HISTORY OF SUBMERGED ARC WELDING

continued-- Page 5 (last)

MULTIWIRE-MULTIPOWER APPLICATIONS

No sooner had weld productivity increased 5 to 10 fold using the Submerged Arc process then even faster welding systems were developed.

In a patent filed in 1944 by Harter ET Al, number 2,436,387, assigned to Babcock and Wilcox Company a 3 wire system was described.  See photo at left of Figure 1 from their patent.  This broad based patent covers the use of various three phase power phasing to control arc deflection and weld power distribution.  A crude analysis of the arc deflection caused by the interaction of the arcs is presented.  They show the benefits of using different phase angles, 60 versus 120 degrees.  The perspective of using an elongated arc verses just a higher current broader arc is defined.  Frankly it was not until the Multi-Cathode Gas Tungsten-Arc Welding paper by Anderson and Yenni in 1965 that the reason for this benefit was described (reference 5). 

The Specification appears quite theoretical with no actual weld data or parameters defined.  However the theory of using a lower voltage on the lead wire then progressively higher voltage on the middle and trial wires is mentioned.    Considering the period when this work was published it is very advanced.

Don Knight published a technical paper in the Welding Journal in April of 1954 entitled "Multiple Electrode Welding by the Unionmelt Process" (reference 8).  He reviews parallel wires, two wire AC-AC Scott as well as Series Arc for weld surfacing.  He presents specific applications and photographs of the applications.

Kubli and Shrubsall presented a technical paper which was published in The Welding Journal in November 1956 entitled, "Multipower Submerged-Arc welding of Pressure Vessels and Pipe" (reference 8).They define the various options for connecting to both DC and AC power.  They describe the benefits of using variable phasing AC to gain a forward deflecting trail arc to control bead shape and eliminate undercutting.  They present welding parameters and show a number of applications in both pipe and vessel manufacture.

With the objective of developing a multiwre welding system that would be viable in production myself (Jerry Uttrachi)  and Joe Messina worked for several years at the Linde Welding Labs and defined an all AC system that achieved welding speeds higher than the objective. This work was published in the Welding Journal June 1968 (reference 10).  An extensive statistically design experiment yielded  unique power phasing and unexpected welding parameters to achieve excellent results.  The system was installed first at the U. S. Steel Pipe Mill in Provo Utah. 

After extensive mill trials an finding none of the grounding or arc blow problems previously experienced with the DC-AC-AC system, the mill manager, Hank Huish, obtained funding to install all new welding heads, controls and power for 5 ID and 4 OD welders (photo is of OD welder).  The results were so successful the American Bridge Division of US Steel pipe mill in Orange Texas also ran a mill trial and installed the three wire system on their 3 continuous OD Berkley cage welders and 4 boom ID welders.  This was an interesting trial since the mill had been equipped with DC-AC.  We first converted the mill to AC-AC Scott then added the third wire.  On second and third shift we shut down our third wire and left a reduced current Scott two wire AC-AC system. On the 0.388 wall pipe we were welding we were able to achieve speeds of 75 ipm (32 mm/sec) verses the maximum the mill had been able to achieve with DC-AC of 60 ipm (25 mm/sec).  This proved the benefit of the forward arc deflection achieved with AC-AC.  This three wire system was subsequently installed in a number of foreign mills, including in Argentina, Italy, Kawasaki in Japan and the British Steel pipe mill in the UK.  Kaiser Engineering in Napa also designed and built two pipe mills using this system, one for Iran and the other for PMT in Mexico.  They were excellent mills; unfortunately Kaiser did not see fit to invest in their own mill in California which remained two wire AC-AC Scott until it was recently closed.  Stelco in Canada employed the system on their new spiral mill designed to produce the Alaska oil and  natural gas pipe lines.  The pipe used for what is referred to as the Alaska Oil Pipeline was made in Japan and the natural gas line was never built! 

Perhaps the most creative use of the three wire system was by the A.O. Smith/ Armco pipemill in Houston.  They had a unique problem of having three massive flash butt welders that they wanted to use as fixtures to make OD Submerged Arc welds.  However three OD welds would not meet the production output requirements. 

After extensive laboratory tests (reported in a Welding Journal article entitled “Hot-Overlap Technique for Three-Electrode Submerged Arc Welding of Line Pipe” reference 11) they successful obtained API approval for a unique welding method.  This was a very sophisticated system for its time. Welds were made starting at both ends of the 40 foot pipe and welding progressed to the center.  At the center, one 3 wire welder stopped and backed up and the other continued to weld through the still molten flux from the first!  All 11 welders (3 OD X 2 plus 5 ID) used our three wire system.   Each welding wire had volt and amp Esterline Angus strip record meters.  The control room with 66 strip recording meters was an impressive site!.  Today that can easily be accomplished with a computer but at the time, paper strip charts were used to troubleshoot potential problems and assure the hot overlap was working properly.

In 1978 Uttrachi published a review of all of the multiwire processes including parallel wires, DC-DC, DC-AC, AC-AC and three wire AC (reference 12).  It presents applications for each.

SIDE BAR

DCCP POWER

A process patent, number 3,659,073, by myself and Messina filed June, 1970 defines a method of Submerged Arc welding with DCCP power (constant potential or voltage). Prior to that invention CP power was not used for slow speed welds made at low current densities.  The main reason for introducing it here is the oscillographs presented give some understanding of what is occurring under the flux!  A technical paper was also presented in the Welding Journal (reference 14).

The oscillograph on the left labeled CC Power is a trace of the amps and volts for a weld made with a 5/32 inch diameter electrode at 360 amps.  That is 18,750 amps per square inch where published data at the time said one could not use CP power below 40,000 amps per square inch.  [For perspective, an 0.045 GMAW weld made at 250 amps is 157,000 amps/square inch.]   Low current densities cause globular drop transfer. Assuming the drop diameter averages the wire diameter that would yield about 5 drops per second.  These traces validate this calculation with the drop rate probably in the range of 6 to 10 per second.

With conventional DCCP power the amperage trace shows large excursions, ranging from almost 0 to 700-800 amp peak (see figure left).  In this trace, it appears the arc is actually extinguishing about 4 times in one second.  That is about the drop transfer frequency.  What is occurring, is as a molten drop leaves wire, the gap between the slowing feeding wire (30 ipm wire feed speed) and the plate is larger than the available power supply voltage! The wire must contact the plate or molten flux to have it reignite.  When it reignites it is probably starting on the hot puddle and leaning the arc backwards.  This creates a peaked undercut weld.

 We found that by putting a very large inductor into the circuit, we were able to store enough energy to deliver it back to the arc as the current tries to go to zero.  This keeps the arc alive and makes the process stable.  Welds look like those made with CC power, flat and uniform.  From the traces made with the system shown at left you'll note  there are no arc outages. Note the voltage trace shows peaks of 45 volts when only 40 open circuit voltage was available.  The extra 5 volts was delivered by the massive inductor.

The advantage of a CP system is in the control.  Maintaining the arc length with CC power requires a voltage control to maintain the arc length.  Prior to microprocessor controls this was not as accurate  or cost effective a means to achieve quality arc performance.  To perform  the control function properly, low speed, low inertia motors and appropriate controls were needed.  Starting was also more difficult and less reliable than achievable with CP power.  Today many of these benefits are less important since arc voltage controls can be made very accurate, inexpensively.

TODAY

Today there are a number of 5 and 6 wire systems in use for various applications.  The use of microprocessor based controls makes these systems viable for use.  In addition there are numerous Submerged Arc flux and wire products that are manufactured in all parts of the world.  Weld performance both physical and mechanical properties can meet the most stringent needs for steel and some other materials.

General References:

1. The History of Welding by  R.D. Simonson, by Monticello Books, Inc, Morton Grove, Ill, Printed April, 1969.
2. “Pullman-Standard Active in Welding Developments” The Welding Engineer, September 1937.
3. “Welding Handbook,” American Welding Society “1938 Edition
4. “Welding in Tanker Construction,” by T. M. Jackson; The Welding Journal October 1937.
5. “Multi-Cathode Gas Tungsten-Arc Welding,” by J. E. Anderson and D. M. Yenni The Welding Journal April 26, 1965.
6. “Characteristics of Submerged Arc Welding with Three Phase Power”, by E. A. Clapp and N. G. Schreiner; The Welding Journal, June 1952
7. “Multiple Electrode Welding by “Unionmelt” Process,” by D. E. Knight published a technical paper in the Welding Journal in April of 1954 entitled
8. “Multipower Submerged-Arc Welding of Pressure Vessels and Pipe,” by R. A. Kubli and H. I. Shrubsall, The welding Journal, November 1956
9. Welding with DC-AC-AC.  Private Laboratory data developed in 1960 and 1961 by Len Mauskaff and private Pipe Mill trial data viewed at Geneva Steel Provo Utah with Welding Engineer, Louis Warren.
10. “Three-Wire Submerged Arc Welding of Line Pipe,” by Uttrachi and Messina Welding Journal, June 1968.
11.  “Hot-Overlap Technique for Three-Electrode Submerged Arc Welding of Line Pipe,” by P.W. Ramsey, T. M. Even and G. R. Wepfer, Welding Journal, October 1972.
12. “Multiple Electrode Systems for Submerged Arc Welding," by G. D. Uttrachi, Welding Journal, May 1978.
13. “Magneto-Hydrodynamic Phenomena During Twin-Arc Welding,” by Mandelberg, Sidorenko and Lopata, Avtomaticheskaya Svarka, Number 2, 1969.
14. “A New DC Power System for Submerged Arc Welding,” by Gerald Uttrachi, December 1970.
15. "Health of Arc Welders in Steel Ship Construction," by US Surgeon General, 200 page review of WWII welders, 1947.

Page  1   2    3    4    (5 last)

Have a Welder?   Improve Weld Starts and Have Shielding Gas Cylinder Last at Least Twice as Long! 

Note: Our Patented GSS is Not Available in "Stores"

A home shop fabricator in Georgia with a Miller ^TM 175 amp welder purchased a 50 foot Gas Saver System ( GSS^TM ) so he could use a larger cylinder and mount it on the wall of his shop.  He wrote:

"The system works great.  Thanks for the professional service and a great product."   Click To See His Home Shop

A Professional Street Rod Builder Had This to Say:

With their standard MIG welder gas delivery hose the peak shielding flow at weld start was measured at 150 CFH. That caused air to be sucked into the gas stream causing poor weld starts.  With the GSS replacing their existing hose, the peak flow surge at the weld start was about 50 CFH.  Total gas use was cut in half.

Kyle Bond, President, quickly saw the improvement achieved in weld start quality as a significant advantage!   Kyle, an excellent automotive painter, was well aware of the effects of gas surge caused by pressure buildup in the delivery hose when stopped.  He has to deal with the visible effects in the air hose lines on the spray gun in his paint booth!  The paint surge is visible and creates defects unless the gun is triggered off the part being painted!  We can’t do that with our MIG gun!

See Video Showing What Users Say About Our Gas Saver System-CLICK Here or Picture

Check Out Welding Math Site