wetted area for column trays

[koshyeng]

Sir,

How do you calculate distillation column wetted perimeter for reflief valve sizing?

Thank You

 

[Jefka]

I assume you're refering to a fire-vapor case?

If so, I generally assume the normal liquid level at the bottom of the column added to the with the total liquid holdup in all the trays. Multiply that combined level by the circumference of the column.

Wetted Area = 2*pi*r*(N*l + L) + [surface area of bottom head of vessel]

pi = 3.14159...
r = radius of column
N = # of trays
l = liquid holdup per tray (I've used ~3" in the past)
L = Highest opertional liquid level in bottom of column.

Anyone out there handle this differently?

 

[whammett]

I've used a formula similar to that, except with our bubble-cap trays, I would subtract the area that the actual caps take up.
something like Wetted area=2*pi*r*(N*l+L)-(average area of bubble caps per tray)+surface area of bottoms liquid.

This is more accurate, but it's better to overcompensate when it comes to safety devices. In which case, your formula is more than reasonable.

"Scientists dream about doing great things. Engineers do them." -James Michener

 

[Jefka]

I do not believe accounting for bubble caps is more accurate.

The formula in my original post is based off the surface area for a cylinder and should only be the wetted *sides* of the column. The trays themselves are not exposed to the fire, but the walls of the vessel are exposed. We're not dealing with cross-sectional areas so simply subtracting the bubble cap areas isn't mathematically valid.

 

[Armen75]

Also, as per API, you can stop counting trays at 25ft of vessel elevation.

 

[CMA010]

Unless the liquid on the trays flows/dumps to the column sump. That's why many designers add the total liquid inventory of the trays to the column sump. For the column sump i would use the highest nozzle of the level measurement. Then add the liquid inventory of the trays.

I think whammett is referring to the volume of the bubble caps.

 

[Jefka]

Again, the equation for sizing a relief valve for a fire vapor case is based on wetted AREA. We're not talking VOLUME. The API equation is not based off volume but area.

Yes you add the wetted area of the sump to the wetted area of the trays. But this area is the outside area of the fluid along the vessel wall. Therefore bubble caps volumes/areas/whatever have no place in the equation.

"Column sump" level is in my equation above as variable "L", the tray level are "N*l". Note they're being added together and multiplied by the circumference of the column.

 

[CMA010]

You do not seem to grasp the concept, but we a similar approach, volume is liquid height, and therefore is wetted area. Subtracting the volume of the bubble caps because the bubble caps do not contain liquid results in a smaller volume increase of the column sump and subsequently in a smaller wetted area (note that often trays are not complete circles, so further volume could subtracted). As stated by whammett this is less conservative approach.

As stated earlier, it is usually assumed that the liquid on the trays ends up in the column sump (see API STD 521, table 5) because the fire scenario is evaluated assuming that the column is blocked in and all mass and heat flow to and from the column has stopped. One could add the liquid height on the trays (all of them) to the level of the column sump but one could also calculate the volume of the liquid on the trays and relate that to column sump liquid height.

Note that in your equation you do not use liquid hold-up (volume in my interpretation) of the trays but liquid height (weir height?) on the trays.

 

[whammett]

Thanks CMA010, that is what I was getting at. After having subtracted the volume of bubble caps, the wetted area actually decreases. More accurate- yes. More conservative- no. When dealing with sizing PSV's, its always better to go conservative. However, some companies may not agree, epecially when it comes to large price differences.

"Scientists dream about doing great things. Engineers do them." -James Michener

 

[Jefka]

Are you saying you add the volume of the combined liquid volume of the trays (subracting the bubble caps) to the liquid volume of the sump and take THAT wetted area? That's not what whammett's equation above implies.

Either way, I would never advise someone to take the bubble caps into account. This "accuracy" adds nothing to the safety of a device that already has some rather arbitrary safety factors embedded in its design equations.

 

[whammett]

Jefka, my formula DID imply that. It uses your previous equation of (N*l + L)+bottoms liquid surface area except i subtract the volume taken up by bubble caps from the total combined liquid volume. Which is the same as saying liquid volume of trays+liquid volume of sump- volume of bubble caps.

"Scientists dream about doing great things. Engineers do them." -James Michener