Pressure transmitter with flange rating below process design P/T

[Steve010]

I am purchasing a 600# 2 inch flanged pressure transmitter. The piping is 600# carbon steel. There is a process condition that could last for days with a pressure of 1250 psig and a temperature of 200 F. According to the ASME tables, a 600# carbon steel flange maximum pressure is 1350 psi at this temperature, so the piping is ok. The problem is that the pressure transmitter 2" flange should be 316 SS, and as per the ASME tables a 316 SS flange maximum pressure is 1240 psi at this temperature. Should i use a pressure transmitter with 900# instead of 600#? or is it acceptable to operate above the flange rating by a slight difference ? it seems weird to me to use a 900# rating instrument on a 600# piping. any advise ?

 

[GenB]

Too close for comfort. It does not meet the ASME Code. You will have to upgrade. Maybe you can pad it with a c/s flange and get away with the 600lbs

 

[georgeverghese]

It sounds like 1250psig at a coincident 200degF is part of the normal oeprating envelope, so the design P and T would be even higher. Flanges should suit the design P/T and not just the operating envelope.

 

[moltenmetal]

To emphasize what georgeverghese said, what matters is the relief pressure of the system, not the operating pressure. I personally don't like the term "design pressure", because you can "design" the piping for any pressure you want, as long as that is in excess of the relief pressure- the suitability of any particular component you add to that piping later (such as this instrument) is determined by the relief pressure.

A pressure transmitter with a pancake style diaphragm seal and a 600# carbon steel backing flange would be your best solution.

 

[Steve010]

There is no relief pressure, this is on the outlet of oil wells. The 1250 psig and the 175 F are the design conditions.

 

[jte]

Not unusual to see this type of circumstance.

Given that in most cases the goal is to avoid a loss of containment, one could look at the various flavors of steel involved and recognize that one of them is less stiff. If one recognizes that one aspect of flange sealing involves avoiding excessive rotation of the flange with the resulting impact on the interaction with the gasket, then it is easier to understand why a geometrically identical flange fabricated of less stiff steel would have a lower flange rating. And this would drive the (correct) decision to use the higher flange class.

The more simplistic reason is that most engineers like a bit of CYA rather than take on the risk of making a decision which could result in bad consequences. In many cases we lean on codes and standards for this coverage. In this particular case, flanging up the lower rated stainless flange into a system which requires the higher rating of the carbon steel flanges would expose the engineer to potential liability by being out of compliance with B16.5. Specifically, B16.5 2013 2.3.2 which states If the two flanges in a flanged joint do not have the same pressure–temperature rating, the rating of the joint at any temperature is the lower of the two flange ratings at that temperature.

I’ve been involved in a situation where this type of situation came up (but far more than a single flange pair) and the decision was made that compliance with the codes of construction would be demonstrated via FEA rather than by compliance with B16.5. That’s one option, but it is only the cheaper option when lots of these situations exist and the steel has already been installed. Far better and cheaper to catch it early and increase the flange rating before the steel is purchased.

 

[Steve010]

I agree with you.
Another question. Does'nt the code (ASME B16.5) has a safety factor? if it has, what is it ? For example if i have a case that my P/T is just above the design conditions by 0.5 psi, should i be comfortable about it?

 

[jte]

Steve-

Most codes and standards have a design margin. This margin is built in to the designs in order to account for the simple fact that as engineers we cannot know and account for everything about a particular design (we leave that to the physicists and other scientists!).

When you infringe on the design margin, you are out of compliance with the standard and thus exposed. To what extent you are exposed is going to be a function of individual circumstance and perspective. You will have to apply your engineering judgment to determine whether or not you are comfortable exceeding pressure temperature ratings for a given component.

For what its worth, I routinely do just that. Much of my business is related to dealing with equipment and piping which has degraded beyond its design intent. So I'm not saying it cannot be done, just that one should be careful and assess the situation (in my case I lean on API 579-1 / ASME FFS-1 and ASME PCC-2 to provide me with decision making criteria).

Specific to your case, my I have two suggestions:
1) Heck no! From the perspective of an owner/user who is purchasing the equipment and expects the engineering to be proper. I paid good money for you to engineer it right, I expect to get what I paid for.

2) Consider the consequence of a loss of containment. If acceptable, put a good gasket in and torque the flanges carefully and monitor periodically for leakage. From the perspective of an owner/user who has already procured said flange pair and is facing a decision based on cost of strict code compliance vs. lost profit opportunity as well as potential cost of a loss of containment - not only in money but far more importantly in lives, significant injuries, environmental damage, negative media exposure, employee morale, etc.

Finally, keep in mind that the decisions you make are yours alone. The advice you receive here, or on any other internet forum, is worth exactly as much as you paid for it.

 

[gwalkerb]

We have an informal rule in place where whenever we are specifying flanged instrumentation with a stainless steel flange, to always bump up the flange rating, for this very reason. We try to always design our packages so that the MAWP of a flange in a line is the lowest rated component. This way, we don't have to worry about our equipment being used improperly by someone who assumes that because a line has, for example, 600# carbon steel flanges, that everything in that line can handle 1480 psig @ 100F. Although field operators *should* always be checking P&IDs and PSV setpoints, we are aware that this is not always the case. So we try to always ensure that our equipment is safe to use in the most naive fashion. Of course this isn't always possible, or practical, which is why I said it was an informal rule. But the cost difference is usually not very significant, and the peace of mind from knowing our package is slightly more idiot proof, is definitely worth it.

 

[Gilmiril]

Is 316 SST an absolute must? Have you considered alternatives such as UNS S31803 (2205 Duplex SST)?

 

[Steve010]

No,Duplex SST is accepted, but isn't it cheaper to use 316 SS flange rating 900# than a Duplex 316 SST flange rating 600# ?

 

[moltenmetal]

You're mounting it to a carbon steel flange, the obvious solution is STILL to use a carbon steel lapjoint-type backing flange on the diaphragm seal! Duplex would be a pure waste.

Increasing the flange class for every instrument in a line just to maintain a flange-limited pipe spec is of questionable sense in my opinion.

 

[LittleInch]

Why does this have to be an integral flanged PT? Can't you use a CS blind flange and tap it to have a tube going to the instrument or just screwed into a #3000 1/2" fitting?

In respect of margins, B1.5 doesn't do this - the code to look at is B 31.3 which does allow certain excess over operating conditions for different lengths throughout a year - see 302.2.4 - with the owners approval
no more than 1000 variations over the life
<33% for < 10 hrs and <100 hrs per year
<20% for < 50 hours and <500 hrs/yr

Be very clear about what is the actual OPERATING temperature. you already quote 200F and 175 F in your responses. If the 175 is the design then this is your issue.

Be also clear about whether the flange will actually see this temperature. The PT flange is not n a flowing branch. What if you made the branch a bit longer and not insulated?

I must admit having said all that, if you want to keep changes to a minimum, just make the PT flange #900. Not uncommon if the rest of your system is C Stl.

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

 

[moltenmetal]

Absolutely LittleInch: if the transmitter doesn't need a diaphragm seal, you can use an impulse line and all these flange related issues go away entirely. I presumed that because the OP said it HAD a flange, it actually needed one.

 

[racookpe1978]

I'll go the other way actually.

OK, so the "expected" pressure (design pressure, or "contract" pressure) is 1250.

It's on an oil well (open environment, last point of a very-politically-sensitive containment.) So what is going to happen when the oil well itself decides to exceed the expected pressure? In a process environment, the presure can and often does - exceed specification pressure by mistake or by control failure. But the relief valves and vent lines and margin provide a measure of "knowledge" about the ultimate pressure that may occur.

What controls the maximum pressure underground? What happens when that expected maximum of 1250 become 1500 or 1800 psig? (Sure, other failures will happen if pressure keeps rising, but it doesn't make sense to me in this case to accept a lower-rated flange with no secondary containment possible on the flange.

 

[moltenmetal]

Nobody is advocating the use of a lower rated flange, unless I missed something. The right thing to do is to get rid of the stainless steel flange- it's a carbon steel line. For a diaphragm seal, the stainless steel flange is unnecessary entirely. The flange itself can be constructed of carbon steel and the diaphragm seal installed wafer-style between flange faces. Problem solved- no need to increase flange class.

As to the use of flange-limited piping without relief, I'll trust that the relief scenarios have been evaluated.