Question Regarding PSV Outlet Piping

[DGrayPPD]

Just another general question. I am getting ready to take a field trip to do a walk down on some PSV lines for a flare study that my company is getting ready to conduct. In gathering the PSV isometrics from a few years ago, i found that a lot of the PSV discharge piping drops underground and travels a sufficient distance to the the flare knockout drum before coming back up out of the ground and tying into the drum.

My question is, are there times when it is okay for PSV discharge piping to drop underground and pop back out further down the line, creating a huge pocket? I have always been taught that PSV discharge piping should slope to the flare knockout drum and that pockets should be avoided.

I have attached a picture for reference. It is a very rough sketch of what the iso shows. Thanks for the tips.

DGrayPPD

 

[pennpiper]

This is very, very unusual.

A- Where is this Plant?
B- What kind of plant is this?
C- How long has this Plant been operating?
D- What is the commodity in this line?
E- Is the Commodity Liquid or Vapor?
F- What would the Operating Temperature be at release?
G- How many other PSV Lines are there?

Sometimes its possible to do all the right things and still get bad results

 

[DGrayPPD]

Pennpiper, I will do the best I can to answer all of your questions.

A - This plant is located in SE Texas.
B - It is a storage and distribution plant for various hydrocarbons (propane, butane, gasoline, etc.).
C - It has been operating since the 70s.
D & E- The commodities in the storage tanks are the various hydrocarbons and are liquid under pressure but at the PSV interface will start to flash to vapor and will continue flashing in the outlet piping until it's all vapor.
F - I am not positive about this one.
G - There are multiple PSV lines that, for the most part, all drop underground and tie into the underground piping at various points.

I asked one of the senior process engineers familiar with the site and he said it is an outstanding issue that has been going on for a long time. Apparently the client has just been too cheap to properly fix it.

 

[mk3223]

IMO, the PSV pressure set point needs to compensate the back pressures due to the long discharge pipe and vertical elevation changes. Depending on the service, the low pocket piping may present some issues, such as surge flow, corrosion, etc.

 

[pennpiper]

DGray,
You have a very interesting situation. Sorry but I can only ask more questions.
H- What are the length dimensions for each of these different segments?
I- What are the elevations for each of the horizontal runs?
J- Is Texas DOT involved in any of these segments(ieo any segments cross under any public roads)?

Regards,
pennpiper

Sometimes its possible to do all the right things and still get bad results

 

[DGrayPPD]

Pennpiper,

Question H & I are a few of the things I have to verify during my field trip. The copies of isometrics we received from the client only showed the piping, the above and below grade change, and a few simple notations. Dimensions and coordinates were not shown on the isometrics. I can't answer why someone would deliberately delete that information from the isometric.

As far as question J goes, no, I do not believe any of the underground segments cross any public roads.

 

[gelsi]

The general rule is to have a slopped line from outlet of PSV to header, but from practical point of view, a line must have no pocket when there are 2-Phase or when it can contain a gas with vapours which can condense.

I think that in the outline of a PSV that discharges vapours, there is no condensate if:
1) the fluid remains over its dew-point AND
2) the dew point is below minimal ambient temperature AND
3) the fluid that back-flows from header or KoD respects conditions 1 and 2
Normally these conditions do not occur.

 

[don1980]

This installation is certainly unusual, and inconsistent with the industry guidance published in API 520 Pt II. That guidance is aimed at preventing liquid accumulation in the tailpipe. It calls for tailpipes to be installed such that they continuously slope downward to the collection header, or installed with a low-point drain valve for preventing liquid accumulation.

Obviously, compliance with this API 520 Pt II guidance isn't possible in your case (you can't install a low point drain on a line that goes underground). The safety risk is that a pocket of accumulated liquid could be rapidly accelerated down the tailpipe. This can potentially cause a mechanical failure (loss of containment). Excessive PRV backpressure is also an issue, but it's a self-correcting one once the liquid is blown from the tailpipe.

It sounds like this facility has a lot of these underground tailpipes. That means it's probably not realistic or practical to modify all these lines. If so, the owner should assess the risk of liquid accumulation in each line. For those that do have a risk of liquid accumulation, I would implement a procedure for periodically blowing the line. That could be done with N2, or by using fuel gas if using N2 poses a risk of extinguishing the flare.

 

[georgeverghese]

With this configuration, it is not only a risk to the people and assets in this plant, but there could be a real possibility that the fire event, if it not contained quickly, could escalate and spread to neighbouring facilities also. Has this client informed his neighbours of these risks?