8 replies to this topic

[streamt]

We are facing a problem of severe vibration in the line connecting CO2 absorber ( UOP Benfield process ) to the flash drum which feeds the regenerator.

 

Absorber operates at ~20 bar g whereas the flash drum operates at ~2 bar g. This line is expected to vibrate considering the amount of flashing/outgassing happening downstream of the valve.

 

Line size is 12". Valve is a Fisher Vee Ball level control valve.

 

Has anyone encountered such a problem ? How was it solved ?

 

What is the sizing criteria used for the valve ?

 

What is the sizing criteria used for the downstream line ?

 

Which type of valve is preferred ? 

[Bobby Strain]

Is this a new unit just started?

 

Bobby

[streamt]

No, the unit was started almost 5 years back. 

No one noticed it since this control valve is located on the top-most platform of a column which is almost 50 m tall.

It is located very close to the nozzle of the flash drum.

[PingPong]

[quote]..... control valve is located on the top-most platform of a column which is almost 50 m tall.[quote]It sounds like the control valve is now located far above the column liquid level from which it gets its feed. As a result the liquid will already flash before it reaches the control valve. Therefor the control valve is handling a mixed phase stream, causing the vibrations.

 

Solution is to relocate the control valve to an elevation well below the column liquid level that is feeding it.

[streamt]

Pressure at the control valve inlet is ~ 15 bar g and the phase is liquid.

Outgassing / Flashing happens across the control valve.

[PingPong]

Pressure at the control valve inlet is ~ 15 bar g and the phase is liquid.

The column bottoms liquid is saturated with gas at 20 barg, so the liquid will start flashing already as soon as the pressure drops below that number, which is the case in the part line above the column bottoms liquid level. So most of the vertical upward line to the control valve will be two phase flow, and the inlet of the control valve will be partly vapor.

[streamt]

This is a Benfield Absorber. K2CO3 + CO2 + H2O -----> 2KHCO3

Some amount of hydrocarbon gas is dissolved in the 20% carbonate solution which is removed by flashing in the flash tank.

Tried simulating using water and PD data since we don't have the electrolyte module for our simulation software. Amount of vapour at the control valve inlet is minimal. 

[Pilesar]

Sometimes line vibration problems can be resolved by adding well-designed pipe supports.

[ychong]

You can consider a more rigid control valve. seems that your control valve throttling down 13 bar which is quite high.

 

Another way is to construct more rigid support for the control valve.

 

You need to check what pressure the liquid starts flashing. (even though there is some pressure loss, flashing might not be too significant)