The double check valve assembly should be installed to protect the public water supply from low or non-health hazards. Some people may argue that the double check valve assembly (DCVA) does not need to be tested as often as the reduced pressure principle backflow preventer.

The problem with this logic is that quite often the double check valve assembly is the wrong choice for the hazard or that the conditions have changed since the initial installation. We often find the double check valve assembly was installed in a fire protection system that has an antifreeze additive. Sometimes we find DCVA's installed on a system that has a fire pumper connection within 1700 feet of a potentially contaminated source. Many times a DCVA is installed where a RP is required. There are also situations where the DCVA should be tested more often than once a year.

The most widely used method for testing a DCVA is the USC procedure using a duplex gauge. This procedure is excellent because it will test the check valve's ability to protect the public water supply from small amounts of backpressure and also test the shut-off valves for leaks. However, with multiple leaks (check valve and shut-off valve), it requires a bit of experience to read the gauges and determine exactly what the problem is.

A major disadvantage to the USC test procedure is that you must close shut-off valve #1 to perform the tests. During the test procedure, shut-off valve #1 is opened and closed several times. The danger to opening and closing shut-off valve #1 is that debris and scale may break lose from the valve and foul check valve #1 whenever the customer again uses water. Check valve #1 may fail immediately after the tester had certified that the assembly was working properly.

This situation may occur. Imagine a pair of 6" DCVA's installed in parallel on a 8" service line to a small community college. While testing assembly #1, shut-off valve #2 is closed and the college is now served through assembly #2. The available area of service pipe is reduced approximately 50 percent. The demand by the customer has not decreased. This means that the velocity of the water through assembly #2 has increased significantly.

Start test using the USC procedures:

1. Flush the test cocks and install brass fittings.
2. Connect high hose to test cock #2 and low hose to test cock #3 to test check valve #1 of the assembly.
3. Bleed air through hose and gauges. The gauges should read somewhere near line pressure.
4. Close shut-off valve #2.
5. Close shut-off valve #1.
   

At this point the needles on the test gauges drop to zero. You cannot find any leaks in the connections of the test kit. If you open shut-off valve #1 the test gauges shows line pressure but drops to zero whenever you close shut-off valve #1.

The problem is: shut-off valve #2 is leaking slightly. The high velocity of the water passing through assembly #2 creates a low pressure area downstream of shut-off valve #2 of assembly #1. The water is being siphoned out of assembly #1 causing the loss in pressure whenever shut-off valve #1 is closed.

The present USC procedure makes it virtually impossible to test DCVA's under these conditions. You need an alternate method to test double check valve assemblies.

Lets look at a few other methods to test the double check valve assembly.

Differential pressure test utilizing a differential gauge: This test is essentially the same as the first differential test performed on check valve #1 of the RP. Using a differential gauge you measure the pressure drop across the check valve. The needle on your gauge should indicate a reading of 1.0 psi or greater if the check valve is holding tight. (see Figures #8 & #9)

Procedure:

1. Flush test cocks.
2. Install brass fittings.
3. Connect high pressure hose to test cock #2.
4. Connect low pressure hose to test cock #3.
5. Close shut-off valve #2.
6. Bleed air from hoses and gauge.(low last)
7. Read gauge. (differential pressure across CV #1)
8. Move hoses. Low to test cock #4, high to test cock #3.
9. Bleed air from hoses and gauge.
10. Read gauge. (differential pressure across CV #2)

The great advantage of this test is that you do not have to close shut-off valve #1. By not closing shut-off valve #1, you minimize the risk of debris and scale fouling check valve #1.

Two major disadvantages: You do not know whether shut-off valve #2 is leaking or not. If shut-off valve #2 is leaking and the customer is using water then the needle on the gauge will read positive or above 1.0 psi even if the check valve is leaking. The second disadvantage is that if the low control valve is leaking while the high control valve is open, then the needle on your gauge will fall to zero. Even if the check valve is holding tight, the gauge will tell you that the check valve is leaking. You must be absolutely certain that shut-off valve #2 is closed tight and that the control valves on your gauge are not leaking.

This is not a good procedure to teach to new testers. It takes some field experience to be comfortable using this test. It always bothers me when "instructors" who have no field experience are teaching these alternate testing procedures.

One further point to keep in mind. You must know how accurate your test gauge really is. Is it accurate at 1.0 psi? If the gauge is slightly off, do you know how much? and which direction? There are not many gauges on the market that actually read 1.0 psi at 1.0 psi.

Another method used to test double check valve
assemblies is the sight tube procedure.

Procedure: (see Figure #1) Throw away the instructions shipped with tube. Flush test cocks. Close shut-off valve #2. Install tube on test cock #2. Fill the tube with water above 28 inches. Close shut-off valve #1. Open test cock #2. Open test cock #3 to atmosphere. Once water has stopped flowing from test cock #3, check tube. If water maintains a level in the tube 28 inches above the point where water was running out of test cock #3, then check valve #1 is reported as holding tight at least 1.0 psi in the direction of flow. Move sight tube to test cock #3. (see Figure #2) Close test cock #2. Open shut-off valve #1 to establish pressure in assembly and fill tube with water. Close shut-off valve #1. Make sure test cock #3 is open. Open test cock #4 to atmosphere.

The great advantage of this test is: A leaking shut-off valve #2 does not affect this test.

• If shut-off valve #2 leaks and the customer is using water, then air may be drawn
  
  into test cock #4 while testing check valve #2. Note: This may change the point
  
  that you consider zero (the level that you measure up 28"). With air in the
  
  assembly the point where you measure the 28" may be changed.

  If shut-off valve #2 leaks and there is backpressure on the assembly,
  

  then water should run continuously out of test cock #4.

  If shut-off valve #2 leaks and check valve #2 also leaks, then water will
  

  run out of test cock #4 and #3. Water may also rise in the tube.

If shut-off valve #1 is leaking, then simply install a "compensating tee" on the test cock before installing the sight tube. (see Figure # 3) This "tee" allows any water that leaks through shut-off valve #1 to escape to atmosphere. Adjust the extra test cock on the "tee" to match the leak by stabilizing the water level in the sight tube above 28" before testing check valve #1.

The location of the base of the tube is critical to the test. In Figure 4, the water coming out of test cock #3 is well above the bottom of the sight tube. The 1.0 psi level is approximately 28 inches above test cock #3. In Figure 5, test cock #3 is below the bottom of the sight tube. The point where water in the tube is allowed to drop may be well below the 1.0 psi mark on the tube.

The major disadvantage of the sight tube test is that you must open and close shut-off valve #1 several times during the procedure. Debris may become dislodged and eventually cause check valve #1 to leak.

A variation of the sight tube test is to replace the tube with a differential gauge. (see Figure #6) You use only the high pressure hose. Hold the differential gauge at the same level as the check valve. Keep the low pressure and by-pass hose at the same level as the gauge.

While testing the double check valve assembly using the differential gauge instead of the sight tube, you will get a reading on the gauge indicating the relative values of check valve 1 and 2. You must use a gauge that is very accurate at 1.0 psi or you must know exactly what the error in your gauge is. You can quickly check the accuracy of your gauge in the field. Connect the high pressure hose to a sight tube. Fill the tube with water and bleed the water down using the high bleed on the gauge. You can check your gauge at 1.5 psi and 1.0 psi. The base of the tube must be level with the center of the gauge.

You still may have the problem of debris breaking loose from shut-off valve #1 and fouling the first check valve #1 after you leave the site. You might try performing a differential test across check valve #1 after you finish the sight tube or differential test. (see Figure #8) Open shut-off valve #1, then test check valve #1 again. This time using the high and low pressure hoses on test cocks 2 and 3, check the differential across check valve #1. This test may tell you if any debris has become lodged in check valve #1. CAUTION: Debris or scale that may have broken loose from shut-off valve #1 may not move to the check valve until a good flow of water has passed through the assembly.

Maybe the right way to test the double check valve assembly is to make use of the best parts of a few different tests. The double check valve assembly is best tested if shut-off valve #1 is not operated. However, we still must know the condition of shut-off valve #2. Lets try something a little different:

1. Flush test cocks.
2. Install sight tube on test cock #4. (see Figure #7)
3. Close shut-off valve #2.
4. Load sight tube with a few inches of water.
5. Close test cock #4, close shut-off valve #1.
6. Open test cocks #2, #3, and #4.
   • If shut-off valve #1 is leaking, water will continue
     
     to run out of test cock #2.

     If shut-off valve #2 is leaking and customer is using water,
     

     then level in tube will drop.

     If shut-off valve #2 is leaking and customer has backpressure,
     

     then water will rise in tube.

     If shut-off valve #2 is NOT leaking, then you can continue.
7. Remove sight tube from test cock #4.
8. Open shut-off valve #1.
9. Flush test cocks again.
10. Install brass fittings.

    Using differential gauge: (see Figure #8)

11. Connect high hose to test cock #2. Low hose to test cock #3.
12. Open test cock #3 then #2.
13. Bleed hoses and gauge.
14. Close high bleed, then close low bleed slowly.
    
    If check valve #1 is holding tight,
    
    needle will then stop at some point above 1.0 psi.

    Close test cocks #2 and #3.

    Move low hose from test cock #3 to #4. (see Figure #9)

    Move high hose from test cock #2 to #3.

    Open test cock #4 then #3.

    Bleed hoses and gauge.

    Close high bleed, then close low bleed slowly.
    

    If check valve #2 is holding tight,
    

    needle will stop at some point above 1.0 psi.

Check valve #1 and #2 should hold a value on your gauge above 1.0 psi, assuming the differential gauge is accurate.

---

As you can see, there are several different test procedures for the double check valve assembly. There are times when a particular procedure will not work effectively. It is a good idea to have a few alternatives from which to choose.