API2000 vent capacity calculation 2

[andyecho]

Hi

I would welcome any insight into how to calculate the capacity of a manway used for emergency relief.

Situation - I was originally asked to calculate counterbalances for vents to be added to existing tanks - hinging and couterbalancing of the manway, (requested by UK HSE). They have now been asked to show some sort of calculation for suitable sizing.

Looking through API 2000 I can calulate the required capcity for outbreathing & fire but am unable to understand how to calculate capacity of my manway.
I assume use of equation 3, (multiplied by 0.5) but what specific heats are used ?, those at inlet to and outlet from manway ? and what value Z ?
vent presure is 7.5 mBar and I doubt the client can easily supply temperatures and molecular weights.
I asume if 'k' = 1 then SCFH = 278000P1.A

Any pointers appreciated, thanks

 

[sshep]

Please allow me to suggest that you may be better off purchasing a designed blowoff manway made by an established company such as Groth (

http://www.grothcorp.com,

ref Model 2000A). These are independently tested devices, with published flow capacity tables, and available to fit to standard manway sizes (16", 20", 24"). As fire sizing is a serious safety application, the minor cost savings from a "do it yourself design" cannot be a serious consideration. I am sure your customer (and their insurance company) would properly consider such a recommendation.

just a thought, SShep

 

[andyecho]

Too late, unfortunately. The client added the hatches and thought about the design later. So still after definitions

ak

 

[TrevorP]

It's not a straightforward calculation, as it depends on how far the hinge will operate. If it will clear right out of the way completely, then you can assume it will be act as an orifice plate, or you can calculate it as a short pipe section with entry and exit loss. If it doesn't clear right out of the way, then it depends on how open it can go. As a rule of thumb, when fully open, the cross sectional area of the gap around the lid in the vertical plane should be greater than the cross sectional area of the opening. If this is the case, then manufacturer data can probably be used to a reasonable degree of approximation. I'm not sure that this will necessary satisfy the HSE though. It probably depends on what size opening you have, and what size opening you would need based on manufacturer's certification.
If you can somehow retrofit certified lids then this would obviously be your best bet.

 

[andyecho]

Lid fully opens to allow full flow, and so may be designed as orifice plate but it would be nice to compare with API 2000 capacity equation 3 so question abot 'k' definition still there.

The end user is stuck with whats there for at least a year until next shutdown

ak

 

[JoeTank]

The flow capacity requirements are based on the vapor properties of hexane (or was it heptane?). I'm not a chemical process engineer, so you should probably verify this by reading all of the info in API-2000. Also, check out NFPA-30 as it has same technical stuff but more commentary.

Steve Braune
Tank Industry Consultants

http://www.tankindustry.com

 

[TrevorP]

API 2000 recommends you use actual physical properties , but does allow for use of the properties of pure n-hexane if actual properties are not known. The ratio of specific heats can usually be used in place of k. For n-hexane this is 1.08. I believe 1.0 is typically used as this gives a more conservative approach. With API 2000, by definition, you are talking low pressures, so you can reasonably assume that Z=1.

 

[andyecho]

I met end user on Friday and confirmed must go with existing manways at present.

Also noticed that equation 3 does not make sense

because k > 1 , P2^(2/k) always less than P2^(k+(1/k)) and hence SCFH value incalculable (root of -ve No.)

Is this a misprint? should it be (P2/P1) raised to power ?

The use of (P2/P1)^ gives flow rate in line with groth model 2400A when manway given coefficient of dischrge = 0.5

clarification much appreciated

ak

 

[TrevorP]

OK, I have an old version, so I hope I'm not leading you astray. Equation 3 in my version is for coefficieint of discharge method for specific design of 3 or more sizes.
In my version, it's not particularly clear, but it appears that the equation inside the square brackets is:
(P2/P1)^(2/k) - (P2/P1)^((k+1)/k)

Note that it looks like the second exponential is (k+1)/k, not k+(1/k) as in your post.

Since P2>P1, this is negative if k<1, positive if k>1. This is then multiplied by k/((k-1)MTZ), which then means that it is always positive, prior to taking the square root.

Note that this equation doesn't work if k=1!

 

[andyecho]

Many thanks

I suspected must be the case but did not think of (K+1)/k
It makes a large difference. So much so that at their low set pressure of 7.5mbar all the existing openings are too small

maybe they will have to investigate other options

ak

 

[aybee]

Protectoseal sell these and will send you a CD which will enable you to calculate the size..Aybee