Page images
PDF
EPUB

329.3. These tests were made on October 8, 1977, before any adjustments were made to the circuit controller. On October 9, 1977, a signal maintainer adjusted two contacts in the circuit controller because the track forces had changed the track alignment through the switch. He did not remember which two contacts he adjusted.

On October 12, 1977, the circuit controller unit was removed from service for further testing and inspection. When the circuit controller was operated a number of times, it was found that contacts Nos. 1, 2, and 3 failed to close completely. The phenolic terminal board, which contains part of the contact assemblies, was loose. The controller box was disassembled for a detailed inspection and a flat spot, not considered significant, was found on the No. 1 operating cam. Excessive play also was observed in the No. 4 contact roller.

The signal block governed by signal 330.7 was 6,336 ft long. The a.c. track voltage operates a three-position polar a.c. track relay. Inspections and tests of signal and track relays at signal 330.7 disclosed that they were all operating within specifications, and there was no evidence of conditions that would cause sticking contacts or improper operation. Tests were made at signal 330.7 to determine if foreign current was present at a sufficient level to operate the signal relays or cause them to fail to operate. There was no measureable foreign current detected. Also, insulation resistance tests of the signal instrument case wiring and the case-to-signalhead wiring was made and no faults were detected.

Other Information

A Southern conductor testified that on May 20, 1977, he was on a train that entered the Spencer Yard through the crossover just ahead of The Crescent. His brakeman had lined the south switch for the running track but had not lined the north switch for the main track when the conductor said he heard The Crescent "rev his engines up" and begin to accelerate as if it had received a "clear" aspect at signal 330.7. conductor immediately went to the north switch and lined it for The Crescent to use the main track. He learned later that The Crescent's engineer said signal 330.7 changed from red to green when the south end of the crossover was lined for the running track. The conductor reported the incident to a trainmaster in a casual manner, but no one investigated the incident to determine what had happened.

During the latter part of May 1977, or the first of June 1977, a freight train entered Spencer Yard through the crossover. A brakeman had lined the south end of the crossover for the running track but before he had lined the north end for the main track, he heard a following train begin to accelerate as though signal 330.7 had changed to a "clear" aspect. He called the engineer of the following train on his radio and

told him that the main track switch was still lined for the yard. The engineer responded that he had a "clear" aspect on signal 330.7, but he slowed the train so that the switch could be properly lined. The traincrew did not properly report the incident to a supervisor. The brakeman testified that he reported the incident to his conductor; however, the conductor testified that he did not recall the report or any action that he might have taken.

Southern signal department employees stated that they did not receive any report of a false "clear" signal in either incident. Federal regulations, section 26, 49 CFR Part 234, paragraph 234.0, require that "all failures of block signal systems...to indicate or function as intended" be reported.

(See Appendix B.)

ANALYSIS

Although they were not properly reported as required by Southern's rules and by Federal regulations, the instances in which engineers said that signal 330.7 indicated "clear" when the crossover switch was in a reverse position must be considered valid. The actions of the engineer of The Crescent in reducing speed from 70 mph to 50 mph in compliance with a speed restriction, the discussion of the infrequent display of a "clear" aspect on signal 330.7, and the engineer's general performance leading up to the accident also support the theory that signal 330.7 was displaying a "clear" aspect while the crossover switch was in the reverse position.

The adjustments made to the two contacts in the circuit controller on October 9, 1977, precluded a conclusive test of the controller's reliability. However, the failure of the contact which the signal supervisor observed after the accident, the contacts which failed during the tests on October 12, and the general condition of the controller assembly (loose terminal board and excessive play in the No. 4 contact roller) indicate a lack of reliability. Also, the circuit could have failed to shunt effectively, and thus permitted a "clear" aspect, because of the foreign matter found on the shunt wire plug and in the rail hole. The substance may have produced a high resistance that prevented the track shunt from "shunting" enough voltage from the track relay to allow it to become deenergized and cause a red aspect to be displayed by signal 330.7. A shunt would not have occurred if the contacts had failed to operate properly, or if the shunt wires had broken.

In spite of the fact that the controller operated properly each time the switches were tested on October 8, the cumulative evidence leads the Safety Board to conclude that the controller was defective and could have failed intermittently. Therefore, the Safety Board concludes. that signal 330.7 displayed a "clear" aspect while the crossover switch was in the reverse position. Because of the "clear" aspect, The Crescent's engineer acted in accordance with operating rules when he entered the signal block.

[ocr errors]
[ocr errors]

The application of the circuit controller in the shunt circuit at the Old Spencer Depot crossover did not meet the requirements of the FRA's RS&I. The installation of a series or break-type circuit a variation of the shunt circuit would have provided a higher degree of protection. Because the track shunt circuit protecting the crossover in the signal block governed by signal 330.7 was not designed to "fail safe" 4/, the Southern should have insured the protection of trains using the crossover through its operating rules and responsible personnel. Although Southern's rule 104 (c) clearly stated, "a switch must not be left open unattended," Southern's interpretation of the rule allowed the yardmaster to assume responsibility for leaving the switch open even though he did not attend the switch. Even though Southern's rule 1203 stated, "yardmasters must require that main track switches be properly lined and locked when not in use," the yardmaster authorized a yard crew to leave the main track switch of the crossover open and unattended, and he then failed to insure that it was closed.

Southern's rule 104 (e) did not specify that both switches of a crossover used during a train movement must be lined in agreement before the route is fouled. If this had been required, the engine crew of train No. 152 would have lined the main track switch of the crossover before fouling the running track switch of the crossover, and the collision would have been avoided. Having signal operations that are not configured to "fail safe" is unreliable enough without the added deficiency of operating rules that do not specify clearly what is required of employees and supervisors in circumstances which rules 104 (c), 1203, and 104(e) are intended to control.

In spite of the signal component failure, this accident would have been avoided if the yardmaster had exercised his supervisory responsibility and fulfilled his assumed responsibility for returning the switches to normal. Whether or not the yardmaster instructed the switch tender to return the switches to normal, the yardmaster was responsible for insuring that the switches were lined properly before The Crescent arrived, because he had relieved the yard crew of that responsibility when they used the route about 11:00 p.m. on October 7.

The engineer and brakeman of train No. 152 failed to exercise good judgment when they operated their locomotive on the running track near the crossover while they knew that the main track switch was in the reverse position. When they occupied the switch area with cars of train No. 152, they assumed the responsibility of leaving the main track switch in the reverse position.

41

A term used to designate a railway signaling design principle, the objective of which is to eliminate the hazardous effects of a failure of a component system. Addendum A, Association of American Railroads Signal Manual Part 55.

CONCLUSIONS

Findings

1.

2.

3.

4.

5.

6.

7.

8.

The adjustments made to the two contacts in the circuit controller
on October 9, 1977, made it impossible to test conclusively the
controller's reliability.

The circuit controller unit did not perform as intended at the time of the accident; it may have been operating improperly as long as 4 months before this accident.

Signal 330.7 was displaying a "clear" aspect as The Crescent approached and the train's engineer operated consistent with operating rules.

The shunt circuit configuration used on the Old Spencer Depot
crossover is not based on a "fail-safe" design. The series break-
type available as an alternate circuit is superior and gives more
reliable protection.

The track shunt protection intended to be provided by the Old
Spencer Depot crossover when either or both ends of the crossover
were positioned other than "full normal" was not provided because
of a discrepancy in the shunt circuitry.

The yardmaster was authorized by the Southern operating rules to
control train movements on main tracks within the Spencer Yard
limits.

The yardmaster's authority to assume charge of a switch in absentia was not clearly defined by the Southern's rules, even though he was charged with the responsibility for all switches within the yard.

The enginecrew of train No. 152 did not discharge their responsibility to line the north switch of the Old Spencer Depot crossover.

Probable Cause

The National Transportation Safety Board determines that the probable cause of this accident was the failure of the switch circuit controller to cause a red aspect to be displayed at the entrance to the signal block. Contributing to the cause of the accident was the Southern's failure to insure by enforceable, specific rules that conflicting routes were not established for the Crescent and train No. 152.

RECOMMENDATIONS

As a result of its investigation of this accident, the National Transportation Safety Board recommended:

to the Federal Railroad Administration:

"Require that the track shunt circuit imposed by contact closure in a circuit controller be phased out as soon as practicable and a series break-type circuit, which will satisfy the requirements of the FRA's Rules, Standards, and Instructions, be used in place thereof. (Class II, Priority Action) (R-78-23)"

to the Southern Railway Company:

"Revise its operating rules to insure that they state as
specifically as possible the action that is intended, and to
enforce those rules pertaining to the operation of switches
and the reporting of malfunctions of the signal system.
(Class II, Priority Action) (R-78-24)"

[blocks in formation]
« PreviousContinue »