U.S. Department of Transportation
Federal Highway Administration
1200 New Jersey Avenue, SE
Washington, DC 20590

Skip to content
Facebook iconYouTube iconTwitter iconFlickr iconLinkedInInstagram

Federal Highway Administration Research and Technology
Coordinating, Developing, and Delivering Highway Transportation Innovations

This report is an archived publication and may contain dated technical, contact, and link information
Back to Publication List        
Publication Number:  FHWA-HRT-14-066    Date:  September 2014
Publication Number: FHWA-HRT-14-066
Date: September 2014


Fatigue Testing of Galvanized and Ungalvanized Socket Connections


Fatigue of structural supports for overhead signs, traffic signals, and highmast light poles has received focused attention from researchers in the last 25 years because of failures that have been reported in welded details. The likely precipitating event was the close consecutive collapse of two cantilevered signal structures in Michigan in 1990, which resulted in property damage, injuries, and one fatality.(1) These lightweight, flexible structures are often susceptible to vibration from wind phenomena such as galloping, vortex shedding, and wind gusts. The dynamic response tends to excite the lower modes of vibration, and with very low damping ratios, large numbers of loading cycles quickly accumulate. These wind-induced vibrations can ultimately lead to fatigue cracking at welded details in the structure. One type of welded connection with poor fatigue resistance is the pole-to-base plate welds that often follow socket connection detailing practice. To make a socket connection, a hole is cut out of the base plate with a slightly larger diameter than the outside diameter of the pole. The tube is partially slid into the base plate, and a perimeter fillet weld adjoins the two pieces, essentially creating a hollow section with a thick stiffening ring welded to the end. The socket connection is the preferred connection method between a pole and a foundation because it can be quickly fabricated.

Fatigue testing of socket connection details began in the early 1980s at Lehigh University but had a resurgence once the 2001 American Association of State Highway Transportation Officials (AASHTO) Standard Specifications for Structural Supports for Highway Signs, Luminaires, and Traffic Signals was published.(2,3) The 2001 specifications were the first to include fatigue design, which  had not been considered in prior editions. In-service structures that historically never had fatigue issues were not meeting the 2001 fatigue provisions. To be compatible with the new provisions, States, designers, and fabricators had to drastically increase pole diameters and wall thickness to comply with the 2001 fatigue provisions. In response to concerns with this situation, several research projects addressed the fatigue resistance of traffic signal structure details, mainly with socket connection details. Some of these projects were conducted at the University of Wyoming, University of Texas–Austin (UT–Austin), Lehigh University, Purdue University, and the University of Minnesota. (See references 4–9.) One of the larger fatigue testing studies published in 2011 (National Cooperative Highway Research Program Project 10-70) included only galvanized details, which provided the lower bound of fatigue resistance based on the research of Koenigs.(6,8) This decision ultimately led to specification changes that neglected any increased fatigue resistance of structures that are not galvanized.


The desired specimen utilized a round tube with an 18-inch outside diameter at the socket connection, 0.25-inch wall thickness, 2-inch-thick base plate, and a 6-bolt, 2-inch-diameter anchor rod arrangement on a 24-inch-diameter bolt circle. A construction drawing was created and sent to two different fabricators. They were asked to provide shop drawings of how the specimen could best be fabricated using their typical processes. Four differences arose between the two manufacturers. Each used different weld profiles, and each had their own galvanizing sources. Fabricator 2 could not make perfectly round poles and instead made round-like tubes by press-braking a flat plate with multiple bends. Fabricator 1 made the tubes from ASTM A595 steel; whereas, fabricator 2 bent ASTM A572 Gr.65 material into a tube. Figure 1 illustrates the detailing of the desired specimen and each fabricator’s shop drawing. Twelve specimens were acquired from each fabricator. Six were in an unfinished state, and six were hot-dip galvanized.

This illustration shows three engineering drawings for the specimens. The top part shows the construction drawing sent out for bid. The specimen is shown to be 169.5 inches in total length, with a 2-inch-thick base plate on the right side and a 1-inch-thick base plate on the left end. The detailing on the right side shows the specimen is a tube with an 18-inch outside diameter, six 2.25-inch diameter holes for anchors rods on a 24-inch bolt circle, and a total baseplate diameter of 30 inches. The detailing on the left side has an unspecified diameter for the tube, but the base plate is 20 inches square, with four 1-inch diameter holes on a 23-inch diameter bolt circle. Generically, both baseplates are shown to be fillet welded to the tube with no specific dimensions. The middle part of the illustration shows the shop drawing received from fabricator 1. It deviates from the construction drawing in the following ways: the tube is tapered, having a 16-inch diameter on the left side, a 3/8-inch fillet weld on the left side, and a 7/16- by ½-inch fillet weld on the right side. The bottom part of the illustration shows the shop drawing received from fabricator 2. It deviates from the construction drawing in the following ways: the tube is tapered, having a 16-inch diameter on the left side, is a 16-side tube with 4-inch inner bend radii, a 5/16-inch fillet weld on the left side, and a 9/16- by 5/16- inch fillet weld on the right side with concave contour.
Figure 1. llustration. Detailing of specimens from each fabricator.


Federal Highway Administration | 1200 New Jersey Avenue, SE | Washington, DC 20590 | 202-366-4000
Turner-Fairbank Highway Research Center | 6300 Georgetown Pike | McLean, VA | 22101