Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
DRILL PIPE HARDBAND REMOVAL AND BUILD UP
Document Type and Number:
WIPO Patent Application WO/1994/015745
Kind Code:
A1
Abstract:
A process for submerged arc welding replacement of hardbanding by material to maintain the strength of a drill pipe is disclosed. For preferred results for build up of drillpipe after hardband removal, process variables that may be controlled are volts per wire, amps, wire feed speed (206), torch travel speed, Eso and flux (204) height. An alloy flux cored wire is used in the submerged arc process comprising nickel, molybdenum and chromium.

Inventors:
GAULT, John, T.
Application Number:
PCT/US1994/000257
Publication Date:
July 21, 1994
Filing Date:
January 04, 1994
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
ADVANCED WELDING TECHNOLOGIES, INC.
International Classes:
B23K9/04; B23K10/02; B23K35/30; B23K35/368; F02B75/02; (IPC1-7): B23K10/00; B23K35/02
Foreign References:
US3596041A1971-07-27
US3890482A1975-06-17
US4432313A1984-02-21
US4673796A1987-06-16
US4214145A1980-07-22
US3999036A1976-12-21
US5192851A1993-03-09
Download PDF:
Claims:
What is claim
1. ed as invention is: The process of rebuilding material into grooves cut in drill pipe, comprising: A. Providing submerged arc welding equipment including a welding torch, the welding torch mounted vertically and offset from the center line away from the direction of rotation of the drill pipe, alloy flux cored wire mounted in the tip, semiactive powdered flux piled on the drill pipe, the weld wire extending into the flux, a constant voltage power supply, a water cooling system, and a preheat system; B. Rotating the drill pipe; C. Moving the torch across the area where the hardbanding was removed while feeding the wire to form a weld.
2. The process of claim l, wherein the rotation speed of Step B is between 34 and 38 IPM.
3. The process of claim 1, wherein said moving of Step C is at a speed between 13 and 15 inches per minute, said moving occurring for intervals between .19 and .21 seconds and maintaining flux height between 15/32 inches and 17/32 inches.
4. The process of claim 1, wherein there is included the step of: D. Maintaining electrical stickout between 31/32 inches and 33/32 inches.
5. The process of claim 1, wherein Step C includes delivering power through the torch to melt the wire with volts per wire between 25 and 27 volts and amps between 275 and 325 amps.
6. The process of claim 6, wherein the delivered power is approximately 16,258 kilojoules of heat.
7. The process of claim 6, wherein the power supply slope is 3 volts per 100 amps.
8. The process of claim 7, wherein the power supply open circuit voltage is 40 volts and the short circuit current is 1200 amps.
9. The process of claim 5, wherein the wire feed speed of Step C is between 50 and 60 inches per minute.
10. The process of claim 1, wherein the weld of Step C overlaps the previous rotation weld by fifty percent.
11. The process of claim 1, wherein there is further included the step of cooling the drilling pipe with the water cooling system between passes of Step C to maintain the interpass temperature between 250 and 300 degrees Fahrenheit.
12. The process of claim 1, wherein there is included the step of preheating with a preheat system the drill pipe to 300° F. and the step of post weld heating of the drill pipe at 700° F.
13. The process of claim 1, wherein said alloy flux cored wire includes: Nickel between .522 and .638% by weight Molybdenum between .477 and .593% by weight Chromium between .027 and .033% by weight.
14. The process of claim 13, wherein said alloy flux cored wire further includes: Carbon between .081 and .099% by weight.
15. The process of claim 14, wherein said alloy flux cored wire further includes: Manganese between 1.845 and 2.255% by weight.
16. The process of claim 15, wherein said alloy flux cored wire further includes: Silicon between .378 and .462% by weight.
17. The process of claim 16, wherein said alloy flux cored wire includes: Phosphorus between .0108 and .0132% by weight; Sulphur between .0135 and .0165% by weight; Copper between .
18. and .022% by weight; Vanadium between .018 and .022% by weight; and Iron to make up the remaining percent to equal one hundred percent by weight.
19. 18 An alloy flux cored wire for use with a submerged arc process to rebuild material into grooves cut in drillpipe, comprising: Nickel between .522 and .638% by weight Molybdenum between .477 and .593% by weight Chromium between .027 and .033% by weight.
20. The wire of claim 18, wherein there is further included: Carbon between .081 and .099% by weight.
21. The wire of claim 19, wherein there is further included: Manganese between 1.845 and 2.255% by weight.
22. The wire of claim 20, wherein there is further included: Silicon between .378 and .462% by weight.
23. The wire of claim 21, wherein there is further included: Phosphorus between .0108 and .0132% by weight; Sulphur between .0135 and .0165% by weight; Copper between .018 and .022% by weight; Vanadium between .018 and .022% by weight; and Iron to make up the remaining percent to equal one hundred percent by weight.
Description:
DRILL PIPE HARDBAND REMOVAL AND BUILD UP

CROSS REFERENCE TO RELATED APPLICATION

This is a continuation-in-part of U.S. Application Serial No. 959,569 filed October 13, 1992 by John T. Gault, entitled Drill Pipe Hardband Removal and Build Up.

FIELD OF THE INVENTION

The present invention relates to welding processes and moire particularly to drill pipe welding for hardband removal and build up.

BACKGROUND OF THE INVENTION

Background of the Prior Art: 1. General

Until recently, hardbanding was done with an E-70S-2 wire and tungsten carbide particles. This process has the benefit of allowing the pipe to have a long life due to wear resistance. A problem arises with this process, as the tungsten carbon particles are put into a matrix and applied to the pipe. Carbon particles actually become small abrasive projections, and the tungsten carbide hardbanding is not smooth. The tungsten carbide hardbanding, although it lengthens the life of the tool joint due to wear resistance, shortens the life of the casing which the drill pipe is running in. Of course, if all drill strings were run in open holes, without casings, this would present no problem. However, most current drilling activity takes place in cased holes where the abrasive quality of the tungsten carbide particles in the hardbanding presents very real, and severe, problems. In some cases, casing wear through has been experienced where the drill pipe rotating inside of the casing has actually ground through the casing during a drilling operation.

Most operating companies today, such as British Petroleum (BP) , Shell, and Texaco, are not allowing any tungsten carbide hardbanded drill strings to be run in their wells because of the severe wear inflicted on the casing by the tungsten carbide actually grinding away the interior walls of the casing.

Carbon Arc Gouging (CAG) of hardbanding and Submerged Arc (SAW) build up have been done the same way for many years. 2. Welding

A. Hardband Removal - Present

Hardbanding removal from the drill pipe is presently done by CAG - which takes considerable time. CAG deposits carbon on the tool joint that is made of 4137 material of 40 points carbon. Steel with 40 points or more is a crack sensitive material and by the addition of more carbon, the material becomes more crack prone. Also, by removing this by hand and not having consistent depth cuts, which determine how much carbide is removed, this can leave hard spots which are difficult to weld over and add to cracking. This process was developed in the 1940's.

Metallurgical Effects:

During gouging and cutting of the air carbon arc cutting process, when the carbon electrode is positive (reverse polarity) , the current flow carries ionized carbon atoms from the electrode to the base metal. The free carbon particles are rapidly absorbed by the melted base metal. Because this absorption cannot be avoided in the prior art, it is important that all carburized molten metal be removed from the kerf (groove) , preferably by an air jet.

When the air carbon arc cutting process is used under improper conditions, the carburized molten metal left on the

surface can usually be recognized by its dull, gray-black color. This is in contrast to the bright blue color of the properly made grove. Inadequate air flow may leave small pools of carburized metal in the bottom of the groove.

Irregular electrode travel, particularly in a manual operation, also will produce ripples in the groove wall that tend to trap the carburized metal. Finally, an improper electrode angle may cause small beads of carburized metal to remain along the edge of the groove.

The effect of carburized metal that remains in the kerf or groove through a subsequent welding operation depends on many factors including the amount of carburized metal present, the welding process to be employed, the kind of base metal, and the weld quality required. Although it may seem that filler metal deposited during welding would dissolve small pools or beads of carburized metal, experience with steel base metals shows that traces of metal containing approximately one percent carbon may remain along the weld bond line. Carbon pickup in the weld metal becomes significant with demands for increasing weld strength and toughness. Increased carbon content can decrease weld toughness, especially in quenched and tempered steels.

B. Plasma Arc Cutting

The plasma arc cutting process severs metal by using a constricted arc to melt a localized area of a workpiece, removing the molten material with a high-velocity jet of ionized gas issuing from the constricting orifice. The ionized gas is a plasma, hence the name of the process. An arc plasma is a gas which has been heated by an arc to at least a partially ionized condition, enabling it to conduct an electric current. The

different gases used for plasma arc cutting include nitrogen, argon, air, oxygen, and mixtures of nitrogen/hydrogen and argon/hydrogen. A plasma exists in any electric arc, but the term "plasma arc" is associated with torches which utilize a constricted arc. The principle feature which distinguishes plasma arc torches from other arc torches is that, for a given current and gas flow rate, the arc voltage is higher in the constricted arc torch.

The arc is constricted by passing it through an orifice downstream of the electrode. The basic terminology and the arrangement of the parts of a plasma arc gouging (PAC) torch are shown in Figures 1 and 2, discussed with respect to the preferred embodiment below.

Plasma arcs operate typically at temperatures of 18,000 - 25,000 degrees F. (10,000 - 14,000 degrees C).

Plasma arc gouging was developed in the mid 1950's and became commercially successful shortly after its introduction to industry. The ability of the process to sever any electrically conductive material made it especially attractive for cutting nonferrous metal that could not be cut by the oxyfuel cutting (OFC) process. It was initially used for cutting stainless steel and aluminum. As the cutting process was developed, the process was found to have advantages over other cutting processes for cutting carbon steel as well as nonferrous metals.

When compared to mechanical cutting processes, the amount of force required to hold the workpiece in place and move the torch (or vice versa) is much lower with the "non-contact" plasma arc cutting process. Compared to OFC, the plasma cutting process operates at a much higher energy level, resulting in faster

cutting speeds and lower overall heat inputs. In addition to its higher speed, PAC has the advantage of instant startup without requiring preheat. Instantaneous starting is particularly advantageous for applications involving interrupted cutting, such as severing mesh.

C. Submerged Arc Welding

Submerged Arc Welding (SAW) produced coalescence of metals by heating them with an arc between a bare metal electrode and the work. The arc and molten metal are "submerged" in a blanket of granular fusible flux on the work. Pressure is not used, and filler metal is obtained from the electrode and sometimes from a supplemental source such as welding rod or metal granules.

In submerged arc welding, the flux that covers the arc plays a major role in that (1) the stability of the arc is dependent on the flux, (2) mechanical and chemical properties of the final weld deposit can be controlled by flux, and (3) the quality of the weld may be affected by the care and handling of the flux.

Submerged arc welding is a versatile production welding process capable of making welds with currents up to 2000 amperes, ac or dc, using single or multiple wires or strips of filler metal. Both ac and dc power sources may be used on the same weld at the same time.

Factors that determine whether to use submerged arc welding include:

(1) The chemical composition and mechanical properties required of the final deposit

(2) Thickness of base metal to be welded

(3) Joint accessibility

(4) Position in which the weld is to be made

(5) Frequency or volume of welding to be performed

SUMMARY OF THE INVENTION

The present invention involves SAW for rebuilding of drill pipe after the removal of hardbanding.

The following equipment and controlled variables are used in the processes of dual wire sub arc for buildup of drill pipe:

PARAMETERS FOR DUAL WIRE SUB ARC BUILD-UP OF 4130 OR 4137 DRILL PIPE

WIRE SIZE 3/32" (.0938)

WIRE TYPE ALLOY FLUX CORED WIRE

FLUX TYPE SEMI-ACTIVE POWDERED

*VOLTS PER WIRE 26

*AMPS (WIRE FEED SPEED) 300 (60 IPM)

*TORCH TRAVEL SPEED 36 IPM

ELECTRODE STICK OUT ("ESO") 1"

HEAT INPUT (KILOJOULES) 13,000

*FLUX HEIGHT .5"

POWER SUPPLY SLOPE 3 VOLTS PER 100 AMPS

OPEN CIRCUIT VOLTAGE (OCV) 40

SHORT CIRCUIT CURRENT 1200 AMPS TORCH POSITION VERTICAL AND 1/2" FROM CENTER LINE

AWAY FROM THE DIRECTION OF THE TURN

INCH CONTROL (FREE FLOW OF WIRE) 50 IPM

BURN BACK CONTROL (TO FREE WIRE AT STOP) .3 SEC.

*INDEX A. SPEED 14 IPM

B. TIME .2 SEC.

INDEX LAP OVER 50% OF PREVIOUS BEAD

WATER VOLUME 30 GALLONS PER HOUR WATER TEMPERATURE 80 DEG. F. MIN.-150 DEG. F. MAX

PRE-HEAT TEMPERATURE 300 DEG. F.

INTER PASS TEMPERATURE 250 - 300 DEG. F.

POST WELD HEAT 700 DEG. F.

RATE OF COOLING SLOW CV POWER SUPPLY

* VARIABLES THAT MAY BE CONTROLLED.

For the above variables, the following tolerances are permitted before the procedure must be shut down:

VOLTS 26 +/- 1 VOLT

AMPS 300 +/- 25 AMPS

TRAVEL SPEED 36 IPM +/- 2 IPM

ESO 1" +/- 1/32"

FLUX HEIGHT 1/2" +/- 1/32"

STEPOVER A. TIME .2 SEC +/- 10%

B. SPEED 14 IPM +/- 1 IPM

By controlling the above variables and their tolerances, a consistent and repeatable quality of weld can be made.

DRAWINGS For a further understanding of the nature and objects of the present invention, reference should be made to the following drawings in which like parts are given like reference numerals and wherein:

Figure 1 is a cross-sectional view of a plasma arc torch;

Figure 2 is a cross-sectional view of a plasma arc torch having electrical circuitry;

Figure 3 is a view, partially in cut away, of a SAW;

Figure 4 is an illustration of how electrical stick out (ESO) controls amperage;

Figure 5 is an illustration of how electrical stick out controls voltage for a given bead width;

Figure 6 is a schematic view of the PAC of the present invention;

Figure 7 is a schematic view of the SAW of the present invention;

Figure 8 is an expanded or partial view of Figure 1;

Figure 9 is a simplified drawing of the indexing system for the PAC torch and SAW welder;

Figure 10 is a rear view of the apparatus of Figure 6;

Figure 11 is a schematic view of the flux system of the SAW;

Figure 12 is a schematic view of the SAW arc portion;

Figure 13 is a schematic view of Figure 12 showing the wire feed motor and feed rolls and liquid flux;

Figure 14 is a schematic view of Figure 13 showing electrical connections for DCEP;

Figure 15 is a schematic view of the cooling water syste ;

Figure 16 is a schematic view of the base metal and weld showing the HAZ;

Figure 17 is a schematic view of the bead and base metal showing penetration;

Figure 18 is a side view of the preheat system; Figure 19 is a partial side view of Figure 6 illustrating the gas system;

Figure 20 is a schematic view of the controller; and Figure 21 is a schematic of the operator station of controller 292.

DESCRIPTION OF THE PREFERRED EMBODIMENTS A. Plasma Arc Gouging (PAC)

As shown in Figures 1 and 2, the PAC process PAC torch 101 operates on direct current, straight polarity between the negative electrode 100 and the positive electrode which is the workpiece 102 both of which are attached, respectively, to the negative and positive poles of power supply 103. Orifice gas 104 is introduced into the plenum 106 of constricting nozzle 108. An orifice 110 is placed at the end of nozzle 108 and has a throat 112 at its end, juxtaposed to the workpiece 102 by a standoff distance 114. An arc 116 is formed which is constricted by passing it through the orifice 110 downstream of the electrode 100. As orifice gas 104 passes through the arc 116, it is heated rapidly to a high temperature, expands, and is accelerated as it passes through the constricting orifice throat 112 toward the

workpiece 102. The intensity and velocity of the plasma 104 is determined by several variables including the type of gas, its pressure, the flow pattern, the electric current, the size and shape of the orifice 110, and the standoff distance 114 to the workpiece 102. The orifice 110 directs the super-heated plasma stream 104 from the electrode 100 toward the workpiece 102. The gas, preferably such as that described below, is heated and expands. As shown in Figure 8, as it expands the gas exits through the orifice at high velocity. When the stream 104 melts the workpiece 102, a high-velocity jet, shield gas or a water shield 119 is provided through a shield cup 121, and blows away the molten metal to form the kerf or cut. The cutting arc 116 attaches to or "transfers" to the workpiece 102, and is referred to as a "transferred arc."

PAC torches 101 are available in various current ranges from power supply 103, generally categorized as lower power [those operating at 30 amperes ("A") or less], medium power level [30- 100 (A)], and high power [from 100-1000 (A)]. Different power levels are appropriate for different applications, with the higher power levels being used for cutting thicker metal at higher speeds.

One of two starting methods is used to initiate the cutting arc: pilot arc starting or electrode (or tip) retract starting.

A pilot arc is an arc between the electrode and the torch tip. This arc is sometimes referred to as a "nontransferred arc" because it does not transfer or attach to the workpiece 102, as compared to the transferred arc which does. A pilot arc provides an electrically conductive path between the electrode 100 in the

torch and the workpiece 102 so that the main cutting arc can be initiated.

As shown in Figure 2, the most common pilot arc starting technique is to strike a high-frequency induced spark between the electrode and the constricting nozzle for which purpose, high frequency generator 118 is used in parallel with power supply 103 as is known in the prior art. One side of generator 118 is connected to the negative side of power supply 103. The other side of generator 118 is connected to constricting nozzle 105 and to power supply 103 by a resistor 120 and a relay contact 122 connected to the positive side of the power supply 103 and the workpiece 102, which is straight polarity or DCEN connection with the positive pole of power supply 208 connected to the workpiece 10 by cable 950 and the negative side connected to the torch 101 by cable 955. A pilot arc is established across the resulting ionized path. When the torch of constricting nozzle 108 is close enough to the workpiece 102 so that the plume or flame of the pilot arc touches the workpiece 102, an electrically conductive path from the electrode 100 to the workpiece 102 is established. The cutting arc will follow this path to the workpiece 102.

Retract starting torches (not shown) have a moveable tip or electrode so that the tip and electrode can be momentarily shorted together and then separated or "retracted" to establish the cutting arc.

The basic equipment for plasma gouging is the same as for plasma cutting, as shown in Fig. 1 and 2. Most plasma cutting equipment can be used for plasma gouging provided that the volt- ampere output curve of the power source permits substantially constant amperage at higher arc voltages and sustains the long

arc used for plasma gouging, such as preferably a model PCM-150 power supply, such as manufactured by L-TEC, Inc. of Florence, South Carolina.

The recommended plasma gas 104 for all gouging is argon plus 30-40 percent hydrogen and preferably 35-40 percent hydrogen. The gas can be supplied from cylinders or prepared using a gas- mixing device (Fig. 19) . There are two gases, compressed air 920 and H-35 925, which is 35% hydrogen and 65% argon, operating at 50 psig and 20 psig, respectively. The gases 920, 925 are supplied through regulators 930, 935, respectively, and hoses 940, 945, respectively, to torch 101. When the gases 920, 925 are mixed at the torch 101 and flow to the 20,000°F. tungsten electrode 100, the gases expand and rush through the orifice. Helium may be substituted for the argon of the argon-hydrogen mixture, but the resulting gouge will be shallower. A secondary or cooling gas, when used, may be argon, nitrogen, or air 117 as shown in Fig. 8. Selection is based on brightness of gouge desired, fume generation, and cost.

In areas where argon/hydrogen mix of gas, such as preferably 65%/35%, is unattainable, a procedure for compressed air may be used, but this procedure will be slower, and cut quality will not be as good. However, satisfactory cuts can be produced. The procedure is as set out in the chart below but with travel speed only 30 IPM and with stepover time at .10 sec. and stepover speed of 15 IPM, with the control variables still kept in a range of plus or minus 10%. The preferred gas for removing the hardbanding is H-35 (Argon 65 percent/Hydrogen 35 percent) . Argon is the best gas for transfer of current (Plasma) . Hydrogen is the best gas for heat transfer.

Air is sometimes used for the plasma gas on air operating systems but is generally limited to carbon steel gouging. Most manual air cutting systems are limited to 100 A output, and this restricts the size and speed of plasma gouging. By taking the major variables such as: volts, amps, travel speed, stepover time and speed and preferably controlling them by computer interface, a cut of quality can be achieved repeatable without the introduction of carbon. Travel speed increases reduce heat input and PAC is then 5 to 10 times faster than CAG. Making heat input very low in plasma gouging reduces the hard spots, and heat treatment that has had to be done in the past, because of carbon arc gouging, is no longer needed.

PAC requires a constant-current of drooping volt-ampere characteristic, relatively high-voltage direct-current power supply 103. To achieve satisfactory arc starting performance, the open circuit voltage of the power supply is generally about twice the operating voltage of the torch 101. Operating voltage will range from 50 or 60 volts (V) to over 200 volts (V) ; so PAC power supplies 103 will have open circuit voltages ranging from about 150 to over 400 volts.

There are several types of PAC power supplies 103, the simplest being the fixed output type which comprises a transformer and rectifier. The transformer of such a machine is wound with a "drooping" characteristic, so that the output amperage drops relatively little as the arc voltage increases, such as the power supply identified above.

In some cases, several outputs are available from a single power supply through a switching arrangement. This switching

arrangement can select between taps provided on the transformer or rectifier of the power supply.

Variable output power supplies are also available. The most widely used units utilize a saturable rectifier and current feedback circuit so that the output can be stabilized at the desired current level.

Other types of controls are available on plasma cutting power supplies 103, including electronic phase control and various types of "switch mode" power supplies. The switch mode power supplies utilize high-speed, high-current semiconductors to control the output. They can either regulate the output of a standard DC power supply, the so-called "chopper" power supply, or they can be incorporated in an inverter-type power supply. As new types of semiconductors become commercially available, it can be expected that improved versions of this type of power supply will appear. Switch mode supplies have the advantage of higher efficiency and smaller size and are attractive for applications where portability and efficiency are important considerations.

By fully automating the gouging process, the cutting or gouging depth heat input, stepover, travel speed and standoff distance can be controlled to closer tolerances, and higher quality gouging can be performed.

When any of the variables such as gas mixture are changed, a new procedure must be developed to have a consistent quality part that has been gouged and is ready for welding.

As discussed above, PAC parameters for use in the process are:

PARAMETERS FOR PLASMA GOUGING OF 4130 DRILL PIPE

ESO (STANDOFF DISTANCE AT THE TORCH ANGLE

PLUS DISTANCE TO TUNGSTEN OR OTHER ELECTRODE) lV

*AMPS 150 AMPS *VOLTS 400 OPEN CIRCUIT

TORCH ANGLE (ANGLE FROM THE DOWN VERTICAL) 30 DEG.

TORCH HEIGHT (STANDOFF DISTANCE AT THE TORCH ANGLE) 1/2"

♦TRAVEL SPEED (ROTATION SPEED OF PIPE) 50 IPM

GASES AND FLOW RATES COMPRESSED AIR - 50 PSI

ARGON-HYDROGEN (65%-35%, respectively) - 20 PSI

♦STEPOVER A. TIME (INTERVAL FOR STEPOVERS) .20 SEC.

B. SPEED (VELOCITY DURING STEPOVER) 20 IPM

CC POWER SUPPLY

* VARIABLES THAT MAY BE CONTROLLED

For the above variables, the following tolerances) are permitted before the procedure must be shut down:

VOLTS 400 OCV +/- 10%

AMPS 150 +/- 10%

TRAVEL SPEED 50 IPM +/- 5 IPM

STEPOVER A. TIME .2 SEC. +/- 10%

B. SPEED 20 IPM +/- 2 IPM

When 400 OCV, 150 Amps, 50 IPM and Stepover Speed and Time are maintained at the above pre-determined preferable tolerances, exact depth and width of the cut can be controlled to produce consistent quality cuts without the introduction of carbon in the cut surface.

Once a process for gouging has been established as set out above in the table, all of the variables have to be maintained to yield a consistent and repeatable part from which the hardbanding has been removed.

Amps are controlled at the power supply 103. They may be remote or controlled by setting through central control system 292 (Fig. 20) . Amps are controlled manually by turning a potentiometer either up or down on power supplies presently

existing in the art to the recommended amperage needed. Voltages are also automatically controlled as a basic part of the operation of a constant current power supply 103 as discussed above. A variation of more than +/- 40 volts will affect the desired results. The longer the arc length for the ESO, Figure 5, the higher the arc voltage, causing a variation in cut quality and depth. Arc voltage is controlled by ESO and can be maintained by control. Manual control is done as the process is in operation by raising or lowering the torch 101 by a knob or hand-wheel 982 connected to a rack and pinion device 983 (Fig. 6) attached to the torch 101. Automatic torch height can be programmed into the controller 292 and raised or lowered to desired ESO by servo motors (not shown) attached to the rack and pinion device 983 on the torch rigging assembly.

During operation of the torch 101 on the drill pipe 10, as shown in Figure 6, the head stock 30 which is connected to a motorized turning device 985 and tail stock rotate the pipe 10. The head stock 30 is connected to drill pipe 10 by sub 987. This rotation is controlled by the head stock 30 and effects cut width, depth and heat input. This rotation variable should be maintained within +/- 2 IPM. If the speed is faster than the procedure set out in the above chart, the cut will be shallower and will not cut out the recommended tungsten carbide from drill pipe 10. If the speed is too slow, the reverse occurs and too much is cut out and more welding has to be performed to fill up the metal removed. The rotation may be monitored by the central control system 292.

As the motorized head stock 30 rotates 360°, as shown in Fig. 8, a raised block 44 engages the limit switch 41. This

causes a signal to the indexing carriage 48 that transmits the programmed time 43 and speed 42 for indexing the torch.

The stepover (time and speed) is controlled by the electrical cross slide 50, as shown in Fig. 6 and 10, which allows the torch 101 to stepover a given amount every 360 degrees of rotation of pipe 10, controlled by a limit switch 41 mounted on the head stock 30, allowing the electrical cross slide to move at a preset controlled speed 42 for a preset time 43, as set out in the table set out above.

As shown on Figure 10, the motorized drive pinion 400 controls movement on the cross slide rack 410 step over which is set for motor speed and time. Power is supplied to the motor of drive pinion 400 which will determine how far the torch 101 travels after a 360° rotation of the headstock 30.

All other variables - the torch 101 height and angle controlled by knob 982 and a swivel with set screws (not shown) , and lateral movement, controlled by torch cross slides 50, and gases 104, pressure and volume, will be controlled in the following manner.

The torch height is controlled by rack and pinion gears 983 that will adjust up or down manually to the correct height set out in the above table. The angle of the torch is manually controlled by adjusting the torch holder to the desired angle. The gases are controlled manually by adjusting the regulators on the gas system for that needed for the cutting or gouging procedure to be preformed, as set out in the above table.

The programmable logic control (PLC) 292 controls and monitors the process through both digital and analog inputs and outputs. The PLC scans inputs as identified in the above table,

calculates new output status based on input status and preprogrammed data and then updates the outputs every 20 to 40 milliseconds. This rapid update rate is needed for reliable control of plasma systems. The man-machine interface consists of a membrane keypad with liquid crystal display (LCD) and a series of operator control buttons. The LCD is used for operator prompting and message display. The key pad is used for data input. Normal operation is via the operator push buttons.

The system is designed to set control and monitor critical process variables and thus ensure repeatability within each production batch.

There are 3 modes of operation of controller 292, EDIT, RUN and DIAGNOSTICS. A. Edit is to change parametric data stored in the controller.

1. Level I: process variables. a. To edit process variables the operator selects EDIT mode and passes security. On the LCD the system prompts the operator to input arc amperage, surface speed, step over width, starting dimension, outside diameter and total cut width via a membrane keypad. This data is stored in the PLC in nonvolatile battery backed RAM.

2. Level II: system variables. a. To edit system variables the operator selects EDIT mode and passes security. On the LCD display the system prompts the operator to input process variable tolerance levels and system

setup data. This data is stored in nonvolatile battery backed RAM. B. Run

1. Part loading.

The operator loads a piece of drill stem 10 into the system and "makes up" the box onto the pin mounted on the headstock 30. Plasma torch angle and radial position are set manually by the operator as discussed above.

2. Cycle start.

The operator presses the cycle start button which begins the process. If the process variables have not been entered (using the EDIT mode) , the system faults and changes to DIAGNOSTICS mode. If the process variable have been entered, the plasma torch slide 50 moves to the programmed start dimension. The PLC converts the programmed arc amperage to a calibrated analog signal interfaced to the plasma power supply 103. The PLC commands arc initiation using a digital output connected to the plasma power supply 103. The plasma power supply begins a timed gas preflow period followed by arc initiation as discussed above. The PLC pauses until a current sensor confirms that the arc is established at the desired amperage. If the arc fails to initiate within a programmed period, the system faults and changes to DIAGNOSTIC mode. Anytime during the process that the arc amperage fails to hold within the tolerance level as set out in the above table, the system faults and changes to DIAGNOSTIC

mode. The PLC monitors the status outputs of the plasma power supply 103. If at anytime during the process a power supply fault occurs (e.g. low gas pressure, overheat or over current) , then the system 292 faults and changes to DIAGNOSTIC mode. The PLC 292 calculates the head stock 30 rotational velocity using the programmed diameter and surface speed data set into the system. This value is converted to an analog value and output 42 to the headstock 30. The PLC monitors the position of the headstock using a digital pulse generator (not shown) mounted on the headstock drive motor. Anytime during the process if the headstock 30 velocity fails to hold within the tolerance level as set out in the above table, the system faults and changes to DIAGNOSTIC mode. 3. In cycle.

Once the process has begun, the PLC monitors head stock position, once a complete revolution has occurred, the plasma torch slide steps over the programmed increment as discussed above. The PLC monitors the slide position using an optical quadrature encoder (not shown) . If the slide fails to move within tolerance, the system 292 faults and changes to DIAGNOSTIC mode. This process continues until the programmed cut width is achieved. At anytime during the cycle, the operator may press either the cycle stop, or emergency stop buttons.

4. Cycle stop options. a. Normal cycle stop.

After the system cuts to the programmed width, the PLC stops the arc and the headstock. The plasma power supply cools the torch with a timed plasma gas post flow period, and then stops gas flow. The drill pipe 10 is ready for unloading. b. Cycle interrupt.

If the operator presses the cycle stop button or a process variable exceeds the programmed tolerance, the system will immediately begin a controlled shutdown. This sequence is identical to the normal cycle stop except that the system will change to DIAGNOSTIC mode and display an appropriate error message. c. Emergency sto .

If the operator presses the emergency stop button, the main control relay is de-energized immediately stopping the arc and all motion. The PLC remains energized. The system changes to DIAGNOSTIC mode and displays an appropriate error message. C. DIAGNOSTIC mode

Diagnostic mode may be entered either by selecting the mode or due to a fault condition.

1. Normal use of DIAGNOSTIC mode.

To enter DIAGNOSTIC mode the operator selects that mode and passes security. This mode is used to manually test and observe automatic functions and calibrate analog outputs.

2. Fault condition.

If the system experiences a fault of emergency stop DIAGNOSTIC mode is automatically started. The LCD will display an error message and prompt the operator to resolve the problem and clear the error. Once the fault is cleared the system returns to RUN mode. B. Submerged Arc Welding

As shown in Figures 3, 12, 13 and 14 in submerged arc welding apparatus 200, the end 201 of a continuous bare wire electrode 202 is inserted below a current pick-up tip or nozzle 203 into a mound of flux 204 that covers the seam or area or joint 214 to be welded. ESO (Figure 12) is defined as the distance between the top of the base metal 212 and the bottom at contact tip 203. The wire 202 is fed by an automatic wire feed 206, as is well known in the art, of a wire feed motor 209 driving feed rolls 211, into a feeder 207. Wire 202 is connected through feeder 207 to one side of a power supply 208 by a contact connector 210 connected to tip 203 by cable 213. The other side of power supply 208 is connected to a base metal holder 212 by a work connection 214.

An arc 252 is initiated using one of six arc-starting methods known in the art. Wire-feeding mechanism 206 begins to feed the electrode wire 202 towards the joint, by drive feed rolls 211 at a controlled rate by controller 292 upon arc 252

initiation, and the feeder 207 is moved manually or automatically along the weld seam 256. For manual or automatic welding, the work may be moved beneath longitudinally stationary wire feeder 207.

As shown in Figures 3 and 11, additional flux 204 is continually fed from flux hoppers 218 through a flux feed tube 216 to welding heads or welding location with flux 204 from a flux hopper 218 in front of and around the electrode 202, and continuously distributed over the joint 224. Heat 255 evolved by the electric arc 252 (Figures 12, 13) progressively melts some of the flux 204 into a liquid 253 that cools to a fuse flux 254, as well as the end of the wire 202 to the weld 256, and also the adjacent edges of the base metal 212, creating a pool of molten metal 258 beneath the layer of liquid slag 253. A flux shelf 260 comprised of unmelted flux deposits 262 and fused flux 254 is used to control flux deposits. The melted bath near the arc is in a highly turbulent state. Gas bubbles are quickly swept to the surface of the pool. The flux 253, 254, 262 floats on the molten metal 258 and completely shields the welding zone from the atmosphere.

The liquid flux 253 may conduct some electric current between the wire 202 and base metal 212, but an electric arc 252 is the predominant heat source. The flux blanket 253, 254, 262 on the top surface of the weld pool prevents atmospheric gases from contaminating the weld metal 222 of weld 256, and dissolves impurities in the base metal 212 and electrode 202 and floats them to the surface. The flux 204 can also add or remove certain alloying elements to or from the weld metal 222.

As the welding zone progresses along the seam 224, the molten metal 258 cools so the weld metal 222 and then the liquid flux 253 cools and solidifies, forming a bead of weld metal 222 and a protective slag shield 226 of fused flux 254 over it. It is important that the slag 226 is completely removed before making another weld pass.

Thus the equipment required for submerged arc welding includes (l) a power supply 208, (2) an electrode delivery system 206, 207, 209, 211, 251 (3) a flux distribution system 216, 218, (4) a travel arrangement of the same sort as used for the torch system discussed above, and (5) a process control system 292 as discussed above and below. Optional equipment includes flux recovery systems as discussed below and positioning or manipulating equipment of the same sort as used for the torch system discussed above.

The power source 208 chosen for a submerged arc welding system plays a major operating role. Several types of power supply are suitable for submerged arc welding. A dc power supply may be a transformer-rectifier or a motor or engine generator, which will provide a constant voltage (CV) , constant current (CC) , or a selectable CV/CC output. AC power supplies are generally transformer types, and may be a transformer-rectifier or a motor or engine generator, which will provide a constant voltage (CV) , constant current (CC) , or a selectable CV/CC output. AC power supplies are generally transformer types, and may provide either a CC output or a CV square wave output. Because SAW is generally a high-current process with high-duty cycle, a power supply capable of providing high amperage at 100 percent duty cycle is recommended.

DC constant-voltage power supplies are available in both transformer-rectifier and motor-generator models. They range in size from 400 A to 1500 A models. The smaller supplies may also be used for GMAW and FCAW. These power sources are used for semi-automatic SAW at currents ranging from about 300 to 600 A with 1/16", 5/64", and 3/32". (1.6, 2.0, and 2.4 mm) diameter electrodes 202. Automatic welding is done at currents ranging from 300 A to over 1000 A, with wire 202 diameters generally ranging from 3/32" to 1/4" (2.4 to 6.4 mm). However, applications for dc welding at over 1000 A are limited because severe arc blow may occur at such high current. As shown in Figures 13 and 14, direct current electrode positive ("DCEP") , which is reverse polarity, allows current flow to have the heat in the base metal 212 giving deeper penetration and allowing a higher deposition rate by increasing wire feed speed at the same amperage than direct current electrode negative ("DCEN") used with the torch.

A constant-voltage power supply 208 is self-regulating; so it can be used with a constant-speed wire feed 206. No voltage or current sensing is required to maintain a stable arc; so very simple wire feed speed controls 206 may be used. The wire feed speed and wire diameter control the arc current, and the power supply controls the arc voltage.

Constant-voltage dc power supplies 208 are the most commonly used supplies for SAW. They work well for most applications where the arc current does not exceed 1000 A, and may work without a problem at higher currents. The CV dc power supply is preferred for high-speed welding of thin steel.

Control of the operating variables in SAW is essential if high production rates and welds of good quality are to be obtained. These variables, in their approximate order of importance, are:

(1) Welding amperage

(2) Type of flux and particle distribution

(3) Welding voltage

(4) Welding speed

(5) Electrode size

(6) ESO

(7) Type of electrode

(8) Width and depth of the layer of flux

Welding current or amperage is the most influential variable because it controls the rate at which the electrode 202 is melted, and therefore the deposition rate, the depth of penetration, and the amount of base metal melted. If the current is too high at a given travel speed, the depth of fusion or penetration will be too great. The resulting weld may tend to melt through the metal being joined. High current also leads to waste of electrodes in the form of excessive reinforcement. This overwelding increases weld shrinkage and causes greater distortion. If the current is too low, inadequate penetration or incomplete fusion may result. Thus:

(1) Increasing current increases penetration and melting rate.

(2) Excessively high current produces a digging arc and undercut, or a high, narrow bead.

(3) Too low welding current produces an unstable arc.

Welding voltage adjustment varies the length of the arc between the electrode and the molten weld metal. If the overall voltage is increased, the arc length increases; if the voltage deceased, the arc length decreases. Voltage has little effect on the electrode deposition rate, which is determined by welding current. The voltage principally determines the shape of the weld bead cross section and its external appearance. Increasing the welding voltage with constant current and travel speed will:

(1) Produce a flatter and wider bead.

(2) Increase flux consumption.

(3) Tend to reduce porosity caused by rust or scale on steel.

(4) Help bridge an excessive root opening when fit-up is poor.

(5) Increase pickup of alloying elements from an alloy flux.

Excessively high-arc voltage will:

(1) Produce a wide bead shape that is subject to cracking.

(2) Make slag removal difficult in groove welds.

(3) Produce a concave shaped weld that may be subject to cracking.

(4) Increase undercut along the edge(s) of fillet welds. Lowering the voltage produces a "stiffer" arc, which improves penetration in a deep weld groove and resists arc blow. An excessively low voltage produces a high, narrow bead and causes difficult slag removal along the bead edges.

With any combination of welding current and voltage, the effects of changing the travel speed conform to a general pattern. If the travel speed is increased, (1) power or heat

input per unit length of weld is decreased, and (2) less filler metal is applied per unit length of weld, resulting in less weld reinforcement. Thus, the weld bead becomes smaller. For example, using the formula for Joules per inch of weld:

Volts (Load Voltage) X Amps X 60 divided by Travel Speed in inches per minutes - Heat Input, yields:

1.) 200 A X 20 V X 60 divided by 20 IPM = 12,000 Joules

(Heat Input) 2.) 300 A X 30 V X 60 divided by 10 IPM = 54,000 Joules 3.) 400 A X 35 V X 60 divided by 5 IPM = 168,0000 Joules 4.) 350 A X 28 X 60 divided by 30 IPM - 19,600 Joules Weld penetration is affected more by travel speed than by any variable other than current. This is true except for excessively slow speeds when the molten weld pool is beneath the welding electrode. Then the penetrating force of the arc is cushioned by the molten pool. Excessive speed may cause undercutting.

Within limits, travel speed can be adjusted to control weld size and penetration. In these respects, it is related to current and the type of flux. Excessively high travel speeds promote undercut, arc blow, porosity, and uneven bead shape. Relatively slow travel speeds provide time for gases to escape from the molten metal thus reducing porosity. Excessively slow speeds produce (1) a convex bead shape that is subject to cracking, (2) excessive arc exposure, which is uncomfortable for the operator, and (3) a large molten pool that flows around the arc, resulting in a rough bead and slag inclusions.

At current densities above 80,000 A/in (125 A/mm), ESO becomes an important variable. At high-current densities,

resistance heating of the electrode 202 between the contact tube 203 and the arc 252 increases the electrode melting rate. The longer the extension, the greater is the amount of heating and the higher the melting rate. This resistance heating is commonly referred to as I 2 R (power) heating. As shown in Figures 4, 5 and 12, if 1/8" wire is used, the ESO should be eight times the diameter of the wire, which is 1" from the end of the contact tube or tip 203. If the height is increased, the corresponding amperage will decrease because the wire is preheated by electrical resistance, requiring less amperage to burn the electrode wire off at the end. The reverse is also a true statement. If the ESO is shortened from 1", the electrical resistance is smaller, and the wire is not preheated as much, causing the amperage to increase to burn the wire off. This raising and lowering of the ESO from one inch, for example, will either increase or decrease the arc voltage. Increase in arc voltage allows the puddle to become more fluid and harder to handle, especially when welding on pipe with 6- " outer diameter. Decreasing arc voltage will decrease the fluidity of the puddle and possibly make cold laps and lack of fusion. This also increases the height of the bead making an adjustment in travel speed and stepover time speed, necessary to stay within the limits of acceptable quality.

Thus, in developing a procedure, an ESO of approximately eight times the electrode diameter is a preferred starting point. The length is modified to achieve the optimum electrode melting rate with fixed amperage.

As discussed above, increased electrode extension adds a resistance element in the welding circuit and consumes some of

the energy previously supplied to the arc 252. With lower voltage across the arc 252, bead 222 width and penetration decrease. Because lower arc voltage increases the convexity of the bead 222, the bead 222 shape will be different from one made with a preferred electrode extension. Therefore, when the ESO is increased to take advantage of the higher melting rate, the amperage or wire feed setting should be increased to maintain proper arc 252 length.

The condition of the contact tube 203 also affects the effective electrode extension. Contact tubes 203 also should be replaced at predetermined intervals, such as daily, to insure consistent welding conditions.

An increase in deposition rate is accompanied by an increase in penetration. Therefore changing to a long electrode extension is not recommended when deep penetration is needed. When melt- through is a problem, as may be encountered when welding thin gauge material, increasing the electrode extension may be beneficial. However, as the ESO increases, it is more difficult to maintain the electrode tip 201 in the correct position with respect to the joint.

The width and depth of the layer of granular flux 204 influence the appearance and soundness of the finished weld as well as the welding action. If the granular layer is too deep, the arc 252 is too confined and a tough ropelike appearing weld will result. The gases generated during welding cannot readily escape, and the surface of the molten weld metal 258 becomes irregularly distorted. If the granular layer is too shallow, the arc 252 will not be entirely submerged in flux 204. Flashing and

spattering will occur. The weld will have a poor appearance, and it may be porous.

An optimum depth of flux exists for any set of welding conditions. This depth can be established by slowly increasing the flow of flux 204 until the welding arc 252 is submerged and flashing no longer occurs. The gases will then puff up quietly around the electrode, sometimes igniting.

During welding, the unfused granular flux 262 can be removed a short distance behind the welding zone after the fused flux 254 has solidified. However, it may be best not to disturb the flux 260 until the heat from welding has been evenly distributed throughout the section thickness. Fused flux 260 should not be forcibly loosened while the weld metal 222 is at a high temperature [above 1100 degrees Fahrenheit (600 degrees Centigrade) ] . The fused material 260 will readily detach itself when allowed to cool. Then it can be brushed away with little effort. Sometimes a small section of fused flux 260 may be forcibly removed for quick inspection of the weld surface appearance. It is important that no foreign material be picked up when reclaiming the flux. To prevent this, a space approximately 12 inches (300 mm) wide should be cleaned on both sides of the weld joint before the flux 260 is laid down. The flux 204 may be recovered, as discussed below. If the recovered flux 204 contains fused pieces, it should be passed through a screen with openings no larger than 1/8 inch (3.2 mm) to remove the coarse particles, as discussed below.

The flux 204 is thoroughly dry when packaged by the manufacturer. After exposure to high humidity, it should be

dried by baking it before it is used. Moisture in the flux 204 will cause porosity in the weld.

The submerged arc welding process has long been recognized for its versatility and excellent economics. It operates well at low or high currents, using single or multiple electrodes on plate from thin gage to unlimited thickness. A key to this performance versatility is the flux, itself — it blankets the arc, eliminates flash, spatter and smoke, controls arc stability, governs bead shape, and influences weld chemistry.

Submerged arc welding fluxes are granular, fusible mineral compounds in various proportions manufactured by several different methods. All of them share certain restrictions in design to assure operability, but differ widely in composition, offering many unique performance features.

Bonded fluxes are dry mixed powders bonded together with a binder such as sodium silicate, differing from prefused fluxes where all ingredients are melted in an electric furnace. The lower temperatures required for bonding permit the addition of metallic deoxidizers and alloying ingredients which can radically change the character of a flux system.

Fused fluxes are characterized by their extremely good chemical homogeneity and non-hygroscopic nature. Fines may be removed without changing the composition of the flux. Changes to the slag/metal consumption ratio will not appreciably affect deposit strength levels. On the other hand, bonded fluxes can be made to exhibit higher degrees of rust and scale tolerance through deoxidizer additions such as metallic manganese and silicon. The combination of fine ingredients mechanically bonded into larger particles provides good performance over a

wide range of applications with one mesh size. Some restrictions in use are necessary to avoid unintentional or non-uniform alloying of the weld metal.

For the past 30 years there has been extensive development of both bonded and fused systems. Originally, these two lines of development were based on proprietary conditions; however, today it is recognized that both bonded and fused have characteristics and advantages in the wide variety of welding application. Of course, there are overlaps in application suitability making property flux selection sometimes arbitrary, but they allow the knowledgeable user to "fine-tune" the job for productivity and economy.

Fluxes perform many key roles in addition to providing atmospheric protection for the molten metal and influencing the mechanical properties of the weld deposit, as discussed above. Their actions in controlling arc stability, bead shape, puddle fluidity, rust and scale tolerance, peeling characteristics, and welding speed capabilities give the submerged arc process its inherent application flexibility. The complex interactions of the oxide, fluoride, and metallic ingredients in a flux give each its own personality. For use with drill pipe it is preferred to use Linde 429 flux, which is semi-active by design.

Contrary to the implied implications of the term "neutral flux", it is believed that all fluxes interact with weld chemistry, and none are totally "neutral". One may define an active flux, at least to its sensitivity for large build up of alloying elements such as manganese and silicon when voltage variations occur. The gray area of defining a neutral flux has been very difficult for both the flux manufacturers as well as

the users and specifiers. All fluxes, to some degree, either lose alloy or pickup alloy, and affect the weld deposit analysis. It is especially true with bonded or agglomerated fluxes because of the method of manufacturing.

Words like "nearly neutral", "semi-active" have been employed, but no clear definition is available from any of the American Welding Society, flux manufacturers, or specifiers. Unfortunately, fluxes are often used for welding without either user's or specifier's sensitivity to the fact that fluxes do add or take away some amount of alloy to the weld deposit and, in fact, can cause excessive (hard spots) , or a reduction in, weld strength.

In automatic SAW welding of drill pipe tool joints, many more factors must be addressed to complete a sound, quality weld. These include the following:

(1) Heat input in Joules - Volts X amps X 60 divided by Travel Speed = Heat Input, discussed above.

(2) Index - (Speed and Time) - To give height of weld and time of completion, as discussed above.

(3) Preheat - Eliminates thermal shock and build-up of stresses. Preheat is usually accomplished using oxygen and fuel gas mixture. Usually a 5000° F. torch preheat tip 990 (Fig. 18) is used to reach a preheat temperature of 300° F. for about three minutes, using fuel gas 992 and oxygen 994 as is known in the art.

(4) Internal Water Volume - As shown in Figure 15 water 400 is fed through a throttle valve 405 into the head stock 30 and there into though a box and pin arrangement 410, using a brass gasket to hold the pin

410. The water 400 is not introduced to the pipe 10, which is the base metal 212 work piece for SAW and past the point of welding where the wire 202 feeders 207 are located. The water 400 then enters a reservoir 415 via a filter 420 and is finally filtered by filters 425 before it is pumped by a pump 430 again to valve 405. The water helps control interpass temperature of the part to be welded.

(5) Interpass Temperature - Controls the finished weld by eliminating the heat effect of varying degrees during welding. The internal water volume set out above is used to control the interpass temperature to a range of 300° F. to 400° F.

(6) Post Weld Heat Treating - Brings the weld heat affected zone ("HAZ") 500 (Figure 16) and base metal 212 that surround the weld metal 222 to a more homogeneous temperature reducing the brittleness of the weld. The HAZ has undergone a slight molecular change because of the temperature of the molten weld metal 222. A post weld heat treatment at 700° F. for 4137 drill pipe will return the HAZ to the original molecular structure.

Drill pipe 10, such as 4137 drill pipe, as the workpiece of the preferred embodiment of the present invention is a high carbon, low alloy (HCLA) and very crack sensitive material that requires the correct variables to give the procedure the consistency that yields crack-resistant welds.

Variables, such as E.S.O., Amps, Volts, Travel Speed, Flux height and type, wire diameter and type, wire feed speed, open

circuit voltage, load voltage, arc voltage, pre-heat, interpass temperature, post-heat, internal water flow, cooling rate, stepover speed and time, torch centering, affect grain structures, size and boundaries of the weld metal and base material and have a major influence on Rockwell C hardness and impacts.

All variables, when controlled to tolerances of the process will determine the total integrity, hardness, structure, testing, and reliability of the weld.

If only one of the controlled variables are changed, the end result of weld integrity may be changed. For example:

A. Amperage - If amperage is lowered from the preferred procedure -

1.) Weld height will decrease,

2.) Penetration 600, as illustrated in Figure 17, will decrease, 3.) Heat input will decrease, 4.) Grain structure will decrease in size, 5.) Weld size will be smaller, 6.) Deposition will decrease, 7.) Production will decrease and 8.) Preheat, interpass temperature and post heat will have to be changed to compensate for the change in procedure.

B. Travel Speed - If travel speed is lowered -

1.) Weld width 610, as illustrated in Figure 17, will widen, 2.) Weld height will decrease, 3.) Penetration will increase,

4.) Heat input will increase, 5.) Grain size will increase, 6.) Weld integrity will suffer, 7. ) Test results will not be as desired and 8.) Preheat, interpass temperature and post heat will have to be changed to compensate for the change in procedure. C. Amps vs Wire Feed Speed:

When using consistent voltage power supplies 208, welding current (amps) is controlled by the electrode wire feed speed of automatic wire feed 206, which is controlled manually by the controller 292 of the wire feed motor 209 controlling feed rolls 211.

The wire feed speed/current relationship can be changed by the electrical stickout ESO (Figure 12) . Preheating of the electrode 202 occurs in the ESO, and it is sometimes called I 2 R heating, as discussed above. This is because the electrode wire 202 extending from the current pickup tip or contact tip 203 to the arc 252 is heated by the amount of current being carried by the wire 202. The power, and thus heat, generated in this position of the electrode wire 202 is equal to the current (I) , squared, times the resistance (R) of the electrode wire, I 2 R. This preheats the electrode wire 202, unspooled from feed rolls 211, and indexed by indexing carriage 50 so that when it enters the arc, it is at an elevated temperature which increases the melt-off rate.

For these reasons, the wire feed speed is controlled at a given ESO, such as preferably 1 inch. An example of this procedure is:

(1) 300 amps (70" IPM wire feed speed) at 1 inch ESO with 3/32 inch diameter wire;

(2) 3/32 inch = 519 inch per pound of wire;

(3) deposition rate is 8.09 pounds of wire per hour Consistency of deposition rate and wire feed speed maintains consistency or quality of the weld.

D. Volts (OCV, Load, Arc) :

Open circuit voltage of the power supply 208 is that voltage without current flowing. Load Voltage is the voltage when the circuit is complete and current is flowing. This is the actual voltage of the power supply 208 under load and is preferably read at the supply 208. Arc Voltage is measured at the arc 252 and is preferably read at manual controls 209 and at any control system 211.

Just as with welding current, the arc voltage increases as the ESO is increased, but will not effect the load voltage supplied by the power supply 208 and is read on the face of the supply 208.

E. Travel Speed vs. Other Variables:

Travel speed of a bus bar and nozzle assembly 90 over the head/sub 987 and drill pipe 10 mounted between motorized head stock 30 and tail stock 45 controls the height and width of the weld and is the third member that influences heat input. At a given volt/amp setting, the faster the travel speed, the lower the heat input. The lower heat input will lower the height of the weld bead and make the weld narrower. The reverse is true for slower travel speed. The weld integrity is controlled by heat input. As discussed above, the following is heat input as determined by:

Volts x Amps X 60 ÷ Travel Speed, measured in joules This is also amenable to control by the central control system 292.

F. Flux:

Flux height is controlled by the amount of flux 204 from the flux feed tube 216 fed by flux hopper 218. If the flux height is too high, the flux 204 holds in the heat of the bead, making it more fluid and flatter. If the flux height is too low, the flux 204 allows the puddle to cool too rapidly, giving a higher crown and the possibility of causing porosity in the weld by not having enough coverage of flux. This is a variable which is also amenable for control by central control system 292.

As shown in Figure 11, the flux may also be recycled by recovery system 80. A flux recovery unit is a vacuum cleaner. It vacuums up the unfused flux through a screen 700, to filter out large particles that could clog the system and impede the flow or flux 204 to a flux catch pan 702. The catch pan 702 is connected by a tube or pipe 704 to a vacuum flux recovery system 2926. The discharge from system 292 returns to hoppers 218. It is controlled by an on and off switch (manually) .

G. Stepover:

During operation of the bus bar and nozzle assembly 90 on the drill pipe 10, the motorized head stock 30 and tail stock 40 rotate the pipe 10, with coolant supplied through a rotary union 50, as discussed above. After 360 degrees of rotation, the indexing carriage 50 will move over. This is accomplished by energizing the motor for a fixed time to traverse the exact distance desired for exact lap over of the preceding pass. This will determine the time to complete a weld and the build up per

pass. The end result of the two variables of stepover time and speed are to cause the following welding bead to lap over the preceding bead approximately 50% and it take a corridanition between speed of carriage and time of rotation to accomplish this task.

All of this may be controlled by the central control system 292, in the following manner:

The system includes a PLC for setting the head stock 40 to rotate the drill pipe 10, two or three submerged arc welding power supplies 208 and arc generating equipment, a precision cross slide assembly 50 to position the arc generating equipment by indexing carriage which is part of the assembly 50. The PLC controls and monitors the process through both digital and analog inputs and outputs. The PLC scans inputs, calculates new output status based on input status and preprogrammed data and then updates the outputs every 20 to 40 milliseconds. The man-machine interface consists of a membrane keypad 890 with LCD 895 and a series of operator control buttons 898. The LCD 895 is used for operator prompting and message display. The key pad 890 is used for data input. Normal operation of the machine is via the operator push buttons.

The system is designed to control and monitor critical process variables to implement repeatability within each production batch.

There are 3 modes of operation EDIT, RUN and DIAGNOSTICS. A. Edit.

1. Level I: process variables. a. To edit process variables the operator selects EDIT mode and passes security. On the LCD the

system prompts the operator to input wire speed, arc voltage, surface speed, step over width, starting dimension, outside diameter and total weld width via a membrane keypad. This data is stored in the PLC in nonvolatile battery backed RAM. 2. Level II: system variables. a. To edit system variables the operator selects EDIT mode and enters the Level II password. On the LCD display the system prompts the operator to input process variable tolerance levels and system setup data. This data is stored in nonvolatile battery backed RAM. B. Run

1. Part loading.

The operator loads a piece of drill stem 10 onto the system and "makes up" the connection with the matching joint mounted on the head stock 30. Welding torch angle, height, radial position and spacing are set manually by the operator.

2. Cycle start.

The operator presses the cycle start button which begins the process. If the process variables have not been entered (using the EDIT mode) the system faults and changes to DIAGNOSTICS mode. If the process variables have been entered, the indexing carriage 50 moves to the programmed start position. The PLC converts the programmed arc voltage and wire speed into calibrated analog signals interfaced to the

welding power supplies 208. The PLC opens the flux gates (not shown) and dwells for the programmed period to permit initial flux 204 build up. The PLC outputs a calibrated analog arc voltage command signal (not shown) to the welding power supplies 208 then commands arc initiation using digital outputs (not shown) connected to the welding power supplies 208. The power supply energizes the welding wire 202. The PLC pauses until open circuit voltage is detected via a analog input. If the PLC fails to detect voltage within the programmed period the system faults and changes to DIAGNOSTIC mode. If the voltage is detected, the PLC outputs a calibrated analog wire feed command signal to the wire feeders 251, to control the wire feed motor 209, then commands wire feed forward using a digital output (not shown) . Once the wire 202 contacts the workpiece 212, the wire feed motor 209 reverses to establish an arc 252. Once current is detected by the PLC, it begins to feed the wire 202 at the preprogrammed welding rate by feed control 251. The PLC pauses to confirm that all arcs are established at the desired amperage. If the arc fails to initiate within the programmed period, the system faults and changes to DIAGNOSTIC mode. If at anytime during the process the arc voltage or wire speed fails to hold within the tolerance level, the system faults and changes to DIAGNOSTIC mode. The PLC monitors the status outputs of the welding power supplies 208. Anytime during the process the if a

power supply fault occurs (e.g. overheat or over current) , the system faults and changes to DIAGNOSTIC mode. The PLC calculates the head stock 30 rotational velocity using the programmed diameter and surface speed data. This value is converted to an analog value and output to the head stock 30. The PLC monitors the position and velocity of the head stock 30 using a digital pulse generator mounted on the head stock drive motor as discussed above with respect to Figure 9. Anytime during the process if the head stock 30 velocity fails to hold within the programmed tolerance level the system faults and changes to DIAGNOSTIC mode. 3. In cycle.

Once the process has begun the PLC monitors head stock 30 position. When a complete revolution has occurred the indexing carriage 50 steps over the programmed increment. The PLC monitors the slide position using an optical quadrature encoder (not shown) . If the slide fails to move within tolerance, the system faults and changes to DIAGNOSTIC mode. This process continues until the programmed weld width is achieved. After a programmed delay, cooling water is forced through the inside of the drill stem 10 in the manner described above for Figure 15 to maintain correct interpass temperature. At anytime during the cycle the operator may press either the cycle stop, or emergency stop buttons.

4. Cycle stop options. a. Normal cycle stop.

After the system welds to the programmed width the PLC stops the arcs 252, the head stock 30 and the cooling water 400. The welding wire 202 is retracted from the puddle 258 and the flux gates (not shown) are closed. The drill pipe 10 is ready for unloading. b. Cycle interrupt.

If the operator presses the cycle stop button or a process variable exceeds the programmed tolerance, the PLC will immediately begin a controlled shutdown. This sequence is identical to the normal cycle stop except that the system will change to DIAGNOSTIC mode and display an appropriate error message. c. Emergency stop.

If the operator presses the emergency stop button, the main control relay is deenergized immediately removing power and stopping the arc and all motion. The PLC remains energized. The system changes to DIAGNOSTIC mode and displays an appropriate error message. C. DIAGNOSTIC mode

Diagnostic mode may be entered either by selecting the mode or due to a fault condition.

1. Normal use of DIAGNOSTIC mode.

To enter DIAGNOSTIC mode, the operator selects that mode and enters the appropriate security. This mode is used to manually test and observe automatic functions and calibrate analog outputs.

2. Fault condition.

If the system experiences a fault of emergency, or other abnormal stop, DIAGNOSTIC mode is automatically started. The LCD will display an error message and prompt the operator to resolve the problem and clear the error. Once the fault is cleared the system returns to RUN mode.

The following equipment and controlled variables are used in the processes of deal wire sub arc for build up of drill pipe:

PARAMETERS FOR DUAL WIRE SUB ARC BUILD-UP OF 4130 DRILL PIPE

WIRE SIZE 3/32" (.0938)

WIRE TYPE AWS #E4130

FLUX TYPE ACTIVE-BONDED

*VOLTS PER WIRE 28

*AMPS (WIRE FEED SPEED) 300 (60 IPM)

*TORCH TRAVEL SPEED 31 IPM

*ESO 1"

HEAT INPUT (KILOJOULES) 16,258

*FLUX HEIGHT .5"

POWER SUPPLY SLOPE 3 VOLTS PER 100 AMPS

OPEN CIRCUIT VOLTAGE (OCV) 40

SHORT CIRCUIT CURRENT 1200 AMPS TORCH POSITION VERTICAL AND 1/2" FROM CENTER LINE

AWAY FROM THE DIRECTION OF THE TURN

INCH CONTROL (FREE FLOW OF WIRE) 50 IPM

BURN BACK CONTROL (TO FREE WIRE AT STOP) .3 SEC.

*INDEX A. SPEED 14 IPM

B. TIME .2 SEC.

INDEX LAP OVER 50% OF PREVIOUS BEAD

WATER VOLUME 30 GALLONS PER HOUR WATER TEMPERATURE 80 DEG. F. MIN.-150 DEG. F. MAX

PRE-HEAT TEMPERATURE 300 DEG. F.

INTER PASS TEMPERATURE 250 - 300 DEG. F.

POST WELD HEAT 700 DEG. F.

RATE OF COOLING SLOW CV POWER SUPPLY

* VARIABLES THAT MAY BE CONTROLLED.

For the above variables, the following tolerances are permitted before the procedure must be shut down:

VOLTS 28 +/- 1 VOLT

AMPS 300 +/- 25 AMPS

TRAVEL SPEED 30 IPM +/- 2 IPM

ESO 1" +/- 1/32"

FLUX HEIGHT 1/2" +/- 1/32"

STEPOVER A. TIME .2 SEC +/- 10%

B. SPEED 14 IPM +/- 1 IPM

By controlling the above variables and their tolerances, a consistent and repeatable quality of weld can be made.

It has been found that a particular type of alloy flux cored wire is particularly useful, is not known in the prior art, and yields unusually good test results for a submerged arc welding. For submerged arc welding for drill pipe, alloy flux cored wire is not used at all in the prior art. Alloy flux cored wire is wire having a sheath of mild steel and an inner core of powdered, uniformly mixed elements. In particular, a formulation having the following elements in the following weight percents, with a maximum tolerance on the percents of plus or minus ten percent, have been found to be a preferred wire. The chemical analysis of the percents by weight of the elements of the preferred wire 202 is carbon .09, manganese 2.05, phosphorus .012, sulphur .015, silicon .42, nickel .58, molybdenum .53, chromium .03, copper .02, and vanadium .02, with iron making up the remaining percent to equal 100%. It is preferred that the chemical analysis given above be varied by no more than plus or minus five percent for optimal results to avoid having to make unusual changes in the process or procedure which may not be commercially acceptable.

The procedure for the actual dual wire, or a three wire, sub arc build-up of 4130 or 4137 drill pipe would be the same as shown previously in this specification but with the following

specific optimal settings that are not exactly as shown earlier in the specification: volts per wire-26 (from 28 volts) , travel speed-36 IPM (from 30 IPM) . This change in parameters is as a result of the difference of the alloy flux cored wire and a desire to build a higher crown because of the unusual properties of the alloy flux cored wire 202 set out above. Accordingly, for this preferred wire 202, the normal variables that would change and have tolerances permitted before the procedure must be shut down would be: volts 26, + or - 1 volt; travel speed 36 IPM, + or - 2 IPM.

As an example of the results of using the above described alloy flux cored wire 202 and a semi-active or semi-neutral powdered flux 204 around the alloy flux cored wire 202, such as L-Tec 429 semi-active powdered flux, such flux 204 and flux cored wire 202 used with base metal 212, would yield the following resultant weld metal 256:

ELEMENT BASE METAL 212 RESULTANT WELD METAL 256

Carbon .34 .13

Sulphur .022 .016

Manganese .86 2.05

Phosphorus .015 .030

Silicon .32 .60

Chromium .84 .22

Molybdenum .17 .44

Nickel .15 .47

Copper .07 .03

Vanadium .002 .020

Columbium .000 .003

Aluminum .040 .011

Calcium .0007 .0000

It should be noted that where weld metal 256 has a lower value of an element than base metal 212, the element is usually lost in the arc 252 and where there is no element in the alloy flux

cored wire 202 to support an increase in the weld metal 256 from the base metal 212, the element is being supplied by the semi- active or semi-neutral flux 204.

With regard to the test results, for an impact test using a 10mm x 10mm Charpy v-notch at 32° F. , comparing base metal 212 with the final weld metal 256:

FOOT/POUNDS

BASE METAL 212 WELD METAL 256

34.0 42.0

33.0 40.0

22.0 45.0

It was also shown that the liquid penetrant examination was satisfactory and the guided bend test was also satisfactory.

Performing the API hardness test on the surface of the weld metal

256, the Rockwell hardness consistently showed at 31. In this regard, it should be noted that two passes are preferred for the process. However, it should also be noted that when the first pass was tested, it still showed a Rockwell hardness between 33 and 34 which is within the API tolerance.

It is therefore preferable to try to use an alloy flux cored wire 202 that has less manganese and carbon, and also to use a flux 204 that has less manganese in order to reduce hardness to avoid cracking of the weld. This is done through the addition in the alloy flux cored wire of molybdenum and nickel and chromium.

It should also be noted that the change in the process with respect to voltage and the inches per minute were caused by a shift from solid wire 202 to a flux cored wire 202, not from the change in the chemistry of the material of the wire 202, and

accordingly any change in chemistry with respect to a flux cored wire would use the same process or procedure.

The embodiments set forth herein are merely illustrative and do not limit the scope of the invention or the details therein. It will be appreciated that many other modifications and improvements to the disclosure herein may be made without departing from the scope of the invention or the inventive concepts herein disclosed. Because many varying and different embodiments may be made within the scope of the inventive concept herein taught, including equivalent structures or materials hereafter thought of, and because many modifications may be more in the embodiments herein detailed in accordance with the descriptive requirements of the law, it is to be understood that the details herein are to be interpreted as illustrative and not in a limiting sense: