8.3 Joint Efficiency (E) Selection, Missing Data & Radiography Factors

8.3 Joint Efficiency (E) Selection, Missing Data & Radiography Factors

API 653 Core Principle: In cylindrical shell calculations, joint efficiency ($E$) directly scales the allowable load-bearing capacity of the vertical weld seams. Because circumferential hoop stress acts perpendicularly across vertical joints, the integrity of these seams dictates shell minimum thickness. API 653 Table 4.2 governs the selection of $E$ based on the original standard of construction, joint design, and degree of radiographic examination, while providing formal engineering pathways to upgrade joint efficiencies when records are absent.

In storage tank design and fitness-for-service evaluation, the vertical longitudinal weld seams represent the highest structural risk. Under internal hydrostatic pressure, circumferential hoop membrane stress is precisely twice the longitudinal axial stress ($\sigma_h = 2 \times \sigma_L$). Therefore, while horizontal circumferential girth seams carry low stress and are rarely governing, vertical weld seams carry maximum tensile load. The weld joint efficiency factor $E$ ($0.35 \le E \le 1.00$) reflects the probability of volumetric fabrication flaws—such as slag inclusions, incomplete penetration, porosity, and lack of fusion—weakening the joint.

---

1. API 653 Table 4.2: Joint Efficiency Hierarchy

API 653 Table 4.2 categorizes joint efficiencies based on the original standard of construction, edition date, joint configuration, and non-destructive examination (NDE) applied during fabrication.

+-------------------------------------------------------------------------+
|                    JOINT EFFICIENCY (E) SPECTRUM                        |
|                                                                         |
|   1.00  ==============================================================  |
|         API 650 Basic Standard / App. A with FULL RADIOGRAPHY           |
|                                                                         |
|   0.85  --------------------------------------------------------------  |
|         API 650 with SPOT RADIOGRAPHY per Section 8.1                   |
|         API 12C Butt-Welded with Spot Radiography / Sectioning          |
|                                                                         |
|   0.70  --------------------------------------------------------------  |
|         API 650 Butt-Welded with NO RADIOGRAPHY (Visual Only)           |
|         API 12C Butt-Welded with NO RADIOGRAPHY                         |
|         UNKNOWN CONSTRUCTION CODE (Default Butt-Weld Assumption)        |
|                                                                         |
|   0.70 - 0.75  -------------------------------------------------------  |
|         API 12C / Unknown FULL DOUBLE LAP-WELDED (thickness limited)    |
|                                                                         |
|   0.50 + k/5  --------------------------------------------------------  |
|         Full fillet weld with >= 25% intermittent fillet opposite side  |
|                                                                         |
|   0.35  ==============================================================  |
|         SINGLE LAP-WELDED ONLY (unknown standard of construction)       |
+-------------------------------------------------------------------------+

Modern Construction: API 650 (7th Edition to Present)

For tanks fabricated to modern editions of API 650 (published from 1980 onward):

• Full Radiography ($E = 1.00$): All vertical butt-welded seams are 100% radiographically examined in accordance with API 650 Section 8.1. This permits the shell to be evaluated at 100% of the material's allowable stress.
• Spot Radiography ($E = 0.85$): Vertical seams are radiographed at specified statistical intervals (typically one spot radiograph in the first 10 feet of vertical weld for each welder, and one spot in each additional 100 feet). Spot radiography provides a high confidence level, carrying an efficiency of 0.85.
• No Radiography ($E = 0.70$): For smaller tanks built under API 650 Appendix A or older base standards without volumetric radiography (visual inspection only), $E$ is restricted to 0.70.

Vintage Construction: API 650 (1st to 6th Editions, 1961–1978)

• Butt welds, basic standard: $E = 0.85$
• Butt welds, Appendices D and G: $E = 1.00$

There is no 0.70 row for these editions in API 653 Table 4.2 — the table gives only the two values above.

Historic Welded Tanks: API Specification 12C (1936–1958)

Prior to the release of API 650 in 1961, petroleum storage tanks were built under API 12C. Table 4.2 breaks API 12C into three edition bands:

• 14th and 15th editions (1957–1958): butt joints, $E = 0.85$.
• 3rd through 13th editions (1940–1956): butt joints $E = 0.85$; full double lap-welded joints $E = 0.75$, limited to plate no thicker than 3/8 in.
• 1st and 2nd editions (1936–1939): butt joints $E = 0.85$; full double lap-welded joints $E = 0.70$, limited to 7/16 in. max thickness; full fillet weld with at least 25 % intermittent full fillet on the opposite side gets $E = 0.50 + k/5$, limited to 1/4 in. max thickness, where $k$ is the percent of intermittent weld in decimal form.

Note that single butt-welded joints with a back-up bar were only permitted from 1936 to 1940 and again from 1948 to 1954 — dates worth remembering when a question hands you a construction year.

Riveted Shell Joints (API 653 Section 4.3.4 & Table 4.3)

Tanks constructed prior to 1940 often feature riveted vertical and horizontal seams. Riveted joint efficiencies are determined in accordance with API 653 Table 4.3 or by detailed mechanical calculation evaluating rivet shear, plate bearing stress, and net ligament tensile efficiency:

• Single-riveted lap joints: $E \approx 0.45\text{ to }0.60$
• Double-riveted lap joints: $E \approx 0.60\text{ to }0.75$
• Triple- and quadruple-riveted butt-strap joints: $E \approx 0.75\text{ to }0.85$

---

2. Policy for Missing Construction Records

A frequent challenge encountered by the API 653 inspector is evaluating an aging tank that possesses zero documentation—no manufacturer data plate, no Form U-1/API certification, no mill test reports (MTRs), and no radiographic film archives.

+-------------------------------------------------------------------------+
|                 MISSING DATA HIERARCHY (API 653 SECTION 4)              |
|                                                                         |
|   If Original Construction Standard & NDE Records Are UNKNOWN:          |
|                                                                         |
|   1. Allowable Stress (Table 4.1, lower two courses):                   |
|                                 S   = 23,600 psi (Product)              |
|                                 S_t = 26,000 psi (Hydrotest)            |
|                                                                         |
|   2. Joint Efficiency:          E = 0.70 (for Butt-Welded Seams)        |
|                                 E = 0.35 (for Lap-Welded Seams)         |
|                                                                         |
|   3. Engineering Action:        Accept derated fill height OR perform   |
|                                 radiography / NDE to upgrade E          |
+-------------------------------------------------------------------------+

Default Values for Unknown Tanks

In accordance with API 653 Section 4.3.3.1, Table 4.1 and Table 4.2:

1. Joint Efficiency: For butt-welded tanks of unknown specification, the inspector must assign a default joint efficiency of $E = 0.70$.
2. Allowable Stress: For unknown steel specifications, API 653 Table 4.1 applies with $Y = 30{,}000$ psi and $T = 55{,}000$ psi assumed, giving $S = 23{,}600$ psi and $S_t = 26{,}000$ psi for the bottom two courses and $S = 26{,}000$ psi and $S_t = 27{,}000$ psi for all other courses.

The NDE Upgrading Pathway (API 653 Section 4.3.3.1 & Section 8)

Operating an unknown tank at $E = 0.70$ imposes a severe operational penalty. API 653 provides an explicit engineering procedure to upgrade joint efficiency through field examination:

• Upgrading from $E = 0.70$ to $E = 0.85$: The owner-user may perform spot radiography (or angle-beam ultrasonic shear-wave examination) on the existing vertical weld seams in accordance with the sampling frequency specified in modern API 650 Section 8.1:
  • One spot radiograph taken in the first 10 feet of vertical weld in each shell course.
  • One spot radiograph taken in each additional 100 feet of vertical weld.
  • Radiographic examination of 25% of all vertical-to-horizontal weld intersections.
  • If all examined welds satisfy the acceptance standards of API 650 Section 8.1, the joint efficiency of the vertical seams in that course may be formally upgraded to $E = 0.85$.
• Upgrading to $E = 1.00$: Requires 100% full radiographic examination (or automated ultrasonic shear-wave testing) along the entire length of all vertical weld seams in the shell course under consideration.

---

3. Mathematical Impact of Joint Efficiency on Shell Thickness & Fill Height

The joint efficiency factor $E$ resides in the denominator of the $t_{\text{min}}$ equation. Consequently, required thickness is inversely proportional to $E$:

$t_{\text{min}} \propto \frac{1}{E}$

Quantitative Sensitivity Comparison

Consider the thickness penalty imposed by reduced joint efficiency relative to a fully radiographed seam ($E = 1.00$):

Economic Takeaway: An un-radiographed tank ($E = 0.70$) requires 42.9% more steel thickness to hold the identical liquid level as a fully radiographed tank. Taking spot radiographs to upgrade an undocumented tank from $E = 0.70$ to $E = 0.85$ increases allowable effective liquid head by $21.4%$, often restoring millions of barrels of throughput capacity without physical structural modifications.

---

4. Master Joint Efficiency Reference Table (API 653 Table 4.2)

API 653 Table 4.2 is organized by standard of construction and edition, then by joint type. Read the rows in that order; the table does not simply grade "amount of radiography."

Table notes (these carry the exam points):

• ᵃ Full double lap-welded — full fillet welds on both sides of the lap.
• ᵇ Full fillet weld with at least 25 percent intermittent full fillet on the opposite side; $k$ = percent of intermittent weld expressed in decimal form. So a joint with 40 % intermittent weld on the back side gives $E = 0.50 + 0.40/5 = 0.58$.
• ᶜ Single butt-welded joints with a back-up bar were permitted from 1936 to 1940 and from 1948 to 1954.
• ᵈ Single lap-welded only — this is the 0.35 row, and it is the worst case in the table.

The 1st–6th edition API 650 surprise. Most candidates expect older editions to score lower. In fact API 653 Table 4.2 assigns $E = 0.85$ to the basic standard for API 650 editions 1 through 6, and $E = 1.00$ to tanks built to Appendices D and G of those editions. Read the row, do not reason from intuition.

---

5. Step-by-Step Worked Upgrading Example

An operating terminal storage tank has the following parameters:

• Diameter $D = 100\text{ ft}$, Design fill height $H = 40\text{ ft}$
• Product: Heavy Fuel Oil ($G = 0.92$)
• All construction and inspection records are missing; the tank has butt-welded seams.
• In accordance with API 653 Table 4.2 default rules: $S = 23,600\text{ psi}$ and $E = 0.70$.
• Ultrasonic thickness inspection of Course 1 reveals an actual measured thickness $t_{\text{act}} = 0.490\text{ in.}$

Step 1: Evaluate Tank with Default $E = 0.70$

1. Calculate $t_{\text{min}}$: $t_{\text{min}} = \frac{2.6 \times 100 \times (40 - 1) \times 0.92}{23,600 \times 0.70} = \frac{260 \times 39 \times 0.92}{16,520} = \frac{9,328.8}{16,520} = 0.5647\text{ in.}$
2. Compliance Check: Actual thickness $t_{\text{act}} = 0.490\text{ in.} < t_{\text{min}} = 0.565\text{ in.}$ The tank fails fitness-for-service.
3. Calculate derated liquid height $H_{\text{max}}$: $H_{\text{max}} = \frac{S \cdot E \cdot t_{\text{act}}}{2.6 \cdot D \cdot G} + 1 = \frac{23,600 \times 0.70 \times 0.490}{2.6 \times 100 \times 0.92} + 1 = \frac{8,094.8}{239.2} + 1 = 33.84 + 1 = 34.84\text{ ft}$ The tank must be derated by more than 5 feet (from 40 ft down to 34.8 ft).

Step 2: Upgrade Joint Efficiency via Spot Radiography to $E = 0.85$

The owner commissions an ASNT Level II radiographer to perform spot radiography on all vertical seams of Course 1 per API 650 Section 8.1. All radiographs meet the weld acceptance criteria, formally upgrading the joint efficiency to $E = 0.85$.

1. Recalculate $t_{\text{min}}$ with upgraded $E = 0.85$: $t_{\text{min,new}} = \frac{2.6 \times 100 \times 39 \times 0.92}{23,600 \times 0.85} = \frac{9,328.8}{20,060} = 0.4650\text{ in.}$
2. Compliance Check: Actual thickness $t_{\text{act}} = 0.490\text{ in.} \ge t_{\text{min,new}} = 0.465\text{ in.}$
3. Engineering Conclusion: By spending a few thousand dollars on spot radiography, the facility eliminated the 5.2-foot derating, completely restoring the tank to its full 40.0-foot operating capacity without executing any structural repairs.

---

5. The Rule That Overrides the Whole Table: $E = 1.0$ Away From Welds

Buried in the variable definitions under API 653 4.3.3.1 is a provision that repeatedly decides real evaluations:

$E$ = original joint efficiency for the tank. Use Table 4.2 if the original $E$ is unknown. $E = 1.0$ when evaluating the retirement thickness in a corroded plate, when away from welds or joints by at least the greater of 1 in. or twice the plate thickness.

How to apply it

1. Measure the shortest distance from the edge of the corroded area to the nearest weld or joint.
2. Compute the qualifying clearance: $\max(1\text{ in.},; 2t)$ where $t$ is the plate thickness.
3. If the corroded area is farther away than that clearance, evaluate the retirement thickness with $E = 1.0$, regardless of what Table 4.2 says about the tank's vertical seams.
4. If it is not, the vertical seam's tabulated $E$ applies.

Example. A 0.375-in. shell plate on an unknown-standard tank ($E = 0.70$ from Table 4.2) has a corroded patch whose nearest edge is 2 in. from the vertical seam. The qualifying clearance is $\max(1.0,; 2 \times 0.375) = 1.0$ in. Because 2 in. > 1 in., the patch is evaluated at $E = 1.0$. Against a $t_{\min}$ that would otherwise be inflated by $1/0.70 = 1.43$, this single rule cuts the required thickness by roughly 30 % and can be the difference between "fit for service" and "cut out the plate."

Counter-example. Move the same patch to 0.75 in. from the seam and the clearance test fails, so $E = 0.70$ governs and the required thickness jumps by 43 %.

Why this makes physical sense. The joint efficiency factor exists to discount the weld metal and its heat-affected zone, where fabrication flaws concentrate. Parent plate far from any weld has no such discount to apply — it is simply sound rolled steel. API 653 codifies that reality instead of penalizing base metal for a defect it cannot contain.