Should I change the size of the vent pipe? 1

[nouanda]

Dear Experts,

I am working on a feasibility study, which requires me to check the compatibility of an installed PSV with new process data.
This valve is 4"x6" and has an orifice of 4" (12.57 sq.in, out of API standards).
The set pressure is 100 barg.
The discharge flange is connected to a 6" vent pipe, to send the gas 163m away (this is the length of the pipe).
At maximum flow, I find a head loss of 67bar through the vent pipe, so the flow is non-critical.
I find a required orifice of 10sq.in, which is quite close to the installed orifice.
If the flow were critical, the required orifice would be 8sq.in, and the margin would be more comfortable.

My question is:
Is it reasonable to enlarge the vent pipe (ie 8"), without changing the valve, using a 6"x8" reducer?
I think it is not a clean solution but it may reassure the client??

Thanks

Nouanda

 

[Latexman]

nouanda,

I think you left out some information. Is it a conventional, bellows, or pilot operated PSV? Have you looked into using a pilot operated PSV instead of changing out all that pipe? It may be less expensive if you can find one that will work.

Good luck,
Latexman

 

[lizking]

Latexman is correct. For conventional valve, you're allowed 10% of set pressure losses on the discharge (10 bar).

 

[nouanda]

right, I forgot to mention it's a pilot operated valve. for those interested, it is an Axelson, Type HF.

And something else I did not forget to mention, but that was badly expressed: the valve already exists, but not the vent pipe.
However, we might have to install one (for safety reason, conformity, etc...), and my first approach was to size it the same as the exhaust flange of the PSV.

Changing the valve is not an option (at least, changing the inlet pipe is not), that's why I was wondering about putting a reducer at the discharge flange, to decrease head loss in eventual future vent pipe.

I think the installed PSV will do the job anyway, since I used *VERY* conservative hypotheses.

Nouanda

PS: what is the opposite of reducer in English? In French, we almost always use reducer, even if its role is to enlarge the pipe...

 

[Latexman]

Opposite of reducer = expander, but everyone calls it a reducer - you just turn it around when you install it per the drawings.

You are on the right track. Dig into the product literature on your PSV and talk to their technical support people. There's a real good chance it'll work with a 6" vent pipe. While the flow in the vent pipe is not sonically choked, it probably is frictionally choked.

Good luck,
Latexman

 

[CJKruger]

nouanda

Assuming the calcs are correct, I think the 6" tail pipe does not break any rules.

However, my view is:

- The pressure drop through the 6" tail pipe is excessive. You can easily get into trouble if the piper puts in more fittings (elbows, etc). I would use a bigger pipe (with expander).

- Is your pressure drop based on rated flow, or required flow ?

- Feel free to post your flow rate (lb/h), mol weight, temperature and k (Cp/Vc) value and we can check if we agree with the pressure drop and absence of choke flow.

 

[nouanda]

right, let's calculate!

the safety valve is installed on an underground gas storage facility. There are 13 wells.

The required flow is 471000 kg/h, relief temp is 40°C (104 °F), compressibility z=0.8536, k = 1.6626
set pressure=100 barg (1450psig).
molar mass is 18.11 g/mol (40 lb/kmol).

upstream pressure is 102barg. I used the maximum flow the wells can provide, but at this flow, there is little chance that pressure will be at 102barg (it should be around 80 barg, but I want to be conservative).

this is a preliminary calc, so I assumed that all the securities failed, and all the control valves are fully opened. I calculated that the maximum flow through the control valves would be 975000 kg/h (if you want to check, Cv=2500, deltaP=2bar, upstream 102barg). However, since the wells can provide only 471000kg/h, I used this value.

The vent would be DN 150 (6"), with 11 90°-elbows.
I used rugosity of 0.06mm, and viscosity is 0.0122 cP.
At set pressure (100barg) sound speed is 490m/s, and I find that for 6"pipe, discharge speed is Mach 0.22.

I hope you would find more or less the same results as me ;-)

 

[nouanda]

BTW, I meant "we laways use the *word* reducer". Expander would be "élargissement", which is far less smooth to say that "reduction" (French for reducer)...

 

[Latexman]

No, the results I got say the exit at the end of the 163 m pipe is sonically choked (Mach = 1).

Is molecular weight 18 or 40?

Is the k = 1.66 the isentropic coefficient or the ideal gas heat capacity ratio?



Good luck,
Latexman

 

[JoeWong88]

You have pilot operated valve and high set pressure (100 barg)... Normally no problem with valve itself...

BUT...Something not very "right"...

67 bar for 163m ==> 411 bar/km !!

What is the rho v2 (density x velocity x velocity) and Mach No along the vent line ?

I am agree with Latexman that the Mach no at the pipe end is 1 (choked) and the pipe rhoxV2 more than 200,000... Exccessive noise & vibration...!!!

JoeWong
Chemical & Process Technology

 

[nouanda]

Thank you all for your guidance, and thanks for the time you spent checking my errors.

Well, I thought about it this week end, and I must say I am lost...
I was wrong when I calculated the speed, since I used the speed at the discharge of the PSV, not at the end of the vent.
At the end of the pipe, with P=1 bar abs, the speed is awesomely high! I didn't realize that earlier...

I am deeply confused. What can I do to make things right? Put a silencer? Change the size of the pipe?

 

[786392]

Dear
I assume both the action needed to thoroughly address the issue!

Best Regards
Qalander(Chem)

 

[nouanda]

sorry to flood, but I post while I'm checking my calcs...

If I want to calculate the pressure at the end of the pipe to have Mach 0,7 (par exemple, zum Beispiel, as an example...), I calculate the flow for u=Mach 0,7. I take my Mach No at the inlet conditions, but it shouldn't change so much, so Mach 0,7 = 0.7*489=342 m/s.

For 6" pipe, volumic flow would be 21000 m3/h. Knowing the normal flow (583300Nm3/h), I find the pressure should be 32.3 bar abs mini. I neglect the change in temperature and k for now.

Then the problem would be just at the opening of the pipe, where my gas will be released at 1 bar?

I am sorry if I seem noob, but it is the first time I have such a problem with a PSV, and I reckon I don't know how to think...

 

[Latexman]

nouanda,

If the velocity at the end of the pipe (which is just at the opening) is Mach 0.7, the static pressure at the end of the pipe = the static pressure at the opening = the pressure of the surroundings = 1 bar.

It is only at Mach 1 velocity and higher there can be any pressure discontinuities. Therefore at Mach 0.7, you cannot have 32.3 bar at the end of the line exhausting into a 1 bar atmosphere.

Good luck,
Latexman

 

[nouanda]

Thanks all

The idea was to put a restriction orifice at the outlet, so that in the pipe, I can have 32.2bar, and 1 bar outside.

 

[Latexman]

nouanda,

Are we still talking about a 6” line? The one that is already Mach 1 at the end, without an orifice! If so, you may want to re-think your last post because that would choke the flow further. Remember, sonic velocity cannot be exceeded in a pipe or orifice, so mass flow rate will be reduced until sonic velocity exists in the vena contracta. I would advise to not have an orifice in the outlet line, or inlet line, of a PSV.

Let’s go back to your original question. Yes, it is reasonable to use a 6” x 8” reducer immediately after the PSV outlet flange to have an 8” vent line. It’s done all the time. It is a good engineering practice to meet pressure drop, velocity, and noise constraints.

It is also good engineering practice for PSV calculations and documentation to undergo a comprehensive review by an engineer experienced in PSV design. I sincerely hope your company requires this quality work process, since you said you were a newbie.


Good luck,
Latexman

 

[nouanda]

The idea of an orifice comes from an experienced guy in my company, who I use as a referee for my calculations. I won't say the name, I don't want to discredit it (I mean, the company, not the guy...)

Even using a 8" pipe doesn't solve the problem completely, because I find it is still sonically choked. At low pressure, low density, thus high volumic flow, and it's my entire problem!

I can feel putting an orifice is not a right solution, even with my little experience, but it's all we can find at the moment.

Anyway, I know for sure that during the next phase of the project (I'm turning my brain all ways round for a feasibility), we will find a very lower required flow, and the client will decide not to put a vent...

Thank you Latexman for your advices, and all of you who read this post.

 

[786392]

It is astonishing as well as ironical to believe that
An orifice can improve the situation
Or rather worsen it.

In all fairness/logical rationale you should upsize the vent pipeline to 8" or even higher diameter.
However learned/experienced the guy might be
NO LOGICAL REASONING SEEMS RELEVANT
Since orifice insertion is
As if you may like to reduce the pipeline diameter to 4".
Hope you realize sincerely.


Best Regards
Qalander(Chem)

 

[Latexman]

nouanda,

With the scenario indicated, the flow will be sonic at the end of the pipe until the pipe diameter exceeds about 26" or close to it. But, having sonic flow at the end is okay Code-wise, you have to meet the back-pressure requirements of the PSV and the flow requirement of the scenario. 8" may do it, but it is borderline. 10" will probably work, but you need to confirm that with the PSV documentation or vendor.

Good luck,
Latexman

 

[JoeWong88]

nouanda,
I did not go in detail of calculation. I presumed what Latexman has indicated is correct (i am 200% confident). You situation is common...

Two approaches have been taken :
i) Common approach : Provide large vent line i.e.26" (as calculated by Latexman) to meet Mach no, Rho x V2, etc.
ii) Challenge approach : Provide small vent pipe to have sonic flow

Approach (i) is pretty straight forward. As long as noise (noise is lower - not sonic flow), high frequency vibration and dispersion problems are solved, then no other issues.

Approach (ii) is complicated. Some part downstream of safety valve is sonic flow. First shall ensure the limited flow is more than the required flow. Second is vibration problem. Ensure the pipe will not failed on high frequency vibration (acoustic induced vibration). Third, large noise level is expected on the vent line. Acoustic insulation may be considered. Next is the vent point. The noise level could be eccesive (since sonic flow), ensure it is high enough to meet the noise requirement. Of course next is the dispersion issue.












JoeWong
Chemical & Process Technology