ROV: Remote Operated Vehicles, Sonar, Underwater, Inspection, Tunnel, Diving

Expanded Menu

Hibbard Inshore - Article

Part of the article "Case Studies in Underwater Survey" from UnderWater Magazine (Sept/Oct 2000) reprinted on the DiveWeb web page. Read the entire article.

Phoenix 360 Visits Hell Hole

PG&E operates a large number of small hydroelectric facilities in the Sierra Nevada Mountains of northeastern California. In June this year, Hibbard Inshore was awarded a contract to inspect the feed tunnel for one of these power stations. Many get their water from the "Hell Hole" reservoir, high in the Sierras, where PG&E has a guest house for visiting scientists, engineers, and technicians. A group from Hibbard Inshore, Deep Ocean Engineering (DOE), and PG&E occupied this upscale "bunk house" for the period of the inspection: they were royally watered and fed in an idyllic location. (Ask them if they prefer inshore to offshore work.)

The tunnel has been inspected by ROVs in the past. The first time was by a MiniRover in the early nineties, and later a Phantom Pipeliner did service. Hibbard proposed using the DOE Phoenix 360 ROV (nicknamed "Magic") for the job on the grounds that a larger, more stable vehicle would support more instrumentation and provide better video and sonar images.

Phoenix 360 is an all-electric light work vehicle, at 1,100 pounds, significantly larger than either the MiniRover or Phantom. For the job, it was equipped with two cameras, two sonars, and two manipulators. The main camera, a DOE Smart Zoom, faced forward mounted on an ROS pan and tilt unit, while the second, a small DSPL black and white camera, faced the rear of the vehicle. Both sonars were Imagenex 881 digital units. One had a profiling head and was mounted rotating around a fore-aft axis to provide a cross section of the tunnel. The other had an imaging head and rotated on a vertical axis; it was mainly used for navigation.

The main manipulator carried by the vehicle was the four-function "Gauntlet" made by Hydro-Lek. This was supplemented by a DOE Phantom three-function unit. The Gauntlet proved its worth by retrieving a two-foot length of one-inch diameter drill steel that had found its way into the tunnel. The vehicle's stability was such that this weight in a manipulator at the front of the vehicle caused only two to three degrees of pitch.

One of the attractive characteristics of the Phoenix family of ROVs is its small-diameter, neutrally buoyant umbilical. Number 360 presently has 8,000 feet of umbilical, but since the diameter is a mere 0.6 inch, no winch or handling system was required. Rob Snider of Hibbard Inshore, the cable tender, might disagree, but he didn't seem to have lost any weight by the end of the operation.

All of the Phoenix vehicles' instrumentation appears as an overlay on the video. This stops the operator's having to take his (or her) eyes off the screen, and means that any video taken of the operation is self-documented. To avoid confusing the operator (and obscuring the screen), the on-screen display has two operating modes: basic navigation mode, with just a minimum of information, and full diagnostic mode. Basic mode displays only depth, heading, date, and time of day. However, should an alarm go off (a water leak or earth leak, for example), a warning flashes and the diagnostic screen can be brought up to pinpoint the trouble. This comprises a complete instrument panel with pitch, roll, elapsed time, umbilical turns, power supply data, water and earth leak status, active camera number, and communication quality all displayed, in addition to the basic navigation data.

Pitch and roll information was important in reassuring the PG&E engineers and consultants that the vehicle was level when interpreting the video and sonar data. Smaller, less stable vehicles are much more affected by external factors like umbilical tension, and can pitch and roll to the extent that it is unclear exactly where the video camera or sonar is looking. With negligible pitch or roll, Phoenix provided superior images.

Hydroelectric power stations make money when they are running, and cost money when they are not, so scheduled inspections are made during periods of low demand. Unfortunately, this usually means at night, and this operation was no exception. The team was allocated three time slots, on Friday through Sunday night, from 8:00 p.m. to 6:00 a.m. The plan was to have everything ready so that the vehicle would be ready to dive as soon as flow through the tunnel had been shut off and the water had stabilized, and to stop the inspection with plenty of time for retrieval in case of problems.

Access to the tunnel is down a 600-foot (182m) vertical surge shaft, its top located in a large surge tank. As the day wore on toward the first evening's operation, the area around this tank slowly filled with vehicles of all shapes and sizes, all bumping along miles of dirt road to get there. A water tanker sprayed the track in an attempt to keep the dust down. The Hibbard team erected a couple of tents, one over the vehicle to protect anyone working on it from the sun, the other to act as a viewing theater during the actual operation. This tent had a projection TV system, seats, and a big screen. It even showed trailers of previous Hibbard attractions while the team members were waiting to get into the water! (A pity the ice cream ran out earlyŠ)

In fact, this tent performed one task admirably: It kept the clients away from the vehicle operator. The pilot, Brad Hibbard, was jammed into a small control van, surrounded by equipment, with barely enough room for one person. When faced with either standing, peering over someone's shoulder at a small monitor or sitting in comfort looking at a big screen, most chose the latter and he was able to get on with his job undisturbed.

The access shaft diameter is 10 feet, a comfortable fit for the Phoenix. However, getting it into the water involved an elaborate piece of juggling, since the crane was outside the surge tank. A couple of people up by the rim were able to coordinate both the inside and outside operations, carefully lifting the vehicle from outside the tank and lowering it down the shaft without stressing the umbilical cable. Finally, with the vehicle in the water, the crane lifted a sheave high enough for the cable to be fed by the handler (outside the tank) through it and down to the vehicle without rubbing on the top of the tank wall. Cable tending is important in any ROV operation, and especially so in a tunnel where the risk of entanglement is high. To avoid this, for the duration of these dives the umbilical was kept under a light tension.

Descending a vertical shaft like this can cause all kinds of headaches. The water level is usually well below the top of the shaft, posing problems for releasing the vehicle from the crane hook, and the presence of steel reinforcing bars can confuse a magnetic compass so that when the vehicle arrives at the bottom, the poor pilot has a 50 percent chance of heading off down the tunnel in the wrong direction. Recent advances in fiber optic gyros offer a high technology answer to this problem. However, the Hibbard team had a simpler and more economical solution: a weighted rope hung down one side of the surge shaft, acting as a visual reference for the vehicle so it could avoid rotating as it descended.

As it turned out, the concrete lining the shaft was unreinforced so the magnetic compass behaved correctly. It proved possible to descend hands-off, using the vertical trim control and maintaining direction with the auto-heading function. No problems were encountered on the way down to the main tunnel. However, the descent took a long time because of frequent stops for a closer look at features on the shaft wall.

After a sheave had been installed to protect the umbilical round the 90 degree bend at the bottom of the shaft, flying along the 13-foot diameter tunnel proved easy. With 750 watts of light distributed over the front of the vehicle, the DOE Smart Zoom camera was able to obtain the best images anyone had seen of the interior. Among other features, a 200-foot rock trap along the base of the tunnel was inspected, and the vehicle also flew up to review the state of the roof. Progress was slow, because the consultants had a steady stream of requests to have a closer look at this rock or that crack. In inexperienced hands, a zoom camera on a pan and tilt mount can be a recipe for disaster, but it is a powerful tool when operated from a stable platform by a skilled operator. The profiling sonar gave a cross section of the tunnel, showing the vehicle's vertical and horizontal position, while the forward-looking sonar kept the pilot informed of his heading relative the walls. It also permitted more accurate measurement of features than could be made by looking at the video image.

Hats off to the pilot, Brad Hibbard, who single-handedly operated all the sophisticated equipment used during the operation. In addition to controlling the vehicle, he had to manage the two sonars, VCRs for the video and sonar images, and a real-time non-linear video editor. To be able to do all this and eat a burger (without dripping sauce on the hand controller) is the sign of a real pilot. His task was made easier by the absence of current in the tunnel, good visibility, the vehicle's autopilots and its intrinsic stability, but it was still a tour de force. Over the course of the three nights the vehicle put in almost 20 hours with him at the helm (not strictly accurate, but it sounds more nautical than "at the stick").

Were there any problems? Do bears go in the woods? It is part of the very nature of underwater work that things go wrong. Get Murphy wet and he goes on steroids. Fortunately there was only one problem of any consequence. Returning to the surface after the first dive, the vehicle encountered floating debris on the surface, which jammed a couple of the vertrans propellers and loosed the clamps holding them to the shaft. This is an intentional weak link to protect the gearbox and the propeller, which is retained on the shaft by a small clip when this happens. No time was lost, and all that was necessary was to retighten the clamps on the vehicle.

If only all operations had as few problems as this one! The teams worked together in a spirit of cooperation that went far beyond contractual obligations, and the client was very pleased, providing the satisfaction of a job well done. UW

Case Studies in Underwater Survey. (2000, September/October). UnderWater Magazine (Houston)