ML20064E851

From kanterella
Jump to navigation Jump to search
Forwards Info Requested by K Desai of NRC Staff During 781106 Meeting Re Svc Water Spray Piping Sys.W/2 Oversized Drawings Available in Central Files
ML20064E851
Person / Time
Site: North Anna  
Issue date: 11/21/1978
From: Brown S
VIRGINIA POWER (VIRGINIA ELECTRIC & POWER CO.)
To: Harold Denton, Parr O
Office of Nuclear Reactor Regulation
References
NUDOCS 7811240111
Download: ML20064E851 (6)


Text

.

gg VIRGINIA ELacTalc Axn Powna COMPANY Ricnxoxn.Vrnorw2A 20261 i.

~

November 21, 1978 J

Mr. Harold R. Denton, Director Serial No. 641 Office of Nuclear Reactor Regulation LQA/ESG: jab Attn: Mr. O. D. Parr, Chief Light Water Reactors Branch No. 3 Docket Nos.

50-338 Division of Project Management 50-339 U. S. Nuclear Regulatory Commission Washington, DC 20555

Dear Mr. Denton:

Attached is the following information requested by Mr. K. Desai of your staff during his meeting with Vepco and Stone and Webster repre-sentatives in Boston on !!ovember 6,1978:

(1) Service Water Reservoir Spray Piping Stress Analysis (2) Service Water Pump Allowable Tilt (3) Discussion of Expansion Joint Margin i

(4) Connection of Service Water Spray Piping to the Pump House Information concerning postconstruction movements of the circulat-ing water intake structure will be provided in the near future.

(

ruly ur I

Ile

//m am C. Brown, Jr.

Vice Pr ident-Power Statio Engine ing and Constructi Attachments i

't 81124M i

5 o ~33 Y ?

So ~331 b

1 I

1 SERVICE VATER RESERVOIR FPRAY PIPING STRESS ANALYSIS L RTH ANNA UNITS 1 AND 2 VIRGINIA ELECTRIC AND POWER COMPANY si The North Anna Units 1 and 2 spray pond piping, stress analysis of both the-

[

carbon steel and fiberglass piping was performed in accordanca with' ND-3600,Section III of ASME Boiler and Pressure Vessel Code. Although ND-3600 of ASMR III,1971 (code in effect when the analysis was performed) did not cover fiberglass pipe, the analysis used the same requirements and thus j

included all loadings considered for steel pipe. The requirements for analysis of piping in accordance with ND-3600 (same as for NC-3600) are tos I

Ccndition 1; the sum of stresses due to effects of pressure, weight, and other sustained mechanical loads within Sh (allowable stress range for material in the hot condition) i Condition 2; the sum of stresses due to effects of pressure, weight, other sustained loads, and occasional loads, including earthquake, j

within 1.2 S

  • h t

Condition 3: 'the stress due to the effects of thermal expansion i

within Sg = 1.25 Sc + 0.25 Sh (where Se = allowable stress range

[

for material at room temperature), or the sum of stresses due to effects of pressure, weight, other sustained loads, and thermal expansion within SA+S*

I h

Besides the above requirements, the stresses in the piping due to the effects i

1 of relative settlements were calculated independently of any other loading or 1/2 S u for fiberglass piping and 3 S for steel and kept within 2 S t

e e

piping.

(Where Sfu for fiberglass pipe is the ultimate flexure strength at j

t room temperature.)

(

The piping analysis computer model starte with a fixed (anchor) point at

[

.the pump house wn11 on 24 in s*. eel pipe. The other supports and restraints I

on steel piping and fiberglass piping were appropriately modeled for com-puter analysi.s as restraining axial, vertical, or lateral motion as appro-t priate for the specific support.

Typical maximum stresses for the fiberglass pipe for the various conditions

[

outlined in the first paragraph of this narrative are provided in the Table on p. 2.

I The thermal flexibility, dead weight, seismic and water hammer analysis were performed for the piping system using the material properties of 3

steel and fiberglass as required. For dynamic analysis of the piping system, the nass of the pipe and fluid was multiplied by 2 3 in order to include the effects of submerged piping in the water.

i (s.

l l

J i

l l

..., ~.

i t

2 The allowable differential settlement value for the fiberglass pipe pro-vided in the FSAR is 1.75 in per 50 ft of pipe. This number is the allowable differential settlement between any two adjacent supports for the most limiting pipe, which is the 24 in dia pipe. The monitoring points were loctrted about 50 ft apart, which is a greater distance than the actual m.

distance between two adjacent supports on the most limiting pipe. The allowable differential settlenent number was established by making the con-servative assumption that the settlement measured could represent that between any two adjacent supports on a pipe, and not just those monitored.

TYPICAL STRESSES FOR FIBERGLASS PIPE Condition 2 Condition 3 Calculated 2911 3959 Allowable 4968 9140 O

W A

e

SERVICE WATER PUMP TILT NORTH ANNA UNIT 2 VIRGINIA ELECTRIC AND POWER COMPANY p.

~

The manuIacturer's requirement for alignment of the Service Water Pumps of

.011 inches per ft (1/2 deg) maximum allowable tilt is a " rule-of-thumb"

{

A to ensure that the pump shaft is plumb. Tilt measurements are taken j

from machined surfaces at the base plate and impeller with respect to a plumb line. The " manufacturer's requirement of alignment" for the 26 ft long pumps corresponds to.29 inch displacement. The magnitude of this.

displacement makes it relatively simple to measure. The manufacturer has indicated that a displacement of.5 inches would not adversely affect f

pump operability. Furthermore, the pump could operate with three or four times this displacement with no significant problem (e.g., excess bearing wear). The long-term results of operating at the maximum allowable displacement would be a slight bowing of the pump shaft.

The above discussion provides additional assurance of pump operability due to building settlement. However, these conditions of service will not be realized as stated in FSAR Response P3.6.

1 I

e' i

EXPANSION JOINT DISCUSSION OF IIARGIN The " allowable equivalent axial compression" is the resultant of the specified. allowable movements of the expansion joints divided by the number of convolutions. This paranater, as determined by the manufacturer

'4 in accordance with industry standards, is.7130 in per convolution for our expansidu joints. The resultant differential movements corresponding to.33 f t of settlement (plus seismic and deadioad) are.4697 in per convolution. At this " compression" the calculated lifetime of the expansion joints is greater than 39,000 cycles. The compression corresponding.

to.33 f t of settlement was conservatively determined by assuming that all differential movements (compression, tension, etc) are maximum at the same circumferentiti point on the expansion joint. This is not the l

case but provides aduAtional conservatism.

It should be noted that l

the only cyclic load on these expansion joints is seismic and that seismic displacements represent less than 10 percent of the resultant " compression."

At " compressions" less than.7130 in per convolution, the expansion joints absorb novaments without significant distortion. To emphasize the degree of safety that does exist, the manufacturer artificially imposed move-ments corresponding to.7130 in per convolution on the expansion joints.

Even with this assumed loading, the expansion joint was still suitable for 2,585 additional cycles. This lifetir.e far exceeds the expected cyclic lifetime for the expansion joints.

The above discussions have assumed movements which do not exceed the elastic limit of the expansion joints. However, tests performed by the manufacturer indicate that movements greater than the " allowable compression" will cause the " convolutions to crack and remain integral.

Cyclic movements superimposed on this condition will cause fatigue cracks to occur." These fatigue cracks are "approximately pinhole in size, and i

in no case did the fatigue cracks propagate to the extent that there was any concern that a guillotine failure would take place." The above comments were based on a test conducted at conditions which exceeded the design linit by 10 percent of the design cyclic lifetime.

'i I

h

i SERVICE WATER RESERVOIR

?

CONNECTION OF SPRAY PIPING TO PUMP HOUSE g,

NORTH ANNA UNIT 2 VIRGINIA ELECTRIC AND POWER COMPANY As shown in drawing 11715-FP-5Y, four 24-in steel pipes extend from the south wall of the pump house to hangers approximately 35 ft to the south that are mounted on a common footing. Two feet beyond the hangers are the flanges to which the fiber glass pipes are attached. The center lines of these pipes were placed at el 307. Settlement-monitoring rods have been placed on two of these hangers: H-569 and H-584.

The block-out in the south wall of the pump house for the 24-in pipes was filled with concrete (thus, fixing these pipes in the wall) between April and June 1975. At the location of the hangers, reinforcing pads were welded to the pipes in June 1975. Therefore, the hangers could not have been assembled around the pipes (thus, fixing the vertical position of these pipes to that of the footing) until af ter June 1975. Because of the times at which the ends of the pipes were fixed to the pump *aouse and the footing, only differential settlement since mid-1975 would have affected thesa pipes.

Settlement of the south wall of the pump house is monitored at points SM-8 (S&W 3) and SM-9 (S&W 4), as shown in Figure 2 of the May 31, 1978, report. Settlement of the footing is monitored by rods acunted on hangers P-569 and P-584, as shown in FSAR Figure P3.9-1.

From mid-1975 to the present, the settlement of the southeast corner of the pump house has increased from 0.195 to 0.275 ft, as shown in Figure 3 q

of the September 8,1978, report, giving an increase-of. 0.080 ft.

In the same period, the settlement of the southwest corner has increased from O.350 to 0.490 ft, as shown in Figure 4 of the September 8,1978, report, giving an increase of 0.140 ft.

Since the westernmost of the four pipes is about 1/4 the distance from the southeast corner to the southwest corner, the increase in settlement at that point is:

3 (0.080) + 0.140 = 0.095 ft.

4 Therefore, the differential settlement along the unsupported 35-f t lengths of the pipes spanning between the pump house wall and the hangers is currently 0.015 ft.

Calculations have shown that the allowable differential settlement across this span is 2.1 in. or 0.175 ft since mid-1975.

..