Supplier Directory Subscribe
Home / Choosing a Butt Weld Joint Preparation

Choosing a Butt Weld Joint Preparation

Which joint configurations provide the best economy? Here is an examination of the “pro-versus-con” arguments that must be weighed out in order to make an informed decision when choosing between U-groove and V-groove joint preparations and small bevel versus large bevel configurations.

Posted: November 26, 2012

Figure 1. AWS D1.1 prequalified joint B-U3c-S utilizing a bevel and a root opening.
(Click on illustration to enlarge it)

Figure 2. AWS D1.1 prequalified joint B-U2-S utilizing a bevel and a root opening.
(Click on illustration to enlarge it)

Figure 3. Two “Double-V” groove joint designs with different included angles. A larger angle can allow for improved penetration through the joint.
(Click on illustration to enlarge it)

Figure 4. Illustration showing how a larger included angle can reduce crack sensitivity.
(Click on illustration to enlarge it)

Figure 5. Example of a U-Groove weld joint configuration.
(Click on illustration to enlarge it)

Figure 6. Example of a compound bevel groove weld joint.
(Click on illustration to enlarge it)

Figure 7. Comparison of Single-V and Double-V groove joint designs. A Double-V joint design utilizes about half of the weld metal of its Single-V counterpart.
(Click on illustration to enlarge it)

Figure 8. Comparison of two different pass sequence procedures. Less plate distortion can be achieved by balancing the weld passes.
(Click on illustration to enlarge it)

Minimizing the volume of weld metal used to fill the joint helps keep costs low to assure profitability. As the thickness of the plates being joined increases, the weld metal volume grows at an even faster rate due to the geometry of the weld joint, that is, the bevel angles. So which joint configurations provide the best economy?

In the process of selecting a suitable butt weld joint preparation for thick plate (greater than 1 in), the most apparent consideration is unquestionably the joint economy – the practice of minimizing the volume of weld metal required to fill the joint. This desire to keep costs low is certainly understandable, as a fabricator would not want to deposit any more weld metal than is absolutely necessary so that profitability can be assured.

Yet as the thickness of the plates to be joined increases, the weld metal volume grows at an even faster rate due to the geometry of the weld joint, that is, the bevel angles. So which joint configurations provide the best economy? This will be outlined in more detail that follows. But before delving into this area, other important factors that must be considered when choosing the appropriate joint design must be discussed first.

One of those factors is the requirement that the weld joint needs to be able to physically accommodate the chosen welding process and any accompanying welding equipment (nozzles, torches, etc.). Consider the case where the joint design incorporates a bevel (on at least one side of the joint) and a root opening. Two examples of AWS D1.1 prequalified joint designs with this type of preparation are shown in Figures 1 and 2.

The B-U3c-S joint design shown in Figure 1 for submerged arc welds shows a 70 deg included angle, which is determined from the addition of the 35 deg bevel angle on both plates. Since the included angle of this joint is relatively large, then the root opening may be fit-up to be narrow without fear of the welding arc not being able to penetrate deep enough into the joint. In contrast, the B-U2-S joint design shown in Figure 2 displays a relatively small included angle of 20 deg. Because the bevel angle is small, the root opening must be increased far enough to allow for complete penetration through the joint.

Although the two joint configurations mentioned above are quite common, there are instances where it would benefit the fabricator to utilize a joint design that features a bevel and a root face (with no gap). The same logic regarding access to the weld joint that was used in the previous paragraph applies in this situation as well.

Because there is little (or no) gap, the bevel angle must be physically large enough to accommodate the welding process and welding equipment. But now there is the additional consideration as to whether or not the root face, also known as the “land”, is ideal when taking into account the depth of penetration of the selected welding process and parameters.

With a specified root face dimension, penetration level will be primarily dictated by the welding current and the travel speed. It is intuitive to see that as the welding current increases, the depth of penetration increases. Moreover, as the welding travel speed decreases, the penetration level increases with all other things being equal. So, if changes to the welding process and parameters are not possible (as is true in the case of qualified welding procedures), then it is even more important that great care be taken to properly size the root face dimension to ensure adequate penetration through the root, as illustrated in Figure 3.

This raises another vital concern in regards to the “bevel and root face” joint design. The joint preparation of the groove has a dramatic effect on the shape of the cross-sectional area of the first-pass weld nugget.  It is imperative to achieve a weld that possesses a width-to-depth ratio of at least 1.25 in order to mitigate the crack-sensitivity of the weld. A width-to-depth ratio of 1.25 can be readily achieved when the included angle is at least 70 deg. The width-to-depth ratio associated with a joint that has an included angle of 60 deg or less can lead to surface or even sub-surface centerline cracks. Once the first pass has been deposited, however, the crack sensitivity associated with the included angle ceases to be an issue. This phenomenon is expressed in Figure 4.

In order to address the inherent crack sensitivity of V-groove type joints, some fabricators opt for the choice of a U-groove joint configuration as shown in Figure 5. The term “U-groove” stems from the application of a radius or partial circle preparation at the bottom of the joint. This radius transitions smoothly to a narrow included angle. Because of the large width-to-depth ratio provided by this U-groove configuration, a natural protection against cracking is provided. When employing a narrow bevel angle with a U-groove, the crack susceptibility is not amplified in the same way that it is with V-grooves.

Another type of narrow bevel joint configuration commonly referred to as a compound joint is shown in Figure 6. In the fabrication of a compound bevel joint, the included angle at the bottom of the joint is made to be large (say 100 deg). Moving up a short distance along the sidewall, the included angle makes an abrupt change to a much narrower included angle (to say 15 deg). Although these two types of narrow gap joint designs decrease crack sensitivity, they are not without drawbacks. One such drawback is that the plates must be machined during preparation rather than torch cut, which adds greatly to the cost of fabrication. However, this objection can be offset somewhat by the extra savings realized because smaller included angles can be used and therefore these joints require less weld metal than their V-groove counterparts.

As if that wasn’t enough to consider, the opportunity exists for the fabricator to make use of double-sided groove joints. The conventional wisdom is that because double-sided groove welds exhibit greater joint economy since the weld metal required is roughly 50 percent of that of their single-sided complements (see Figure 7), one may expect that double-sided joints are nearly 50 percent less expensive to construct.

Another perceived benefit to using double-sided joints is that because half the weld metal is required, then the chance for weld defects is also cut in half. And yet another logical assumption commonly associated with double-sided groove welds is that because fewer weld passes are used, there is a lower level of residual stresses present in the base material surrounding the weld. So considering these apparent benefits of double-sided groove welds, why would anyone want to use a single sided groove weld?

Although it is true that double-sided welds improve joint economy based simply on the geometry of the weld joint, it is not necessarily true, however, that they cut the cost of production in half. Because joint preparation now must be carried out from two sides of the plate rather than one, labor and cutting costs will double.

Another cost concern with double-sided groove welds is that two root passes must be deposited instead of one. As welding fabricators know, root passes are notoriously difficult to deposit – inadequate penetration, difficult slag removal, and sidewall undercut are just a few of the problems associated with the root pass. Furthermore, root passes generally take more time to deposit than the fill and cap passes, which counters the savings generated from the reduction in weld metal attributed to double-sided groove welds.

The same problems encountered in root pass welding of double sided groove welds also present themselves during overhead welding, which is obliged to be performed if the weldment cannot be re-positioned for downhand welding. And even if re-positioning of the weldment can transpire, it is not a one-time occurrence for each double-sided weld joint. On the contrary, the weld passes need to be deposited in a counter-balancing method similar to that shown in Figure 8 to achieve optimum distortion control.

While double-sided groove welds do offer improved joint economy over single-sided welds, the savings can be wiped out if the proper mechanization and positioning equipment is not used to reign in preparation and handling costs. And even if such equipment is utilized, the savings over single-sided welds may not necessarily be two-to-one. The same type of “pro-versus-con” arguments must be weighed out in order to make an informed decision when choosing between U-groove and V-groove joint preparations and small bevel versus large bevel configurations.

  • Pradeesh Kumar VK wrote:


    I am having one doubt as to why the standard defines the bevel angle as 37.5 degrees. Is there any reason for that?

    Pradeesh Kumar VK

  • Beatriz wrote:

    I have some doubts about the number of passes needed for square, single-V and double-V welds. I considered 1, 2 and 4 respectively but I am not sure if it is correct.
    Can you help me?

    • Regis Geisler wrote:

      Hi Beatriz, the joint design and the number of weld passes needed will depend upon a number of factors, including base metal type and thickness, the welding process selected, joint access, the weld metal properties required, and cost considerations. Without knowing the details of your joint design, it is not possible to provide specific advice. However, these welding design “rules of thumb” may help guide you in the process:
      • Select a joint design requiring the least amount of weld filler metal. For instance, on thick plate consider using double-grooves instead of single-groove joints.
      • Avoid utilizing weld joints and depositing weld beads that create extremely deep grooves.
      • Design the joint for easy accessibility of welding.
      • Plan out the welding sequence and use appropriate preheat and interpass temperature control to minimize distortion.
      • Be aware that welding procedures leading to significant base metal pick up into the weld can alter weld metal properties. Also, pass sequence – and hence the number of passes per weld layer – can have a dramatic effect on weld metal strength and toughness.
      • Use sound workmanship practices, including wire feed speed, current, voltage, and travel speed procedures appropriate for the process and as recommended by the filler metal manufacturer.

  • Muhammad Furqan Iqbal wrote:

    I am working on Mild Steel Sheets of 3.45mm thk. Welding process is MAG i.e. CO2 shielding gas, ER70s-6 filler wire. Is the I joint feasible for it?

  • John Forbes wrote:

    We are looking at double “V” preparations for heavy wall piping in Duplex piping , CRMo piping & A106B piping This is piping 48″ & above with walls 50mm & above.

    The intent is to have 3/4mm landings – with – initially no gap – so that the passes can be deposited virtually with no root pass until the back side is gouged.

    Preparations will vary between 50/50 preps to 70/30 preparations

    Do you know of any instance where this violates any code???

  • Nori VSN Murthy wrote:

    Thanks to Regis Geisler for providing several valuable insights into weld-groove designs. This article is a commendable contribution to the worldwide community of welding engineers, consultants, fabricators and technologists. Let us join hands to congratulate him for his good work.

  • Usama Salman wrote:

    Hello Sir,
    What standard angle or length will be required for beveling a Double-V butt joint if the beam size more than 65mm ??
    Will it be suitable for a 65 mm Beam Length to be beveled a Double-V joint??
    Awaiting your responce.

  • Abhishek Pathade wrote:

    I have a situation wherein I’ve to butt weld a gusset to the web of a girder, the angle of the gusset is very acute to the web and has accessibility issues, the angle available is 15 degrees to be specific. Is there any codal reference suggesting minimum angle required to achieve full penetration butt weld which can pass the test?

    Thanks & Best Regards


  • Renjith wrote:

    Can anyone pls let me know the max thickness up to which we can bevel the pipe to standard single bevel( 37.5 deg) angle and beyond what thickness should we opt compound beveling( 37/10deg) as minimum. Thanks

  • Dan Ondiek wrote:

    Hi sir. I need to weld a 12 mm plate with a butt weld. Should I bevel the joint or just put a gap between the plates? Is a gap of 6 mm OK with it?

  • krishna wrote:

    I want to weld aluminum 6061 & 6082. In order to optimize tensile strength, what configuration should I select. I also want to study if there is any strength difference in weld joints when different configurations are used. Could you please help?

  • Ralph Varney wrote:

    I need to design a weld prep to weld a 13″ diam pipe with a wall thickness of 5/8″ to a tapered cone that has a wall thickness of 3/4″. 316 SS. I appreciate having to machine back the excess thickness to make the root even. Is there a specified angle?

    But.. the biggest issue is to keep the end of the cone (3 feet long) in line with the rest of the pipe. My tolerance is +/- .005″ Lots of luck with that when dealing with stainless. But.. I must try…

    Any thoughts or suggestion on how to best design such a weld prep? Eliminating the traditional gap in the root comes to mind.. but I need a 100% weld and access to the inside is impossible.


  • santokh singh wrote:

    How do I calculate the bevel degree in relation to steam pipe thickness?

  • Prasad rode wrote:

    How much is allowed in plate distortion for square butt joint? Specially for storage tank annular plate joint.

  • mohan Potpelwar wrote:

    Which joint suitable for pressure vessels (for ss material)

  • John Forbes wrote:

    Please advise if there is a code listing the MAX allowable weave distance for FCAW Gas Shielded wire in the VERTICAL position ?
    Advise if there is a LIMIT on the included angles for vertical plate welding – single & double V ( all material Carbon steel)

  • John Forbes wrote:

    Provided that a weld can be successfully done with a reduction in weld included angles ( say 45 degrees instead of 60 degrees included angles & qualified by WPS complete mechanical & NDT? NDE testing in say 50 mm Carbon steel plate are there any CODES that prohibit this? API 650 ASME VIII???

  • Saravanan wrote:

    Good morning. I work in structural steel fabrication and need one clarification regarding a Column to Column Butt joint (V-joint). We are doing an FCAW process where the thickness of the column flange is 30 mm. Room temperature is 22 deg C. Can you explain whether preheat is required or not? Thank you.

  • Joseph Williams wrote:

    How can I determine the minimum root face of an I-beam flange thickness 25.4 mm welded to a 50.8 mm base plate? Are there any formulas regarding this?

  • Kevin Chapman wrote:

    Regis, I am bidding a project on a some vertical well shaft pipe that will be inspected/U.T tested to ASME section 9. The shafts pipes are 72” in Dia. x 1” thick. The owner doesn’t want obstructions like a backing bar or any manpower inside the shaft for back welding. So that eliminates the double V grove joint. The joint has to be welded from the outside and we are considering a open buttweld w/ 6010 root 5/32 and 0.52 Fcaw-g dual shield ( Hobart ) W/striat c02 for fill and cap. The ASTM is A-36. There are 39 well shafts with 10 welded joints in ea. ( 390 ) welds. I realize that the double v grove buttweld would be more cost effective, but as I mentioned we are not allowed inside the pipe. So I am planning to use the 22-1/2 degree ( 45 degree ) bevel W/ 1/8” land and for our joint configuration I am coming up with takes about 29.0136 lbs. deposited material for 18.84 liner foot of weld with an 3/32” cap restraint, fill, and root ( 6010 ) 1/8”landing with a deposition rate of approx. 81% in addition of incorporating the slow deposition rate of the 6010 ( 29.0136 lbs. deposited material ) for ea. weld. The welds would be in the horizontal position. We basically have to complete the fit up, the weld, and the U.T in 12 hours x’s 2 welders on days and 2 welders on nights total of 22 man hours with breaks, lunch, etc. to complete 2 joints between day and night shift with total of 4 welders. We always run the Hobart in excess of the manufactures recommendations so @380 IPM avg. and per our qualified procedure we ran we can run as high as 485 IPM if need be, of course turning it down for the cap from 33 to 28-29 volts. I was able to get 13.88 IPM in which =’s 2.74 lbs. of material per foot.
    My question to you is would you know what would be our optimum joint configuration to gain speed and save consumables? 15 degree bevel with 1/8” land on one side and J bevel with 1/8” land on other side but what would be the concavity on the J bevel? Or 10 degree bevel on one side and 30 degree bevel on the other side W/1/8” landing and get some tapered cups? Or single bevel with 40 degree bevel or other side W/ I/8” land? This seems like it would be difficult to get the root in? Hopefully you can provide your expertise and answer my questions by tomorrow. I have to revise and resubmit our quote be Mon. afternoon.
    Thank You Kevin Chapman
    Chapman’s Welding Inc

  • Mohamed wrote:

    Dear Sir,
    I would like to ask you for my knowledge that, Two vertical weld joints can meet in same junction (Plus). If i remove (Flush) by grind one vertical joint it is acceptable.
    Thanks is advance

  • Reza wrote:

    May I ask where this ratio of 1.25 comes from?

  • YASH RAJ wrote:

    Please give weld design for double-bevel full penetration welded joint for 60 mm thick plate.

  • kiran wrote:

    Does a single Vee butt joint preparation for GTAW will differ from that of GMAW in the same thickness of similar metals?

    • Rob wrote:

      GMAW (MIG) might have a larger Root Opening (width between plates) but doesn’t have to.

      GTAW (TIG) fitup is likely to minimize the gap due to the higher skill level (for TIG) and the ability to get the Electrode to the Root (and heat correctly before making a deposit).

      TIG could guarantee Root penetration (if you weld correctly) but with Wire you don’t want to run over the Joint (super fast) and not be certain of penetration of the Backup Strip.

      TIG you can see, Wire might be run too cold (cold lap, see: ) and not have fusion in the ‘corner’ of the Root and the Backup Strip.

      Your question makes me think the answer is not too technical and I tried to make it simple. Did I answer your question?

  • Ramakrishna wrote:

    Dear Sir,
    I have a question regarding the fillet weld. For a T-joint of 5 mm thick aluminum plates, what is the safest weld size if I have to specify the fillet weld on both sides? Is it Z5 or Z3? Thank you.

  • Alfredo Guevara wrote:

    I am butt welding by laser/robot for high volume using DP980 material with thickness around 2.8 mm. Do you have some experience with this steel? I think the butt welding by laser is the best option for heating this material. Our main goal is to reduce the weight, reduce thickness and eliminate the overlap steel.

    Thank you. I appreciate your comments!

    Alfredo Guevara

  • Sivakumar wrote:

    My question: The material thickness is 25 mm. It has two different joint configurations which are one joint in single bevel & another one is single V groove joint. Will both joints give the same strength or will any difference will show in the result? It is full penetration butt weld joint.

  • Ahmed Abdelaal wrote:

    I am going to use an automatic welding machines in pipe spools fabrication and the manufacture say that,a special “J” bevel must be machined on the pipe end. It is not necessary to machine the standard “V” bevels on the fittings. is this type of joint is accepted or not?

  • Ananth wrote:

    Is there any reason for using a double V butt joint when vibratory tig welding in AL 5052 metal?

  • Motilal G wrote:

    In SAW welding process, which plate thickness should consider a Double V Butt Joint? What is the filler wire and flux consumption for 3.9 m dia and 20 mm thick?

  • Senaka wrote:

    What standard angle or length will be required for beveling a Double-V butt joint if the beam size more than 50mm?

  • Robert Resendez wrote:

    What is the welding procedure for welding 14 gauge plate butt joint flat with MIG?

  • haragopal wrote:

    Sir, We have a Converter Plate of thickness 60mm which is to be repaired. To do 45deg Double V welding, there is no access. So we are planning to do a single V 45deg. Can you suggest/advice us if it is ok or we have to change the included angle.

  • Josef crawley wrote:

    Making a full pen weld using v groove – root opening -0 – root face – 1/16 on 1/2 well thickness pipe – cement lined – is it still a full pen weld – if cement is not removed on inside of pipe ? Thanks

  • bishnu kumar wrote:

    25+32 mm plate joint with backing strip butt weld.
    please tell me the testing procedure.

  • Suzhou Omax Building Materials Co., Ltd wrote:

    Awesome article! I want people to know just how good this information is in your article. It’s interesting, compelling content. Your views are much like my own concerning this subject.Expansion Joint Plate Cover

  • Jon wrote:

    Can I ask what size v prep would you put on 10 mm thick plate ?
    Using mig

End Comment -->
Calendar & Events
Greater Chicago Design-2-Part Show
August 3 - 4, 2021
Schaumburg Convention Center – Schaumburg, IL
MD&M West
August 10 - 11, 2021
Anaheim, CA
rapid + tct
September 13 - 15, 2021
Chicago, IL
September 13 - 16, 2021
Chicago, IL
Wisconsin Manufacturing & Technology Show
October 5 - 7, 2021
West Allis, WI
October 19 - 21, 2021
West Springfield, Massachusetts