Section 3 Electrodes for manual and gravity welding
Clasification Society 2024 - Version 9.40
Clasifications Register Rules and Regulations - Rules for the Manufacture, Testing and Certification of Materials, July 2022 - Chapter 11 Approval of Welding Consumables - Section 3 Electrodes for manual and gravity welding

Section 3 Electrodes for manual and gravity welding

3.1 Grading

3.1.1 Dependent on the results of the mechanical and other tests, approval will be allocated as one of the grades from Table 11.1.1 Welding consumable grades appropriate to structural and low temperature service steel grades.

3.1.2 Approval of an electrode will be given in conjunction with a welding technique indicated by a suffix `m' for manual welding, `G' for gravity or contact electrode and `p' for deep penetration electrode.

3.1.3 If the electrodes are in compliance with the requirements of the hydrogen test given in Ch 11, 3.4 Hydrogen test, a suffix `H15' or `H10' or `H5' will be added to the grade mark. Table 11.3.1 Minimum low hydrogen approval requirements for manual and gravity electrodes shows the mandatory levels of low hydrogen approval for the various approval grades.

Table 11.3.1 Minimum low hydrogen approval requirements for manual and gravity electrodes

Approval grades Low hydrogen grade required
1 (1N), 2 (2N), 3 (3N) NR
2Y, 3Y, 4Y H15 (see Note 2)
2Y40 to 5Y40 H15
3Y47 H10
3Y42 to 5Y42 H10
3Y46 to 5Y46 H10
3Y50 to 5Y50 H10
3Y55 to 5Y55 H5
3Y62 to 5Y62 H5
3Y69 to 5Y69 H5
3Y89 to 4Y89 H5
3Y96 to 4Y96 H5
1 Ni H15
2 Ni H15
3 Ni H15
5 Ni NR (see Note 3)
9 Ni NR (see Note 3)

Note 1. NR – Not required. Approval may be obtained when requested.

Note 2. Optional in this case. If low hydrogen approval is not obtained, there is a limitation on the carbon equivalent of the steel which is permitted to be welded.

Note 3. Assumes use of an austenitic, non-transformable, filler material.

3.1.4 For each strength level, electrodes which have satisfied the requirements for a higher toughness grade are considered as complying with the requirements for a lower grade.

3.1.5 Electrodes approved for normal and higher strength levels up to and including 'Y' are also considered suitable for welding steels in the three strength levels below that for which they have been approved.

3.1.6 Electrodes approved for strength levels Y40 to Y50, but excluding Y47 are also considered suitable for welding steels in two strength levels below that for which they have been approved.

3.1.7 Electrodes approved for strength levels Y47, Y55 and above are also considered suitable for welding steels in only one strength level below that for which they have been approved.

3.1.8 Electrodes approved for strength level Y89 are considered suitable for this strength level only.

3.1.9 Electrodes approved for strength level Y96 are also considered suitable for welding steels in one strength level below that for which they have been approved.

3.1.10 The welding current used is to be within the range recommended by the manufacturer and, where an electrode is stated to be suitable for both a.c. and d.c., a.c. is to be used for the preparation of the test assemblies.

3.1.11 Where an electrode is submitted only for approval for fillet welding and to which the butt weld test provided in Ch 11, 3.3 Butt weld test assemblies is not considered applicable, approval tests are to consist of the fillet weld tests as given in Ch 11, 3.5 Fillet weld test assemblies and deposited metal tests with chemical analyses as given in Ch 11, 3.2 Deposited metal test assemblies.

3.2 Deposited metal test assemblies

3.2.1 The deposited metal test assemblies are to be prepared in the downhand position as shown in Figure 11.3.1 Deposited metal test assembly, one with 4 mm diameter electrodes and the other with 8 mm diameter electrodes, or the largest size manufactured if this is less than 8 mm diameter. If an electrode is available in one diameter only, one test assembly is sufficient. Any of the grades of steel in Table 11.1.1 Welding consumable grades appropriate to structural and low temperature service steel grades may be used for the preparation of these assemblies, up to a strength level which is not more than two levels above that for which approval is sought.

Figure 11.3.1 Deposited metal test assembly

3.2.2 For Y47 grades, as an alternative to Figure 11.3.1 Deposited metal test assembly, the thickness of the plate used for the test assembly may be taken as 50 mm.

3.2.3 The weld metal is to be deposited in single- or multi-run layers according to normal practice, and the direction of deposition of each layer is to alternate from each end of the plate, each run of weld metal being not less than 2 mm and not more than 4 mm thick. Between each run, the assembly is to be left in still air until it has cooled to less than 250°C, the temperature being taken in the centre of the weld, on the surface of the seam. After being welded, the test assemblies are not to be subjected to any heat treatment, except in those higher strength grades where it is considered necessary to use the welded joint in the stress-relieved (tempered) condition. In those cases, the code `sr' will be added to the approval grading.

3.2.4 The chemical analysis of the deposited weld metal in each deposited metal test assembly is to be supplied by the manufacturer and is to include the content of all significant alloying elements. The results of the analysis are not to exceed the limit values specified in the standards or by the manufacturer, the narrower tolerances being applicable in each case.

3.2.5 One tensile and three impact test specimens are to be taken from each test assembly as shown in Figure 11.3.1 Deposited metal test assembly. Care is to be taken that the axis of the tensile test specimen coincides with the centre of the weld and the mid-thickness of the plates. The impact test specimens are to be cut perpendicular to the weld, with their axes 10 mm from the upper surface. The notch is to be positioned in the centre of the weld and cut in the face of the test specimen perpendicular to the surface of the plate.

3.2.6 The results of all tests are to comply with the requirements of Table 11.3.2 Requirements for deposited metal tests (covered electrodes) as appropriate.

Table 11.3.2 Requirements for deposited metal tests (covered electrodes)

Grade (see Note 3) Yield stress N/mm2 minimum Tensile strength N/mm2
(see Note 1)
Elongation on 50 mm % minimum Charpy V-notch impact tests
Test temperature °C Average energy
(see Note 2)
J minimum
1N, 2N, 3N 305 400 – 560 22 +20, 0, –20 47
           
1Y, 2Y, 3Y, 4Y 375 490 – 660 22 +20, 0, –20, –40 47
           
2Y40, 3Y40, 4Y40, 5Y40 400 510 – 690 22 0, –20, –40, –60 47
3Y47 460 570 – 720 19 –20 64
3Y40 400 510 – 690 22 –20 47
3Y42 420 520 – 680 20 –20 47
3Y46 460 540 – 720 20 –20 47
3Y50 500 590 – 770 18 –20 50
3Y55 550 640 – 820 18 –20 55
3Y62 620 700 – 890 18 –20 62
3Y69 690 770 – 940 17 –20 69
3Y89 890 940 –1100 14 –20 69
3Y96 960 980 –1150 13 –20 69
4Y40 400 510 – 690 22 –40 47
4Y42 420 520 – 680 20 –40 47
4Y46 460 540 – 720 20 –40 47
4Y50 500 590 – 770 18 –40 50
4Y55 550 640 – 820 18 –40 55
4Y62 620 700 – 890 18 –40 62
4Y69 690 770 – 940 17 –40 69
4Y89 890 940 – 1100 14 –40 69
4Y96 960 980 – 1150 13 –40 69
5Y40 400 510 – 690 22 –60 47
5Y42 420 520 – 680 20 –60 47
5Y46 460 540 – 720 20 –60 47
5Y50 500 590 – 770 18 –60 50
5Y55 550 640 – 820 18 –60 55
5Y62 620 700 – 890 18 –60 62
5Y69 690 770 – 940 17 –60 69
11/2Ni 375 490 – 640 22 –80 34
21/4Ni 375 490 – 640 22 –90 34
31/2Ni 375 490 – 610 25 –100 34
5 Ni 400 540 – 740 25 –120 34
9 Ni 400 640 – 790 25 –196 34

Note 1. Single values are the minimum requirements.

Note 2. Energy values from individual impact test specimens are to comply with Ch 11, 1.4 Approval procedures 1.4.3.

Note 3. Grade 1Y is not applicable to SMAW consumables referenced in Ch 11, 3 Electrodes for manual and gravity welding.

3.3 Butt weld test assemblies

3.3.1 Butt weld assemblies, as shown in Figure 11.3.2 Butt weld test assembly, are to be prepared for each welding position (downhand, horizontal-vertical, vertical-upward, vertical-downward, and overhead) for which the electrode is recommended by the manufacturer. In the case of electrodes for normal strength and higher strength steels (up to 355 N/mm2 minimum specified yield strength), electrodes satisfying the requirements for downhand and vertical-upward positions will be considered as also complying with the requirements for the horizontal-vertical position. In all other cases, approval for the horizontal-vertical position will require a butt weld to be made in that position and fully tested.

Figure 11.3.2 Butt weld test assembly

3.3.2 For Y47 grades, as an alternative to Figure 11.3.2 Butt weld test assembly, the thickness of the plate used for the test assembly may be taken as 50 mm.

3.3.3 Where the electrode is to be approved only in the downhand position, an additional test assembly is to be prepared in that position.

3.3.4 The grades of steel used for the preparation of the test assemblies are to be as follows:

Grade 1 (1N) electrodes A
Grade 2 (2N) electrodes A, B or D
Grade 3 3(N) electrodes A, B, D or E
Grade 2Y electrodes AH32, AH36, DH32 or DH36
Grade 3Y electrodes AH32, AH36, DH32, DH36, EH32 or EH36
Grade 4Y electrodes AH32
AH36, DH32, DH36, EH32, EH36, FH32 or FH36
Grade 2Y40 electrodes AH40 or DH40
Grade 3Y40 electrodes AH40, DH40 or EH40
Grade 4Y40 electrodes AH40, DH40, EH40 or FH40
Grade 5Y40 electrodes AH40, DH40, EH40 or FH40
Grade 3Y47 electrodes EH47

Where Grade 32 higher tensile steel is used, the tensile strength is to be not less than 490 N/mm2. The chemical composition, including the content of grain refining elements, is to be reported in all cases where higher tensile steel is used.

3.3.5 For all other grades, the steel plates used are to be selected by reference to Table 11.1.1 Welding consumable grades appropriate to structural and low temperature service steel grades, and are to have at least their chemical composition and tensile properties within the limits specified for that grade in Ch 3 Rolled Steel Plates, Strip, Sections and Bars. The strength grade used is to be the same as that for which approval is sought, and the toughness grade is to be no higher than that for which approval is also sought.

3.3.6 The test assemblies are to be made by welding together two plates of equal thickness (15 to 20 mm), not less than 100 mm in width and of sufficient length to allow the cutting out of test specimens of the prescribed number and size. The plate edges are to be prepared to form a single V-joint, the included angle between the fusion faces being 60° and the root gap 2 to 3 mm. The root face is to be 0 to 2 mm.

3.3.7 The following welding procedure is to be adopted in making the test assemblies:

Downhand (a). The first run with 4 mm diameter electrode. Remaining runs (except the last two layers) with 5 mm diameter electrodes or above according to the normal welding practice with the electrodes. The runs of the last two layers with the largest diameter of electrode manufactured or 8 mm whichever is the lesser.

Downhand (b) (where a second downhand test is required). First run with 4 mm diameter electrode. Next run with an electrode of intermediate diameter of 5 mm or 6 mm, and the remaining runs with the largest diameter of electrode manufactured or 8 mm whichever is the lesser.

Horizontal-vertical. First run with 4 mm or 5 mm diameter electrode. Subsequent runs with 5 mm diameter electrodes.

Vertical-upward and overhead. First run with 3,25 mm diameter electrode. Remaining runs with 4 mm diameter electrodes or possibly with 5 mm if this is recommended by the manufacturer for the positions concerned.

Vertical-downward. If the electrode being tested is intended for vertical welding in the downward direction, this technique is to be adopted for the preparation of the test assembly using electrode diameters as recommended by the manufacturer.

3.3.8 For all assemblies, the back sealing runs are to be made with 4 mm diameter electrodes in the welding position appropriate to each test sample, after cutting out the root run to clean metal. For electrodes suitable for downhand welding only, the test assemblies may be turned over to carry out the back sealing run.

3.3.9 Normal welding practice is to be used and, between each run, the assembly is to be left in still air until it has cooled to less than 250°C, the temperature being taken in the centre of the weld, on the surface of the seam. After being welded, the test assemblies are not to be subjected to any heat treatment, except in those higher strength grades where it is considered necessary to use the welded joint in the stress-relieved (tempered) condition. In those cases, the code `sr' will be added to the approval grading.

3.3.10 It is recommended that the welded assemblies be subjected to a radiographic examination to ascertain if there are any defects in the weld prior to the preparation of test specimens.

3.3.11 The test specimens as shown in Figure 11.3.2 Butt weld test assembly and Figure 11.3.3 Butt weld test assembly position of impact test specimens are to be prepared from each test assembly.

Figure 11.3.3 Butt weld test assembly position of impact test specimens

3.3.12 The results of all tensile and impact tests are to comply with the requirements of Table 11.3.3 Requirements for butt weld tests (covered electrodes) as appropriate. The position of fracture in the transverse tensile test is to be reported.

Table 11.3.3 Requirements for butt weld tests (covered electrodes)

Grade (see Note 3) Tensile strength
N/mm2
Bend test ratio:
Charpy V-notch impact tests
Test temperature
°C
Average energy
(see Note 1)
J minimum
All positions
(see Note 2)
1N, 2N, 3N 400 3 +20, 0, –20 47 (34)
         
1Y, 2Y, 3Y, 4Y 490 3 +20, 0, –20, –40 47 (34)
         
2Y40, 3Y40, 4Y40, 5Y40 510 3 0, –20, –40, –60 47 (39)
3Y47 570 - 720 4 –20 64
3Y40 510 3 –20 47 (39)
3Y42 520 – 680 4 –20 47
3Y46 540 – 720 4 –20 47
3Y50 590 – 770 4 –20 50
3Y55 640 – 820 5 –20 55
3Y62 700 – 890 5 –20 62
3Y69 770 – 940 5 –20 69
3Y89 940 6 –20 69
3Y96 980 7 –20 69
4Y40 510 3 –40 47 (39)
4Y42 520 – 680 4 –40 47
4Y46 540 – 720 4 –40 47
4Y50 590 – 770 4 –40 50
4Y55 640 – 820 5 –40 55
4Y62 700 – 890 5 –40 62
4Y69 770 – 940 5 –40 69
4Y89 940 6 –40 69
4Y96 980 7 –40 69
5Y40 510 3 –60 47 (39)
5Y42 520 – 680 4 –60 47
5Y46 540 – 720 4 –60 47
5Y50 590 – 770 4 –60 50
5Y55 640 – 820 5 –60 55
5Y62 700 – 890 5 –60 62
5Y69 770 – 940 5 –60 69
11/2Ni 490 3 –80 27
21/4Ni 490 3 -90 27
31/2Ni 490 3 –100 27
5 Ni 540 4 –120 27
9 Ni 640 4 –196 27

Note 1. Energy values from individual impact test specimens are to comply with Ch 11, 1.4 Approval procedures 1.4.3.

Note 2. Values in ( ) apply only to welds made in the vertical position with upward progression.

Note 3. Grade 1Y is not applicable to SMAW consumables referenced in Ch 11, 3 Electrodes for manual and gravity welding.

3.3.13 The bend test specimens can be considered as complying with the requirements if, after bending, no crack or other open defect exceeding 3 mm in dimensions can be seen on the outer surface.

3.4 Hydrogen test

3.4.1 The hydrogen gradings are specified in Ch 11, 3.1 Grading 3.1.3. The hydrogen grading required determines the method of testing permitted as shown in Table 11.3.4 Permitted methods for obtaining low hydrogen grading. Four test specimens are to be prepared and tested, and all four hydrogen test results must be below the maximum value for the hydrogen mark required.

Table 11.3.4 Permitted methods for obtaining low hydrogen grading

Hydrogen Grade Permitted Method
H15 ISO 3690 (Mercury or Thermal Conductivity Detector Method) Or Glycerine (See Note)
H10 ISO 3690 (Mercury or Thermal Conductivity Detector Method)
H5 ISO 3690 (Mercury or Thermal Conductivity Detector Method)

Note ISO method preferred.

3.4.2 The minimum holding time at a given test temperature for Thermal Conductivity Method should be as shown in Table 11.3.5 Temperature and minimum holding time.

Table 11.3.5 Temperature and minimum holding time

Measuring Method Test Temperature (°C) Minimum Holding Time (h)
Thermal Conductivity Method (see Note 1) Gas Chromatography 45 72
150 6
Note 1. The use of hot carrier gas extraction method will be specially considered subject to verification of testing procedure to confirm that collection and measurement of the hydrogen occurs continuously until all the diffusible hydrogen is quantified.

3.5 Fillet weld test assemblies

3.5.1 Fillet weld assemblies as shown in Figure 11.3.4 Fillet weld test assembly are to be prepared for each welding position (horizontal-vertical, vertical-upward, vertical-downward or overhead) for which the electrode is recommended by the manufacturer. The grade of steel used for the test assemblies is to be as detailed in Ch 11, 3.3 Butt weld test assemblies 3.3.4. The length of the test assembly, L, is to be sufficient to allow at least the deposition of the entire length of the largest diameter electrode being tested. Where an electrode is submitted for approval of both butt and fillet welding, approval tests are to include the deposited metal tests as given in Ch 11, 3.2 Deposited metal test assemblies, the butt weld tests as given in Ch 11, 3.3 Butt weld test assemblies, and only one fillet weld test as given in subsequent paragraphs of this sub-Section welded in the horizontal-vertical position.

Figure 11.3.4 Fillet weld test assembly

3.5.2 For Y47 grades, as an alternative to Figure 11.3.4 Fillet weld test assembly, the thickness of the plate used for the test assembly may be taken as 50 mm.

3.5.3 The electrode sizes to be used are the maximum and minimum diameters recommended by the manufacturer for fillet welding. The first side is to be welded using the maximum diameter. The second side is to be welded only after the assembly has been allowed to cool below 50°C using the minimum diameter. The size of these single run fillet welds will, in general, be determined by the electrode size and the welding current employed during testing and should represent the range of fillet weld bead sizes recommended by the manufacturer.

3.5.4 Each test assembly is to be sectioned to form three macro-sections, each about 25 mm thick. These are to be examined for root penetration, satisfactory profile, freedom from cracking and reasonable freedom from porosity and slag inclusions. Any undercut is not to exceed 0,5 mm in depth. Convexity or concavity of the profile is not to exceed one-tenth of the fillet bead throat dimension. All such observations are to be reported.

Figure 11.3.5 Hardness tests for fillet weld test assembly

3.5.5 Hardness measurements are to be made on the central macro-section only, as shown in Figure 11.3.5 Hardness tests for fillet weld test assembly. The results are to be reported.

3.5.6 One of the remaining sections of the assembly is to have the weld on the first side gouged or machined to facilitate breaking the fillet weld on the second side by closing the two plates together, subjecting the root of the weld to tension. On the other remaining section, the weld on the second side is to be gouged or machined and the section fractured using the same procedure. The fractured surfaces are to be examined. They are to show satisfactory penetration, freedom from cracks and reasonable freedom from porosity and this should be reported.

3.6 Electrodes designed for deep penetration welding

3.6.1 Where an electrode is designed solely for the deep penetration welding of downhand butt joints and horizontal-vertical fillets in normal tensile strength steel, only the tests detailed in Ch 11, 3.7 Deep penetration butt weld test assemblies and Ch 11, 3.8 Deep penetration fillet weld test assemblies are required for approval purposes.

3.6.2 Electrodes designed solely for the deep penetration welding technique will be approved as complying with Grade 1 requirements only and will be given the suffix `p'.

3.6.3 Where a manufacturer recommends that an electrode having deep penetrating properties can also be used for downhand butt welding of thicker plates with prepared edges, the electrode will be treated as a normal penetration electrode, and the full series of tests in the downhand position is to be carried out, together with the deep penetration tests given in Ch 11, 3.7 Deep penetration butt weld test assemblies and Ch 11, 3.8 Deep penetration fillet weld test assemblies.

3.6.4 Where a manufacturer desires to demonstrate that an electrode, in addition to its use as a normal penetration electrode, also has deep penetrating properties when used for downhand butt welding and horizontal fillet welding, the additional tests given in Ch 11, 3.7 Deep penetration butt weld test assemblies and Ch 11, 3.8 Deep penetration fillet weld test assemblies are to be carried out.

3.6.5 Electrodes approved for both normal and deep penetration welding will have the suffix 'p' added after the appropriate grade mark for normal penetration welding.

3.6.6 Where the manufacturer prescribes a different welding current and procedure for the electrode when used as a deep penetration electrode and a normal penetration electrode, the recommended current and procedure are to be used when making the test assemblies in each case.

3.7 Deep penetration butt weld test assemblies

3.7.1 Two plates of thickness equal to twice the diameter of the core of the electrode plus 2 mm are to be butt welded together with one downhand run of welding from each side. The plates are to be not less than 100 mm wide and of sufficient length to allow the cutting out of the test specimens of the correct number and size as shown in Figure 11.3.6 Deep penetration butt weld test assembly. Grade A steel is to be used for these test assemblies. The joint edges are to be prepared square and smooth and, after tacking, the gap is not to exceed 0,25 mm. The test assembly is to be welded using an 8 mm diameter electrode, or the largest diameter manufactured if this is less than 8 mm and the assembly is to be allowed to cool below 50°C between runs.

3.7.2 The test specimens as shown in Figure 11.3.3 Butt weld test assembly position of impact test specimens and Figure 11.3.6 Deep penetration butt weld test assembly are to be prepared from each test assembly.

Figure 11.3.6 Deep penetration butt weld test assembly

3.7.3 The results of tensile and impact tests are to comply with the requirements of Table 11.3.3 Requirements for butt weld tests (covered electrodes) for Grade 1 electrodes. The position of fracture in the tensile test is to be reported. The bend test specimens are to be in accordance with Ch 11, 3.3 Butt weld test assemblies 3.3.13.

3.7.4 The discards at the end of the welded assemblies are to be not more than 35 mm wide. The joints of these discards are to be polished and etched and must show complete fusion and inter-penetration of the weld beads. At each cut in the test assembly, the joints are also to be examined to ensure that complete fusion has taken place.

3.8 Deep penetration fillet weld test assemblies

3.8.1 A fillet weld assembly is to be prepared as shown in Figure 11.3.7 Deep penetration fillet weld test assembly with plates about 12,5 mm in thickness. The welding is to be carried out with one run for each fillet with plate A in the horizontal plane during the welding operations. The length of the fillet is to be 160 mm and the gap between the plates is to be not more than 0,25 mm. Grade A steel is to be used for these test assemblies.

Figure 11.3.7 Deep penetration fillet weld test assembly

3.8.2 The fillet weld on one side of the assembly is to be carried out with a 4 mm diameter electrode, and that on the other side with the maximum diameter of electrode manufactured. The welding current used is to be within the range recommended by the manufacturer, and the welding is to be carried out using normal welding practice except that the assembly is to be allowed to cool below 50°C between runs.

3.8.3 The welded assembly is to be cut by sawing or machining within 35 mm of the ends of the fillet welds, and the joints are to be polished and etched. The welding of the fillet made with a 4 mm diameter electrode is to show a penetration of 4 mm (see Figure 11.3.7 Deep penetration fillet weld test assembly) and the corresponding penetration of the fillet made with the maximum diameter of electrode manufactured is to be reported.

3.9 Electrodes designed for gravity or contact welding

3.9.1 Approval for welding using the gravity, `G', technique is available for welding only normal strength and higher tensile steels up to and including Grade 36.

3.9.2 Where an electrode is submitted solely for approval for use in contact welding using automatic gravity or similar welding devices, deposited metal tests, butt weld tests and, where appropriate, fillet weld tests similar to those for normal manual electrodes are to be carried out using the process for which the electrode is recommended by the manufacturer.

3.9.3 Where an electrode is submitted for approval for use in contact welding using automatic gravity or similar welding devices in addition to normal manual welding, butt weld and, where appropriate, fillet weld tests, using the gravity or other contact device as recommended by the manufacturer, are to be carried out in addition to the normal approval tests.

3.10 Annual tests

3.10.1 For normal penetration electrodes, the annual tests are to consist of two deposited metal test assemblies. These are to be prepared and tested in accordance with Ch 11, 3.2 Deposited metal test assemblies. If an electrode is available in one diameter only, one test assembly is sufficient.

3.10.2 Where an electrode is approved solely for deep penetration welding, the annual test is to consist of one butt welded test assembly. This is to be prepared and tested in accordance with Ch 11, 3.7 Deep penetration butt weld test assemblies.

3.10.3 Where an electrode is approved for both normal and deep penetration welding, annual tests as detailed in Ch 11, 3.10 Annual tests 3.10.1 and Ch 11, 3.10 Annual tests 3.10.2 are to be carried out.

3.10.4 Where an electrode is approved solely for gravity or contact welding, the annual test is to consist of one deposited metal test assembly using the gravity or other contact device as recommended by the manufacturer.

3.10.5 Where an electrode is approved for both manual and gravity welding, annual tests as detailed in Ch 11, 3.10 Annual tests 3.10.1 and Ch 11, 3.10 Annual tests 3.10.4 are to be carried out.

3.10.6 A hydrogen test is required for Y89 and Y96 grades.


Copyright 2022 Clasifications Register Group Limited, International Maritime Organization, International Labour Organization or Maritime and Coastguard Agency. All rights reserved. Clasifications Register Group Limited, its affiliates and subsidiaries and their respective officers, employees or agents are, individually and collectively, referred to in this clause as 'Clasifications Register'. Clasifications Register assumes no responsibility and shall not be liable to any person for any loss, damage or expense caused by reliance on the information or advice in this document or howsoever provided, unless that person has signed a contract with the relevant Clasifications Register entity for the provision of this information or advice and in that case any responsibility or liability is exclusively on the terms and conditions set out in that contract.