Pressure Testing: An Update on Fact and Fiction
Print Date: 12/18/2014 12:16:16 PM
General Meeting Presentation Feature
The following presentation was delivered at the 79th General Meeting Monday afternoon session, May 3rd, by George Galanes. It has been edited for content and phrasing.
George Galanes was graduated from the University of Illinois with both Bachelor's and Master's degrees in Metallurgical Engineering. From 1982 to 1999, Mr. Galanes was employed by Commonwealth Edison. During his tenure he served as metallurgical engineer, principal metallurgical engineer, senior metallurgical engineer, and boiler expert. In 1999 he joined Midwest Generation where he now serves as Manager of Metallurgy and Quality Assurance. In addition to being licensed as a professional engineer in metallurgy for the State of Illinois, Mr. Galanes has served as a member of the NBIC Main Committee and Chairman of the Part 3 Subcommittee on Repairs and Alterations. He is active in a number of professional societies including AWS and TMS. A member of ASME, he has served as a member of a number of groups, including the Boiler and Pressure Vessel Section 1 Committee and AG Piping and SG Materials under Section 1.
MR. GALANES: I appreciate the invitation by the National Board to give this brief presentation. I want to lay down some definitions for what pressure testing is in comparison to hydrostatic testing. Over the last couple of years it seemed to me people have been interchanging the terms hydrostatic testing and pressure testing. So if anything, I would like to lay out some definitions and also stir some thinking with the members.
I developed this presentation about three years ago after writing an article for the National Board. When the National Board staff asked me if I was interested in giving a presentation at this meeting, I thought this would be a good time to update the original presentation and talk about some ideas I have down the road. These are my ideas. They don't represent the NBIC Main Committee – let me make that perfectly clear. These are things you can start thinking about to lay the groundwork for future endeavors. So with regards to pressure testing we will talk about what it is, why we need it, and advantages and disadvantages.
Pressure testing, what is it? It's basically a method where fluid is used to pressurize a component either internally or externally, and the fluid can be compressible as in a gas or it can be a liquid. Most of the time we prefer to go with the liquid because it's compressible – not a lot of stored energy – so it's really the safe way to pressurize things. In fact, there are two types of pressure tests. And this is where it's important to understand the industry terminology.
We have hydrostatic tests where water is used to provide pressure that is above normal operating or working pressure, typically in accordance with the original code of construction. And then you have air and gas or water pressure tests which are substantially below maximum allowable working pressure. Keep these separate because they both do different things.
For pressure testing, my interpretation is that we are limited to proving the integrity of a weld or mechanical repairs to pressure-retaining items, and to look for gross defects or workmanship issues associated with repair welds. And I'm talking about inservice equipment.
What's nice about it is the NBIC is very flexible and we wrote it in the code to give us a wide range of pressures, anywhere from a static head fill up to operating pressure in some cases, if it's required.
With hydrostatic testing we are doing a couple of things. One, we are verifying the leak tightness of a pressure-retaining item. It could be new construction or it could be an inservice item where perhaps NDT methods are impractical to evaluate whether the component can operate safely in service. Two, we can test after fabrication to assure adequacy of design. In this case, of course, this is the basis for code of construction for hydrostatic tests. And three, we test to strengthen material at locations of stress concentration. This is a good benefit for new material where you have new construction, and in some cases you may have very small defects. And in the case of doing a hydrostatic test, in areas of high stress concentration, the material begins to slightly deform or work-harden, and then the stresses are redistributed. And that's truly a benefit, because in that case we strengthen the material locally.
We know pressure testing, like hydrostatic testing, can be used to look for gross defects and to assure the integrity of weld repairs, which is the foundation for the wording in the NBIC.
The advantage pressure testing has over hydrostatic testing is that it can be performed at or well below working pressure. Let's say you’ve got a vessel and you are concerned about relief devices. Certainly the NBIC gives you maximum flexibility to go to a lower pressure than operating. And in some cases we fully exercise that on some of our power boilers.
As far as hydrostatic testing proving adequacy of design, that's really the foundation for that. We look at the advantage by introducing tensile membrane stresses by hydrostatic testing and assuring the component will not rupture. Of course, going back to the original code of construction, and then also locally work-hardened material at either 1.3 or 1.5 times, which again is a fact.
On the topic of strengthening of new material there are two issues. One is with the hydrostatic test by locally strengthening the material area around regions of stress concentration that contain small flaws, which is obviously a big benefit for new construction. The second is reinforcing the fact that the benefit is derived from local yielding at regions of stress concentration. This is only limited to pressure-retaining items with service temperatures below time-dependent properties. In other words, when you get into service temperatures where you begin to creep the time-dependent effects, then that benefit goes away. You can try to reintroduce the benefit, but in most cases it's really not a benefit anymore.
For hydrostatic testing, normally it's a one-time benefit to be gained versus one that can be used after an item has been in service, and this applies to aged equipment. Once you have done a hydrostatic test during an original code of construction, in my opinion, doing any type of hydro test beyond that (after it's been in service) really does no good, because you are trying to squeeze this thing and have the adequacy of design, but you have done that already with the first hydrostatic test. Trying to gain it back is very risky. On our equipment, our power boilers, we will not do any hydrostatic tests once it's been in commission and in service. Attempting to regain this benefit on aged pressure-retaining items is risky and could lead to rupture or cracks, especially with aged material that has low toughness.
In regards to the disadvantages of pressure testing, exposure to aged equipment with low toughness could result in brittle fracture. And this is really by far the greatest threat on aged equipment. The reason pressure testing was never required, or it shouldn't be required, is because testing operates at elevated temperatures.
So when you are doing repairs on boilers, it's good to keep in mind what you are trying to accomplish with a pressure test. Are you looking to prove the workmanship of the welds, the integrity of the welds, or are you trying to squeeze this thing to either cause it to fail or to look for leaks you may think are there but can't see. These are things to keep in mind.
It is a false premise pressure testing or hydrostatic testing will allow for determination of remaining useful service life. This is a key point. Some inspectors and owner/users feel comfortable a hydrostatic test or an elevated-pressure test will determine if the item can remain in service trouble-free (or not develop leaks) over a period of six months or a year.
During a pressure test today, whether it's at operating or below, or if one does a hydrostatic test, all you can really claim is that the component can hold water and pressure. But six months, a week, or ten days from now, if the damage mechanisms or condition of the boiler is not understood, you can certainly have a leak or a rupture.
So I want to make sure we understand what a pressure test is doing for you. In my opinion, when you look at a pressure test, understanding you are trying to look at repair welds and integrity of repair welds, then perhaps NDT in combination with pressure testing is really the way to go.
And that is something I pushed in our company with our inspectors after we conducted repairs on our boilers. Nondestructive testing has a significant place today with repairs on aged equipment, and I truly believe that lower-pressure tests in conjunction with NDT are the safe way to go. The reliance of only doing a pressure test is risky in my opinion.
There are some concerns with pressure testing. If an aged boiler or pressure-retaining item is to be subjected to a pressure test, owner/user should carefully evaluate the material of construction. The low-toughness/fracture toughness steels I'm concerned about are SA-212, SA-515, and SA-455. These steels have been around for many years, but they also have a potential for low toughness due to their coarse grain structure. Since the pressure tests will be conducted at or below 70 degrees, you need to know what's going on with that material. And if that's the case, maybe more NDT needs to be used or you could do a lower-pressure squeeze and introduce a longer hold time; some other possibilities to think about.
In the NBIC 2010 edition we introduced a cautionary statement to make sure owner/users are fully aware some of these boiler and pressure vessel steels that we use, mainly boiler steels, could have low toughness. The fact that they have not had elevated temperatures is good, but from a pressure testing standpoint you have to be on guard.
Next we will talk about low temperature pressure testing concerns. Obviously you want to know what the ductile or brittle transition temperature is. That's the temperature where steel may go from a brittle fracture behavior with a notch or a crack to ductile behavior. Of course, you always want to stay in the ductile region because that has higher toughness. However, Charpy V notch tube head testing is one method that is somewhat inexpensive, but it's intrusive. You have to look at other options if that one doesn't work.
One option might be to do scoop samples EPRI (Electric Power Research Institute) has developed to determine toughness. It's a very easy method to use. You get actual material and run punch tests to give you some idea of what the ductile or brittle transition is. On the other end of the spectrum, there is fracture toughness testing, which is even more intrusive than Charpy V notch testing. So if you are unsure about steel characteristics, there are some options. We are also looking at other methods. One might be to pre-warm the water to above 70 degrees. You may have to go to 80 or 90 degrees depending upon the steel you have in construction.
EPRI has provided a presentation to the chiefs meeting. I think it was a year ago when Ken Coleman mentioned some of the steels they were looking at where pre-warming would have to be at 100 or 125 degrees. That raises flags because there are safety issues trying to conduct a pressure test at those temperatures.
An alternative approach to fracture toughness testing is to review the materials of construction or increase the water temperature, as I just mentioned, or perform pressure testing at reduced test pressure to avoid harmful tensile membrane stresses that can result in brittle fracture. This is an item that is really ripe for us to take a look at on the NBIC. A gentleman I'm working with on the petrochemical side shared some information about some of the low-toughness steels they deal with. He said they do a pressure test that's about 40 percent of operating pressure. They determined it based on testing they did.
This is interesting for the NBIC to look at as an alternative to pre-warming water, for example, for these low-toughness steels. Currently Part 3 of the NBIC will open an item for consideration to address the above alternatives.
But that's something I think will help inspectors and owner/users in the field. There is some good information out there that we can try to gather in. And one of the thoughts I had was to look at using a table that has material thickness and pre-warming requirements based on the steel.
So for the two steels I mentioned initially, the 515 and 212, a table would have some suggestive pre-warming temperatures per pressure test based on the thickness of the component. The other option might be to look at having the temperature, it's going to be held at 70 degrees, and reduce the pressure test to a percentage below operating pressure of the vessel to make sure we are out of harm's way.
One last item to talk about is pressure testing effects on boiler water wall tube circuits, and this is something that I can speak specifically about because of my background. I have been in the power generation industry for 30 years, so I have seen a lot of boiler tube failures over the years, especially corrosion fatigue failures. One of the things we found – and this applies to power boilers –is if you do repairs on water wall tubes and we come back, we do not allow pressure to exceed 500 pounds on our power boilers. What we found is the magnetite layer (the protective oxide layer on the water- touched tubes) may fissure. When it fissures, it allows introduction of oxidated water to retouch the tubes, and it actually promotes more corrosion fatigue damage.
We discovered this about ten years ago, and now on our power boilers, at least for the fleet at Midwest Generation, most of our post-repair pressure tests will be typically whatever we can get below 500 pounds. Sometimes it may be a static fill after the repairs, if they are routine repairs, or we may run a condensate booster pump to get up to 250 to 300 pounds, but that's the limit. And our corrosion fatigue leaks have decreased dramatically. Our biggest hitters right now in our fleet are simple erosion and fly ash erosion versus ten years ago when 90 percent were early fatigue. That was a big effect for us.
In conclusion, the hydrostatic testing is a useful test method for new construction to ensure no gross material defects, adequacy of design, and to strengthen new material.
I would not recommend it for inservice components or aged pressure-retaining items unless, as I said, careful consideration has been given to ensure the materials of construction are suitable – for the fracture toughness issue is the big deal there.