mig controlwasteTOMindexlogo 



Electroslag Welding

  1. Experience With Electroslag & Consulting Services

  2. Background and History of Electroslag Process

  3. Aluminum Electroslag

  4. High Speed Electroslag Process Welds at 10 ipm Vertical Travel


Extensive Experience with Both Consumable Guide and High Speed Conventional Electroslag Welding

  1. Two patents granted relate specifically to Electroslag welding:

    a) Patent # 3,778,587 is for a metal cored wire of a specific chemical composition to achieve high weld toughness.

    b) Patent # 3,854,028 is entitled "High Speed Electroslag Welding," and defines welding parameters and slag control procedures that allowed   Electroslag welding at travel speeds up to 10 ipm.  Plate thickness range   welded is from 3/8 to 2 inches.  An article was published in the Welding Journal defining the product and application in National Steel And Shipbuilding shipyard in San Diego California.


  2. Extensive R&D and production experience with proper control of critical variables.  For example, weld hydrogen can be a problem even in lower strength steels.  Sources of hydrogen are not only the wire and flux.  Other more subtitle variables are even more important and can cause cracking (See Background Details Below). Yes you can get Hydrogen Cracking in even A36 steel with this high heat input process!  Paul Kockol, then with US Steel Research, wrote two excellent papers on the subject entitled; "Causes of Grain Boundary Separating in Electroslag Welds," published in The Welding Journal June 1979 and "Effects of Electrode Composition...Grain Boundary Cracking in Electroslag Welds;" in The Welding Journal March 1983.

  3. Design of copper shoes is critical.  Light weight shoes, designed to reduce production cost when these items are offered for sale, are NOT the way to go.  Shoe designs can be provided which are more tolerant of weld parameter variations, less dependent on circulating water temperature and can be constructed in-house.  Excessive cooling can caused condensation inside the weld joint which can lead to Hydrogen Cracking. 

  4. Welding equipment for Consumable Guide Electroslag can be made from Heavy Duty MIG feeders (today's digital designs allow precise control) or existing Sub Arc equipment can be used by inserting the guide tube into a specially modified busbar.  See Accompanying photo. A design for a guide tube holder can be provided.  Simple power source recommendations avoid expensive equipment purchase.

  5. Operating parameters for making Consumable Guide Electroslag welds can be provided.

  6. Recently attended Federal Highway demonstrations on the NEW High Speed Electroslag System.  It has many features similar to the High Speed Electroslag patent referenced above.   This demonstration showed the development work by a friend, Dr. Jack Devletion who presented the lecture.


Background Details/History:

Some references site Robert Hopkins for having invented the Electroslag welding process in the 1930's.  Most of his patents relate to Electroslag melting for ingot manufacture, not welding.  However one US patent, number 2,191481 filed in June, 1939 does describe the surfacing of one material on another.  The illustration, however looks more like a melting furnace than welding.  In fact the fellow who invented Submerged Arc Welding, Harry Kennedy, was granted a US patent in October of 1950, number 2,631,344, assigned to Linde Division of UCC that more closely related to Electroslag welding.  However it too falls short of defining what we know today as this simple welding process. 

A unique version of the Consumable Guide Electroslag Welding Process was developed and patented (US Patent Number 2,868,951 filed in March 1957, assigned to the Linde Division of UCC) by a colleague, Harry Shrubsall.  The figure on the left is from that patent.  It clearly shows the process as it is known today.  Harry's patent includes the use of a flux coated tube to replace part of the flux that plates out on the copper molds.  It also insulates the metal guide tube from the work.

The Paton Institute in Russia introduced the process and did a great deal of development work.  They published a book, "Electroslag Welding" in 1959 with an English translation published by The American Welding Society in 1962.  Of the 92 references sited in that text, most range in dates from 1955 to 1959.  The first mention of Electroslag  is in a paper entitled "Electroslag Welding,"  published in Avtomaticicheskaya Svarka in 1953 by Voloshkevich. 

Harry Shrubsall and I worked as engineers in the Linde Development Labs (although I started after his invention!).  Harry had several subsequent patents on the use of the process to butt weld Railroad Rails.    Patent numbers 3,192,356 and 3,291,955 filed on September 1962 and February 1963 respectively describe  Railroad Rail welding with Consumable Guide Electroslag.

I worked on a Plate Electrode Electroslag Welding development with Harry,  It used a triangular shaped electrode with two holes that spread two electrodes at the bottom to weld the Rail Base then narrowed them to weld through the Rail Web and the Rail Head on top.  Welds were excellent and passed all required Railroad tests.  The resulting welds were far better quality than the Thermite deposits often employed but did take longer to make.  This extra time ultimately caused resistance to its use.  If a train happens to be coming down the track, all workers want to be able to move fast!  Can't blame them.

Refer to the schematic below copied from the original Linde Publication 51-220 "Procedures for Consumable Guide Electroslag Welding." In the Electroslag Welding process current transfers from the guide tube to the welding wire and in turn though a molten flux bath.  The resistance of the flux bath creates the heat and temperature that melts the wire and surrounding base material.  The guide tube itself will also contact the molten flux bath periodically which will cause an increase in welding current as the end is consumed.  The molten flux will solidify on the copper retaining molds which are placed on the sides on the weld joint.  This flux layer protects the copper molds from having direct contact with the weld.

The Linde Flux Coated Guide Tube defined in Harry's original patent looked like a big stick electrode!  They were made with a hole down the center and a thin flux coating that protected the tube from shorting to the plate side walls and  also automatically compensated for a portion of the flux that plated out on the copper retaining shoes.  The tubes were purposely coated with a quantity of flux that was less than the amount that plated out on the copper molds (see schematic on left showing flux layer that plates out on the copper molds )  This avoided excess flux build-up in the joint as welding progressed.  Therefore additional flux was always needed and was added by the operator but far less frequently than when uncoated tubes were used.   ( The concept of adding flux by listening to the sound, works great in a Laboratory environment.  However in a noisy shop it is not effective.  There are some simple tricks at measuring flux depth.)  For multiple tube welding of heavy sections (up to 4 tubes for 8 inch thick plates) a special very thinly flux coated tube (referred to as Type M) was used so as not to produce excess flux burden as welding progressed up the joint.  Special slip-on metal guide tube clips centered the tubes and prevented vibration and shorting.  This was possible since the tubes flux coating was an insulator.  Little preparation was needed to center the tubes.  Those fabricators that employed bare tubes could have shorting problems that caused equipment failures and the worse thing that can happen during an Electroslag weld - - having to restart! 

Linde's patented Flux Coated Guide Tube Process was very successful for the user since a great deal of welding time and plate preparation were saved.  It was also very successful for Linde who obtained a substantial patent royalty (> $1/foot of tube) built into the sale of the Flux Coated Guide Tubes.  They were used extensively with some the notable applications being:

1.      Over 100,000 feet of Flux Coated Consumable Guide Tubes were used by Kaiser Steel to weld massive columns for the Bank of America Headquarters in San Francisco.  This process was very successful and made some difficult to do any other way, transitions joints.  Thin coated tubes were used in locations where there were only captive steel weld dams. The operator at left is placing a copper dam on a box beam reinforcing plate.  Four Electroslag welds will secure the plate to the heavy wall box beam.  An article entitled "Electroslag Welding with Consumable Guide on the Bank of America World Headquarters Building," published in The Welding Journal in 1968 by Tommy Agic of Kaiser Steel and Jim Hampton of Linde,  mention that a total of 23,000 welds were made for the job. Plate thickness ranged from 1/2 to 5 inches.  The massive columns ranged from 26 to 42 feet long.  The operator on the right is making a beam splice joint with the consumable guide Electroslag process.

2.      Thousands of feet of Electroslag welds made on the John Hancock building in Chicago:

      (Above photos from a Linde publication F-51-220.  On right are two simultaneous welds being made on beam flanges splices on the edge of the John Hancock building under construction.  Each weld is being made with two Linde Flux Coated Guide Tubes.  The photo on the left shows the weld joint being prepared)

3.      12 foot long seams made in 1 ½ to 2 inch steel for the 72 cubic yard bucket for Texas Gulf Sulfur dragline.  A standard sub arc tractor was used to make these welds with the simple addition of a guide tube holder, see below:

      When welded with manual stick electrodes these 12 foot seams required 150 pounds of electrodes and 40 man hours to complete.  With Consumable Guide Electroslag the electrode requirements reduced to 40 pounds and the weld was completed in 1 3/4 hours. 

4.      Detailed application information was described in a Welding Journal Article entitled “Vertical Submerged Arc Welding,” by Paul Masters from American Bridge Division of US Steel and Bob Zuchowski from Linde (Linde Publication 52-539).

5.   In addition to the high speed conventional Electroslag used at National Shipbuilding in San Diego (mentioned above), they also made many hundreds of welds splicing deck stiffeners on Roll-On-Roll-Off ships with Flux Coated Consumable Guide Tubes.  The process was also used for other applications and other shipyards.

The royalty income derived from the sale of the Patented Guide Tubes allowed my Welding Process R & D Group to develop numerous applications and procedures for the Consumable Guide Electroslag Welding Process.  At one time we had 7 engineers/technicians working on the process in our Laboratory. After successfully welding steel anode bars in an aluminum pot line (with the line in operation) and a market evaluation, we decided to do extensive research on the welding of massive aluminum busbars used in aluminum production.

Other companies also introduced Electroslag systems in the US.  Arcos marketed both conventional Electroslag and Consumable Guide Electroslag products.  They produced a plate electrode, some of which were dip coated with flux as I recall.  Arcos also marketed products for an Electrogas process. The Electrogas process utilized a cored wire with flux added to replace that which coated the weld retaining copper shoes.

Hobart developed a simple oscillator which allowed the welding of heavier plate without the need for additional guide tubes and related equipment.

Airco marketed  an Electrogas process which used a solid wire and argon based gas mixture.  In was used by the Litton Shipyard in Erie Pennsylvania to make the vertical hull welds in ship tankers.

Lincoln introduced a vertical welding process which used a self shielded flux cored wire and operated with moving shoes.


Aluminum Electroslag Development

The Consumable Guide Aluminum Electroslag Welding process was developed in our Laboratory and produced  welds in 2 inch think (50 mm) and 4 inch thick (100 mm) busbar material.   Welds were made at a very rapid rate of vertical travel speed not possible with steel welding.  A sample of a weld made with the process is shown on the left. Unfortunately the main application for the process was for joining heavy aluminum busbars.  These are mostly employed in aluminum production facilities and the market for aluminum had significantly deteriorated.  The development work was therefore terminated and the process was not commercialized.

The demand for Aluminum is now high and new plants are under construction.  A company who works in the area asked if it were possible to weld over 10 inch thick by 4 foot high busbars by completing the early development work and extending it to these much thicker sections.  After considerable additional development work and cost, refining the flux, welding parameters and equipment; the objective was achieved.  The process was used on a production application over 10 inches thick with welds made at very high vertical travel speeds [>2 inches (50mm) per minute.]

Photo left shows the equipment system welding a >10 inch thick section. 

Center photo is the finished weld.  Welding speeds were very high, much higher than in steel welding.  Weld surface is excellent

Photo right is a cross section showing good fusion and defect free weld.

Need to Weld Aluminum in a High Magnetic Field?  

Of interest, we developed the Aluminum Electroslag Welding Process because we had successfully welded steel anode bars in an aluminum pot line that remained in operation.  The aluminum producer asked if we could develop the process to weld their aluminum busbars.  The Electroslag process relies mostly on resistance heating of the flux and is only partially affected by high magnetic fields.  Work on welding in magnetic fields continues.

If you would like to know how to obtain a service of Aluminum Electroslag welding, please E-mail.

High Speed Steel Electroslag

We developed a "High Speed Electroslag Process" that  was used in production by a number of companies.  We were granted a US patent (# 3,854,028) on the process.  It utilized a unique Metal Cored Wire that was a low carbon, high Mn, Si alloy but had an oxygen content of from 500 to 1000ppm versus the low oxygen we were usually working to achieve!  The figure on the left is from the patent.  The wire was also patented (US Patent # 3,778,587.)  An article entitled "Electroslag Welding Speeds the Welding of Ships" was published in the American Welding Society's Welding Journal (April, 1974) defining the procedures and application at what was then named National Steel And Shipbuilding in San Diego California.  The authors were Ray Parrot (Welding Engineer at the yard), Sid Ward (Manager of Engineering, West Coast Region of Linde) and myself, Jerry Uttrachi.  A schematic of the moving device is shown on the upper right.  The picture on the left from the AWS Journal article and shows the Plate Crawler as it was called mounted on a ship hull butt seam.

The process operated at vertical speeds up to 10 ipm.  A weld listed in Table II of the patent shows it was made in 1 inch thick plate at 8.5 ipm vertical travel speed.   Welding parameters and weld properties are presented for a number of welds in the Table.

One of the key parameters discovered as necessary to make welds at these speeds was keeping the copper shoes hot!  The patent simply mentions to accomplish this one needs "properly designed shoes."  This proprietary shoe design feature was not described or shown.  It is not easy when using water cooling to have the water exit the shoes without boiling on the cooling passage surfaces and therefore limiting the exit water temperature.  Our shoe design allowed very hot water to exit. 


  If you'd like a copy of the patent, Email.  We'll send it FREE.


If you'd like the shoe design features that allowed high exit water temperatures; a report is available for purchase- Email 


Email If you are interested in our Electroslag Consulting Service

See Capabilities and Background