September 6, 2008
Advanced Search

Login:

Password:

Forgot your password?
Register now

Home / Issue Archive / 2006 / September #9 / Estimating Cyclic Strength of ESP Intersection Connections

№ 9 (September 2006)

Estimating Cyclic Strength of ESP Intersection Connections

Oil production equipment unification and interchangeability of assemblies and parts of various manufacturers are urgent matters for repair services, oil companies, and equipment producers.

By A. Komarov, Yu. Uliyanov

Design of duplicate parts manufactured in accordance with oil companies’ specifications is quite diversified, which complicates inventory stock accounting, storage, and shipping. This variety results in issuing of numerous design documentations, continuous realignment of equipment, and fabrication of additional processing and measuring instruments. Unification should be achieved by selecting the most preferable equipment option in terms of enhanced reliability and constructability.

Thus, computations have been made to analyze the cyclic strength of bolted flange connection bases of four different designs.

This analysis has been made using a finite element method and design standards established for nuclear power engineering. It has been assumed that the connection bears an internal pressure of 42 MPa that corresponds to 3,000 m setting depth, and a 1,400 N/m3 density of formation fluid with the pressure cycle changing from zero to the maximum value at a temperature of 100 C. Equipment submersible assemblies weight and case bending have been ignored while setting and at the setting depth.

Fig. 1 shows that the maximum tensile stresses are within fillet junction between the flange and the wall. These stresses are smaller in assemblies designed with eight bolts as compared to assemblies with six bolts, because in the former the flange is more equally pressed down to the counterpart.

Data on cyclic strength comparative analysis of the bases is given in Table 1.

It is worth mentioning that analysis has been conducted ignoring bending stresses, which affect cyclic strength parameters.

For checking computations, base specimens have been forwarded to IMASH Resurs LLP and tested at MTS-50, an electrohydraulic test machine by MTS Systems Corporation (USA).

Special high strength bolts manufactured by IMASH Resurs have been used to connect the head to the base. The bolts’ parameters are M12x1,25, 6gx35 10.9 029p (HC.073.02); steel grade is 30 HGSA. The bolts have been tightened by a wrench with a torque of 55 N•m. The loading pattern has been cyclic tension stress with a zero-to-tension stress cycle, a cycling frequency of 2 Hz, a maximum testing pressure of Pmax= 169,600 N, and minimum testing pressure of Pmin= 0…2,500 N.

While testing, crack initiation has been registered visually by means of a lens with sevenfold magnification.

Design features of specimens and test results are presented below (Fig. 2).

Fracture pattern is presented on Fig. 3.

Fig. 2 displaying the base drawings, shows that designs have varied in a number of tension bolts, flange thickness, shape of neck-to-flange junction, and diametric dimensions.

Testing has proved that reliability of bolted flange connection is subject to a base design. No destructions of high strength fasteners have been observed in connections with 6 or 8 bolts. Bolt destruction in specimen No. 1 (Fig. 3) has occurred only after initiation of fatigue cracking in the flange and its reaching the hole.

Test results have proved that destruction of any base occurs within the fillet junction. In some cases (specimens 1 and 3, Fig. 3) cracking may reach the hole. Specimens 3 and 4 have been registered to have minimum durability as compared to specimens 1 and 2 with maximum durability.

Based on experimental data of base cyclic strength, it can be assumed that connections with six bolts are more preferable than those with eight bolts providing different flange thickness and shape of neck-to-flange fillet junction. However, it is necessary to increase the number of tests in order to make more affirmative conclusions. On the basis of research study [1] that describes the results of similar testing, cyclic strength of duplicate connections was endured for 698,300 cycles, 997,200 cycles and 1,225,400 cycles at a less pressure of Pmax = 135,640 N. Thus, the results obtained in this study are to be considered as a probabilistic estimate.

At the same time, cyclic strength of bolted flange connections is about 2.5 times higher than that of the connections manufactured previously, the test results of which were presented in study [1].

Conclusions:

1. Reliability of a flange bolted connection tightened with high strength fasteners at a specified torque is limited by the base fatigue properties.

2. The base design applying connections with eight bolts (M12x1,25) does not result in enhanced cyclic strength.

3. BORETS Production Company applies the obtained results for selecting high strength bases.

4. Further testing is required with an increased number of life test specimens for analyzing fatigue resistance of bolted flange connections, including those made by various manufacturers.

The authors thank N.I. Smirnov, Candidate of Engineering Sciences, for assistance in the article preparation.

References:

1. N.I. Smirnov, N.N. Smirnov, Durability and Wear Resisting Properties of Pumps (Designs, Tests, Technologies). Data of All-Russia Engineering Conference on ESP (Electrical Submersible Pumping Systems) Manufacturing and Operation; Nov. 1-4, 2000, Almetjevsk.
Copyright © 2007 Eurasia Press, Inc. (USA). All rights reserved.
Copyright © 2007 Eurasia Press (www.eurasiapress.com)