Pressure reduction for 12 bar gasoline flow to low pressure vessel

[hammy1554]

If the pump discharge is 12 barg , line size 2"and if we connect 12 barg gasoline to 0.5 barg vessel(Big volume 700 M3)does it require pressure reduction a Pressure controller. or it is okay of discharge 12 barg to this low pressure vessel?

 

[LittleInch]

Not enough detail here.

Needs a system description and where you have vents etc. How long is your 2" line. Is this something new? Why 12 bar?

Just because your pump says its duty is 12 bar, doesn't mean it will be 12 bar. If your 2" line isn't very long your pump will go off the end of the curve and probably trip or overheat as it will flow a lot more, but at a lower pressure until it reaches equilibrium with the pressure drop in your 2" line.

If your venting is adequate for the flow and can't be blocked then there might not be an issue,but if there is any chance of getting a blocked or inadequate vent then this is a recipe for a disaster.

This all sounds very loose and potentially dangerous. You don't mess around with Gasoline - it can very easily kill a lot of people if it all goes wrong..

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[TiCl4]

Your question is rather vague and a bit worrying. What do you mean " it is okay of discharge 12 barg to this low pressure vessel?"

Centrifugal pump here? If so, 12 barg will likely not be the actual pump discharge pressure. You will need to use the pump curve and the piping design (isometrics and other data) to determine the actual pump discharge pressure and flow rate for the full-open line. Yes, it is normal for a pump with a high discharge pressure to empty into a low-pressure tank. However, this does present an over-pressure scenario which will have to be evaluated (pump on, tank outlet's blocked case, likely 1 vapor-relief case and 1-liquid phase relief case for tank partially filled scenario and tank full scenario) for the pressure relief design.

Give us the pump curve and any piping information you have (length, relative elevation, number of bends), and we may be able to see if your pump will run off the end of its curve.

You also have failed to provide the design point. What volumetric flow rate are you trying to achieve?

 

[danschwind]

Question is not clear.

Please include a P&Id, pump curve, pipe lengths and elevations.

Daniel
Rio de Janeiro - Brazil

 

[hammy1554]

thanks everyone for the response. actually situation is that we want to contact 2 inch line of gasoline having 12 bar pressure, to a vessel with big volume of 700 M3, vessel is properly vented and connected to flare.

@littleinch line is very long, say 500 m, vessel is properly vented. our concern was it can overpressure the tank or not? however tank has vent and PSV to flare.

@tiCL4 it is not about pump can deliver for length of 500 m but we are more looking into if we can have high pressure of vessel in this case. we have 300 pound pipe spec which is sufficient i guess, however cannot attach PID due to confidentiality reason. One thing i take good from you to consider vessel relief valve to be design at all these scenarios.

 

[LittleInch]

Your issue is that the tank is at our close to atmospheric pressure and the other end of your pipe at some higher pressure (12 bar in your case). If your pipe is only 500m long this will result is a v high velocity in your pipe which will exit into your tank like a fire hose.

This is not a good idea.

If there is no other control or restriction in flow in this 2" line then the end pressure will just be whatever the pressure is in the tank. Pressure drop will be even along the pipe.

Adequate venting for all possible flow rates is a requirement. If you don't get that right then yes you can overpressure the tank.

It is quite alarming that you mention nothing about flow or flow velocity. The flow will increase until the frictional losses in the pipe equal the pressure difference from one end to the other. If your pump can't supply the flow rate at the 12 bar then the pressure will fall until the frictional losses equal the pump pressure.

Does that make sense?

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[hammy1554]

Regarding your first point, we are connecting pipe to the vessel at bottom side under the liquid level. so it will not be exposed to the vapor space of the vessel. In this case we can avoid hammering impact on the vessel.

Flow is 5 M3/hr , we estimate pressure drop along the line is .5 bar only. Vessel overpressure is very less likely as it is connect to PSV and a 16" equalization line is available to act as pressure regulating. vessel is connect to column(operating 0.7 barg) through equalization line.

In this case my suggestion no need of pressure reducer to reduce the supply pressure, even restricted orifice is not required as we connecting it under liquid level.....

 

[LittleInch]

hammy,

You're either not understanding what I am saying or you're ignoring it. Not sure which.

However I'll give it another go.

Lets say at one end of your 500m long 2" pipe you have a large pump which may be servicing other users ( we don't know as you won't give us any details) where you have 12 bar pressure regardless of the flow rate in your 2" pipe.

At the other end you have a tank with a fixed pressure of not more 0.5 barg ( though if the connecting vessel is operating at 0.7 barg then so will your vessel which has a design pressure of 0.5 barg??) Anyway let's use 0.5 barg arrival pressure.

So, if this 2" line has no other pressure or flow control valves or other restrictions then you take the known and fixed data (pressure drop, line size, length and DETERMINE flow rate, not the other way around.

In your case a 500m long 2"ID pipe with 11.5 barg pressure drop will flow at about 25m3/hr with a velocity of 3.5 m/sec (I used a simple online calculator).

Your pump may or may not be able to handle that - I don't know. Equally your nozzle into the tank may be on the side, but if it's pointing upwards then at 3.5m/sec, if the liquid level is not high you could easily get a fountain inside.

To get only 5m3/hr you need some sort of device to take away 11 out of the 11.5 barg pressure drop. This could be a valve or an orifice plate, but if you just plug in a 2" line into your 12 bar system then you'll get a lot more flow than you thought.

But it begs the question - why are you feeding a 700m3 tank at 5 m3/hr??

Answers on a postcard please.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[hacksaw]

Suggest you check the API standards interms of safety. You risk blowing the plant.

 

[TiCl4]

Hammy1554,

You will NOT have a pressure drop of 0.5 bar. You WILL have a pressure drop of 11.5 bar if your source pressure stays at 12 bar. If your source of pressure is a centrifugal pump, you do not have a constant pressure source

LI is correct here - depending on the roughness of your pipe, you will flow somewhere in the neighborhood of 25 m3/hr through your proposed line (assuming no significant elevation change) if your source pressure is actually 12 bar. However, unless your pump curve encompasses a duty point of 25 m3/hr @ 12 bar, you will not have a 12 bar source and will not actually flow 25 m3/hr.

There is a reason everyone is asking for the pump curve and pipe routing information. You say your goal is 5 m3/sec, and that your pressure source is 12 bar. You need to answer the following questions for us to understand your piping/pumping proposal and provide feedback.

[ul]
[li]What is the source of the 12 bar pressure? If it is a centrifugal pump, provide the pump curve and inlet pressure.[/li]
[li]If the pressure source is a pump, please confirm this pump services this line ONLY.[/li]
[li]What is the total line length, elevation change, and rough number of bends of the piping?[/li]
[/ul]

 

[don1980]

Hammy1554, I assume that the flow to this tank is being controlled, and not just freely flowing without any flow control device. That is, I assume there's a control valve between the outlet of the pump and the tank, and that 12 barg is the pressure at the pump discharge, upstream of this control valve. If so, there's no relationship between the pump discharge pressure (12 barg) and the pressure at the terminal end of the pipe (at the tank). The pressure at the tank is entirely a function of the pressure in the tank plus the static head of the tank's liquid. The fact that the pump can put up 12 barg or 100 barg doesn't change the pressure at the end of the pipe.

 

[LittleInch]

The OP says "IF we connect. ...", implies this is an idea and not something currently working.

About 10% of the required information supplied.

Only good thing is that even at 25 m3/hr, it would take quite a while for the pressure in the tank to actually rise to something that would be dangerous even if blocked, if this is the only inlet.

But the lack of design knowledge is frightening.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[TiCl4]

The OP does not sound like a native English speaker. LI, it may be more of a language barrier issue than lack of design knowledge - we may be talking at cross purposes. I'll wait for the OP to clarify the question and provide additional information.

 

[hammy1554]

Guys sorry for the providing the limited information initially , as i was just looking for some answer related to high pressure in tank.

Just back to question by LI, i agree with you, may be increase in pressure will not be likely scenarios but high velocity could be the concerning to in this case.

We are going to use circulating pump for oil skimming, Pump is centrifugal, rated discharge is 50 m3/hr and pressure is 12-13 barg. Flow from pump passing through a coalescer(C) then there is level controller(LV) which is controlling level of Pump suction vessel(A). Actual purpose of the pump is to pump to gasoline unit. However we are making one tie in at d/s of level controller(LV) which going to large tank(V) which has design pressure of 2.5 bargs and there is PSV at 2.5 barg on this tank(V) and tank(V) pressure is controlled through 16" equalization line to a column (C2) which is operating at 0.7 barg , hence there is no issue of high pressure in tank (V) however i got point from LI that high velocity could be concern. Coalescer (C) operating pressure is around 11.5 bargs as per designer. it is protected with a PSV. Hence pump is also protected with PSV. but my only issue is not high velocity of the pump.

I am attaching one block diagram for you to understand. Here is answer for TiCL4 question.
What is the source of the 12 bar pressure? If it is a centrifugal pump, provide the pump curve and inlet pressure.(centrifugal pump, it will be feeding other unit through coalescer but we will use this new tie in discussion intermittently )
If the pressure source is a pump, please confirm this pump services this line ONLY.(No pump is feed to some other plant, pump feeding two places, incase of offspec we will close on side and send to waste drum (V)
What is the total line length, elevation change, and rough number of bends of the piping?(This tie in line length is 700 m approx. after LV. One engineer did the hydraulics and found pressure drop for this whole line is just 0.5 barg)

To LI Question, But it begs the question - why are you feeding a 700m3 tank at 5 m3/hr?? ( We are not feeding, there is existing waste oil drum, we are just making tie in to the (V) incase of offspec. actually coalescer capacity is 5 m3/hr hence we estimated same flow to tank (V)

 

[TiCl4]

Thank you for the information! So it looks like you are looking to install this as a secondary flow path in the case of off-spec material. Flow will be diverted from the line that is currently exiting the left side of your diagram to tank V via manual valves. I need a further point of clarification. It looks like you tying into an already existing line rather than directly into the vessel itself - is this the 16" equalization line you referenced earlier?

Based on the design point of the pump (13 barg at 50 m3/hr) I think you are correct in expecting a reasonably constant source pressure, especially given that you are simply diverting flow to vessel V rather than introducing additional flow - the duty point of the pump shouldn't change when going to vessel V.

In your proposed arrangement, you will be taking ~10.5 barg pressure drop across your level control valve at 5 m3/hr (11.5 barg source pressure - 0.5 barg from line losses - 0.5 barg vessel V pressure). You will need to confirm that this level control valve is appropriately sized to handle 5 m3/hr at 11 barg differential pressure. I say this for this reason - I don't know where you are sending the flow right now. If you are sending to a higher pressure user (higher than vessel V operating pressure), or if the line length/restrictions are much greater than your newly proposed line, your valve may be over-sized for this new tie-in and may not be able to control flow down to 5 m3/hr. Look at your Cv characteristics and confirm the valve will be in good control range at your desired operating point.

If the valve is fine for control, then I don't see any issues with your setup for normal operations. However, you WILL need to re-evaluate the PSV design for vessel V (and interconnected column through the equalization line) for the case of the level control valve failing open. You will want to ensure that the flowrate to vessel V does not exceed the current design capacity of the existing PSV on vessel V (or connected column). That may not be the worst-case scenario and may not change the PSV design, but you need to check and confirm that it will not alter the over-pressure case.

There may be other safety concerns with failures in the coalescer that may affect vessel V in other ways, but that is something that should be addressed in the project hazard review process by those knowledgeable in the process.

 

[LittleInch]

Hammy, Thanks for the information and the response can now be varied by the understanding of the system.

SO depending on what level there is in your pump suction vessel and presumably flow into that vs flow out, plus back pressure from whatever is downstream of the coalescer, the pressure at the tie in point could be anywhere from ~12 barg to < 2bar??

Until you know the max and normal operating pressure at the tie in you won't know your flow rate.

The engineer was correct in that 700m of 2" in gasoline flowing 5 m3/hr only needs about 0.5 barg.

The question he should have been asked was for a 700m 2" line with a pressure drop of X (your tie in pressure -0.7 barg) what is the flow rate / velocity?

But as noted above the max it is going to be is ~25m3/hr.

Now it comes to vessel V - what is the max liquid content versus total capacity?
If it's < say 0.7 then you've got quite a long time before you could overpressure the vessel in the event of blocked outlet.
If it's >0.95 or it's full that's a different story

You haven't really told us much about what vessel V actually is and how full it etc.

Hopefully you can go through all the possible scenario's in the change HAZOP.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.