(1) Rock pockets.

(4) Broken bells containing circumferential reinforcement.

(5) Impact damage over less than 45° of circumference except for spigots.

(7) Out-of-round bells, if not so far out of round that reinforcement steel will be exposed after the repair.

(8) Air holes and roughness in gasket bearing surfaces of bell and spigots (Exh. 5N, p. 7).

The Bureau's letter of May 13, 1965, provided in part:

3. Experience has been that extensive repairs to bells and spigots has impaired the function of the joints. Pipe having imperfections or damaged areas that extend over six inches of gasket area in the bell 133 or four inches in the spigot will be rejected. *

4. Extensive repairs to rock pockets in bells and lack of consolidation of the concrete that will result in poor bond between the concrete and the steel will not be permitted.

Although it appears that rocky bells result from a lack of complete

133 Instructions precluding repair of rocky bells in excess of six inches in gasket areas and fallouts in excess of one square foot were in effect as early as April 30, 1965 (note 26, supra). This restriction was reiterated in the Tentative Instructions to Concrete Inspectors dated May 7, 1965 (Id.).

consolidation of concrete (Deposition, p. 75) and it is clear that the Bureau regarded rocky bells as repairable,134 it is far from clear that rocky bells are normally repairable in accordance with the Concrete Manual. However, the Government's argument that rocky bells are not repairable appears to be based solely upon Cen-Vi-Ro's alleged failure to take known corrective action and we assume that rocky bells are normally repairable as rock pockets, exposed steel, or as roughness in gasket bearing surface of bells. Although the Manual does not provide any area or size limitations on the repair of the above defects, we hold that all of such defects are not repairable and that some discretion may be exercised as to the extent of the area that is properly repairable. Bureau instructions in effect as early as April 30, 1965 (note 133, supra), precluded the repair of rocky bells in excess of six inches. This restriction was continued in the May 13 letter which, as we have seen, provided that pipes having imperfections or damaged areas of bells in excess of six inches would be rejected. Since the Manual allows the repair of impact damage to bells extending over less than 45° of circumference (45° on the circumference of a 54-inch pipe is in excess of 21 inches), the May 13 letter clearly prohibited repairs to

134 A Pipe Rejection Certification, dated April 2, 1965 (App's Exh. E), clearly indicates that the Bureau regarded rocky bells as repairable. An Inspectors Daily Report, dated July 29, 1964, reflects that only bells regarded as "too rocky" were not repairable.

impact-damaged bells which were permissible under the Manual.

While the Tentative Instructions to Concrete Pipe Inspectors of May 7, 1965, allowed the repair of damaged spigot gasket areas of six inches and below, the May 13 letter precluded repair of imperfections or damage to gasket areas of spigots in excess of four inches. Since breaks entirely through the shell in gasketed spigots which extend into or beyond the gasket bearing area and extend more than four inches around the circumference under the gasket are not normally repairable under the Manual, it is obvious that restrictions on repairs to spigots were much less extensive than the restrictions to repair of bells. However, read literally, the May 13 letter precluded the repair of defects in spigot gasket areas in excess of four inches irrespective of whether the break extended entirely through the shell and to that extent was contrary to the Concrete Manual.

We find that some unconsolidated areas in barrels and spigots were normally repairable as rock pockets in accordance with the Concrete Manual.

In its letter of June 10, 1965 (Exh. 5N, p. 15), Cen-Vi-Ro stated that its experience was that properly repaired bells and spigots have not impaired the function of the joints and that pipes so repaired were equal to pipes not requiring repairs. Cen-Vi-Ro also stated that a certain number of rocky bells was inherent in manufacturing pipe by the spinning process and that prop

erly made repairs to rocky bells and damage to spigot ends should be allowed in accordance with the specifications without the restrictions in the May 13 letter. The contracting officer found that the Bureau, in the early stages of the work, permitted extensive repairs to bells and spigots but withdrew this concession when it was determined that such repairs impaired the function of the joints (par. 37, Findings of Fact). The only evidence supporting the contracting officer's finding that repairs to bells and spigots impaired the function of the joints is some rather vague testimony by Mr. Rippon to the effect that repairs to bells and spigots which were not not accomplished within specification tolerances on other contracts resulted in leaking joints (Tr. 1752), and some testimony equally lacking in specificity by the resident engineer of reports to the effect that the laying contractor experienced difficulty in joining pipes with rough bells.135 Hydrostatic tests on joints produced satisfactory results (Page 4 of Special Report, dated May 21, 1965, note 24, supra). This is some evidence that joints were not defective. Roughness in gasket bearing surfaces of bells and spigots, is, of

135 Tr. 1888, 1889. A letter from R. H. Fulton to Cen-Vi-Ro, dated October 7, 1965 (Exh. 26) states that because of the marginal quality of the bells Cen-Vi-Ro had consigned repair crews to the field in order to avoid laying problems. However, a memorandum written by Mr. Herrera, dated October 11, 1965 (CenVi-Ro Correspondence) states that he has made periodic checks in the field, that pipes were above average requirement and indicates laying difficulties were attributable to prac tices of R. H. Fulton's laying crews.

During the May 15 inventory 278 pipes were rejected for damaged bells or spigots which includes 233 rejected for rocky bells, and 31 pipes were rejected for rock pockets.136 136 There is no evidence that any of these pipes had previously been accepted. It appears that prior to May 15 inventory 143 pipes had been rejected for bell and spigot defects and 12 pipes had been rejected for rock pockets (Exh. 59). In early August 1965, the Bureau relaxed the criteria for repair of rocky bells stated in the May 13 letter so that rocky bells which extended less than one-quarter of the circumference of the bell and which did not extend beyond the reinforcing steel when chipped out were repairable (Tr. 1289; memorandum of Bureau Meeting, dated July 9, 1965, Cen-Vi-Ro Correspondence; Inspectors Daily Report, dated August 4, 1965). The relaxation applied to pipes undergoing review in the reclaim program as well as to current production. Hydrostatic testing of these pipes does not appear to have been required (Inspectors Daily Report, dated August 4, 1965). The criteria in the May 13 letter for the repair

of bells was not further relaxed and the Bureau continued to reject pipes with bell defects or damage other

136 Pipe Units Rejected on May 15, 1965 (Exh. 60), and tabulation attached to memorandum, dated May 27, 1965 (note 27, supra). The tabulation indicates that 233 pipes were rejected for rocky bells and 25 for rock pockets during the May 15 inventory.

inches in bell area. 137 The Summary of Pipe Units Reclaimed (Exh. 146) indicates that only 34 pipes previously rejected for bell or spigot defects were accepted in the reclaim program during the seven-month period November 1965 through May of 1966. However, since there were 146 pipes finally rejected for rocky bells (Exh. 5Q), which includes at least 22 rejects manufactured subsequent to May 15, 1965, it is apparent that a minimum of 109 pipes previously rejected for rocky bells must have been accepted. The above figure does not include any pipes rejected for rocky bells prior to May 15, 1965, and we have no doubt that a majority of the 143 pipes rejected for bell and spigot defects prior to May 15 were for rocky bells. We have previously found that pipes marked for special hydro in the May 15 inventory were not listed as rejects in the summaries resulting from that inventory but that some of such pipes were, nevertheless, included in the tabulations purportedly representing the disposition of all rejects in the May 15 inventory (note 95, supra). We conclude that most, if not all, of the 233 pipes rejected for rocky

137 Mr. Herrera is reported to have inquired of Mr. Thomas why the Bureau was still rejecting pipes with rocky bells which could be repaired. The response was that the extent of defective concrete could not be determined until it was chipped out, the pipes were rejected until satisfactorily repaired, all such pipes were not repairable and in any event, Cen-Vi-Ro would never repair all of such pipes. (Inspectors Daily Report, dated August 19, 1963.)

་་

bells in the May 15 inventory were subsequently accepted.

Our examination of the Final Inventory of Rejected Pipe (Exh. 152) indicates that it contains 88 final rejects for broken or impact damage to bells, of which five were charged to Fulton, and 72 pipes rejected for broken or impact damage to spigots, of which four were charged to Fulton. It is not clear how many pipes were rejected for these reasons prior to the May 15 inventory or how many, if any, of the 45 apparently rejected for this reason in the May 15 inventory were accepted during the reclaim program.

The Hydrostatic Test Study (Exh. 64) reflects that 25 percent of 16-foot pipes subjected to special hydrostatic tests for rock pockets or unconsolidated areas failed the tests and that approximately 16 percent of 16-foot pipes subject to special hydrostatic tests for miscellaneous reasons failed the tests. Only three 20-foot pipes appear to have been specially hydrostatically tested for rock pockets or unconsolidated areas other than gyro areas of which two passed and one failed. Approximately 32 percent of 20-foot pipes tested for miscellaneous reasons failed the tests. While it may well be as Cen-Vi-Ro asserts that some of the above pipes classified as failures would have healed within seven days, we find infra under the heading of "Testing Criteria" that the contract placed the burden of proof that dripping pipes would heal on Cen-Vi-Ro.

We turn to the question of CenVi-Ro's alleged failure to take known corrective action. Dr. Davis attributed rocky bells to the failure to have sufficient concrete in the bell due to lack of vibration during the spinning process (Deposition, p. 75). The project engineer and the resident engineer considered that rocky bells were caused by difficulties in properly filling the bells due to the relatively dry Cen-Vi-Ro mix (Tr. 1889, 1890, 2043). This reason is supported by Mr. Murray's memorandum of July 8, 1965, (note 127, supra). Irrespective of whether one or the other or both of these reasons may account for rocky bells, we conclude that there was a corrective action for rocky bells which Cen-Vi-Ro either knew

or

as a reasonable skilled contractor is chargeable with knowing, namely, proper mix and proper filling of the bells. However, Dr. Davis testified that in spinning concrete pipe, segregation of aggregate tends to occur (Deposition, p. 75) and we accept as accurate Cen-ViRo's assertion (letter of June 10, 1965, Exh. 5N, p. 15) that a certain number of rocky bells is inherent in manufacturing concrete pipe by spinning methods.

The record reflects that 124 of the 146 final rejects for rocky bells were manufactured prior to May 15, 1965 (Exh. 5Q). Ten of the final rejects for rocky bells were manufactured in September (approximately 1.4 percent of pipe production), 24 in November (approximately 2.4 percent of pipe.