Is one backflow prevention assembly test better than another? This question brings the passion out of so many people in the cross-connection industry. Do any or all of them rigorously test a backflow prevention assembly? My one request before we go any further is that individuals reading this article keep an open mind as they take in the information below. When you finish let me know your thoughts at sean.cleary@iapmo.org.
Let us start with why it is important to test backflow prevention assemblies. We test to ensure the assemblies are providing the required protection and are working properly. Any backflow assembly is a mechanical device that is subject to failure. With the exception of the spill-resistant vacuum breaker assembly, all backflow assemblies have independently acting parts and a built-in redundancy to prevent a total assembly failure from occurring. Testing assemblies allows us to discover failures of individual components; it also allows us to ensure components are exercised on a regular basis. For example, at least once a year shutoff valves are exercised, air-inlet and differential relief valves are opened, test cocks are flushed, and the entire assembly is examined. If we do not test the assemblies on a regular basis we have no idea how they will perform when hydraulic conditions occur. The reliability of any cross-connection control program requires regular assembly testing performed by qualified individuals.
A qualified tester needs far more knowledge than simply remembering the testing steps to a specific test procedure. They need to understand what backflow is and the differences between the two types of backflow. They need to understand how water flows in a system and within all types of backflow assemblies and devices. They need to understand plumbing and mechanical code requirements, and how to establish a specific degree of hazard for the cross-connection on which they are working. They need to understand both containment and isolation protection and the requirements in the specific jurisdiction in which they are performing testing. The range of knowledge includes everything from confined space requirements to thermal expansion, to fire protection systems, to testing equipment, and the list goes on. They need to know far more information than they can learn in a 32- or 40-hour training class. They also need to realize that they will never stop learning, as change is a continuous part of the cross-connection industry.
Whenever you test a backflow prevention assembly there are questions a tester needs to ask; they should have a mental checklist in their head. It should contain the following questions:
WHAT IS THE DEGREE OF HAZARD?
WHICH TYPE OF BACKFLOW (BACKPRESSURE OR BACKSIPHONAGE) IS POSSIBLE?
WHICH TYPE OF SYSTEM IS THE CROSS-CONNECTION PART OF?
WHAT ARE THE SYSTEM PRESSURE REQUIREMENTS?
WHAT IS THE LOCATION?
IS IT FOR CONTAINMENT OR ISOLATION PROTECTION?
WHAT ARE THE INSTALLATION REQUIREMENTS?
The answers to these questions are needed before we even begin a testing process. In my 46 years working in this field, I have met some incredibly talented people. I have worked with so many who were talented at installing systems — craftsmen who could take a blueprint or a shop drawing and install the system perfectly, rolling offsets, large and small piping correctly done each and every time. Quite a few of these people, however, lacked the natural curiosity needed to try to understand the way the systems they were installing actually worked. If the shop drawing or blueprint were incorrect, then the system would be installed incorrectly. In the crossconnection world, we need technicians and testers who look a little further into the systems. People who ask the question, “If I were water, in which direction would I or could I flow?” They look at the backflow prevention assembly they are testing as a part of a larger system, as one simple but vital component within that system. The field test procedure is a critical part of our evaluation, but it is not the only part.
Once we have the training, certification, and knowledge to test backflow prevention assemblies, we need to decide which field test procedure we will use. There are a number of field test procedures used in the United States by certified testers and it varies depending on the jurisdiction in which you are testing. It may also depend on the certification a tester possesses. Testers should always use the field test procedure to which they have been trained and tested. I teach backflow tester certification classes in a number of areas of the United States, and as a result I am remarkably familiar with numerous different field test procedures used in the areas in which we conduct classes. In most cases, people tend to think that the procedures they were trained in are the best procedures out there and the only ones that should be performed anywhere. This is the part in my article where I must again ask you to keep an open mind as you continue reading. Many people in the industry become like the Hatfields and the McCoys when someone makes the following two statements: The first is that there are a number of field test procedures used in the United States that will rigorously evaluate the condition of the backflow assemblies being tested. The second is that backflow assemblies can be field tested with five-, three-, or two-valve test kits. At this point we need to talk about the field test procedures themselves.
The most common field test procedures I see in classes are the ASSE Field Test Procedures, the One Hose Field Test Procedure, and the USC 10th Edition Field Test Procedures. We also encounter the New England Water Works Field Test Procedures in some classes we do in the northeast part of the country. Some certification programs require a specific procedure to be performed in all their practical examinations and training. The same is true for certain jurisdictions that require a specific field test procedure or certification program to be used. Other jurisdictions may allow for multiple certification programs and multiple field test procedures to be used. It is important to understand the regulations in place in any area in which you may be testing.
The test equipment you use may change the field test procedure slightly since a five-valve test kit allows for the bleeding of air without using the manifold valves as you would with a three-valve test kit. The two-valve kit makes the bleeding or air removal process even more complicated. Our discussion below will focus on the test procedures without making them specific to the test equipment being used.
When we look at the field test procedures themselves, we can break things down into two groups — single hose testing and multiple hose testing procedures. We can also look at the differences in check valve testing, either in the direction of flow or by using backpressure to test the check valve. Some of the test procedures are remarkably similar in certain areas and vastly different in others. To start with we can look at the spill-resistant vacuum breaker assembly (SVB). This is the valve that upsets many test takers during recertification classes. They look at them like Bigfoot — they know they exist but may never have seen one in the field. Since there is only one test cock this assembly will be tested using a single hose test procedure no matter which certification the tester possesses. The SVB requires the testing of numerous assembly components in a specific order. All three field test procedures we are looking at test the check valve first, followed by the air-inlet opening point. While there is no specific testing of the No. 1 or No. 2 shutoff valve included in the test, the check valve and airinlet valve would allow the tester to pinpoint problems with the shut-off valves and ensure a valid test is conducted. Proper elevation of the testing equipment is an important part of the SVB test. The USC 10th Edition Field Test Procedure requires the installation of a bleed valve arrangement on the high side test cock when testing any assembly using a one-hose procedure in case of a leaking number on shutoff. Other one-hose procedures do not require this prior to test assembly, but in case of a shutoff valve failure would allow it to be added and the testing procedure restarted. The pass/ fail criteria on an SVB are a minimum of 1-pound differential pressure on the check valve and the airinlet opening at a minimum of 1 pound and also that the air inlet fully opens.
When we look at the pressure vacuum breaker assembly (PVB), we see similar test procedures if we compare the One Hose and the USC 10th Edition Field Test Procedures. With the exception of the 10th Edition requiring the bleed valve arrangement being installed as part of every field test, the other steps, and the order of those steps in testing the air-inlet and the check valve in the PVB are almost identical. The air inlet is tested first with the check valve being tested next in both procedures. Both the No. 1 and the No. 2 shutoff valves are closed when testing the air inlet and the check valve in either the One Hose or the 10th Edition Procedures. The ASSE Field Test Procedure is vastly different in several ways. Using the ASSE procedure, the first test is to ensure the valve is in a static or no flow condition. The No. 2 shut-off is closed, and since the check valve is tested under line pressure in a direction of flow using both the high and low side of the test kit, it is critical to confirm the assembly is in a no-flow state.
Flow through the assembly will cause a differential pressure across the check and could cause a tester to pass a failing check valve. So our first test is the no flow state, the second test is the check valve differential, and the third test is the opening point of the air-inlet valve. The third test is a one-hose test similar to the testing done to open the air-inlet valve in both the USC and One Hose procedures. All three procedures will allow the tester to prove the assembly is working correctly or is a failure mode.
If we again look at the three field test procedures used to evaluate the double check valve assembly (DCVA), we will again see that, with the exception of the 10th Edition requiring the bleed valve arrangement being installed as part of every field test, identical in both the One Hose and in the USC 10th Edition Procedure. The ASSE Field Test Procedure is completely different. The other procedures require the tester to isolate the assembly by closing both the No. 1 and No. 2 shut-off valves and testing the two check valves in the direction of flow and comparing the pressure upstream of the check valve with atmospheric pressure. In the ASSE procedure, the assembly remains under line pressure throughout the test, with the No. 2 shut-off being closed and the No. 1 shut-off remaining open throughout the entire test procedure. The valve is first tested to prove the assembly is in a static or no-flow condition, and then a direction of flow test is conducted on the first check valve followed by a similar test on the second check valve. Which procedure is more accurate? Which procedure is easier to use? What are the advantages and disadvantages of each? Let us look at those questions.
This is where the differences of opinion take place. Do all three procedures work? I think they do; each will evaluate the assembly. If we are testing valves 2 inches and smaller, any assembly made in the past 35 years will have ball valves. The odds are good that these shut-off valves will close tightly, or in the case of a leak on the No. 1 shut-off, it will be a manageable leak that can be compensated for using a bleed valve arrangement. It is also true that if you are testing an assembly using the One Hose or the USC 10th Edition Procedure a leak on the No. 2 shut-off will not be a significant problem unless the assembly is under backpressure. When using a one-hose type procedure, the height of the test cocks and the test kit is a critical factor in conducting a valid test. If the assembly is installed in a horizontal orientation and the test cocks are the highest part of the assembly, it is relatively simple to determine the proper height for the test equipment. If the test cocks are not the highest part of the assembly or if the assembly is in a vertical upflow or vertical downflow orientation, it is more complicated to ensure the testing equipment is placed at the proper height. A reference tube will be necessary in these situations. However, what if it is a larger-size DCVA?
If I am testing a 30-year-old DCVA with gate valves, do I really want to attempt to get both the No. 1 and No. 2 shut-off valve to hold in a static condition? Do I want to close the No. 1 shut-off, disturbing dirt and debris that, when I complete my testing, will flow directly into the assembly I just tested? Does it make more sense to test DCVA under line pressure in the way the ASSE Field Test Procedure does? Using line pressure on both sides of the test kit allows me to place the kit above or below the assembly and removes the concerns related to reference tubes or test kit height. These are questions people with an open mind can consider. Is it possible that in certain situations it may make sense to qualify and test backflow testers in more than one specific procedure? Consider this: If a 10-inch No. 1 shut-off on a DCVA has a large leak that you are unable to compensate for, should a tester consider testing the valve under line pressure leaving the No. 1 shut-off valve open if the No. 2 shut-off valve holds? Or should they simply tell the facility owner they need to replace the gate valve and fail the assembly? The same goes for a situation when a tester testing the assembly under pressure has a badly leaking No. 2 shut-off. Would it make more sense to use a one-hose-type test if the No. 1 shut-off were holding tight? You could compensate by running a jumper hose from test cock one to test cock four, but by doing so you are creating an unprotected crossconnection around the assembly. Also, what if the leak is too great to compensate for the leak? Logical people may consider a different proven test procedure. I am authoring this article to hopefully get people to think about these issues and consider if there are reasonable options out there.
The last and possibly most contentious assembly we need to talk about is the reduced pressure principle assembly (RP). It is the most complex assembly with by far the largest differences in the three field test procedures we have been speaking about. Both the USC and ASSE procedures test the assembly under line pressure with only the No. 2 shut-off being closed. The One Hose Procedure will test the assembly with both the No. 1 and the No. 2 shut-off closed. This is the other steps, and the order of those steps, is almost the assembly with which we will really look at the differences in all three procedures and attempt to explain why the procedures used to test the assembly are so different.
The magic of the RP is the differential pressure relief valve. Its job is simply to sense the pressure in the reduced pressure zone and ensure that the pressure downstream of the first check remains lower than the pressure upstream of the first check. As long as we have lower pressure downstream than upstream, backflow is impossible. The differential pressure relief valve does not sense the condition of the check valves; it does not communicate with any other component in the assembly. The relief valve is only concerned with the pressure in the reduced pressure zone.
The ASSE and the USC procedures test the differential pressure relief valve in a comparable way under line pressure. They both use the test kit high and low control valves to slowly add high pressure water from upstream of the first check valve into the reduced pressure zone until the relief valve begins to open. The USC 10th Edition procedure tests the relief valve opening point first in the test procedure. The ASSE Field Test procedure tests the relief valve last. Both procedures test the first check valve in th
direction of flow under line pressure and the second check valve under backpressure. The USC procedure tests the differential pressure relief valve first and uses that test to ensure the assembly is in a static condition. The procedure then tests the second check valve with backpressure and finishes the procedure with a direction of flow test on the first check valve. The ASSE procedure tests the second check valve with backpressure and confirms the assembly is static first. It then tests the first check in the direction of flow and, in the final steps, tests the opening point of the differential pressure relief valve. The procedures are remarkably similar but also vastly different.
All three test procedures start the flushing procedure for the test cocks by opening the No. 4 test cock and establishing flow through the assembly to prevent the differential pressure relief valve from opening before having our testing equipment ready to record the opening point. This is important because if you are conducting an annual test you want to get the most accurate reading possible. If you exercise the relief valve before you can record the opening point correctly you will not get the accurate reading you are looking for. The USC 10th Edition Procedure has a number of steps aimed at ensuring the relief valve is not exercised prematurely. One such step is opening test cock No. 4 and establishing flow, then opening test cock No. 3 and letting it flow, and repeating the process with test cocks No. 1 and 2. The tester then closes test cock No. 1, 2, 3, and then 4. The USC 10th Edition Procedure also takes it a step further by using a specific process to hook up the testing equipment to the assembly flowing both the high and low bleed valves, while shutting the No. 2 shut-off. This ensures a failing second check or disc compression on the second check — as a result of closing the No. 2 shut-off valve — cannot prematurely open the differential pressure relief valve. The ASSE and One Hose Field Test Procedures do not contain those requirements. They open test cock No. 4 to establish flow through the assembly and then open and close test cock No. 1, 2, and 3, with the tester then closing test cock No. 4.
Which procedure is better for flushing test cocks? Is it slightly more likely that the differential pressure relief valve could discharge prematurely using the ASSE or One Hose Procedure? The answer is yes to that question. Is it a large problem? I would say no. If we do every step correctly using any of these three procedures and the relief valve discharges anyway, do we continue with the test procedure and test the assembly? What if, as I walk up to the assembly and before I begin the flushing procedure, the relief valve discharges? Can I still continue and do a valid test? The answer to these questions is yes. If you are running water though the assembly by opening test cock four and you carefully open the remaining test cocks, the relief valve will remain closed in most cases. I suggest you experiment with all three of these procedures when you are testing in a class or in your lab. If you are testing in the field, use the procedure the jurisdiction requires and/or you have been certified in. We should not experiment on assemblies in the field that protect the public.
The last thing I would like to discuss is how we test the assembly — specifically how we test the differential pressure relief valve when using the One Hose Field Test Procedure. When using the One Hose Procedure we need to think about what we discussed related to shut-off valve type and assembly age when we looked at the DCVA testing. When testing an RP with the One-Hose Procedure, some of the same issues need to be considered. If we look at the actual procedure, we are isolating the assembly by closing both the No. 2 and the No. 1 shut-off valves. We then do a direction of flow test on the first check by opening the No. 3 test cock and dropping the downstream pressure to atmospheric. We then use the high bleed to drop the pressure on the high side of the relief valve diaphragm slowly and record the opening point of the differential pressure relief valve. Once we have that number, we open the high bleed completely, which allows the relief valve to fully exercise, completely opening the relief valve and draining the zone completely. For the second check valve test, we must repressurize the assembly either by reopening the No. 1 shut-off or by running a jumper hose from test cock one to test cock two and repressurizing the assembly. We then do a direction of flow test on the second check valve, isolating the assembly and dropping the pressure downstream of the second check valve, opening the No. 4 test cock and dropping the pressure on the downstream of the check to atmospheric pressure. Remember that since this is a one-hose procedure, the height of the test kit is a critical factor.
A number of people in the industry question if the one-hose procedure accurately tests the differential pressure relief valve; they are concerned that not using line pressure to test this component allows the relief valve to open at a higher pressure than if it is tested under line pressure. They are correct that it is possible that using the One-Hose Procedure will give you a higher opening point on the relief. The question we need to ask ourselves is whether that means we can pass an assembly using one procedure, but that same assembly will fail under another procedure. What is the difference in the reading? Is it 0.1 or 0.2, or is it significantly higher? Am I looking at a relief valve that opens at 3.2 using the One Hose Procedure, but opens at 3.0 when tested using the ASSE or USC Procedure? Do we have the required data to make that judgment call? Is it available for review? I would encourage everyone to take some time and compare procedures. In your test lab, test an assembly using the same test kit with all three procedures and compare the results. Keep an open mind while you do it.
As a person who sees all three procedures used in different classes I instruct, I see advantages and disadvantages in each. The One-Hose Procedure for the differential pressure relief valve is the only one that fully opens and exercises the relief valve. The USC and ASSE will open and cause the relief valve to open, but not completely. If we are only doing annual testing, does the complete opening of the relief valve make a better test? What about check valves? Is a direction of flow test more accurate that a backpressure test? I know as an instructor I can remove the spring from a check valve assembly and it will pass a backpressure test but fail a direction of flow test. Should we always test the check valves with a backpressure test and then a direction of flow test whenever possible? These are questions reasonable people can ask. We should always be trying to build a better mousetrap.
All our procedures have improved over time. We no longer use duplex gauges; the USC 8th Edition Field Test Procedures are different from what you find in the 9th and 10th Editions. At some point, we will no doubt see a USC 11th Edition, which will contain improvements and changes to meet the current situation. The ASSE Field Test Procedures have been updated six times since the first Series 5000 Standard was completed in 1990. I hope at some point we can get the industry together at a neutral venue to have a frank discussion about test procedures in which we can compare and contrast each procedure with hands-on testing in a laboratory setting. The industry is full of great people who work hard to protect water systems everywhere. Everyone has their own opinions based on their own experiences; I know I do. Please consider what I have written here and let me know your thoughts. I am truly blessed doing the work that I do. Don’t forget what President Teddy Roosevelt once said: “Far and away the best prize that life has to offer is the chance to work hard at work worth doing.”
