Summary This paper reviews the planning and implementation of a flow test of a gas reservoir, including the general goal of the test, equipment used, problems anticipated, and the steps to minimize or to avoid problems. The well successfully tested at rates of 22.5 MMcf/D and 2,670 B/D condensate with a 2,800-psi flowing tubing pressure through a 46/64-in. choke for 4 days. Introduction The test of Viosca Knoll Block 957 Well No. 1 Sidetrack No. 2 was Amoco Production Co.'s deepest test from a floating rig. Viosca Knoll 957 is 115 miles southeast of New Orleans in 3,500 ft of water. The test, at a record water depth for the Gulf of Mexico, also set a world water-depth record for testing a gas reservoir. Safety to crew and the environment were top priorities during the planning. A team consisting of drilling, completion, reservoir, and facilities engineers and a foreman were assigned to plan and implement the test. Early planning involved field, service company, and engineering groups. Every effort was made to identify potential problems and to design the system to handle them. The goals of the test were to determine reservoir properties and reservoir limits. Several significant challenges were involved in the well test. The reservoir was gas with a potentially significant condensate yield. The ability to dispose of the large volumes of produced fluids safely without polluting was critical to maintaining uninterrupted flow. Potential shut-in surface pressure was 6,500 psi. Seafloor temperature in 3,500 ft of water was 39F. Preventing gas hydrates was a major problem. The inhibition was Preventing gas hydrates was a major problem. The inhibition was done by chemical injection into the flow stream before the gas and condensate reached ambient seafloor temperature. The ability to acidize was necessary. Optimum reservoir evaluation required varying the lengths and rates of the flow periods and shut-in times. During startup, hydrate inhibition was critical and a large volume of fluids in the surface tanks, both the tubing volume and produced fluids, required significant handling and proper disposal. produced fluids, required significant handling and proper disposal. The well tested was the third sidetrack from a wellbore spudded in Jan. 1989 by the drillship Discoverer 534. Initial plans were to test the second sidetrack from the drillship in the spring, but a third sidetrack was added to the drilling program and the drilling was prolonged into the summer. Rather than begin testing operations in July, a decision was made to move the drillship off location and to bring back the ship or another rig after hurricane season. The time required to secure the ship for a hurricane when in the testing mode would require beginning to secure when the potential storm was at least 750 miles away (mid-Cuba). This was not considered practical; therefore, Penrod 78 was moved to location in late October after Hurricane Penrod 78 was moved to location in late October after Hurricane Jerry.