Vacuum Ramp: OP vs SP 1

[controlnovice]

Just about every where I worked (on batch reactors) where there is a vacuum to be pulled on the vessel, it is usually done by slowly pulling vacuum so the product doesn't foam or be brought over into the vacuum system.

More often than not, this is done by ramping the output of the vacuum control valve, not ramping the SP of the vacuum controller.

Why would this be?

______________________________________________________________________________
This is normally the space where people post something insightful.

 

[Compositepro]

I am now quite sure I understand your question as it depends on the details of the vacuum system and the control system.

In any case, the effect of vacuum on the process is very non-linear. Vacuum will have very little effect until you get close the the partial pressure of the volatile components, which depends on vapor pressure, concentration, and temperature. In a typical batch vacuum process full vacuum flow rate is desired to get down to the pressure where the material begins to foam. First air bubbles begin to expand in a linear relationship with pressure. As you approach the partial pressure of volatiles, the air bubbles will expand at at exponential rate as pressure decreases. Theoretically, an air bubble will expand to infinite volume when the vessel pressure reaches the volatile partial pressure, limited only by the amount of volatiles present and the rate of mass and heat transfer causing the volatile to vaporize.

At this point the vacuum flow rate must be greatly restricted to prevent foam from expanding into the vacuum system. The foam must be given time to collapse due to bubbles popping. The amount of time depends on the viscosity and elasticity of the material, and type of stirring. Given enough headspace in the vessel, the foam can collapse almost instantly due to the very high expansion of the foam. But the problem is the the vessel may only be 5-10% full in order to have this headspace. In most case of degassing a mixture in a vacuum chamber (no stirring) the container should no more than 30% full, in order to allow enough expansion so that the foam collapses in a reasonable time.

After air bubbles have been removed the mixture will boil. At this point full vacuum flow can again be used. The rate of boiling will usually be controlled by the rate of heat flow into the material (vaporization requires heat).

Many industrial batch processes do not get to really high vacuums (>25"Hg) and therefore never get to the full boiling range. This will remove the majority of air bubbles but some will remain and get smaller when vacuum is released. Stirring under atmosphere mixes air into liquids. Under partial vacuum you still mix air in. It is just a smaller amount.

 

[EdStainless]

Most vacuum sources (pumps) are not adjustable, They simply move what they are able to given the pressure ratio. So you need a control element in the system to provide the ramp function.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[georgeverghese]

When a control loop is brought into operation under load, the PID controls tend to over react to a large delta between SP and OP, and push the final control element to a maximum possible position. Hence it is preferable not to bring a control loop onstream by putting it on AUTO, but by putting the controller on MANual and ramping the control valve slowly. Shift the controller into AUTO only when you see that the delta between SP and OP is small and the OP is not fluctuating much.

 

[controlnovice]

to a large delta between SP and OP

I think you mean delta between SP and PV....

My question may not have been very clear.

I understand why one would want to slowly bring a vessel to vacuum (don't create foam or bubbles that can be carryed over into the vacuum system).

I understand why there is a control valve/element (the vacuum pump is not adjustable or on a VFD).

I understand why a control valve would be put in MAN first, until the PV is close to SP, then switched to AUTO (to limit the bump in control valve output).

My question is:

In most of the processes I've seen when pulling vacuum on a vessel....even an empty vessel, what is the reason that the Output is Ramped (in MAN mode), and not the Setpoint (in AUTO mode)?
- Is it because it is a difficult loop to tune, so they just give up and control the position of the valve based on operator experience over the years?
- Is it because of the non-linearity of vacuum, so ramping the OP helps stabilize it?
- Is it because the valve characteristics: Equal Percentage vs Linear vs Quick Opening, etc...
- Is it because that's how the operator did it before the control system was put in place, so when they designed the system, they just kept the same operating philosophy?

______________________________________________________________________________
This is normally the space where people post something insightful.

 

[georgeverghese]

Yes, I meant PV, not OP.

I've seen both cases programmed into the startup sequence, but more commonly the ramp up on controller SP whilst the controller is on AUTO. I can see alternatives where a bit of both strategies could be applied also.

Think one of more of the reasons you've stated could be valid in most cases, particularly reason 1

 

[Compositepro]

Both methods can work. Both are methods are actually just different methods of manual control by the operator. The operator is required to provide the visual feedback of the amount of foaming and then take a control action.

Most vacuum control loops do not have sufficient accuracy or speed of response to control foaming. In my experience, a separate manual valve is used to shut the vacuum flow completely when foaming starts. Then this valve is cracked open to expand the foam but not allow it to overflow the vessel. When the foam has broken the valve is fully opened to allow the controller to take over. The manual valve is located on the vessel where the contents can be observed while regulating the vacuum.

Again, foaming is a very non-linear process. Think about how you open a bottle of soda that has been shaken versus not shaken.