Tapline Pipeline Repairs.
Case histories of some of the major pipeline leaks and their repair.
By J. Makkinje, October 1999
(lIllustrations accompanying this article.)
The pipeline is 1213 kilometers long, of which 604 kilometers is above ground. These above ground sections are all located in Saudi Arabia, where the line traverses a sparsely populated rocky desert and no manmade structures were in existence at the time of its design and construction.
It was the first pipeline to be built without expansion loops to accommodate thermal expansion and contraction. Instead in places where the pipe was above ground, it was supported and restrained by ring girders that were bolted to small concrete piers at 20 meter intervals. Additional anchorage was provided for horizontal and vertical angles in excess of 45 minutes.
Vertical angles were kept to a minimum by varying the pipe to ground clearance. In places where dry river beds and short depressions had to be crossed, the ring girders were mounted on simple steel A frames that could well be 30 feet high. A picture showing both types of piers can be found in the article: Kilometer 1213:Tapline Today. Although this totally restrained type of construction had previously been used on large diameter thin wall water lines, Tapline was the first company to use this type of construction for their oil pipeline and the reasons were basically one of economics.
The Right of Way in Saudi Arabia in places had many large sections where lime stone rock was encountered and ditching machines could not be used, therefore having to bury the line would add to the construction cost . The only thing that was needed in the longer above ground spans were short buried sections about 25 feet wide where Bedouins with their camels could cross.
If large bends were required, very heavy anchors were installed to restrain the line at these points. These anchors consisted of very heavy sleeves around the pipe, which then were welded to very heavy H beams that were anchored in massive concrete blocks that in turn were anchored in the ground.
Such anchors were also installed at each of the stations near the scraper traps as well as in some of the longer above ground sections. I was always fascinated by the size of these anchors and wondered why they had to be that large. Only recently I was told how this came about.
Many of the old pipeliners were not convinced that this totally restrained pipe theory would work. J. L. Culbertson, Tapline's Chief Engineer at the time, was apparently one of them. Rhea Putnam who was the first engineer hired by Tapline, all the others were on loan from the parent companies, told me that he took the original design drawings from Bechtel to Culbertson for his approval. Culbertson took one look at the anchor drawings and flatly refused to approve them. He then told Rhea to return them to Bechtel and to tell them to double the size of all main anchors.
The pipe is subjected to many stresses but the longitudinal stress is the one that was always foremost in the mind of those who had to repair the line after its construction was completed. In simplified form the net longitudinal stress is when the combined effect of operating pressure and temperature varies from the combined effect of atmospheric pressure and installation temperature for both buried and above ground line.
The design operating temperature range was approximately 40 to 140 degrees F. and 80 degrees F. was taken as the tie-in temperature to keep the longitudinal stresses within a reasonable range . At the time of construction this temperature could not always be attained, without incurring long delays in construction. Therefore the nominal tie-in temperature was allowed to vary between 70 and 90 degrees F.
In addition the actual tie-in temperature for buried sections was allowed to vary by +/- 10 degrees F. and for above ground sections, including the first kilometer of buried line adjacent to it was allowed to vary by +/- 3 degrees F.
Normally tie-ins would be made of slightly more than one kilometer long sections, more frequent tie-ins were made in rough terrain. An accurate record was kept of the tie-in temperature of each section.
Detailed repair criteria and procedures were tested and adopted and it appeared that everything was going well. No major deficiencies in the pipe or welds were found during hydrostatic testing. The field welds were visually inspected and about 5 percent were inspected by gamma-ray. This type of inspection was used to check and quite probably more to encourage quality control.
Engineering made a survey of all leaks that had been reported during the period November 1950 through 1960. It lists Round Seam, Long Seam leaks by pipe size, month and loss of oil. etc. etc. One type of leak immediately catches one's eye, namely bullet holes. The reason for these bullet holes was due to the environment where the above ground pipe was located.
Consider the following:
From the above description how the line was constructed, it is not difficult to visualise that the above ground sections in particular were finely balanced and disturbing this balance could create many difficult problems. From my very frequent trips along the pipeline I can still remember seeing many ring girders in the Km 390 area that are leaning, indicating that the pipe at some time has moved longitudinally. What happened there, I don't know, but I recall that in 1954 or 1955 there were some major problems in that area.
(For a story about that job see the note at the end of this report) However well all aspects of the above ground pipe were engineered, as time went on the conditions in the area were changing. Some things had more effect than others and the most damaging ones were the increased amount of commercial trucking traffic along the pipeline road and the traffic normally associated with the increasing population of the townships in the pipeline area. This resulted in a high number of trucks hitting the line and then for a time sabotage by politically motivated groups. The damage from sabotage was mainly limited to the buried section in Jordan and therefore the repair was pretty straightforward. However for a period they were very frequent, once in one night we had three acts of sabotage all resulting in a fire at locations about 5 miles apart.
The damage caused by vehicles colliding with the above ground pipe were often more serious. Such collisions usually resulted in extensive damage mostly depending on the speed of the vehicle and the angle of impact. They also more often than not resulted in a fire, which depending on the location, could be a difficult and time consuming job before the actual repair could be started.
The collisions could be classified as follows:
Deep scratches were normally covered by a sleeve made up of 5/8th wall thickness pipe that was cut into two half shells. One of the shells would have a butt strap made of a 10 inch wide strip cut lengthways from another 5/8th length of pipe that was then welded along the longitudinal cut with a half width overlap.
These half shells would then be clamped over the damaged section with a chains, that were tightened by hydraulic jacks. The butt straps would then be welded to the other half sleeve followed by the circumferential welds between the sleeve and the pipe. This type of sleeve was also used to cover sections of pipe that were damaged by corrosion. The leak would be covered with a patch of neoprene rubber and the sleeves were then clamped over this patch and then welded.
As all sleeves of the type described above were fabricated in the field, they were sometimes fabricated to "order" and the length could vary between a few feet to about 20 feet each. If longer sleeves were needed they were installed end to end with the longitudinal welds offset from the next section and then welded together using another butt strap to cover the circumferential weld.
The repair sleeves as described above could not cope with a severely deformed pipe.
If the deformed section was less than about 24 inches long and the pipe was punctured a Plidco sleeve was used.
These Plidco sleeves came in two halves that were bolted together along the longitudinal split. The middle section of the sleeve had an inside diameter of 36 inches and the ends were 30/31 inches in diameter, the sleeve was positioned so that the deformed section would be under the middle of the sleeve.
The seal on the pipe was obtained by compressing a neoprene ring with jacking bolts against the pipe. After all the bolts were tightened, they would be cut flush with the sleeve and backwelded. Then after flow was re-established, the circumferential and longitudinal welds were completed. If the pipe was not punctured it was not strictly necessary mostly depending on the operating pressure at that particular location. However Tapline welded them routinely.
If the deformity was so severe that the Plidco sleeve halves could not be closed up, a so-called DoughNut sleeve could be used provided that there was no leak associated with the deformity. This type of sleeve was also made of 5/8th wall thickness pipe but the diameter of the center section was 39 inches. Like the normal repair sleeves, it was clamped around the pipe using chains and a hydraulic jack. Then the longitudinal joint would be welded followed by the circumferential weld on the pipe. Once installed it looked like a doughnut around the pipe, hence the name. At one location a leaking 1 foot repair sleeve had a faulty weld that started to leak. The DoughNut sleeve was perfectly suited for this type of repair.
If the damaged section was too long for a Plidco sleeve or so severe that I could not be covered by a DoughNut sleeve and there was also a leak, the damaged section would have to be cut out and a complete spool piece installed.
This was still a major repair job and the normal procedure used would be the same as when the pipe was actually cut by a vehicle collision.
The ring girders are very strong, so the impact did not normally break the ring girder but it could shift the concrete pier and the ring girder could then do some damage the pipe which would be repaired with a short sleeve if necessary.
If a collision resulted in a rupture followed by a fire, the damage could be considerable. Much depended on the amount of oil that fed the fire which depended on two things, first the terrain because if the leak was on a high spot, the flow of oil could easily be limited by draining the line in either direction and second, where the location of the leak was in relation to the nearest upstream main line check valve.
If for example, the pipeline was going up hill and the fire was located just upstream of the check valve, the amount of oil feeding the fire would be the column of oil between the check valve and the fire. If however the fire was just downstream of the preceding check valve, this column of oil could be several tens of kilometres long. As a kilometre of pipe holds an average of 2500 bbls. of oil, one can get an idea of the size of the fire and how long it could take to extinguish it. A good example is the accident at Km 245 on the 5th of April 1975 when a truck hit and broke the pipeline and the escaping oil caught fire. The nearest check valves are located at Km. 210 and Km. 256. However the location of the impact and fire is a low spot between a point at Km. 232 and the Km. 256 check valve. Therefore one can assume that a 24 kilometer column of oil, i.e. 60,000 bbls of oil will feed the fire. And the resulting fire was spectacular as was the damage. When the fire was finally extinguished and the damage surveyed a 523 feet section of pipe had to be cut out and replaced.
As explained before, this is where the problem of tie-in temperatures arises. First of all the movement of the pipe at each end must be controlled as much as possible by covering a long section of the pipe with sand. Then the area where the new section would be installed was cleared and graded.
The exact length of the damaged section cannot be measured accurately so the distance between the good ends was measured and used as a guide to fabricate the spool piece.
A Weld Plus end was then slipped over the pipe at each end. These Weld Plus ends were one piece sleeves one foot wide. On the inside diameter at each end was a machined groove. A neoprene ring seal and a steel jacking ring were inserted side by side in this groove with the latter on the side facing the sleeve end. At each end were two rows of bolts. One set was set in diagonally tapped holes and these were the clamping bolts, their purpose was to clamp the sleeve tightly on the pipe so it would not slip.
Another set of bolts was set in longitudinally tapped holes around the edge of the sleeve and they extended to the groove holding the steel ring and neoprene seal. When tightened they pushed the steel ring against the neoprene ring and thus provided a tight seal around the sleeve and the pipe. The distance between the grooves was about 6 inches and the sleeve could accommodate an opening between pipe ends of maximum 4 inches.
This property made them very useful in the repair of above ground pipelines as it would allow a shortening of the spool piece when the pipe temperature was below the tie-in temperature.
When the spool piece was ready, it was trimmed to the exact length required and as the original tie-in temperature was 75 degrees F. this was done when the actual pipe temperature reached that point. Since it was April, the nights were still very cool and this temperature was reached at about 1 AM.
The spool piece was rolled into place and placed on wooden skids and after coarse alignment it was also covered with sand leaving only uncovered spaces where the new piers and ring girders would be installed .
The Weld Plus ends was slipped across the end of the spool piece so that the cut was at the center of the Weld Plus end and the seal bolts were then tightened to the required torque to establish a good seal. Now the upstream station would resume operation and pump at a very slow rate to pack the line. The 2 inch vent valve located at the check valve at Km 256 would be opened to vent the air trapped in the line. Packing could take a long time depending on the oil losses and could well last about 10 to 12 hours.
Only when the line was packed and no leaks were evident on the Weld Plus end seals, could final tie-in of the ends be started. This involved welding the ends to the pipe and cutting the bolts and back welding them.
The whole procedure was not as easy as it sounds. First of all a lot of the work such as lining up the spool piece had to be done at night. This could not be done when the sun was shining as the empty pipe heats up very quickly. The side facing the sun then expands more than the cool side and the pipe ends start moving away from the direction of the sun. If the ambient temperature at the time was lower than the tie-in temperature an empty spool piece, especially a very long one could contract so that the clamping bolts could not hold it and the spool end could slip out. Packing an empty line can be a tricky business. If the line has been down for a number of days, the temperature of the upstream pipe and oil may well be lower than the tie-in temperature. When pumping is resumed this cold oil will enter the empty section and it will tend to contract. If the spool piece has been cut too short this cold oil can then shorten the spool piece even more and it could slip out of the Weld Plus end.
If low temperatures were anticipated the Weld Plus end would not be placed centrally on the joint, instead it was moved so that the joint was more on the side of the main line end, this would allow a few extra inches of movement of the spool piece.
Two years prior to this incident at Km. 245 there was a similar leak and fire at Km 187. At this location the downstream check valve was at Km. 210.403, uphill and 23.5 kilometres away. The ensuing fire was considerable but the amount of oil that drained towards the fire was considerably less because of the following conditions.
The check valve at Km. 210.500 was holding , which was not always the case with other check valves. There was a high spot at Km. 196 that effectively split the column of oil that could drain back in half. The final calculation confirmed this as only 24.000 bbls. of crude was lost instead of a possible 61.000 bbls. if the check valves had been leaking. All conditions were similar and there were no indications that there would be any major problems. The damaged sections were cut out, the site cleaned up and the spool piece was prepared following the normal procedure. The daytime ambient temperatures had been about 10 degrees above the tie-in temperature, the tie- in temperature was easily reached a few hours after sundown and there were no major problems.
The time things went almost completely out of hand was at Km. 545. A truck hit the pipeline and a fire ensued. After the fire was extinguished and the site was cleaned up, the 285 ft. damaged section was moved away and stringing the spool piece started. The weather was bitterly cold and getting colder by the day.
Problems were anticipated because of this. As mentioned previously, the Weld Plus ends could each only cope with 4 inches of shrinkage. When the spool piece was set in place and the Weld Plus ends at each end were secured, the Oil Dispatcher was advised that line packing could start at a very low rate.
By this time the temperature had dropped very low but it was assumed that the oil would be at about 70 degrees and the Weld Plus ends would cope with shrinkage if the temperature did go below that. Several hours into the line pack, John Hendley, Pipeline Supervisor, was standing near the downstream Weld Plus end when he heard a suspicious hissing noise. He turned around and noticed that the spool end had slipped out of the Weld Plus end to a point just past the seal, and the noise was the air escaping through the gap. He directly called the Oil Dispatcher in Turaif to stop packing the line.
The damage was surveyed and it appeared that the upstream Weld Plus end was still securely centered over the cut. But something had to be done to cope with the excessive amount of contraction of this long spool piece at the freezing temperatures that were prevailing at the time.
The downstream Weld Plus end was cut out along with a 5 foot section of the spool piece. A new spool piece was fabricated to fill the gap and a Weld Plus end set at each side.
The upstream Weld Plus end was welded in The downstream Weld Plus ends then were secured with the clamping bolts , but the 5 foot spool piece was left floating with only the seal bolts torqued up. However the Weld Plus ends were prevented from moving apart by four straps made of 6 inch schedule 80 pipes that were cut in half along their length and then welded between the Weld Plus ends. The fact that two Weld Plus ends without the straps would now give about 8 inches of travel was recognised, but with the very low temperatures that prevailed at the time it was decided not to take another chance.
As it turned out, the packing started, the straps held and no leaks were encountered.
The pipe temperature was closely monitored and after many hours of packing when the tie-in temperature was approached, the straps were cut and the pipe started to slide into the Weld Plus ends. The joints on the 5 foot spool piece were welded up and the job was complete.
Following this experience the manufacturers of the Weld Plus ends were contacted to find out if they could fabricate Weld Plus ends that were 24 inches wide as this would increase the maximum opening between pipe ends to about 10 inches.
The manufacturer confirmed that they could be made to this length and orders were placed to be used in future repairs if there was any reason to believe that they had to be installed when the pipe temperature would cycle widely below the required tie-in temperature.
I do not know if these longer ends were ever received or used after the closure of the line.
If we had them available at Km. 545 it would have saved us all a few anxious moments.
The two employees that come to mind with the repairs at Km 390 are Jim Hughes and Willie Burgess.
In 1953 when I arrived in Turaif, the permanent bachelor quarters were not yet constructed, we all lived in 16 X 16's and I happened to live in the one behind the one they shared. I shared one with Jim Bos, who was the Radio Operator at the time.
Jim Highes was the Pipeline Foreman and Willie was his assistant. In those early days we still worked 6 days a week , Friday off and the ban on alcohol was not yet in effect.
They would come home on Thursday afternoon and their week-end ritual would start. Take a shower and go to the old Messhall across the "street" for dinner. They returned home, strip to the waist and each one would have a bottle of whiskey standing on the floor next to their bed.
Above Willie's bed was a shelf that he could just reach while laying on his bed and on it was a small RCA Victor 45 automatic record player. On it was a stack of 6 or so 45 rpm. Hillbilly records. Most of them were records by Hank Williams Sr. and his famous songs like " Hey good lookin, " and " Always late with your kisses ".
So Willie would have a sip of whiskey, reach up to the player and start it. Once the stack was played , he would reach out and turn it over, start it and go back to whatever he was doing, reading books mostly. These same 45's would be played over and over until he fell asleep but the ritual would start again when he woke up and continued all through Friday.
Whenever you knocked on their door, one of them would holler "Come in" and on entering you would see the two, Jim with a very big grin on his wide face and Willie peeking at you over his reading glasses and he always welcomed me with " Hello Joohwn!."
Two of the most pleasant fellows one could ever meet.
Both were of course involved with the repairs at Km. 390. I had to install an ET-10 400 Watt transmitter inside a portable shed that would be used to establish communications between the repair site and the dispatchers in Beirut.
The location was an area of very hard limestone rock that prevented installing a proper ground and Willie complained that whenever he picked up the handset and keyed the transmitter , he would burn his fingers on the transmitter frame, receiver and everything metal in the room was "hot". So initially we advised him to cover the handset with a rag before keying the transmitter. This muffled his voice so he poked his face into the rag while he spoke. Anyway it was decided to move the "transmitter shed" a few feet. Willie did not see this as a problem and volunteered to "ease" it over with the bulldozer blade. He did just that and a little bit more, his "easing" was a bit rough, he pushed it and it turned on its side, effectively wrecking the shed, the transmitter inside it but he solved his problem of burned finger tips.
Another thing that reminds me of them as well as some of the others is the weekend poker games that were held in Bunkhouse "A" in 1956 -1957.
They were friendly games and the regular players were Jim and Willie, Tip Adams , Al Kleeman who were both IBI employees and a young Government Relations Man Jim, whose last name escapes me and then anyone else who happened to pass by and wanted to join in.
I used to pass by occasionally and just sit back and listen to the running commentary that Willie used to provide. One of his most colorful remarks must be whenever he or anybody passed wind , Willy would followed it up with " It is better to f... and bear the shame than not to f... and bear the pain!"
Nobody at the table would look up or respond and the game simply went on.