API RP521 5th Ed - 4.3.2

[marvalar]

Hi all,

I just came across section 4.3.2 - Closed outlet on vessels, of API RP521 - 5th edition.

I'm a little bit confused.

First it is stated: "If the pressure resulting from the failure of administrative controls can exceed the corrected hydrotest pressure, reliance on administrative controls as the sole means to prevent overpressure might not be appropriate."

Then it has an example where there's a sample calculation of the corrected hydrotest pressure.

But after the example it is stated that reliance on administrative controls might not be appropriate if the gauge pressure caused by closure of the outlet valve exceed 672 kPa, which is the uncorrected pressure.

What is the correct statement?

Thank you

 

[marvalar]

Help...

 

[don1980]

The next sentence following the one you cite, explains this apparent contradiction..."This assumes the overpressure occurs while the vessel is at the design temperature." Therefore the corrected and uncorrected values are the same in this particular example problem.

The document is simply saying that you need to be cautious in relying on administrative controls to ensure that intervening valves remain open. It suggest that you not rely on admin controls if a failure of those controls could cause the risks to be too high, and it recommends using the corrected hydrotest pressure as the threshold for "too high".

 

[marvalar]

Hi Don and thank you for your reply.

The corrected hydrotest gauge pressure is always at ambient temperature. This is independent of the design temperature.

The uncorrected hydrotest gauge pressure will always be lower than the corrected one, except if the allowable stress at the design temperature is similar to the allowable stress at ambient temperature, which will make them equal. This, of course, will always be the case if the design temperature is close enough to ambient temperature.

For the cited example, it is stated:
Corrected hydrotest pressure: 713 kPag
Uncorrected hydrotest pressure: 672 kPag

I agree with you about the document message. It applies the same principle that we use for the 10/13 rule for heat exchangers. But the 10/13 rule is based on the uncorrected hydrotest pressure.

If the equipment has a high design temperature (leading to a corrected hydrotest pressure different than the uncorrected one), than it makes no sense to assume that it will be at ambient temperature when the closed outlet occurs.

This way, I personally believe that the value to consider is the uncorrected hydrotest pressure.

 

[EGT01]

In my opinion, API did a less than adequate job covering this subject.

By ASME UG-99 the standard hydrostatic test pressure is...
"at least equal to 1.3 times the maximum allowable
working pressure (the maximum allowable working pressure may be assumed to be
the same as the design pressure when calculations are not made to determine
the maximum allowable working pressure) to be marked on the vessel multiplied
by the lowest ratio (for the materials of which the vessel
is constructed) of the stress value S for the test temperature
on the vessel to the stress value S for the design temperature
(see UG-21)."

Using API terms, this would translate to be...

Phydro = 1.3 * Pdesign * Stest/Sdesign

In one respect, the hydrostatic test pressure is already a corrected test pressure.

In any case, I believe API botched their definition for "corrected hydrotest pressure"...
hydrostatic test pressure multiplied by the ratio of stress value at design temperature to the stress value at test temperature
NOTE See 4.3.2.

I see the API definition for "corrected hydrotest pressure" as being represented as...

Phydro * Sdesign/Stest

or, in other words,

(1.3 * Pdesign * Stest/Sdesign) * Sdesign/Stest = 1.3 * Pdesign

This seems to me that the API definition for corrected hydrotest pressure gives the same result as what API references as the "uncorrected hydrotest pressure" in their example in section 4.3.2. As further confusion, API uses the term uncorrected hydrotest pressure without giving a definition for it.

What I think API really should have done is define "corrected hydrotest pressure" as...
hydrostatic test pressure multiplied by the ratio of stress value at relief temperature to the stress value at test temperature

Which would be represented as...

Phydro * Srelief/Stest

or, in other words,

(1.3 * Pdesign * Stest/Sdesign) * Srelief/Stest = 1.3 * Pdesign * Srelief/Sdesign

I believe the issue of the relief temperature having an impact on material strength is the issue they are trying to cover. In this case, I think you can say you are only concerned with applying a temperature correction factor when the relief temperature is above the design temperature. Unless you happen to have a material whose allowable stress value decreases with decreasing temperature.

 

[don1980]

Marvalar & EGT01,
Good points made by both of you. This existing API text is anything but clear. Hopefully this will be fixed in the next version. In the meantime, I agree with EGT01's final comments. It's only a concern when the relief temperature is above the design temperature.

 

[marvalar]

HI EGT01,

The only problem that I see is that, the API definition for "corrected" test pressure is:

Phydro * Stest/Sdesign

And not:
Phydro * Sdesign/Stest

So, to summarize. If we don't want to be in trouble, we should keep the operating pressure bellow 1.3xPdesign. This works as long as the relieving temperature is lower than design temperature.

But, than again this is not what they say in text, but it is what they show in the example.

 

[ChEMatt]

Forgive my ignorance, but shouldn't a PSV on a vessel be set at the design pressure of the vessel, and not the hydrotest pressure of the vessel? The PSV itself, depending on the case, allows 10% or 21% overpressure in the vessel, but the PSV is still set at the design pressure, always.

No?

The inference is that the operating pressure will always be below the design pressure; you should never exceed the design pressure under "normal" operating conditions.

I always refer to the design pressure as the MAWP of the vessel. Yes, it varies with temperature, but your PSV cannot take that into account.

Am I wrong? Did I read something wrong in this thread?

Onwards,

Matt

 

[EGT01]

A relief valve does not always have to be set at the design pressure but it must not be set any higher than the design pressure/MAWP for a single valve installation. So I wouldn't say you are wrong on that point but just not completely right.

As to the discussion in the thread, it is not about the set pressure of the relief device but about under what cirsumstances an overpressure case may be excluded from the sizing basis of the relief device. You should review the referenced sections in API Standard 521 for additional information.