The Utilization Of Escort Tugs In Restricted Waters
By Captain Gregory Brooks and Captain S. Wallace Slough

In the United States, the mandatory escorting of laden tankers is required by Federal Regulation in Puget and Prince William Sounds, and by the State of California in the ports of San Francisco, Los Angeles/Long Beach, San Diego, and Port Hueneme. Both the Federal and State regulations require the master to operate the vessel within the performance capabilities of the escort vessels.

To truly realize the benefits of escorting vessels within our ports, it is important for masters, pilots, and operators to understand the basic principles involved in escorting. They should also know the capabilities and limitations of the escort tugs being used.

This article was prepared to share the basic principals involved in escort operations and to illustrate the limitations in the performance of the various types of tugs currently utilized for escorting. The opinions expressed herein are based on the combined experience of the two authors. One, a licensed ship’s pilot in San Francisco Bay who has studied the methods of escorting vessels within the port firsthand, and the second a tug captain who has actively participated in most of the recent series of escort tug trials conducted in the United States. For the purposes of this article, the term restricted waters means waters where the narrowness of the channel restricts the ability to turn the vessel around, the article is not intended to address the escorting issues in wide open sounds such as Prince William Sound or Puget Sound.


Pivot Point: The pivot point is the center of lateral resistance. It is the point around which the vessel will appear to turn. When a vessel is dead in the water, the pivot point will be near amidships if the vessel is on an even trim. As a vessel gains sternway, the pivot point will move toward the stern. Conversely, as a vessel gains headway, the pivot point will move toward the bow. When tugs are utilized, they will be most effective in changing the vessel’s heading when they are placed as far as possible from the pivot point. If a vessel has headway, an escort tug will be most effective on the stern due to the leverage realized, and least effective on the bow due to the reduced leverage.

Kinetic Energy: As a vessel’s speed increases, it’s kinetic (stored) energy increases geometrically. The kinetic energy is equal to one half the weight of the vessel multiplied by the velocity squared (KE = ½ WXV2 ). Simply put this means that at ten knots the kinetic energy that the tugs must control in an emergency is 100 times greater than that generated at one knot.

Tug Force Vectors: A tug influences the movement of a vessel by applying a force to the vessel at some angle called a force vector. At zero speed a tug can basically apply its full bollard pull to the vessel in any direction. As the vessel gains headway, the tug’s ability to apply forces is reduced by two factors. First, it is using some of its own power to propel itself through the water. Second, if the tug force is desired at an angle to the vessel, the tug will bleed off more power using its rudder(s) and thrusters to change the angle of the tug to the vessel and attain the thrust vector the pilot desires. A pilot or master must keep in mind this reduction in available thrust as the speed of the vessel increases. (Note: Tractor tugs can produce high force vectors at higher speeds due to the shape of their hulls, which will be discussed later in this article).

In reviewing these basic principles, several key points become apparent:

  1. The escort tug will render the greatest assistance to the vessel with headway if the tug is utilized at the stern. When the tug is on the stern, it is at the greatest distance from the pivot point of the vessel and thereby has a greater lever to work with.
  2. The speed of the vessel through the water must be kept as slow as possible. As the speed of the vessel increases, the kinetic energy increases, and the ability of the escort tug to affect the direction of the vessel decreases.
  3. In the event of a failure, the vessel’s quick recognition of and reaction to a casualty, coupled with the tug’s rapid response is critical in reducing off track error. If the vessel were allowed to swing off course unrestricted, the kinetic energy associated with the vessel’s swing will rapidly overpower the tug’s ability to control it. The escort tug should therefore be tethered (made fast) on the stern to be most effective in an emergency situation by significantly reducing the response time of the tug.


Conventional Tugs:

Conventional tugs are either single or twin-screw designs. It is the authors’ opinion that conventional tugs are only effective escorts if they are able to make up on the stern of the vessel. This means that the vessel must be deep enough in the water that the escort tug’s bow fender can maintain solid contact with the vessel’s transom and not slide off when applying force against the vessel’s hull. When made up in this manner, a tug acts as a “rudder” and can steer a vessel very effectively in the event of a propulsion and or steering failure, provided the vessel’s speed is not excessive. They can also reverse their engines to act as a retarding force. It should be noted that a tug might not be able to make up to or work at the transom in unprotected waters with excessive seas.

Figure 1

Figure 2

Figure 3

Tractor Tugs:

Tractor tugs are available in three basic types (although the hard line distinction between them continues to become hazier as designers seek to use the best features of the different types in their new designs). A “true” tractor is a tug with its propulsion units [either Z-Drive or cycloidal (Voith-Schneider)] located under the forward part of the hull, and reverse tractors with their propulsion units mounted aft.

”Z-Drive” tractors, (figure 2) utilize two steerable propulsion units located forward that revolve 360 degrees to provide thrust in all directions. Since these units usually utilize kort nozzles, they produce very high bollard pull compared to the installed horsepower (about 26 pounds per horsepower although the latest designs are now realizing over 27).

Cycloidal, or Voith-Schneider, tractors (pictured on the following page) utilize a pair of five bladed vertically mounted propellers located under the forward part of the hull. These constant speed propellers have the advantage of near instantaneous thrust in any direction and provide a very precise application of thrust. This rapid control of thrust can be an advantage when maneuvering ships (although a well operated Z-Drive can obtain similar control and response by keeping their drives engaged in gear and opposed). The disadvantage of the cycloidal system is the low mechanical efficiency of the units with a bollard pull rating per horsepower of about 19 pounds) and their high construction costs. [NOTE: The latest Voith Schneider propeller designs, using longer blades, are now producing approximately 22 pounds per horsepower.]

Both of these "True" tractors are fitted with a large skeg aft for directional stability and use the stern of the tug as the "working end." While the skeg provides advantages in the "indirect towing" mode mentioned below, it does present problems with older tractor designs which have a "flat plate skeg" (literally a flat piece of steel). When this type of skeg is driven stern first during tethered escort operations, the turbulence in the vessel’s wake hits the skeg and causes the tug to constantly sheer from side to side. This can cause fatigue for the operator and to the possibility of tripping the tug during high-speed indirect towing operations if the skeg digs in too hard during the initial turn. This problem has been reduced in some tractors by the placement of a secondary tie-down (or staple) at the tug’s stern. This shifts the pivot point to the leading edge of the skeg, and reduces this hazard to the tug. Many tractors built before 1996 were fitted with this flat plat skeg.

A recent solution to this skeg problem was designed by the Neptun Company of Oslo, Norway. They equipped their new cycloidal tractors Bess and Boss with a skeg designed with a very full airfoil shape. This design can be operated at much higher angles of attack to the water flow without stalling (loss of laminar flow). It has proven stable enough to allow the tugs to be operated fin forward all the time, solving the instability problem referred to above. This "full bodied skeg" has proven so stable that several “true” tractors designed with bi-directional controls in their wheelhouses are now being operated by their crews fin first (stern first) all of the time.

"Reverse" tractor tugs (pictured below) are essentially conventional tugs with ”Z-Drive” propulsion units mounted aft. Unlike the “true” tractors, reverse tractors work like a regular tug, i.e. bow first. Because these boats are not equipped with the skeg they are more stable during a tethered escort. This enhanced directional stability can be an advantage for escort operations versus flat plate skeg equipped tractors due to the crew fatigue issue. The latest reverse tractor designs have been equipped with “box keels” along the bottom of the hull to increase lateral resistance, and work somewhat like a "true" tractor’s skeg.

Figure 4


Described below are some of the unique ways that a tractor can be utilized in escort work that greatly increase their value over a similarly sized conventional tug.


While a tractor can be used as a conventional tug, their true value for escorting is in the unique ways that they can create tow line forces while working on a long line (200’ - 300’ astern of the vessel). At slower speeds, the tractors can be used in the "direct mode" where the tractor simply pulls on its line in various directions to create retarding and / or steering forces for the vessel. This mode can also be used to create drag for the vessel so that it, in turn, can increase its main engine rpm’s and gain increased rudder power or avoid stopping and starting the ship’s diesel engine. This is normally used in narrow waterways to allow the vessel to increase its maneuverability while limiting its speed. A conning officer should note that as the speed of the vessel increases, it takes the tractor longer and longer to “back” out to either side of the ship in order to provide steering forces.


Another maneuver all tractor tugs are designed to perform on a long line is the “indirect” towing mode. To perform this maneuver, the tractor turns its "working end" in the direction the force vector is desired and, like a water skier, rides it’s line out to a predetermined angle to the vessel’s heading (usually the towline will be placed at a 45º angle to the ship’s centerline). Once in position, the tractor places its hull at about a 30-degree angle to the water flow. The water flow against the tug’s skeg and hull create greater line pull than the tug could produce with its engines. These forces are a function of the speed through the water, the design of the tug’s skeg and hull form, and the forces applied by the tug’s propulsion units. The forces generated with indirect towing increase with speed through the water, first exceeding the tug’s mechanical forces at about 7 knots. At about 10 knots it can produce up to 2 to 2 ½ times the tractor’s rated static bollard pull.

Significant heeling can occur during indirect towing. The tug operator must be extremely careful not to trip the tug as it heels over and nears deck edge immersion. Tractors with flat plate skegs must be extremely careful in conducting this maneuver at speeds over eight knots as the skeg can dig in too hard and potentially cause the tug to trip before the operator can recover from the maneuver.


A recent refinement of indirect towing has been referred to as “Powered Indirect Towing”. This maneuver seeks to combine the hull’s ability to create towline forces with the mechanical ability of the tractor in order to improve performance while operating in the five to seven knot range. At these lower speeds, the tethered tractor tug veers out as if in the indirect mode, but then pushes into her line at full power. Depending on the speed of the vessel, the tug may be able to get out to a position where her tow line is at a 90º angle to the vessel’s centerline (all line forces would then be steering forces) with the tug at perhaps up to a 70º angle to the water flow. In the 3-7 knot speed range, tests have shown that the forces created using this "powered indirect" mode are greater than could be attained with by the tractor in the "pure indirect". Further, as the tractor can very quickly perform this maneuver the steering forces created are faster to apply than with the direct mode.


One of the unique maneuvers that a Z-drive tractor (or reverse tractor) can perform is a method of stopping a vessel called "Transverse Arrest". Instead of placing the thrusters pointing ahead to stop the vessel, the thrusters are directed outward and full power is applied. The "wall of water" created on each side of the tug actually slows the vessel. Manufacturers indicate the transverse arrest can achieve 2.5 times the bollard. During our live testing we have only been able to record line pulls equal to 1.5 times the tractor’s rated bollard pull when operating in the transverse arrest mode.


Over the past two years, BP Oil Shipping Company, USA, ARCO Marine, SeaRiver Maritime Inc., and Hvide Marine, Inc., have conducted a number of trials to measure the true performance of various types of tugs. They have made the information collected during these trials available to the maritime community. Shown below are a portion of the results obtained from these computer simulations and live tests that illustrate some very important information on assist and escort work that all professional mariners should be familiar with.

Test No. 1: (S/R Baytown 57,720 DWT; Tug S/R California 7200 HP)

In this test, a 57,720 dwt vessel loaded to 33 ft. and a very powerful conventional twin screw tug (7,200 Hp / 220,000 pounds bollard pull) were utilized to illustrate the relative effectiveness of a tug working at the bow, quarter, and transom of a ship making headway. With the vessel at a speed of 5 knots, on a steady heading, and the rudder amidships, the vessel’s engine was stopped and the tug ordered to push at full power at each of the three positions. During each of the three runs the wind was on the port bow at approximately twenty-five knots. Measurements of the rate of turn in degrees per minute were noted. As shown on the illustration below, the rate of turn was 3 degrees/minute with the tug on the bow, 12 degrees/minute with the tug on the quarter, and 24 degrees/minute with the tug on the transom. This clearly demonstrates the superiority of utilizing a tug on the transom during an escort. It also shows that a single conventional tug is of limited use as an escort vessel at higher speeds if it cannot make up on the transom.

During this test, we noted that when the tug was working at the port bow, it could not get out to more than a 30º angle to the ship. This meant much of its force vector was an accelerative force versus a steering force. Very quickly, the vessel’s speed increased to 5.6 knots. After initially establishing a rate of swing of 3 degrees per minute, at 5.6 knots the swing of the vessel to starboard stopped, and the vessel started to swing back to port! The tug was still working at full power, but the vessel was responding in a fashion that was completely opposite to what was expected. While the forces involved in creating this action are not totally understood, the reaction of the vessel was clearly documented. This is an extremely important finding, as placement of an escort or assist tug at the bow is considered by many as the most prudent action.

Figure 5

Computer Simulation No.1: (John Young 188,000 DWT
7,200 HP conventional tug and 3,000 HP Voith-Schneider Tractor)

The following computer simulations were presented in the San Francisco Bay Tanker Escort Study and clearly illustrate the critical importance of tethering and limiting the speed of the escort. In each set of simulations shown below, the 188,000 DWT vessel is simulated being escorted by a conventional 7,200 Hp tug in the plate on the left, and in the plate on the right by a 3,000 HP cycloidal tractor. In each simulation, the rudder fails locked hard right, and there is a 30-second time delay to recognize and respond to the rudder failure. Each plot illustrates the results attained by using the tug in three distinct ways: using the escort tug to stop the vessel’s swing (oppose) and return the vessel to her original track; to accelerate the vessel’s swing (assist) to turn the vessel around; and to stop the vessel (retard).

In the first set of plates the vessel is making six knots and the tugs are responding from an untethered position. For the purposes of the simulation, a tethering time delay of two minutes was simulated. This is, in the authors’ opinion, highly optimistic. The conventional tug was able to limit the off track error to about 1,200’ by retarding and the tractor’s off track error was about 1,500’ using the assist method.

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Figure 15

In the next set of plates, we can see that the same tugs can reduce this off track error at six knots to approximately 150’ for the conventional tug and 250’ for the tractor by simply tethering the escort tug to significantly reduces the tug’s response time. These two plates clearly illustrate the advantage of tethering the escort tug in restricted waters.

In the last set of plates, the same vessel and tugs are simulated at a speed of 10 knots with the tugs tethered. In these plates, we can see how the speed has clearly overpowered both tugs. This simulation illustrates how a vessel’s kinetic energy will overpower even the largest tug if the escort speed is excessive.

Trial No. 2: (S/R North Slope 165,000 DWT; Dr. Jack 7,200 HP)

SeaRiver Maritime Inc. conducted full-scale trials using the S/R North Slope (165,000 DWT) loaded to 50’ and the conventional tugs Sea Voyager and Dr. Jack (7,200 HP each) tethered on the stern. Each of these trials simulated a simultaneous rudder and propulsion failure. At the start of the test, the rudder was placed hard right, 15 seconds later the engine was stopped, and 15 seconds after that the tug was ordered to oppose the vessel’s swing. During all of these trials the wind and seas were calm.

The results of these trials at speeds of 5, 6, and 7 knots are shown in the illustrations. While the tug was capable of controlling the vessel at a speed of 5 knots, the two-knot increase in speed to seven knots made the tug lose control of the vessel. The combination of the increase in kinetic energy (KE = ½WXV2), the tug’s more limited capability at the higher speed, and the vessel’s quicker reaction due to the increased water flow against the vessel’s rudder, overpowered the escort tug.

Trial No. 3: (S/R North Slope 165,000 DWT; Dr. Jack 7,200 HP)

In Trial No. 2 (above), it was assumed that both the vessel’s rudder and engine failed simultaneously. Since a simultaneous failure of propulsion and steering is not as likely, the six-knot test was repeated as described above with the rudder failed again in the hard right position. The tug was ordered to oppose the swing of the vessel with full power at 30 seconds into the test as in trial No. 2, but this time the vessel’s engines were also backed full. Due to the torque of the propeller backing, the resultant off track error was much worse than when the engine was not used. This information suggests that mariners should not assume that they should use the vessel’s engine to correct the hard over rudder failure scenario.

It should be noted that the S/R North Slope is fairly short for its beam (906’ x 173’), which may have influenced this result. The point to understand is that each mariner must know how his or her vessel may handle in this type of situation before the failure occurs.

Trials No. 4

A series of live trials were conducted during April of 1997 using Foss’ 8,000 HP Voith-Schneider tractor Lindsey Foss as part of the evaluation process prior to constructing new escort vessels for Prince William Sound. For this series of tests, Glosten and Associates of Seattle, WA were hired to “hard wire” the vessel and the tug in order to be able to record many aspects of the boat’s performance during the test. Not only were the line tension forces monitored, but also the vessel and tug headings, tow line angle, control and power settings on board the tug, angle of tug heal, speed over the ground, etc. For the first time, we were able to accurately review after the fact the actual performance of the two vessels during a trial versus relying on observer’s impressions.

Trial No. 4 (Arco Juneau 125,000 DWT; 140,000 Disp.;
VSP Lindsey Foss 8,000 HP)

This series of trials was designed to explore a tractor’s performance while operating in the five to eight knot range. The tractor was used in the indirect towing mode to oppose the swing of the vessel during three tests at eight knots (a one minute time delay was used as a recognition / reaction time delay). The wind conditions recorded for the plot shown below was 25 knots on the port quarter, sea conditions were nominal.

The off track error recorded was 945’ in the test to port, and 1700’ and 1715’ in two tests to starboard [one illustrated]. While these tests results appear quite good at first glance, it should be noted that this is the largest tractor currently available in the world, and was teamed with only a 125 kdwt vessel. Further, few ports have channels wide enough to tolerate this type of off track error.

It is also very enlightening to review the line pull graph of the test to starboard shown above. It will be noted that after an initial peak of 400 KIPS (thirty seconds after the tug was ordered to work) as the tug came into her line in the indirect mode, the line tension fell back to about 125 KIPS thirty seconds later. What actually happened was when the tractor aggressively entered the indirect towing mode, the tug literally “cracked the whip” and reached a position 90º to the stern of the vessel, at a rate of speed faster than the vessel she was escorting. To avoid passing the vessel’s stern, the tractor captain flattened the angle of attack to the water flow (which significantly reduced the tug’s indirect towing line pull). At approximately 220 seconds into the test, the line pull begins to come up again. This is the point where the tractor captain had reduced his speed and was again attaining the proper 30º angle to the water flow to create the high indirect towing forces. At four minutes into the test, the tug swung around to pull directly on the line (direct towing), increasing the force to 200 KIPS. Please note that during this test, the tug was most effective in the direct towing mode (pulling on her line) versus the indirect mode.


Figure 16

Figure 17

This article has been written so that all mariners can study and learn from the trials that have been recently conducted in the US. A careful review of this data will show that we have much to learn on this subject and that some of our past assumptions need to be re-evaluated. Until such time as these important issues can be tested further and agreed to by recognized experts in the field, the prudent mariner should accept all performance claims for escort tugs, at any speed, with some caution until proven as fact through live trials.

This series of trials using conventional tugs as well as the latest designs in tractor tugs suggest that the actual results obtained during a live trial vary greatly from computer simulated results. At this time, it is not obvious why this variation exists.

We can, however, based on our experience of 24 years of piloting and intensive study on the subject of escorting, offer the following recommendations.

  • The escort tug should be tethered on the transom of the escorted vessel to be effective. A conventional tug is generally only acceptable if its bow fender is high enough to make up on the transom of the vessel and not slide off when working.
  • Vessel transit speeds should be limited to that which the escort tug has proven it can save the vessel within the channel width available. The authors recognize that the transit speed must be balanced against the difficulties of conning the vessel at slow speeds in areas with strong currents. To do so may expose the vessel to a different and more immediate higher risk of grounding due to the current, rather than a possible mechanical failure. However, in considering the question of currents, the prudent mariner should also consider adjusting his time of passage to minimize the currents encountered. This is why a thorough risk assessment is the preferred method of developing escort strategies.
  • The pilot, vessel’s crew, and tug captain and crew should be extremely alert and respond immediately to any casualty before the kinetic energy associated with the vessel’s swing is allowed to build. We recommend that the Master and Pilot specifically discuss the potential emergency maneuvers that may be required during the escort as part of the Master / Pilot Conference prior to the pilot taking the con.
  • Vessel pilots, masters, and tug crews may need training (preferably simulator training) in the proper utilization and response of escort tugs.
  • The vessel’s bitts and chocks at the transom may need reinforcement to withstand the forces that can be exerted by large escort tugs.
  • The vessel’s crew (Engine and Deck) should be regularly drilled in emergency response to a steering failure. In the event of a rudder failure, it is imperative that the rudder be returned to amidships as soon as possible, and hopefully restored to full operation soon thereafter.
  • An anchor detail should be manned during escort transits and properly trained in dropping the anchor short to enable the use of this tool at higher speeds.

During the live testing, there were many new findings uncovered. By sharing this information, we hope to generate a dialogue between the various ports of the world where professional mariners will share openly their experiences so that we may all learn and improve our skills. The authors would greatly appreciate hearing from other mariners who can add to the knowledge of this subject.

To truly realize the benefits of escorting vessels within our ports, it is important for masters, pilots, and operators to understand the basic principles involved in escorting.

Also of Interest