Rupture Disk Pressure/Temperature Setting 5

[ccfromtx]

The inspector in our plant refuses to allow rupture disks that have a pressure rating at ambient that is greater than MAWP. I say that the ASME only requires that the rupture disk pressure rating is less than MAWP at the temperature it is expected to rupture. Our vendor always stamps the disk with the specified burst pressure and temperature and the burst pressure at ambient conditions.

Example:
A vessel has an MAWP of 200 PSIG at 300 Deg F. The operating conditions are 150 PSIG at 250 Deg F. The rupture disk is specified to burst at 200 PSIG at 250 Deg F. When the disk is received from the vendor the disk will be stamped 197 PSIG (due to manufacturing tolerance) at 250 Deg F and at 215 PSIG at 72 Deg F. Our inspector will reject this disk because of the 215 PSIG rating at ambient.

I think our inspector is wrong, can anybody give me supporting evidence? Also, can anybody tell me why the vendor stamps with both the specified and ambient conditions. I know the vendor has to test at ambient but why stamp the disk?

 

[TD2K]

Your inspector may be thinking if the system ever operated at a low temperature, then the rupture disk would open at a pressure greater than the MAWP of the system and thus not meet code. However, if you used this arguement, then you'd be restricted to a lower operating pressure at the actual operating temperature to avoid the disk failing prematurely.

To my mind, having the disk stamped at the desired relieving pressure at the expected temperature is no different than specifying a cold test pressure on a PSV (which will be higher than the equivalent 'hot' set pressure) in order to ensure when in operation at operating temperature, the PSV opens at the design set pressure.

UG-127, 3.c.3 seems to address what you are looking for and it states "The stamped bursting pressure of the rupture disk ... However, in no case shall the stamped bursting pressure of the rupture disk at the coincident temperature plus any pressure in the outlet piping exceed the maximum allowable working pressure of the vessel ...".

The biggest problem is assuring him that the coincident temperature at rated set pressure is realistic. Often, the location of a rupture disk can be such that it can see much lower temperatures than the vessel or if it is steam traced, estimating the temperature is a factor of the how much steam tracing is installed (eg # number of loops) AND the ambient temperature at the time of relief (which I just finished working through for one case and is not fun). In that case, if the disk temperature is lower than specified, the disk would fail at a higher than intended pressure.

 

[Exploguard]

Specifying a rupture disc can be more complicated than one might expect, in my opinion. For one, dependent upon the design/style of disc and it's material of construction, its relieving pressure at two different temperatures over a broad range can be substantial. In some cases, the change in strength may be far greater than the change in vessel strength given the same range. So, in practice, for such a case barring corrosion resistance considerations, inconel or nickel would be much better choices for disc construction than stainless steel for example. There can be cases where tightening up the manufacturing range tolerance may be wise for such situations as well.

 

[RPG]

I have to agree with your inspector, I have had vessels fail a state fire marshal inspection for just that reason. Remember that discs have a tolerance of +/-5% over 40 psig, so a 200 psig disc can be stamped as high as 210 psig at ambient and pass the inspection for a MAWP vessel of 200 psig.

 

[CHD01]

The inspector is indeed incorrect. The disk is rated at the coincident disk temperature at the relieving pressure for the overpressure event. Only if a single rupture disk is provided for more than one overpressure event; and if the relieving pressure for one of the events is at a coincident atmospheric temperature is the inpsectots observation correct. TD2k has given you the required reference in the ASME COde Section VIII Division I. To prove your inspector wrong, you could requeset a Code Clarification. The more you learn, the less you are certain of.

 

[RPG]

The purpose of a rupture disc or relief valve on a pressure vessel is as a last line of defense to protect the vessel from a pressure greater than the MAWP. Although I agree that a vessel may never see an elevated pressure unless the temperature is also elevated, the potential still would exist from possibly an unknown or not thought of scenario. That is why it must be protected even at ambient temperatures, from an over pressure event.

 

[CHD01]

RPG: Its still not required by the Code. If personal preference is to apply the policy you state for the plant so be it; but its NOT REQUIRED BY CODE. Actually I don't have a big problem with what you state, since so many of our engineers miss the boat when they size a relief system; and because a stamped burst pressure greater than MAWP does confuse operating personnel - who we often ask to audit the process. But we need to stay clear on what is plant preference and what is stated in the Code. Normally I can avoid this kind of problem during the design of a plant by making sure the margin between max OPERATING and DESIGN pressures are adequate. But after a plant is built and ages; trying to up capacity can get you into this type of problem. Then the inspector and the plant need to be able to communicate and work together; there is no excuse for undefined overpressure scenarios.
The more you learn, the less you are certain of.

 

[CHD01]

RPG: On rupture disks - For your example, a rupture disk cannot legally be stamped for a burst pressure above 200 psig at its coincident burst temperature. Its actual burst pressure is allowed to exceed the stamped burst pressure by a maximum of 5% - but the stamped burst pressure can NEVER exceed the MAWP (except for the stamped equivalent burst pressure at ambient temperature if you have an elevated relief temperature, which we discussed earlier).

Remember there are two kinds of tolerances built into the design of a rupture disk: One is the Manufacturing Range (varies with type of disk, cost, material and burst rating) AND the second is the burst TOLERANCE (+/-5%).

If the MAWP is 200 psig for your example and the Mfg Range is +6 psi and -4 psi; then the specified max STAMPED burst pressure should be = 200-6=194 psig and the min STAMPED burst = 194-4=190 psig. Thus you should order the disk to be STAMPED from 190 to 194 psig. The manufacturer may stamp the disk anywhere in this range (unless you request a tighter mfg. range - you can even get a 0% range sometimes but it costs a lot more). Once the disk is stamped, THEN the 5% code tolerance applies (if above 40 psig). So if the disk is stamped for example at 192 psig (between 190 and 194) then the code guarantees actual disk rupture between 180.5 and 201.6 psig. The other check is that the burst and oeprating margin be adequate; a tension disk requires a 70 % margin; while a reverse buckling type disk cna withstand up to a 90% margin. In your example, if you used a tension type disk, the maximum operating pressure must be limited to 0.7 x min burst pressure = .7x190= 133 psig (I prefer including the 5% code tolerance which is equal to 0.7x190x.95=126.35 psig).

Disk burst pressure is not determined like relief valve set points are (many people make this mistake).

Apologize if you alrady knew all this; but it may help others. The more you learn, the less you are certain of.

 

[CHD01]

RPG: On rupture disks - For your example, a rupture disk cannot legally be stamped for a burst pressure above 200 psig at its coincident burst temperature. Its actual burst pressure is allowed to exceed the stamped burst pressure by a maximum of 5% - but the stamped burst pressure can NEVER exceed the MAWP (except for the stamped equivalent burst pressure at ambient temperature if you have an elevated relief temperature, which we discussed earlier).

Remember there are two kinds of tolerances built into the design of a rupture disk: One is the Manufacturing Range (varies with type of disk, cost, material and burst rating) AND the second is the burst TOLERANCE (+/-5%).

If the MAWP is 200 psig for your example and the Mfg Range is +6 psi and -4 psi; then the specified max STAMPED burst pressure should be = 200-6=194 psig and the min STAMPED burst = 194-4=190 psig. Thus you should order the disk to be STAMPED from 190 to 194 psig. The manufacturer may stamp the disk anywhere in this range (unless you request a tighter mfg. range - you can even get a 0% range sometimes but it costs a lot more). Once the disk is stamped, THEN the 5% code tolerance applies (if above 40 psig). So if the disk is stamped for example at 192 psig (between 190 and 194) then the code guarantees actual disk rupture between 180.5 and 201.6 psig. The other check is that the burst and oeprating margin be adequate; a tension disk requires a 70 % margin; while a reverse buckling type disk cna withstand up to a 90% margin. In your example, if you used a tension type disk, the maximum operating pressure must be limited to 0.7 x min burst pressure = .7x190= 133 psig (I prefer including the 5% code tolerance which is equal to 0.7x190x.95=126.35 psig).

Disk burst pressure is not determined like relief valve set points are (many people make this mistake).

Apologize if you already knew all this; but it may help others. The more you learn, the less you are certain of.

 

[RPG]

You are correct on the stamped burst pressure of the disc. The other issue, surrounding the pressure rating at ambient temperature, the code does not allow an exception just because you are operating at higher temperatures. Otherwise one might experience an overpressure event, exceeding the MAWP, due to fire exposure to the bottom of the vessel, where the rupture disc, located above, is at ambient temperatures. I have referenced the code and interpretations already and found no exceptions. If it is needed to have a rupture disc that is that close to the MAWP, I would suggest that they use a composite material disc which is less temperature dependant and has a tighter tolerance.

 

[CHD01]

RPG:

If the disk is at top of tank and fire is beneath tank, the designer should have the disk rated to burst at the actual disk temperature (ambient in your example). The disk actual burst temperature must be specified. If I did not feel I knew with certainty what this was; then I would use ambient temperature. That's common sense. The example you state is an example of an invalid disk specification.

The code says what I've stated:

UG-127(a)(1)(a): "Every rupture disk shall have a stamped bursting pressure within a manufacturing range at a specified disk temperature (note 44),....."

Note44: "The specified disk temperature supplied to the rupture disk Manufacturer shall be the expected temperature of the disk when an emergency condition exists and the disk is expected to rupture."

UG-127 (a)(1)(b) goes on to say how many tests are done at the elevated temperatures and how many are done at room temperature; and how to interpolate between these temperatures to rate the disk for its actual stamped burst pressure at its coincident burst temperature.

The burst pressure at ambient that is included on the disk namplate is a requirement and is provided to state the base on which interpolation is made for coincident burst pressure and temperature.




The more you learn, the less you are certain of.

 

[RPG]

You just argued in agreement with the inspector who failed the disc in the original posting, which I am in agreement with. The disc was stamped at a pressure greater than the MAWP at ambient 72F (215 vs 200). A worst case scenario can be fire exposure, so the disc will need to be at 200 psig at ambient and less than that at operting temperatures.

 

[CHD01]

Not so! But I give up; the sky is any color you perceive it to be. The more you learn, the less you are certain of.

 

[CHD01]

RPG:

Lets try this again, I may have been too quick to respond and not in the best of moods. Sorry!

I agree with your ambient temperature position IF it is the actual temperature of the disk at relief conditions. But then it is an incorect disk specification (the elevated temperature should not have been specified in the first place) and we should not even be having this discussion.

But if it is a fire basis for overpressure even as you stated (here is where I've been re-thinking the problem); then the vessel internals will have to come up to relief conditions at its equilibirum vapor pressure and temperature - thus it would relieve at its elevated temperature. That being the case, all that TD2K and I have discussed is true and the inspector should not disallow the disk because of the ambient equivalent pressure rating being greataer than the MAWP of the vesel at its elevated temperature.

After all, even the vessel has a higher rating at ambient conditions versus its elevated relieving condition.

The more you learn, the less you are certain of.

 

[RPG]

Your scenario is too idealistic, many rupture discs are on uninsulated vent pipes far enough from the vessel that they will never see a rapid temperature rise from a fire case. If you go back to the main issue, an inspector will not likely pass a vessel that has a stamped rating higher than the MAWP, even if it is at a lower temperature than where the vessel operates. Nothing in the code (or code interpretations) allows him to overlook this.

I do not plan on responding any further to this issue which has been beat to death and probably no longer being looked at by the originator.

Think twice before responding once.

 

[ccfromtx]

I am still reading this post and have enjoyed the well informed and thoughtful discussion. The disagreement in this thread reflects the confusion and uncertainty that I have run into when talking with other people on the subject. The way I see it, a disk rupture pressure is required to be specified at a coincident temperature at expected rupture conditions. It is up to the sound engineering judgement of the disk specifier to decide what those conditions are. The fact that our vendor chooses to stamp the disk with an 72 deg F rupture rating as well as the specified rating is the cause of the confusion here. If the cause of the overpressure is a fire case, regardless of how far the disk is away from the vessel, I would argue that the vapor pressure is dependent on the vapor temperature and the disk would be at the same temperature as the vapor touching it. I would argue the same for a hydraulic expansion case where the liquid would be touching the disk before rupture. There is no way of knowing what ambient conditions will be during an overpressure condition so, IMHO, the ambient (72 deg F) rating is worthless as far as a disk specification is concerned. I do understand the vendor tests at 72 deg F so that an interpolation can be made for elevated temperature ratings but no where in the ASME do I see that the vendor must stamp that rating on the disk.

 

[CHD01]

ccfromtx:

Thanks for input. I'll say this, I have actually measured disk temperature much less than vessel operating temperature. So it is possible and not even unusual. In my case the operating temperature was over 300 Deg F, but the disk temperature only SEVERAL FEET AWAY from the hot vessel was cool enough to TOUCH without discomfort! So I can envision a rupture disk spec rated to burst at a temperature between atmospheric and the elevated relief temperature. I look for this all the time. I don't have a problem with that. But once you investigate and state what that temperature for the disk is at the required burst pressure - THEN you should not be held by the inspector to the requirement that burst rating be less than the MAWP at the atmospheric temperature that the disk vendor includes. So I still agree with you in principle - it all depends how detailed the engineer is nad how much information he has regarding the installation. In some cases where little is known, I might even agree with RPG; but its still a position of preference, NOT A CODE REQUIREMENT. RPG claims that it is but the code cannot be interpreted that way. The more you learn, the less you are certain of.