAR pro-vided these do not involve a change in the Technical Specifiestions or an unreviewed safety question.
An unreviewed safety question involves (1) the increase of probability of occurrence or the increase of consequences of an accident or malfunction of equipment important to safety compared to that situation previously evaluated in the SAR, or (2) the possibility for an accident or malfunc-tion of a different type than previously analyzed in the SAR. or (3) the reduction in margin of safety as defined in the SAR.
With regard to the request for approval described herein, the above points have been considered and no reason has been found to prevent a satisfac-tory finding on this application of 10 CFR 50.59.
I 1
GENERAL ATOMICS TRIGA Reactors Facility Approval For Change Under 10CFR50.591 Special Experiment Ucense Amendment
Title:
Personal Computer Based Console Monitoring System for the TRIGA Mark F Reactor.
APPROVALS
_1 c
-in-Charge i
Date Physicist-in-Charge
)\\'
I/.21[ P/
Date /
' 39 ie
(%
L,d. Ovkum A Date
$ S.7/yg C p a e 6,
m Date l
l Reactor Safety h
b[d/hf
/
Committee (CSC):
/
b
%h
/
Date N
Other:
/
Date i:
RETURN TO TRIGA REACTORS FACILIT'l (MS-21) AFTER APPROVAL.
Routed to SRos/R0s on l
l l
7 FACILITY lODIFIGTIQi UNDER IOCFR50.59 PER90%L 034EUIER IESED OtNSOI.E MNI'ITRING SY!ffEM for t}m TRIGA }&RK F REACIIR L
1.
INIRODUCI'IQ4 This application under 10CFR50.59 is being subnitted to permit the 'IRIGA Reactors Facility to proceed with the installation of a persona'l ccatputer in the Mark F control rocxn. The ccxtputer will be interfaced to the reactor console as a system for nonitoring, recording and annunciating various pa-rameters during reactor operations.
The haniware and software have been developed so that a nultitude of reactor operating parameters can be centrally nonitored, displayed, and recorded - on a pemanent medium; and any abnormal conditions pronptly displayed to the operator for proper action.
At present, no systen exists on the Mark F which can perforin such a valuable function.
It was therefore decided that, with the powerful personal conputer based systems now available, such a system would be a significant aid in carrying out a variety of nonitoring activities associated with reactor operations, especially in the round-the-clock operating node presently necessary for Mark F op.arations.
To that end, we undertook a program-to specify and procure the necessary hardware, develop the software, and design the proper interface between the signal generating channels (both safety and non-safety related) and the ccarputer to be able to provide a nonitoring system for the reactor operator without any inpact on the control and-safety systems as presently installed.
The systen described and evaluated in this document is a result of this ef-fort, and approval is being sought to install and use this system on the TRIGA Mark F reactor.
l
Page 2 10CTR50.59 Application Personal Caputer Monitoring Systern for Mark F 2.
STSTEM IESGIPTIN The _ Personal Ccrtputer _ Console @nitoring System (PCCMS) is a cmputer based reactor console nonitor, trend recorder, and alarm annunciator. The syste is based on the rack nounted IBM 7532 cmpLtet* (IBM AT equivalent).
An analog and digital input and output board (Advantech FCL-714 Super-Lab data acquisition card) is used for input of data frcan each of the nonitored channels for processing by the nonitoring system.
POCMS continuously scans sixteen reactor operating paramters and displays the current values on a CRT.
The system also checks to ensure each input is within preset limits (windcws).
If an input falls outside of the opera-tor selected window, PCCMS will sound an audible alarm and post a " pop up window" on the Poms display, alerting the reactor operator to the changing condition.
PCCMS also logs data for trend displays on the CRT or printer, and offere the operator the ability to archive this data to a floppy disk.
The ccanputer system performs none of the license required control and safety functions; its sole purpose is to aid the operator in the muitar.ing and reporting of the various operating parameters. As such, it is an oper-ator aid during startup, shutdown, as well as long-term steady state opera-tions.
PCCMS operation has been designed to be cxmpletely hve=r=rv6=t of the operation of any and all control and safety systems, and further, is not designed to be a substitute for any of the real-time data displays presently provided on the reactor console.
- 3. HARDERE IESCRIPTIN In addition to the Super-Lab analog input board, the IBM 7532 cmputer is equipped with 640K of system mencry, an AST SixPackPlus card (extended
j Page 3 10CFR50.59 Application Personal cmputer Monitoring System for Mark F matory, parallel and serial ports, battery backed-up clock), 20 16 hard disk, a 1.2 Mb floppy drive, a 2400 baud nodem, an EGA video card and display, and a dot strix printer.
(see Appendix A for caaplete hardware p nent specifications, and Figure 1 for a system block diagram.)
All analog interfacing with the Mark F console is through isolation anpli-fiers (Analog Devices AD204 Miniature Isolation Artplifiers) with a
- 2000 volts cmrton node isolation. Thus, each input channel is galvanically iso-lated frcan the signal source for eximum instrunentation isolation and pro-tection.
'Ihe isolation anplifier outputs are differentially fed into the analog input board, where the A to D conversion is perforned.
Four analog isolation boards with four input channela each will be nounted near the cmputer. Figure 2 shows input isolation details.
The super-Lab analog input board has sixteen fully differential inputs with fourteen bit resolution and
- 0 25% accuracy.
The channel assignments are shown in Table 1.
It also contains sixteen digital inputs and sixteen dig-ital outputs (fifteen D/Os and all sixteen D/Is are available for future expansion), plus two fourteen bit analog outputs (both currently uncarnit-ted). The audible annunciator (used too warn the operator of an input out-side of the expected window) is also located on this board.
- 4. SYS'ITH CINFIGURATI0t4 The PCCMS nonitors sixteen individual reacter parameters (Table 1), which are displayed on the CRT.
Two basic interfacing techniques are used to maintain isolation between the Super-Lab card and the reactor instrumenta-tion signal source.
The first method converts a 4 to 20 mA signal into an isolated 0.2 to 1.0 volt output, which is fed into the input card.
The second method converts a 0 to 1.0 volt signal into an isolated 0 to 1.0 volt output, which is fed into the input card.
Each of the sixteen input l
~ - - - ~ ~ -
.m Page 4 o
e 10CTR50.59 A; plication Personal Ccrtputer Monitoring Syste for Mark F 1
channels are configurable for either type of input.
Signal sources frm the Mark F console are generally signal level nonitors already existing on the reactor instrumentation, which are nin directly to
)
the input isolators.
The only exceptions are those signals which fall un-der the 4 to 20 nA category (see Table 1). A resistor is placed in series with the signal source, and the resulting voltage generated across the re-sistor is fed into the ieolator.
It should be noted that the techniques utilized for the Poms data acquisition are basically the same as those used in the Mark I Digital Control Console, with the major difference being that PCCHS utilizes fully differential inputs, while the Mark I ICS uti-lizes single-ended inputs. The advantages of using differential inputs in-clude inproved rejection of conmon mode voltages and a superior signal to noise ratio.
In order to maintain naxistum isolation betwten the PCCMS and all reactor nonitoring and Safety Syst es, the Analog Devices AD204 Isolation Acplifi-era are used in conjunction with buffered outputs wherever possible.
For exanple, each Keithley safety channel has a buffered "nonitor out" output (analog signal, o to 1 volt) which is fed into a Soltec chart recorder for long term trend recording.
The Soltec chart recorder input nodule has a buffered "nonitor out" output also; this buffered output is than fed into I
the AD204 Isolation Anplifiers.
Thus, the Safety Syste is twice removed
(
frm the PUCMS, which further reduces the possibility of the PCCMS inter-I fering with the Safety Systs.
Analyzing two possible failure nodes for the Keithley channel inputs to the Poms show that if the inputs were to either short together or open, the PCCMS would annunciate a " signal low" l
alarm; neither the soltec chart recorder nor the Keithley Safety Channel l
will be affected.
Interface between the POCMS and the fuel tenperature channels is slightly different.
The Fuel Tmperature Safety System utilizes a 4 to 20 nA cur-
Page 5 j
i 10CFR50.59 Application i
Personal CaTputer Monitoring System for Mark F 1
rent loop (which is floating with respect to ground) as an output signal.
Sensing the current flow (and thus the measured tenperature) is accom-plished by measuring the voltage drop across a resistor. Analyzing failure nodes show that if the signal input leads were to beccane shorted together, the indication on the Pcms will fall to -250 degrees centigrade (generat-ing an alarm on the POCNS); indications on the console would not change, and the Safety Systen would not be affected.
Should the signal input leads open, the PCCMS output will again fall to -250 degrees Centigrade, as would the indication on the console, but in this case, the Safety System would not be able to trip on the open signal leads.
Even if this highl'y unlikely event were to occur, two of the three channels are still available to the operator for nonitoring and control.
Further, if a high comnon mode voltage were to fall across the floating current loop, the safety System would still be able to function, as the entire current loop is isolated fran ground.
- 5. SOFDERE DESCRIPTICH AND IHOGRAM OPERATION The PCms software has been designed to be versatile and sinple to operate through the use of menu-driven routines to perform the various functions.
All user interface is through the use of numerous menu-driven software rou-tines initiated by the reactor operator frcxn the main program screen. This accesses separate sub-menus to perfonn a variety of functions.
Five such separate sets of screens and utility programs are available for use.
Each is described below; Appendix B includes a series of flow diagrams which detail the program logic and information flow in the PCCMS software l
p3Ckage.
l 5.1 PROGRAM HEtU l
Four utilities are available under the Program nenu, including the nonnal Operate display for the continuous monitoring and display of all inputs (Figure 3).
During normal operation, the program scans and converts each l
l'
Page 6 o
10CFR50.59 4 plication Personal Ccrtputer Monitoring Systen for Mark F input into neaningful engineering units for display on the monitor.
It also checks each input to ensure that it falls within the expected operat-ing window, and alerts the operator through an external audible alam and warning wind >w on the display if the input falls outside of the window.
Current operational data is logged to an amhive file each minute, and alams and alam setpoint changes are logged to
" tattle tale" files as they occur.
Also available frm the PROGRAM menu are the Startup and Shutdown utill-ties, in which the nonitor program sets the operator up for a progranmed reactor startup or shutdown.
The program prontpts the operator for the desired ranp up or ranp-down (in kW per ntinute), stopping points (break points), and start time. The program then displays the next " target power" and " target time"; the operator adjusts the rod bank and ranp rate to match the target time and power. The final utility under the Program nenu is the Power calibration program, which logs time vs. testperature, and perfonns a least squares fit of the data.
It also calculates actual themal power, and anps per watt for each power channel, and, will generate a tine vs.
testperature graph of the power calibration.
5.2 ITTILITY ME21U During Poms operation, data is logged to an archive data file every minute for future analysis or data reduction.
This data may be copied fran the hard card to a floppy disk under the Utility Menu's archive backup utility.
Also included with this archive is a " tattle tale" file which contains all records of warnings to the reactor operator, and any changes made to the alarm windows.
Any time an input is found to be outside of its expected window, an entry similar to the normal archive is made in the tattle tale file, and the operator is alerted via screen displays and audible alarms.
The sec,ond program found under the Utility Menu is a screen dunp of the op-erational display to the printer.
F-Page 7
+
10CFR50.59 Application Personal Conputer Monitoring Systen for Mark F 5.3 CCNFIGURE ME24U Four utilities are available under the Configure Menu, including a program i
to change the screen, text, and background colors.
Also included are pro-grams to adjust the current power channel calibration factors (change the anps per watt constant), and to adjust the operational hours and kilowatt hours as necessary (due to shutdowns, end of the nonth, end of the quarter, etc.).
The last utility program under the Configure Menu allows the opera-tor to change the alarm windows for each input as necessary.
Any changes made to the alarm setpoints are also recorded in a tattle tale also.
5.4 GRAP11 ME21U The graph menu allows the oparator to plot the operational data for any in-put over any one hour, two hour, four hour, er eight hour time interval of data in the archive (the archive contains the last twenty four hours of op-eration data).
A previous archive may be loaded over current archive (af-i ter saving the current archive) for plotting of old data.
The graphs can then be dunped to the printer for hard copy display.
5.5 MESSAGE SCREE 24 l
The final screen is the PCCMS message boarti, where operators may leave mes-sages pertaining to reactor operations.
All messages are "stanped" with a date and time, and offer a centralized message center for c.mnunications between operating staff on all shifts. This feature is particularly useful during three-shift, round-the-clock reactor operations for relaying ccmiu-l nications from one shift to the next.
1 i
1
g t.
Page 8 10CFR50.59 AFplication Personal Cmputer Monitoring Systen for Mark F 6.
UltER PFA'ItRES AND PUIURE EKPANSIN i
6.1 RDDIE 10NITORING Since Poms will run on any emputer equipped with an ira display, data be-
)
ing logged and displayed on the control room emputer by PCms may be played back on any properly set up DGA system that is equipped with a nodem j
and telephone line.
After PCQ4S has been installed and experience has been gained with its op-eration, software will be developed to allow ruote monitoring of the Poms through a 1200/2400 baud nodem.
6.2 ADDITIONAL INPUTS The Super-Lab data acquisition card allows us to add additional inputs -
both analog and digital - for nonitoring and display. These additional in-puts are presently being reserved for future use as the nead arises.
7.
SAPETY EVAURTIm The proposed upgrade / addition of a cmputer based nonitoring system to the TRIGA Mark F reactor control console (i) does not introduce any unreviewed safety questions, (ii) does not cause an increase in the probability of oc-currence or an increase of the consequences of an accident, and (iii) does not inpair the ability of any of the control and safety systans to function as intended.
As sphasized earlier in this document, the proposed upgrade does not have any inpact on the existing control and safety systens of the reactor.
Any renote probability that the failure of a cmponent in PCWS will cause a simultaneous failure in the ability of any of the required safety systems to function as intended is further reduced by the use of
Page 9 o
10CFR50.59 Application Personal Cmputer Monitoring Syste for Mark F isolation anplifiers and buffered outputs where applicable (Section 4) to interface the input frm the power, fuel tenperature and other channels to the I/O board that interfaces these channels with PCQ4S.
Thus, the system has been designed to function independently of the operation of all of the control, safety as well as non-safety systems asscciated with Mark F reactor operations.
F PCCMS therefore, neets all criteria for facility nodifications as autho-rized under 10CFR50.59. We expect that the addition of this systen to the Mark F control room will allow greater operational flexibility and effi-ciency in the way operating parameters are nonitored and recorded on a real-tine basis, especially during round-the-clock operations.
The use of such a system can only be expected to enhance our ability to store and re-view operating data, both for routine operations as well as the review of any abnornal conditions that may occur at any time during operations.
The capabilities designed in PCINS - and additional capabilities that can be exploited in the future - nust be viewed as an aid to the reactor operator as well as facility management in conducting our operations efficiently and within the bounds of safe reactor operation as defined by the reactor li-cense.
8.
'IESTING AND INS'IRIIATICE The design of the system has been performed by 'IRIGA Reactors Facility per-sonnel who are licensed Senior Reactor Operators as well.
Further, these staff cambers are intimately familiar with operating requirements as well as design and operation of the reactor I&C system.
Specification and procurment of hardware cmponents, which are standard, off-the-shelf ites, was by William E. Hood, TRIGA Reactor Engineer with specialization in reactor instrunentation.
All the required software de-veloptent was performed by Phillip Rieke, SRO at TRIGA who also possesses i
{
l l
Page 10 10CTR50.59 Application Personal Cmputer Monitoring System for Mark F considerable expertise in software developtent and personal emputer appli-cations. Thus, the developtent of PCois has been totally indigenous to the TRIGA Reactors Facility.
The systere has been undergoing bench testing at the facility using signals frcrn test equipent for the past six weeks, during which software " bugs" have been found and eliminated.
We propose to install the system on the console during a two week shutdown of the !%rk F reactor, planned for the period June 26 - July 9, 1989.
The installation will be carried out early during this shutdown period, so that any additional bugs found in the sys-tem can be corrected before the resunption of round-the-clock reactor oper-ations.
4 l
e
Page 11 10CFR50.59 Application Personal Ccarpater Monitoring Systen for Mark F Table 1 INNr 09ME1 ASSINo Input Channel Function Signal 1
Fuel Tarperature 1 4 - 20 nA 2
Fuel 'Ibitperature 2 4 - 20 nA 3
Fuel 'hrtperature 3 4 - 20 nA 4
Power Channel 1 0 - 1 Volt 5
Power Channel 2 0 - 1 Volt 6
Power Channel 3 0 - 1 Volt 7
Self Powered Detector 1 0 - 1 Volt 8
Self Powered Detector 2 0 - 1 Volt 9
Self Powered Detector 3 0 - 1 Volt 10 Self Powered Detector 4 0 - 1 Volt 11 Pit Tutperature 0. 1 Volt 12 Pit Conductivity 4 - 20 nA 13 Log Power Channel 0 - 1 Volt 14 Danineralizer Flow 4 - 20 nA 15 Continuous Air Monitor 0 - 1 Volt 16 Start Up Rate 0 - 1 Volt
E 758L 602.86 Mrs MICROMoCUOR.
/4fNoty
\\
k U
\\7 PCCMS SY37Eb1 BUS A
MK F CONTRot contes MM PCCM3 Aun JAlfvT*
ost DAE 44.
3
(,,' 4 p A
PoweA. ch Ets
/
4\\
FuGL Tsv 9
- h (1SeonA)
ANAt.oc
.SdLS Powtuo f
DATA p ersc. Ton,3
(,t(, cu)
L DEM it4 P LovJ 6 za M m
1
/@r
- c. A R
?
N l'** *
(laEm)
E CoNooctuirt
- (41o54)!
I
[
406 C#4m6L srur ve un Ptr TL w a arix.e f
11k17-PantTt0 ON see.10eoM CLEAnpaiwt e e% e
?
ArtAlTER.
4b-l 5
cr wwu sea,a un
- i. z a NA T
%AT HAe DI;k R.oppy otsk l
Mr CAAD4\\
A A
U V
U a
V V
NOOEM 66A VIDEO y
EGA VIDEO ese M00 snuo CAea pisrt.A y 3
i"'
st APElfrURE paoac ua#
1 CARD Also Annabic on
(
Aperture Card 9Aost40 k.3-O n SYS-EV 3_00< J AGRAV
{
OAT.: 4 /5 99 nevisro DCCMS DRAWING NUM.ER I
l.
,w f
~
-. -.. _.. _... ~.. ~..... ~ ~.... _. -.. ~
t j
)
i 4
1-l l
Y t
t t
0 EO i
t e
a 4
3-m g
y v.
o o
4 lH U
q
! U bo o
i
~n>
i, n e
u I
A k
M i
a a
i
\\
/
?
(
e g
/
N H
i
\\.
x E
I J
P w
v g
R-
, r-c e
a g
a v
=
~
d 4
4 5
Q n
u O
,g
+00E t,t, w
w w
we 3
I:{s
?.
>j m
w-w W
l le!!
1 l
c a
g -w..we e
G
36 F8
/
oter #s g3
.I
3 O-c setr Orr
,a~a.~,
y
~r -
v.u a LOGf 1, < COM 2,
W p'.
nr 3%,+yo, N
i 2'o %
Ap2%
AD 204
's
- can
+ sse c rm i
tt w con c O TO I VT 4 -20 m A ISOLATOR (ama J
== 6-s1-99 3CCMS 999 9D
- 14. i_ _ __
e Fip< a cJ) W=4 by. Pccur ngn ta,~ p cuwe~e am
~ - - - -
- - ~ - -l
Progran lit i 1 i ty Cotrigure Graph
- liessaga ALT-X to Exit GENERAL ATOMICS l
TRIGA Reactor Facility PC Console Monitoring System l
i 1
I 15:04:55 Jun 5,
1989 l
l i
l l
l
Progran Uti1itg-Configure Graph flessega
' ALT-X to Exit
~l~; P D 1 l
SPD2 SPD3 SPDel 4 LOG K1 K~9 K3 ' -
100 120- -
1200- -
1280 8
8 0
0 JH 110-1100-1100-
- ~
100-1800-1808-g F'I'2 H3 POOL 1
0 0
0 0
-l 0~
9 900~
gg f
88-HH8-888-
_p 0~
Y00~
DEMIN POOL CAM
_3 FLOW COND 10 '
66-600-600-8.88 0.00 8
-4 58-500-500-gg l
400-400- ~
-S 43-gg Date:
Jun 5,
1989
-6 38-300-300-Tine:
15:09:53 10 28-2' %
200-IO~
100~
100~
i Rx Hoitrs Todag; 8h an
_g Rx Hours Total:
8h 8n 18 8-0-
0-l Kw Hours Todag:
0h 0m 1E-007 32.5 w
165 w
220 w
l Kw Hours Tota 1:
0h 0m
-11 DPt1 1.00E-009 1.00E-009 1.00E-009 i
F+1e a$)Isa Opa~ kf s' ?ROGRA+t H%..
,ev--
w m-
-c-u y-%+-
,,m.
___,w_e c
+ - - -
I h
- f-
+: Page 12 -
10CFR50.59 Application.
e Personal Ca puter Monitoring System for Mark F.
j i
rf.
r APPENDIX A l
~ Design Specifications for a Ca puter Based Monitoring Systen for the l
'IRIGA Mark F Reactor l
l 1
. I
~
I i
t'I i
-I t.!
-v
(
b
?
.?
i i.
5 t
r J
?
I
't t
9
's e
f Fl s
i h.
l
' E
\\
~
r..
b l -
?.
F 2
'I i',.
~
,_.;__m
.r.,..
1 l
Page 13.
10CrR50.59 Application Personal Cartputer Monitoring Systern for Mark F Design Specifications for Mk F PC Monitoring Systern Revision Date: 6-15-09 I. Hart 3 ware Ccatponents A. Catputer
- 1. IBM 7532 Industrial Cortputer (or equivalent PC/AT cartputer)
- 2. 80286 CIU
- 3. 8 MHz
- 4. 640K system mertory (256K base)
- 5. AST SixPackPlus Carti
- 304K ituttory
- serial port
. parallel port
. battery backed internal clock B. Data Storage
- 1. 1.2 ItK floppy drive
- external 2, 20 Mb harti disk C. Video
- 3. ITA adapter / display D. I/O
- 1. RS232 serial port
- 2. Parallel printer port
- 3. 16 channel I/O Doard
- PCL-714 Super _ Lab (14 Bit) A/D + D/A + DIO + Counter (American
. Analog Input (A/D converter):
Mvantek) 16 differential channels 14 bit, 25,000 sanples / see Input range: 45V to -5V, +1V to -1V
- Analog output (D/A converter):
2 D/A channels 14 bit, +/ SV full range
_ Digital I/0:
Input: 16 channels, TIL certpatible output: 16 channels, TIL cattimtible
. Counter:
1 channel of tiner/ counter Prograttmable pacer function
- P.O. 9122965
7 4
Page 14 10CFR50.59 Application Personal Conputer lenitoring System for Mark F D. Cmputer/ console hardware interface
- 1. 16 individual miniature Isolation Anplifiers (AD204KN Dip Package)
- AD246 Clock Driver 1
- P.O. #122984 (Analog Devices)
- 2. Variable gain and offset nulling or, each anplifier
- 3. Isolation Anp outputs differentially fed into I/O board for A/D J
Conversion.
E. Future Upgrades
- 1. Modem (for renote nonitoring)
- 2. WA adapter / display II. Hardware Installation A. Ccriputer is to be installed in the lower left hand section of the Mark F console. It will be on a sliding shelf to facilitate ease of access. Keyboard will be tenporarily located in the auxiliary console, and printer will be on sliding shelf facing the back of the console.
B. The nonitor will sit on top of the console approximately where the Thermionics printer now sits. The security nonitor and printer will be noved as necessary.
C. Isolation Anp circuit box will also be located in lower left hand drawer t>f console where cmputer is located. This will minimize cable runs Detvun this box and conputer chassis.
l
Page 15 4
10CFR50.59 AFplication Personal Ccmputer Monitoring Systen for Mark F III. Software A. Monitor Program
- 1. Interrupt-driven (PC tiner)
- 2. Reads / scans data frm I/O board continuously
- 3. Saves scanned data to mencry once each minute
- 4. Saves accumulated data (60 sets) to disk once each hour (HCTIE: The I/O board has the capability for using DMA, which means the translated analog data could be directly transferred to memory for faster access; as of this revision it is not known if we want/need this feature.)
B. PCCMS Program
- 1. Continuously display and update input data (as defined in the Input Configuration File "INPlTI'.CFU" and read in by the Monitor program)
- 2. Additionally, display the following: (as a minimum)
Reactor Operating Hours for current day Reactor Operating Hours for current month KW Hours for current day Kw Hours for current nonth
- 3. Menu drive operator input:
- a. Program menu: select one of the following programs:
- Startup (reactor startups during Thermionics operation)
- Power Cal (standard power calibrations)
- Operate (continuous long-term operations)
- b. Utility menus select one of 5 utility options:
- Copy archive file frcan hard drive to floppy disk
- Restore archive file frm floppy disk to hard drive
- Copy configuration files frm hard drive to floppy disk
- Restore configuration files frm floppy disk to hard drive
- Graph data - on screen line graphs of nost inputs for for probably day / week /nonth periods of time
- c. Monitor menu: enable or disable nonitor program functions:
- Monitor On/Enbabled: nonitor maintains displayed time, performs continuous scans, stores data once each minute, saves data once each hour
- Monitor Off/ Disabled: nonitor only maintains displayed tine I
43
.--__~-_.--_---____..n
Page 16 4
10G R50.59 Application Personal Cmputer Monitoring System for Mark F
- d. Configure menut select one of the following options:
- Input list /nodify settings frun input Configuration File
- Display set screen display attribates
- Renote ont enable retote nonitoring of input data
- Renote off: disable renote monitoring
- 4. Alert operator to any I/O channel value outside the upper and lwer limit setpoints by activating external audble signal and flashing value on screen.
C. Startup Program
- 1. Monitor reactor startups
- 2. Display the follwing during startups:
- Actual power levels
- Decired power levels (baoed on ranp setting)
- Keithley anp readings corresponding to actual and desired power levels
- Current time
- Time since startup began
- Current ranp setting
- Current Keithley range settings
- 3. Allow limited operator input BE NRE startup
- When startup actually cmmences
- Desired ranp(s) to full / operating power
- Keithley calibration factors
- Keithley range settings
- 4. Allow limited operator input APIER startup cmnences
- Redefine ranpa if conditions change
- Keithley range settings
- 5. Optional operator input once startup is COMPLETE:
- Run Power Calibration program
- Run Display program
1 Page 17 10CPR50.59 Applicat3cn Personal Cm puter Monitoring System for Mark F D. Power Calibration Program
- 1. Monitors reactor power and pit tenperature during power cals
- 2. Constant display of reactor ] power and pit tertperature
- 3. Save tatperature data at 2 minute intervals (to mencry)
- 4. Recalculate least squares fit after each tenperature save
- 5. Display updated lef inforation after each calculation
- 6. Save accurmilated data to hard drive after power cal empletion
- 7. Allow lbnited operator input:
i
- When power cal comtences
- When power cal cmplete
- 8. Display final results of power cal and percent error for K1/K2/K3 3
)
v 4
9
Page 18 10CFR30.59 Applio9 tion
. Personal Ca puter Monitoring System for Mark F APPDOIX B Flow Diagrams for POQ4S
Figure 1. PC Console Monitoring System - General overview PC-DOS Failed Initialization (Fig. 2)
Passed
?
Response Loop (Fig. 3)
Pronram Menu (Fig. 4)
P Key A
Startup/ Shutdown / Power Cal / Operate Utility Menu (Fig. 5)
U W
Key b
Copy Archive To Floppy / Print Screen Configure Menu (Fig. 6)
C Warning Setpoints/ Cal Factors Operating Hours / Display Colors Graph Menu (Fig. 7)
G
+--
Key n
K1/K2/K3/SPD1/SPD2/SPD3/SPD4/FT1/FT2/FT3/
l-POOL TMP/ POOL COND/DEMIN FLOW / CAM /SUR/ LOG Messate Screen (Fig. 8)
M Key Exit PCCMS7 ET-X No b
Yes Keys 1r PC-DOS l
l:
l
~
i Figure 3. Response Loop Yes Keyboard Input Pending?
I No U
Operate Mode Active?
w Yes u No y
Update operational Update Time /Dats Display on Logo $creen 9
Read Data (PCL-714)
Translate Data p-Check Data Against Setpoints - Warning On If Exceeded O
Update Channels Update Time /Date Update Rx/Kw Hours v
i Retrieve Keyboard Input u
Perform Appropriate Routine (Fig.1)
{
I Figure 4. Program Menu Response Loop I
e
'P' Key Pressed i
IP Display Program Menu And Wait n
H For Response qr Ir ir 1r u
Startup/
L'ower Cal Operate ESC Key
,p Shutdown ir 1r 1r t
Startup/
Power Cal Clear Window Shutdown Routine Routine t
l 1r i
operate Mode Flag Set?
No Yes ir V
Set Operate Flag Roset Operate Flag i
i y
l Display Logo l
IP 1r IF 1r w
e.
o,- &
Figure 5. Utility Honu Response Loop I
ir
'U' Key Pressed u
Display Utility Henu And Wait For Response ir IP tr qr copy Archive Print Screen ESC Key File To Floppy ir F
Copy Routine Print Screen Routine ir 1r
)
Figure 6. Configure Menu Response Loop i
ir
'C' Key Pressed 1r I
Display configure Menu And Wait t
For Response ir IP 1r IP 1r 1r
- Warning Cal Factors Operating Display ESC Key Setpoints Bours Colors a
v 1 V 4
1r ir Switch Active Video Page List Applicable Data Edit. Data IP Switch Active Video Page
s.
w-Figure 7.-Graph Manu Response Loop 1
n
'G' Key Pressed n
Il Display Graph Menu And Wait For Response k
t t
(All.16 Channels ESC Key Listed In Menu) u Graph Routine -
Graph 1 Hour of Archive Data For Selected, Channel Figure 8. Message Screen Response Loop t
n
'M' Kay Pressed V
Display Message Screen And Wait For Response u
Read / Enter Messages
x
- . =... :.
h GENERAL RfDMN*C -
i i
ENCLOSURE 3 TO GA LEITER 67-1554 7
This enclosure responds to the following (i.e., item 3 from referenced letter):
3.
Please provide additional. detail on the GA Environmental Surveillance-Program as applied to the TRIGA Mark F.
10955 JOHN JAY HOPKtNS DRIVE, SAN DIEGO. CA 921211194 PO. BOX 85608. SAN DIEGO. CA 92136-5600 (619) 455-3000
a
,i. - -
.x
)
GENERAL ATOMICS ENVIRONMENTAL MONITORING PROGRAM TRIGA Reactors pedlity l
1.
Emergency air sampling - Five (5) emergency air samplers are situated on the roof and around the facility. One air sampler is on the roof of the reactor facility, the other four are located in the area immediately surrounding the facility (within 50 feet)in the North, South, East and West directions. Each air sampler has filter paper (for particulates) and charcoal filters (for iodine collection). The samplers are not in routine operation but can be turned on immediately if a release is suspected. The samplers are checked operationally on a quarterly basis.
2.
Environmental ait sampling - Environmental air is sampled at no less than 15 locations on, adjacent to, and near the site in accordance with OA's SNM-6%
license requirements. The sampler filters are changed weekly, and analyzed for long-lived alpha and beta radioactivity. They can also be counted by gamma spectroscopy if a positive result is obtained. GA maintains high resolution gamma-ray spectrometers (HpGe and Ge(Li) detector systems).
For routine gamma analysis, the samples are composited and gamma scanned monthly.
3.
Liquid Effluent Monitoring - Effluent water is sampled daily from GA's main pump house. The xmple is then prepared and analyzed for gross alpha and gross beta concentrations.
Where radioactive materials are authorized for release to the municipal sewerage system, the liquid is filtered and held in a tank until samples have been taken and analyses performed to assure their release is in compliance with applicable laws and regulations.
l i
---,,l--,.
a-
6-,.l-
. -. "i','...,..,
- .,, o
- 4.
Annual soit vegetation and water samples - An annual environmental survey is conducted by collecting typical samples of soil, vegetation, and water.
Gross alpha, gross-beta, and gamma spectral analyses are performed on each sample. Gamma radiation is measured at each station. There are 16 stations used in the annual environmental survey. Some of these stations are near the TRIGA reactor facility (including the canyon drainage north of the facility), others are in other nearby canyons, in Sorrento Valley and in other offsite locations.
5.
Externni Radiation - Meter surveys are conducted around the facility-periodically. In addition, there are 26 area dosimeters located in: 1) each of the 15 environmental air samplers locations 2) in four (4) locations - one each in each of the four cardinal directions from the reactor facility (dosimeters are placed on the fence surrounding the reactor facility) 3)in seven (7)" site boundary" locations along the perimeter of the canyon north of the facility; this direction is in closest proximity to GA's " site boundary" but the dosimeters are actually located well within GA's site boundary. Much of the site boundary is located in unoccupied canyon areas.
6.
Room Monitoring - Both reactors have air samplers in the room to routinely sample room air.
7.
Exhaust Ventilation Monitoring - Each reactor has in place a beta / gamma detector (GM) next to the ventilation plenum. The detector output is to an alarmed ratemeter set to alarm at 5 mR/ hour.
8.
Pool Water Monitors - Each reactor has in place a beta / gamma detector (GM) in the pool water treatment system flow. The detector output is to an alarmed ratemeter set to alarm at 45 mR/ hour.
9.
Radiation &ga Monitors - Each reactor has at least two (2) radiation area monitors. The alert monitor alarms at 20 mR/ hour, and the high (alert) alarms and sounds the evacuation alarm at 5 R/ hour.
l 2-
,,-,,,,,...-,,,,---5
a
.,:a.
Mark i MONITORING ONLY 1.
Continuous Air Monitors - A Continuous Air Monitor samples the air above the Mark I reactor pool. These monitors have two alarm set points (alert and alarm) which are currently set at 10,000 counts per minute and 35,000 counts per minute, respectively.
Mark F MONITORING ONLY 1.
Continuous & Monitors - A Continuous Air Monitor samples the air above the Mark F reactor pool. These monitors have two alarm set points (alert and alarm) which are currently set at 10,000 counts per minute and 50,000 counts per minute, respectively. The Mark F reactor will scram at the alarm set point of 50,000 count per minute.
2.
Sisk Monitor - The monitor is continuously sampling air from the Mark F reactor room. This monitor uses a beta / gamma detector (GM) for to measure activity gas sampling. The stack monitor is calibrated for Ar-41, 3.
Sigh & Sampler - The Mark F exhaust is filtered through a prefilter and a high efficiency particulate air (HEPA) filter before being sampled and released to the stack. Two types of samples are collected; a glass fiber filter for particulate monitoring and a charcoal filter for iodine monitoring. The samples are collected weekly and analyzed for gross alpha and gross beta activities. The samples can be counted by gamma spectroscopy.if a positive result is obtained.
4.
CellIop Monitor - A beta / gamma detector (GM) is located above the pool within the reactor shroud. This will alert the operator of a potential problem with the in-core fueled experiments and is included here to provide information on readiness to respond.
.....-,.--...,,..m
yy
- '6=,C_4
- 5. _
Emergency & Filtering - The TRIGA Mark F revised technical speci-fications have strict criteria for reactor room air confinement, Further, to meet the R-67 Tech Specs, an activated charcoal filter was installed during I
1989 to function as an iodine trap during emergency situations related to a potential release of radioactivity from the in-core fueled experiments.
I l
4 9
f l
. _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ -