Sometimes the movies have bonus footage of things that add to your enjoyment, such as bloopers, or simply a behind the scenes look at how certain scenes were made.
Today is your lucky day, as I am going to share some of my bonus footage that occurred during the research involved in preparing the air pressure test quiz.
Segment 1 – How do you select the appropriate gauge range for use on a project?
I really do not know the answer to this, as it has never been brought up in my career until I started to do research for this blog. The gauges I’ve observed on jobsites typically range from 0-60 psi through 0-100 psi (such as Gauge 1 shown here).
Gauge 1
The best I could find were recommendations by several manufacturers that suggest the maximum gauge operating scale be double the operating range of the test for the best accuracy, and of course for a factor of safety. Commonly, 30 psi is used as an approximate starting point for pressurization of HDPE and LLDPE seams, so using the logic of double the operating range, you would expect the gauge to read between 60 and 70 psi on the maximum side.
So, what if you are using a gauge that only reads to 40 psi? Obviously, there is a safety factor to consider concerning the gauge itself. But does it have any impact on the accuracy of the test? How about a gauge that reads to 100 psi for the same 30-35 psi test range? And, don’t forget that per ASTM D7177-05 (2015), the required pressure varies on sheet temperature and that range runs from a pressure of 19 psi (at 40o F) to 60 psi (at 110o F) per the ASTM.
Segment 2 – Gauge Accuracy
Gauge accuracy is not addressed in the ASTM other than calibration being required once per year. In looking into gauge accuracy, ANZI/ASME B40.1 seems to be the governing document. I have not yet had time to research this document, but data available on manufacturer’s websites suggests that there are at least 8 common grades where grade is based on gauge accuracy. The following table was obtained from winters.com showing the +/- % accuracy in each portion of the scale:
|
Accuracy Grade |
Lower ¼ of scale | Middle ½ of scale | Upper ¼ of scale |
|
4A |
0.1 | 0.1 |
0.1 |
|
3A |
0.25 | 0.25 |
0.25 |
|
2A |
0.5 | 0.5 |
0.5 |
|
1A |
1 | 1 |
1 |
|
A |
2 | 1 |
2 |
|
B |
3 | 2 |
3 |
|
C |
4 | 3 |
4 |
| D | 5 | 5 |
5 |
Ashcroft suggests commercial gauges begin with the classification as Grade B, which have an accuracy of +/- 3-2-3% of span. However, Grade C (accuracy of +/- 4-3-4% of span) are extremely common on geosynthetics projects based on my research of models I have seen used. I have come across many that I cannot classify, as there is no manufacturer identification on them.
It took a bit more research on these sites to understand exactly what the “4-3-4% of span” means. Basically, span is the maximum value of the range of the gauge, so if using gauges with a maximum value of 60 psi (such as Gauges 2 and 3, shown here), the numerical value of the span is 60.
Gauge 2
Gauge 3
For a gauge with a range of 0-100 psi (such as Gauge 1), the numerical value of the span is 100. In all of the examples I have seen online, the accuracy is calculated off of the maximum range value. So, +/- 4% accuracy on a gauge with maximum range of 60 psi is an accuracy of +/- 2.4 psi, whereas 3% accuracy for this same gauge would be +/- 1.8 psi. How do these numbers apply?
The “4-3-4%” accuracy correlates to specific portions of the range itself. The first value (in this case 4%) represents the first quarter of the range. The second value (in this case 3%) represents the next half of the range. The third value (in this case 4%) represents the last quarter of the range. Using the maximum value of 60 psi, the accuracy is therefore: +/- 2.4 psi from 0-15 psi and from 45-60 psi whereas the accuracy is +/- 1.8 psi from approximately 16-44 psi.
So, using a Grade C gauge ranging from 0-60 psi, the test accuracy is +/- 1.8 psi in the common test range of 30-35 psi when testing materials like 60 mil HDPE and 40 mil LLDPE. Often, the maximum allowable drop for these tests is 3 psi, which means that the potential error within the test itself is approximately 60%. Is this a big deal? As a note, IAGI’s “Guidelines for Installation of HDPE and LLDPE Geomembrane Installation Specification” found on their website uses a drop of 4 psi, where the potential error within the test itself is 45%.
Again, strictly for continuity testing, I am not sure how much of an issue this really is. But, for the PVC testing following ASTM D7177, it would seem to be much more critical. ASTM D7177 uses the results of the air pressure test to correlate whether the common 15 ppi peel test requirement is being met, thus reducing the number of destructive samples required. From the table in ASTM D7177, the following have been taken and compiled:
|
Sheet Temp (oC) |
Air Pressure (psi) |
|
18 |
40 |
|
21 |
36 |
|
24 |
34 |
NOTE: It is my understanding that Environmental Protection, Inc. was influential on the development of this procedure and the creation of ASTM D7177. However, the current ASTM values differ slightly from the ASTM table shown on Environmental Protection, Inc.’s (EPI) website, as of February 13, 2018.
Keep in mind that at each sheet temperature, the air pressure requirement is supposed to represent an equivalent of 15 ppi of seam strength. Is it fair to assume that a Grade C gauge is accurate enough when it’s accuracy is only +/- 1.8 psi?
Furthermore, you will note that this ASTM D7177 procedure uses pressures from 19 psi to 60 psi. If the installer uses the same 0-60 psi gauge (assume Grade C for example) they have on hand for HDPE testing, the accuracy of certain ranges of this test is now only 4% (+/- 2.4 psi). Would a gauge reading from 0-100 psi be more appropriate?
Actually, if you would be applying ASTM D7177 to the lower sheet temperature range such as 40o F requiring 60 psi for the test, this logic would dictate that you actually need a gauge capable of reading to 120 psi.
How often have you seen huge swings in PVC sheet temperature? I’ve been on fall and spring projects where the sheet temperature variance is extreme: 40o F in the morning, 100o F in mid-afternoon, and then back close to 40o F in early evening. Assuming you are on one of these project, now what? Is a gauge with a scale ranging from 0 psi to 120 psi acceptable? If so, what class is needed?
As shown in the first table above, there could be different accuracy levels throughout different segments of the scale. Does the installer need to have gauges of varying scales on hand, more suitable to the varying sheet temperatures? HINT: Try keeping track of different scaled gauges on a project and ensuring the correct ones are always being used – it is going to be very difficult for both the installer and QA to get it right consistently.
The best answer I have for gauge accuracy is that there needs to be some open discussion about it, especially in applications where the results of a continuity test are used in lieu of physical destructive seam samples.
How’s that for an ending?! This bonus footage really ends with a cliffhanger! Of course, where else would you have the opportunity to write the ending (don’t thank me all at once here)? The ending (our future) really is up to you! I have asked the questions, but I do not have the answers. It is up to all of us to push for these answers.
I love the ending….
Of course all of your observations and are spot on as usual. You are very good at presenting published facts that are well documented on the web, just like Rush Limbaugh (I like him too). The problem is that the deeper you look, the more inconsistent the data gets.
I think a revamping of the air pressure test procedure and min/max pressures is in order. I’m not saying I want to chair the committee though.
Dave – as always, thanks for the feedback!
Getting people looking at some of these details and getting the industry talking is the goal and this blog seems to be stirring up some good conversations. I have had several people ask me privately for my thoughts on what should be done and like you – I don’t necessarily want to chair the committee but hope to be a part of the solution.
A few of the questions that need to be answered as part of the solution:
1. Other than for PVC, are we really trying to correlate seam continuity to seam strength?
2. Should the testing pressures vary by material type and/or thickness?
3. What accuracy is required for any of the gauges used (vacuum box, air pressure test, air lance, etc.)?
4. How should the calibration issue be dealt with, and in that regard, what defines a calibration, what voids a calibration and how can calibration be checked/confirmed in the field.
As a note, it is my understanding that although less frequently referenced, GRI Test Method GM6 is still an active test method that also deals with testing pressurized air channels. I have not been able to obtain a copy of this document so I am curious if perhaps the GRI Test Method addresses any of these issues referenced in this blog series.