When should we be using a bypass line for a valve? Is there any rule for this?
Answers
|
04/02/2011
|
A:
|
keith bowers, B and B Consulting, kebowers47@gmail.com
Mr Ragsdale's answer is a good one. In general, use a 'valve bypass' ONLY when necessary to a) safely start-up a unit; b) safely shut down a unit; c) enable operations to continue-safely- while the by-passed valve is repaired
There is no 'rule' or 'recommendation' concerning when to install piping and valves enabling another valve to be by-passed. The RULE is to minimize points of leaks to the environment and minimize mechanical complexity consistent with maintaining desired operability. One should argue WHY is this valve needed? What happens if we delete this valve/component/pipe/etc. from the plant?
Only when a comprehensive HAZOP has been conducted for the system can one assess the need/desirability of a valve by-pass.
|
|
04/02/2011
|
A:
|
Ralph Ragsdale, Ragsdale Refining Courses, ralph.ragsdale@att.net
If you mean without blocks and bypass valves, I do not remember ever seeing a gate or globe valve with a bypass. Several valves would be required to provide removal during operation, such as a control valve manifold or a PSV manifold. But a single gate or globe, no. Only FCC slide valves and other valves known to wear out during the run would have a bypass, but they would have blocks and bypass valves. I suppose you could have a situation where the gate valve would be closed during normal operation, and a globe valve in parallel would be used for control, but that globe valve should be a control valve. For example, the bottoms flow from the last column in a cyclohexane unit was only a few BPD. We had a control valve manifold with a 2 inch bypass for startup and line-out, and the control valve was so small, it was actually a laboratory pilot plant type, and was used for lined-out continuous flow.
|