Reactor Pressure Control Design 1

[chemegirl]

What is the best method of controlling back pressure on a reactor that contains some absorbed gas in the liquid?

 

[Latexman]

More details please. Is set pressure fixed? What is the value? Is set pressure variable? What is the range?

Good luck,
Latexman

Technically, the glass is always full - 1/2 air and 1/2 water.

 

[chemegirl]

We are trying to maintain pressure at around 2000 psig with little variation due to affects on the process.

 

[StoneCold]

chemegirl
One way of holding back pressure on a reactor is a simple back pressure regulator. That works if your outflows are only gas and you are not entraining any solids or liquid slugs.
Another way to do it is a control valve and a pressure transducer. The advantage of the control valve is that you can handle more solids or liquids entrainment.
If you have liquid and gas leaving your reactor you may want to set it up so that liquid flows through one control valve and the gas flows through another control valve.
That would make pressure control and sizing of the control valves much easier.

If this is not what you had in mind maybe you could add more details and get better help.

Regards
StoneCold

 

[chemegirl]

Currently we separate the gas and the liquid in a separator the pressure control valve is located on the vent gas. We are having issues with pressure swings that upset the process and cause big issues. I was looking at a back pressure regulator on the liquid side (prior to separator) vs using the traditional pressure control valve scheme. Any recommendations on back pressure regulators? The gas in the liquid is very small.

 

[Latexman]

Are these pressure swings new, or have they always been with you?

Good luck,
Latexman

Technically, the glass is always full - 1/2 air and 1/2 water.

 

[moltenmetal]

If the reactor is liquid full, controlling pressure in it will be a bear irrespective of the dissolved liquid.

If the reactor has a gas headspace, which presumably would be the reason for desiring the pressure, you use a separator and control pressure on the gas and level via throttling liquid on the separator.

It's when you try to send both gas (separate phase) AND liquid out through the same control valve that you have control problems. Don't do it. Dissolved gas that comes out of solution in the trim of the liquid control valve is manageable via correct trim selection and sizing.

 

[chemegirl]

Pressure swings have always been there. Process tolerance to the swings has changed and for our recent improvements we can no longer tolerate the swings.

 

[Latexman]

Sounds like both gas and liquid saturated with gas flow out the same nozzle and pipe and goes to the separator. Is that right? If not, please explain. Two phase flow is almost always noisey and unstable from a flow and pressure stand point. A PFD or P&ID may help, if that is possible. It would answer many questions.

Good luck,
Latexman

Technically, the glass is always full - 1/2 air and 1/2 water.

 

[moltenmetal]

Sorry- meant "irrespective of the dissolved gas"

 

[chemegirl]

Yes, liquid with small amount of gas absorbed out of the reactor (same line) into a separator, pressure control valve on vent and level control valve on bottom of separator.

 

[sshep]

If the net vapor from reactor is small, disturbances like temp, level, feedrate, and variations in dissolved gas rate in the feed can be upsetting the controls.

Where does the vent go (flare, vapor recovery system, downstream process)?
Is the vapor flowrate measured?
Can an additional flow of gas (or inert) be used as a blanket?

best wishes,
sshep

 

[moltenmetal]

LPS to sshep for identifying, and solving, the root cause of your problem. Your problem isn't with the dissolved gas- it's with having too little gas to properly control pressure. A small amount of inert pad gas will help immensely with pressure control by giving the pressure control valve something to do when pressure goes high. Otherwise your pressure and level loops will fight one another.