Vacuum Breaking Safety Device for a Tower 1

[sshep]

My Friends,

We have a tower rated for 3barg and not rated for vacuum. The vacuum protection is set of 3 pressure switches which trip solinoid activated 6" HV. This HV floods the the tower with N2 coming from a high pressure supply (i.e. more than 4barg) which lifts the tower relief valve. This is the existing system.

Is there a vacuum breaking device with the reliability of a relief valve which can blanket the tower without overpressuring it- i.e. like a pilot operated regulator designed to serve a safety function. We would want something that would meet the probability of failure on demand criteria of a relief valve. Any advice is appreciated.

best wishes,
Sean

 

[Latexman]

Sean,

Must it be inert, like nitrogen? Or, can ambient air be used?

Good luck,
Latexman

 

[sshep]

Hey Latexman,

It pretty much has to be inert due to the temperature and flammability.

These type problems drive me crazy because if we just rated these towers for vacuum at the start there wouldn't be any issue at all. Instead there are project people who are incentivized to "save" on cost and pass the problems on to us.

best wishes,
Sean

 

[Latexman]

Can you pull up the process data during one of these vacuum trips and from the P vs. time plot and knowing the volume of the system estimate the orifice size needed to slow the flow down so the PSV doesn't open. You'd have to use this new flow on the P vs. time plot before the N2 vlave opened to convince yourself the column would not go under too much vacuum. On the surface, it seems doable to me.

Hopefully, the data archive has the data you need.

Good luck,
Latexman

 

[Latexman]

Does one of those pressure switches close the HV when vacuum has been reduced and pressure has not gotten too high, say 1 barg? Is it faulty? I'm struggling to understand what the problem is - too high a flow, faulty pressure switches, bad pressure switch settings, a combination of the aforementioned issues.

Good luck,
Latexman

 

[sshep]

Latexman,

Thanks for the help. You get a star. The main problem is that the vacuum breaking N2 supply exceeds the column design pressure. I hear your orifice idea to slow the incoming flow to a reasonable rate. Although the relief valve is sufficiently sized to handle the incoming flow, we don't want it lifting everytime there is a steam trip (for example).

I was hoping to get away from instrumenting out the problem by use of a device which could regulate itself once the pressure has come to an appropriate level. There are several options to keep the relief valve from lifting either by modifying the existing system; or installing another system in parallel (which won't have the same reliability standards) but will kick in first and stop the flow before the relief valve lifts.

Thanks again.

best wishes,
Sean

 

[Latexman]

"A device which could regulate itself once the pressure has come to an appropriate level" - put a self-contained pressure regulator on the nitrogen line set for, say, 1 or 2 barg. When the pressure rises to that point it closes. Or, put a pilot operated PSV in the nitrogen line with discharge going to column and the sensor on the column. Again, the set pressure can be selected to do what you want.

Good luck,
Latexman

 

[SeanB]

Is the tower designed for any vaccum at all? I would be surprised if it wasn't. Typically it would be designed for at least 1/2 FV for maintenance (steam out) purposes.

 

[sshep]

Thanks SeanB,
Even per our own company standards what you say should be true (steam out -> FV/2), but this depends on someone in the design group recognizing that a tower might need to be steamed out and taking the trouble to note this. It appears our designers don't think about these things. As it is there is no record that the tower is designed for any vacuum. Some mechanical engineers are doing the hoop stress etc to figure out what kind of rerating could be done.
best wishes,
Sean