mig controlwasteindexlogo 

 

     

Technical Details of Welding Normalized 4130 Chrome Moly (Chromoly) Steel for Race Cars

Technical Details of Welding Normalized 4130 are Presented Here; for Other Information:

Click for BASIC 4130 WELDING DETAILS

Click for  EQUATIONS Defining Weld Cooling Rate in Tubing

Click for WELDING HEAT TREATED 4130 CHROME MOLY

Click for METALLURGICAL DEFINITIONS

Click for WELDING A BETTER STEEL;  HY 130

Click for DEFINING "EFFECTIVE HEAT INPUT"

 

Video-Welding 4130 CrMoly

Selecting Filler Metal

Selecting the proper welding filler metal and techniques requires an understanding of the metallurgy of the base material and the weld metal that develops from an admixture of the 4130 and the filler metal.

The graph on the left was taken from a US Steel "Atlas of Isothermal Transformation Diagrams."  These graphs define the metallurgical structure that occurs in various steels when they are cooled.  This one defines that we can expect to have high hardness, brittle Martensite form when 4130 is cooled at a particular rate. 

The graph at the bottom of the page shows the structure that formed when a bar of 4130 was heated to 1550 deg F then one end spayed with water.  Note that from about 3/16 inch from the quenched end the microstructure is almost all Martensite.  The hardness in this area is 50 RC.  The equivalent tensile strength for that hardness is about 250,000 psi.  Very strong but also brittle meaning a small crack will easily propagate.  If one made a chisel from that material it would break on the first blow!  If we wanted a chisel we would heat it after quenching to "Temper" the Martensite to a lower strength and make it far less brittle.

The reason for understanding the 4130 IT diagram, particularly the End Quench Hardenability Test  (Called a Jominy Bar, after the inventor, Walter E. Jominy) will become apparent as we present additional information below.

Many years ago, for a Masters Thesis, I made a Jominy bar from a weld deposited in a steel that had about 105,000 psi yield and about 120,000 psi ultimate strength.  I also plotted the cooling curves at numerous points along the Jominy Bar.  (No Excel at that time, had to use graph paper!  Note that Walter Jominy only sited cooling rates at 1100 deg F, not sufficient for my Professor!  Had to use a fast response Light Beam Visicorder, thermal coupes at 1/8 inch intervals along the bar etc!) Calculations were also made defining the cooling rate in welds made in 1 inch thick plate with the submerged arc welding process.  The accompanying graph was developed from that data.  The cooling rates for TIG welds made in thin material are also noted on the graph.

Estimated cooling rates for several TIG welds in tubing are shown in the accompanying graph in red and blue lines.  The cooling rates  are fast since welding volts and amps are relatively low and the energy efficiency of TIG is  low, about 50%.  Calculations for cooling rate are presented on this page: CLICK.

From the IT diagram we can estimate a continuous cooling diagram using a technique suggested by Grange and Kiefer.  It is shown with the green line superimposed on the IT diagram.  Also shown is the cooling rate for a TIG weld made on 0.040 inch thick material which is about 50 deg F/sec at about 1100 deg F.  For 1/16 inch thick material the TIG cooling rate  would be about  62 deg F/sec, between the two curves. The fast 80 deg/sec cooling rate would occur if small fillets are made on 0.093 inch thick material. This information is directly related to the weld heat affected zone (HAZ).  These cooling rates can produce some amount of Martensite in the HAZ.  The structure in the weld metal will depend on the weld rod or wire selected as well as the amount of admixture of 4130.  For example, very small fillet welds may have mostly melted 4130 in the deposit which can create potential cracking problems.

The table on the left shows the estimated weld metal structure of: 1) a weld made in 4130 without filler metal (called an autogenous deposit), 2) with 30% ER80S-2 diluted into the 4130 and 3) ER70S-2 rod diluted the same 30% into the 4130.  The resulting chemistry is a simple ratio of the materials assuming TIG welding with Argon shielding gas and minimum carbon or other element loss.

The data showing "Critical Diameter" was developed from a book on Steel Hardenability by Crafts and Lamont.  It defines the diameter of a bar, that when heated to 1500 deg F and quenched in water will have 50% Martensite in the center.  Notice the deposit with 30% ER80S-D2 filler rod is much more hardenable than even the 4130.   An austenitized and  quenched bar 3.3 inches in diameter of this material would have 50% Martensite in the center versus only a 2.4 inch bar for 4130 .  The deposit has significantly higher Manganese and Moly than the 4130 or the deposit made with ER70S-2.

A deposit made with ER70S-2 in 4139 tubing will most likely have a slightly lower tensile strength than Normalized 4130.  When mixed with the melted 4130, it will probably be 85,000 to 90,000 psi versus the ~95,000 psi in 4130 depending on how the material was processed.  This joint strength can be increased for intersecting tube joints by making the fillet size slightly larger. 

I have seen  comments on forums about ER70S-2 and ER70S-6 only having a tensile strength of 70,000 psi. These statement may have been made based on the AWS designation which indicates a MINIMUM of 70,000 psi is needed to label the product. That is NOT what is typically found. The following data is from a TIG weld deposit made with and ER70S-2 rod with essentially no dilution into the base plate. The information is from published ESAB data: Tensile Strength was 82,000 psi with a very ductile 31% elongation and 170 ft-lbs Charpy "V" notch impacts at -20 degrees F test temperature. That is very ductile and tough! When diluted into the high carbon 4130 the strength will increase.

Conversely, using ER80S-D2 because it contains moly (more than twice that contained in 4130) may provide a higher strength than needed or desired. For example, an undiluted weld made with this alloy produced a tensile strength of 110,000 psi and only 22% elongation. ER80S-D2 is often used to weld a Q&T alloy developed many years ago by US Steel called T-1. It is a structural steel which has a minimum tensile strength of 110,000 psi, more than normalized 4130 which is typically 95,000 psi.  It also has only about 0.15 carbon making it easier to weld.  The 80 in the ER80S-D2 designation does not define what the actual strength of a deposited weld made with that alloy will produce-only the minimum to be able to label the product with that designation. 

WELD QUALITY

It is very important to check weld quality and understand the types of defects that could be encountered when welding 4130.  Check your weld procedures and keep them consistent.  You should make some sample welds and bend them to destruction to assure failure occurs only after considerable bending has taken place.  Look for porosity or cracks that may have been present in the weld.  It would be a wise investment to hire the services of an American Welding Society (AWS) Certified Welding Inspector (CWI).  There are over 20,000 registered.  In fact many of them are members of the  60,000 member AWS.  They can advise on procedures and what to check for such as small undercuts at the weld toe of fillet welds that can lead to premature failure.

Consistently following the proper weld procedures and knowing how to check for possible weld problems is of major importance. Be sure to employ the skills of a qualified welder who has experience welding this material.  Also inspection of the final welds by an Certified Welding Inspector (Certified by The American Welding Society) is highly recommended.

This is a photo from an interesting article in the March 2007 issue of "Hot Rod Magazine."  The whole ridged elaborate cage was made from 4130 tubing.

Note they used ER70S-2 welding rod and made all joints with TIG welding.  Have to believe this was a mocked up photo since the description says the tube ends were ground to almost a "press fit."  This is far from that and unacceptable for making any type of weld!

Cleaver idea to use the drilled hole to relieve hot gases in closed tubing joints. 

An article in the April 2010 issue of the AWS Welding Journal entitled "Best Practices for GTAW (TIG Welding) 4130 Chrome-Moly Tubing" defines welding parameters and practices need to make quality welds in race car tubing joints.  It defines practices used by a race car shop.  They also recommend  ER70S-2 welding rod for most welding and when high strength is needed, with a sacrifice in ductility, ER80S-D2.  The article states that excellent fit up is essential.  They provide a rule of thumb for setting welding amperage of using 1 amp per 0.001 inch of wall thickness.  This suggests 35 amps for 0.035 inch and 80 amps for 0.080 inch wall thicknesses.  Gas flow of 15 to 20 CFH is suggested.  They mention the use of a "Gas Lens" and a small cup size so the tungsten electrode can protrude out further for visibility and access to the joint.  Welding in 4 quadrants; two 90 deg segments opposite each other than completing the remaining two segments is recommended.  They suggest using smaller rod diameters, generally no larger than the wall thickness.  The article presents a number of other suggestions regarding welding conditions from a company with successful experience.  Slow pulsing is suggested either with a foot control or pulse settings built into the welder.

TIG welding (also called Heliarc welding an ESAB trade name) or the official AWS designation, Gas Tungsten Arc Welding (GTAW) requires more skill than MIG welding.  However it allows separate control of welding heat and metal addition.  Currents can be set at very low levels for thin material allowing the operator to watch the puddle and assure complete weld penetration.

If a Searc Engine Found This Page 1st- - - We'd Suggest a Visit to The Basic Welding 4130 Page; Then Return.  Click Here

This page presents the technical information that supports the reasons for the suggestions found on the "Basic, Welding 4130" page.

Like Math?  Want to see the calculations for weld cooling rates and references?  CLICK HERE.

Check Out Welding Math Site

"NEW" Book:

"Advanced Automotive Welding"

Author: Jerry Uttrachi

(President of WA Technology)

This 176 page book includes detailed metallurgical information about welding Chrome-Moly

CLICK for Book Details

The "Gas Blast" at MIG Weld Starts Causes Problems:

1) Shielding Gas Waste - typically 50% of gas used

2) Inferior Weld Quality - due to air pulled in gas stream

See Patented Gas Saver System (GSS) Solution

GSS  Works for TIG Welders with Foot Pedal or Thumb Control

This Page Presents Technical Details of Welding 4130 for Other Information:

Click for BASIC 4130 WELDING DETAILS

Click for  EQUATIONS defining weld cooling rate in tubing

Click for WELDING HEAT TREATED 4130 CHROME MOLY

Click for METALLURGICAL DEFINITIONS

Click for WELDING A BETTER STEEL;  HY 130

Stop Wasting Shielding Gas!!

See YouTube Videos

Including Our Patented MIG Shielding

"Gas Saver System" (GSS)

 

Why MIG Gas Waste

Detailed Review of GSS

Short Overview of GSS-90 Seconds

What User Say About GSS

Welding Race Cars

Rat Rods-Then and Now

Welding: Go Green

Engineering A Pro Street / Street Rod

bullet

ENGINE - CLICK

bullet

TUNING HOLLEY CARB - CLICK

bullet

ENGINE COOLING- CLICK

bullet

WIRING - CLICK

bullet

CHASSIS - CLICK

MIG Gas Delivery and the Small Block Chevy Evolved in a Similar Way and Time!

 Both had Advances and Setbacks Before

They Were Optimized!

 

Click for a PDF Report on the Similarities

Thanks to those "Car Crazy" folks who have purchased the Gas Saver System and commented on the improved starts and reduced gas usage they have achieved.  Many purchased our 6 foot prefitted GSS, Part Number WAT FB6, a 3 and 4 foot system, FB3 & FB4, are also available.

ONE PAGE SUMMARY

PURCHASE PRODUCT

 

 

FREE Corvette Accessory Installation Info

 

 

"WARNING: "Weld Safely"