Evaluation of the Highland Tank Periodic Interstitial Vacuum

Test Procedure for Double-Wall Underground Tanks

(Titan®, STi-P3®,ACT-100U® and HighGuard®)

Final Report

Prepared for: Highland Tank and

Manufacturing Company, Inc.

December 20, 2010

Revised September 14, 2015

Ken Wilcox Associates, Inc. 1125 Valley Ridge Drive, Grain Valley, MO 64029, USA

Phone (816) 443-2494, Fax (816) 443-2495 E-mail kwilcox@kwaleak.com

Evaluation of Highland Tank Periodic Interstitial Vacuum

Test Procedure for Double-Wall Underground Tanks

(Titan®, STI-P3®,ACT-100U® and HighGuard®)

Final Report

Prepared for:

Highland Tank and Manufacturing Company, Inc.

One Highland Road Stoystown, PA 15563

Revised September 14, 2015

Preface

This report presents the results of an independent third-party evaluation of vacuum leak detection system provided for the Highland Tank Models Titan, Sti-P3, ACT-IOOU and HighGuard double wall underground tanks. These tanks differ primarily in the construction of the outer wall. Both walls of the Sti-P3, ACT-IOOU and HighGuard listed tanks are constructed of mild carbon steel and have advanced polyurethane coatings applied to the outer shell for correction protection. The Sti-P3 tank also has sacrificial anodes as part of its cathodic protection process. The outer wall of the Titan is constructed of polymeric materials that protect the inner metal tank from corrosion and forms the interstitial space.

Al l tanks are tested by placing a 14" Hg vacuum on the interstice and monitoring for the greatest times shown in Table 1 of this report for a specific volume. I f the vacuum decreases are greater than 5" Hg over the test period, the tank is considered to be leaking.

This report is based upon testing previously performed by KWA Associates. This report is also based upon tanks constructed after the effective date of December 20, 2010.

Technical questions regarding this system should be directed to Mr. Charles Frey at the contact points listed below.

H. Kendall Wilcox, Ph.D., President September 14,2015

Contact regarding this report:

Highland Tank and Manufacturing Company, Inc. One Highland Road Stoystown, PA 15563 http://wvvw.highlandtank.com/Companv/ Phone 1-717-664-0600

Approved.

u

Table of Contents

Preface ii Table of Contents i i i Executive Summary iv

Introductions 1

Description of Leak Detector and Operational Principles 1

Evaluation Methodology 2 Vapor Pressure Tests 3 Displacement Tests 3 Calibration of Flowmeter 4 Water Leaks Through Cracks in Outer Wall 4 Calculation of Test Times 5

Test Results 5 Vapor Pressure Tests 5 Displacement Tests 7 Calibration of Flowmeter 7 Water Leaks Through Cracks in Outer Wall 11 Calculation of Test Times 11

Discussion and Conclusions 11

Test Procedure 15

Appendix - Alternative EPA Results Forms

iii

Executive Summary

The Highland Tank interstitial monitoring method was evaluated for determining liquid leaks into the interstitial space of a doublewall tank. The interstice is evacuated to 14" of Hg at the start of the test period. I f the vacuum does not decrease more than 5" of Hg over the test period specified for each tank, the tank is considered to be "tight". The test times range from less than one hour for a gasoline leak into the interstice of a 550 gallon tank to 45 hours for a diesel leak into the interstice of a 20,000 gallon tank.

Once the tank is installed, the interstice may be monitored using a liquid or vapor sensor installed in a riser that extends into the interstice. In general, the liquid level will reach the threshold of a liquid sensor in less than 4 hours.

iv

Background

The Environmental Protection Agency (EPA) requires that in-tank leak detectors be tested to determine i f they meet certain performance standards. In general, methods are required to detect a leak of 0.1 gal/hr with a probability of detection (PD) of at least 95% and a probability of false alarm (PFA) of 5% or less. The regulations for external monitors are much less well defined and there is no EPA protocol, for interstitial testing such as that developed by Highland Tank. To meet the specialized requirements for the Highland Tank system, it has been necessary to develop alternative evaluation procedures for this system. The requirements for alternative protocols are discussed in the introduction to all of the EPA evaluation protocols. The procedures described in this document meet these requirements.

It should be noted that the measurements conducted for this project are based on a limited range of petroleum products. The physical characteristics of fuels vary with geographic location, climates and brands. Physical measurements reported in this document should be taken only as indicative of general behavior and not as absolute values which might apply to other fields.

Description of Leak Detector and Operational Principles

The leak detection system developed by Highland Tank is based on the loss of vacuum in the interstitial space formed between a steel, primary containment vessel and a steel or HDPE secondary containment system that is formed around the outside of the tank. A vacuum gauge is used to monitor the vacuum which must be at least 20" Hg when the tank leaves the factory and 15" Hg when the tank is installed in the ground. (The initial vacuum at the factory may be as high as 27" Hg.) The leak detection process is based on the fact that the vacuum cannot be maintained is a leak is present in either the primary or secondary shell.

The Highland Tank interstitial volumes are relatively small. The tanks are constructed of an inner steel tank, and an HDPE jacket (Titan tank) or secondary steel wall wrapped tightly on the primary tank (Sti-P3, ACT-IOOU and Highguard). The interstitial spacing between the inner and out wall is 0.018 inches. The tank also consists of a 3" riser that extends through the inner steel wall into the interstitial space. A variety of liquid or vapor sensors may be installed in the riser as an additional means of leak detection.

Loss of vacuum can be due to several factors. These include:

* air leaks in either the inner or outer shell

* product leaks in the inner shell

* water leaks through the secondary containment i f the tank is below the water table

1

Highland Tank & Mfg. Co.

In all cases, the resultant loss of vacuum will alert the owner/operator to the presence of a leak. I f a leak is present, the vacuum cannot be permanently restored, and in the case of a liquid leak, the interstitial space will contain liquid.

In some instances, very small leaks in the outer wall or in the fittings of the tank will result in a very slow loss of vacuum. Although air flow through such leaks will be relatively rapid, liquid flow will be extremely slow and will not present any threat to the environment. This slow loss of vacuum may occur over a period of several days or months.

Evaluation Methodology

The data and conclusions for most of the testing conducted for this project were obtained in a laboratory environment using a test cell with a volume of 5.1 gallons. This is approximately the same volume as the interstitial space of a 4,000 gallon tank. This approach assumes that the behavior of the vacuum is not dependent on the shape of the interstitial space. That is, vapors and liquids will flow freely throughout the interstice and the secondary containment is not bonded to the inner wall. Tests for which volume was not a factor (e.g., vapor pressure measurements) were conducted in a smaller test chamber.

The loss of vacuum in the interstitial space will depend on the type of product (or air) that leaks into the interstitial space and the temperature of interstitial environment. Air leaks from the outside of the tank or vapor leaks above the liquid in the tank will reduct the vacuum to zero relatively quickly. The variations in vacuum with liquid leaks are different than those produced by an air or vapor leak. It is important to note these differences when interpreting leak test data. This report describes the testing conducted to determine the characteristics of liquid leaks into the interstice.

The effects of various liquids on the vacuum will depend on the vapor pressure of the liquid which in turn is dependent upon the temperature. Testing has therefore been designed to demonstrate the characteristics of each type of leak using several liquids. These liquids were selected because they represent a range of materials that are likely to be stored in underground tanks.

2

Highland Tank & Mfg. Co

The following tests were conducted to establish the performance characteristics of interstitial vacuum monitoring.

1. Measurement of the effect of vapor pressure on the vacuum

2. The effects of displacement on the vacuum.

3. Flowmeter characteristics under varying vacuum

4. Flow of water through cracks in outer wall

Each of these tests is described in this section of the report and the results of each are discussed in the following section.

Vapor Pressure Tests

As the temperature of the interstice increases, the vapor pressure of liquid present in the interstitial space will also increase, resulting in a loss of vacuum. To determine the effects of varying temperatures on the vacuum in the interstitial space, several tests were conducted using water, gasoline and diesel fuel and air. Vapor pressures were measured for gasoline and diesel over the temperature range of 32°F to 120°F. This was accomplished using a small test chamber equipped with a vacuum gauge and a thermocouple to measure the product temperature. The empty test chamber was evacuated to a minimum vacuum of 14" Hg. It was then placed in the temperature bath. Ice was added to bring the temperature to a nominal 32°F . Product was then introduced into the chamber via the valve system at the top of the chamber. Adequate liquid was added to fill the test chamber to approximately 1/3 capacity. The liquid temperature was allowed to stabilize and the temperature and vacuum were recorded.

The temperature was then raised slowly by heating the water in the bath while the bath was stirred vigorously. Temperature and vacuum readings were recorded at regular intervals until the temperature reached a nominal 120°F.

Displacement Tests

Liquids with low vapor pressures will produce little change in vacuum when small amounts of liquid are present in the interstitial space. The vacuum will be reduced, however, as the liquid fills the interstitial space. A series of tests were conducted accordingly to determine the volume of liquid required to reduce the vacuum from 14" of Hg to 9" of Hg.

3

Highland Tank & Mfg. Co.

Volatile liquids will produce an initial large change in vacuum as soon as liquid enters the interstitial space. The vacuum will continue to drop until the test cell atmosphere is saturated. The initial drop will be dependent on temperature. Once the cell atmosphere is saturated, the vacuum will drop slowly as the interstitial space fills with liquid.

Displacement tests were conducted by evacuating the empty test cell to 14" of Hg. The temperature was maintained at a constant value during the addition. Liquid was added incrementally and allowed to stabilize between additions. The vacuum and volume were recorded after each addition. This process was repeated for each liquid.

Calibration of Flowmeter

Liquid flow rates were calibrated for each of the liquids considered during the evaluation. The flowmeter was attached to the vacuum chamber and the flow rate was set to 0.1 gal/hr with a vacuum of 8" of Hg. This corresponds to a new pressure of 4 psi against the orifice. (Four psi represents the pressure against a leak at the bottom of a full tank with a diameter of 8 ft.) The temperature was maintained at approximately 70°F. After the initial rate was established, the vacuum was varied from 20" of Hg to less than 2" of Hg. A graph of the flow rate vs vacuum was then constructed for each liquid.

Water Leaks Through Small Cracks in the Outer Wall

One of the possible sources of loss of vacuum in the interstice is a water leak through the outer containment. This could be a result of damage to the out wall prior to or during installation. While air leaks through these cracks can be large enough to cause loss of vacuum, the flow of water into the interstitial space could be extremely slow.

A special test apparatus was designed to test the effects of water leaks through small cracks in the out shell. The test apparatus consisted of a steel plate (inner wall) with a sheet of outer wall bounded around the out edges of the plate so that an interstice similar to that for a Highland Tank was created. Twelve small star cracks were produced in the out wall to simulate possible damage to the tank. A vacuum was then applied to the interstice to determine that air would in fact flow through the cracks. The vacuum vs time was monitored until the vacuum disappeared. The volume of the interstice was also estimated by creating a vacuum followed by quickly allowing water to fil l the interstice.

The interstice was then connected to a water column that could be elevated to a height of 12 ft. The water flow rate through the cracks was measured at a different water head pressures to determine the behavior of the cracked outer wall material.

4

Highland Tank & Mfg. Co.

Calculation of Test Times

The time necessary to achieve a vacuum change of 5" of Hg was calculated for the interstitial volume for each tank using the equation

Vtank Test Time = x tceii

Vcell

where Vtank is the volume of the interstitial space for the largest tank (18.2 gallons), Vceii is the volume of the test cell (5.1 gallons for this evaluation) and tcdi is the time necessary for the vacuum to decrease from 14" to 9" in the test cell. When the leak rate is 0.1 gal/hr at a temperature of 70°F, the value for tcell is fixed at 1.0 hrs for gasoline, 12.1 hrs for water and 12.4 hrs for diesel fuel.

Test Results

Vapor Pressure Tests

The results of the vapor pressure tests are shown in Figure 1 for water, gasoline and diesel fuel. For water and diesel fuel, the initial vacuum in the test chamber prior to addition of product was nominally 27" of Hg. For unleaded fuel, two levels of initial vacuum were tested 27" and 22" (designated as "low P" on the graph).

In each case, the initial vacuum drops very rapidly as the first few ml of liquid enter the test chamber. More volatile liquids, such as gasoline, produce a large drop, while less volatile materials such as diesel fuel, produce much less initial loss of vacuum.

As the temperature of the test chamber increases, the vapor pressure of any liquid present will also increase, resulting in an increasing loss of vacuum in the interstitial space. For low volatility liquids, the decrease in vacuum is small. For both water and diesel fuel, the decrease in vacuum was approximately 2" Hg over the entire temperature range. Temperature will not be a significant factor for these liquids.

For gasoline, the vapor pressure increases much more rapidly as the temperature rises. For this reason, a gasoline leak will be much more readily detected. The initial decrease in vacuum will be rapid and should be easily identified by tank operators i f consistent records are kept. The initial drop will be dependent on the temperature of the environment. At 32°F, the nominal drop was from 27" to 18" and from 22" to 9" for the low pressure test. At 60°F, the nominal increase will be from 27" to 14" or from 22" to 5". At temperatures around 100°F, the vacuum will decrease to near zero. At temperatures above 100°F, the interstitial space will be pressurized.

5

Highland Tank & Mfg. Co

30

20

Highland Tank & Mfg. Co

Displacement Tests

The second factor that caused a decrease in vacuum i f the displacement process that occurs when liquid is added to the test chamber. This displacement process effective decreases the volume of the test chamber by the volume of liquid that is added.

The time required to detect a leak for nonvolatile liquids will depend directly on the volume of the interstitial space and the size of the leak. At an initial vacuum of 27", approximately 10% of the air at atmospheric pressure is still present in the interstitial space. When the liquid volume reaches approximately 90% of the total volume, the vacuum in the interstitial space will be reduced to near zero. For an initial vacuum of 15" Hg, approximately 50% of the air at atmospheric pressure is still present in the interstitial space. The vacuum will be reduced to near zero when the total volume of liquid added is approximately 50% of the interstitial volume.

This behavior is illustrated in Figure 2 for water and gasoline. The vacuum changes only very slowly as water is added to the test chamber. When the volume of water reaches approximately 75% of the chamber volume, the decrease in vacuum is rapid until the vacuum reaches zero.

For volatile liquids, the behavior of the vacuum is dependent on temperature. At low temperatures, the behavior is similar to that for nonvolatile liquids. At higher temperatures, the vacuum initially decreases rapidly followed by a slow decrease until the interstitial space is approximately 75% full. At 75% capacity, the decrease is again rapid. This behavior is also illustrated in Figure 2 for gasoline.

The behavior of the vacuum under the test conditions of 14" of Hg is shown in Figure 3. At 70°F, the volume of liquid required to decrease the vacuum from 14" to 9" is approximately 1 gallon for both water and diesel fuel and 0.075 gallons for gasoline.

Calibration of Flowmeter

The results of the calibration of the orifice at different vacuum levels is shown in Figure 4. Under the test conditions of this evaluation, the variation in leak rate with vacuum is expected to be linear and uniform for all liquids. The line represents the theoretical variation assuming that the leak rate is a direct function of the vacuum level. The theoretical curve was used for the calculations.

7

Highlaod Tank & Mfg. Co 30

20

10

0

I t

a Water (60 deg F) eULGasol i r

1

I !

i

IB (52 deg F)

1 I

1 l.

1

•

*

\ L 0 5 10 15 20

Thousands Volume Added fm!)

Fig 2. Vacuum vs volume of water and gasoline added to interstitial space.

8

ffighiand Tank & Mfg. Co

Highland Tank & Mfg. Co

Orifice Set @ 0.1 gal/hr @ 4 psi

1 I 1 1 1 i i i 1 1 !

i ! i

1 I

1 1 ! i : ! 1 1

1 1

! 1 : 1 i l y . i i ^

i 1 y

A 1

1 1 / 1 i

/ 1 1

/ 1 ! ( 1 1

' i I ! 1 / 1 i 1

/ ] \

/ '— 1 I i

1 / t i 1 i

1 i / 1 1 !

J, / i !

/ i 1 / f \ 1 j 1

>K i ! '. I 1 1 1

y 1 1 1 ! 1 ! t 1 •- 1

y\ i 1 I 1 — \ •

1 1 i

0 4 8 12 16 20

Vacuum (in Hg)

Figure 3. Flow Characteristics as a funcition of Vacuum.

10

Highland Tank & Mfg. Co

Water Leaks Through Small Cracks in the Outer Wall

The results of these tests are shown graphically in Figures 5 and 6. The volume of the test cell was determined to be 68 ml. Figure 5 shows the effects of vacuum on the air leak through the test cell. The time required for the vacuum to drop from 14" of Hg to 9" of Hg was only 15 seconds. This corresponds to an air leak rate of 1.8 gal/hr.

When water Is applied against these cracks, the leak rate as a function of head pressure is show in Figure 6. These results indicate that water will flow through these cracks at a rate of less than 0.001 gal/hr under the test conditions.

Calculation of Test Times

The calculated test times for liquid leaks into the interstice is shown in Table 1 for tanks up to 20,000 gallons. The test times are provided for water, gasoline and diesel fuel for leak rates of 0.1 gal/hr and 0.05 gal/hr.

If a liquid sensor is located within the riser with a threshold of 1", an alarm will occur within 4 to 5 hours for the largest tanks. The interstitial vacuum will be at approximately 13.5" of Hg for diesel or water. The vacuum for gasoline will decrease to around 10" of Hg.

Discussion and Conclusions

The Highland Tank monitoring method included a position-sensitive liquid leak sensor installed in the lowest point of the interstitial monitor tube, which will monitor for liquid leaks through both the primary and secondary tank walls when installed in accordance with Highland Tank's Interstitial Leak Detection Sensor Installation instructions.

While higher blends of ethanol in gasoline were not specifically tested, this report has stated that the more volatile liquids, such as ethanol, will produce larger and faster changes of vacuum in the interstitial space. Therefore, we conclude that ethanol blends of gasoline, up to and including E85, can be stored in these double wall tanks, and the testing protocol should follow the periodic testing procedures and test times noted for gasoline.

Periodic vacuum testing as required by the NO DENR will be performed as outlined in Highland Tank's Periodic Interstitial Test Procedure before start-up, six to twelve months after startup, and every three years thereafter.

11

Highland Tank & Mfg. Co

Interstitial Volume = 68.2 mL

Highland Tank & Mfg. Co.

12

I

Leak Rate (mL/hour)

Figure 6. Water Leaks Through Cracks in Outer Wall

13

Table 1. Calculated Test Time for Tanks.

Tank Size

Diameter Volume (in) (gal)

• •"

48 550 48 1000 64 1000 64 1500 64 2000 64 3000 64 4000 72 4000 72 5000 72 6000 84 4000

Reference 96 4000 96 5000 96 6000 96 8000 96 10000 96 12000 96 15000 120 8000 120 10000 120 12000 120 15000 120 20000 126 10000 126 12000 126 15000 126 20000

Interstitial Volume Multiplier

1.8 0.4 2.8 0.6 2.4 0.5 3.2 0.6 4 0.8

5.7 1.1 7.3 1.5 6.4 1.3 7.8 1.6 9.7 1.9 6 1.2 5 1

5.6 1.1 6.6 1.3 7.7 1.5 9.9 2 12.2 2.4 14.2 2.8 17 3.4 9.2 1.8 10.6 2.1 12.3 2.5 14.7 2.9 18.8 3.8 9.9 2 11.5 2.3 14.1 2.8 18.2 3.6

HIGHLAND TANK Mfg. Co

Time for Decrease of 5" of Hg Leak Rate = 0.10 gal/hr

Water Gasoline Diesel

4.8 0.4 5 7.3 0.6 7.5 6.1 0.5 6.2 7.3 0.6 7.5 9.7 0.8 10 13.3 1.1 13.7 18.2 1.5 18.7 15.7 1.3 16.3 19.4 1.6 20 23 2 23.7

14.5 1.2 14.9 12.1 1 12.5 13.3 1.1 13.7 15.7 1.3 16.2 18.2 1.5 18.7 24.2 2.1 24.9 29 2.5 29.9

33,9 2,9 34.9 41.1 3.4 42.5 21.8 1.8 22.5 25.4 2.1 26.3 30.3 2.5 31.3 35.1 2.9 36.3 46 3.8 47.5

24.2 2.1 24.9 27.8 2.4 28.6 33.9 2.9 34.9 43.6 3.7 44.8

Time for Decrease of 5*' ofHg Leak Rate = 0.05 gal/hr

Water Gasoline Diesel

ma 9.7 0.8 10 14.5 1.2 14.9 12.1 1 12.5 14.5 1.2 14.9 19.4 1.6 19.9 26.6 2.3 27.4 36.3 3.1 37.4 31.5 2.7 32.4 38.7 3.36 39.8 46 3.9 47.3 29 2.5 29.9

24.2 2.1 24.9 26.6 2.3 27.4 31.5 2.7 32.4 35.3 3.1 37.4 48.4 4.1 49.8 58.1 4.9 59.8 67,8 5.8 69.7 82.3 7.1 84,7 43.6 3.8 44.8 50.8 4.4 52.3 60.5 5.3 62.3 70.2 6.1 72.2 92 8 94.6

48.4 4.1 49.8 55.7 4.7 57.3 67.8 5.8 69.7 87.1 7.4 89.6

Highland Tank & Mfg. Co.

Test Procedure

1. The vacuum shall be a minimum of 14" of Hg at the start.

2. The test period for each tank shall be that stated in Table 1.

3. Vacuum decreases of less than 5" of Hg during test period indicate that the tank is right. Record vacuum change, i f any, in the space provided on installation checklist.

4. A vacuum decrease of more than 5" of Hg during test period required further investigation.

5. I f further investigation is required, the tank manufacturer shall be contacted. After investigation is complete, this test shall be repeated.

6. This test procedure meets the EPA requirements for tightness testing of operating underground storage tanks containing product.

15

APPENDIX - ALTERNATIVE EPA EVALUATION FORMS

Results of U.S. EPA Alternative Evaluation Interstitial Monitoring Method

This form documents the performance of the interstitial monitor described below. The evaluation was conducted by the equipment manufacturer or a consultant to the manufacturer according to the U.S. EPA's requirements for alternative protocols. The full evaluation report also includes a report describing the method and a description of the evaluation procedures, and a summary of the test data. The results forms were modified from the Vapor-Phase Out-of-Tank Product Detectors. The evaluation procedures are included in Attachment A of this report.

Tank owners using this leak detection system should keep this form on file to prove compliance with the federal regulations. Tank owners should check with State and local agencies to make sure this form satisfies their requirements.

Method Description

Name Highland Tank & Mfg. Co.

Version

Vendor Highland Tank & Mfn. Co.

One Highland Road

Stoystown, PA 15563

Detector output type: (x ) Quantitative ( ) Qualitative

Detector Operating Principle: loss of vacuum over time

Detector Sampling Frequency: ( ) Intermittent (x ) Continuous

Evaluation Results

The detector described above was tested for its ability to detect losses in vacuum over a period of time. The following parameters were determined:

Accuracy - Ability of the detector to respond to small leaks.

Detection Time - Length of time required to detect a leak of known size.

Lower Detection Limit - Smallest leak which could be readily detected.

Ambient Conditions - Effects of temperature and product type on behavior of the leak detector.

Criteria for Declaring a Leak

The tank is declared to be tight when the vacuum decreases less than 5" of Hg over the test times

specified in Table 1. The vacuum prior to the test must be a minimum of 14" of Hg.

Interstitial Monitor - Results Form

Interstitial Tightness Testing Method

Version

Highland Tank & Mfg. Co

Page 1 of2

Compiled Evaluation Results

Tank sizes range from 500 gallons to 20,000 gallons. Interstitial volumes range from 1.8 gallons to 18.2 gallons. Test times to detect 0.05 gal/hr and 0.1 gal/hr liquid leaks are shown in Table 1 for water, gasoline, gasoline/ethanol blends up to and including E85 and diesel fuel.

Accuracy - System has a probability of detection of 100% for leaks of 0.1 gal/hr or greater when all of the testing criteria are met. The false alarm rate for a tight tank is <5%. It is impossible to maintain a steady vacuum i f a leak is present.

Specificity - This test procedure is intended to detect liquid product or water leaks.

Lower Detection Limit - Est 0.01 gal/hr with an extended test time.

>Safety disclaimer: This test procedure only addresses the issue of the method's ability to detect leaks. It does not test the equipment for safety hazards.

Certification of Results

I certify that the interstitial monitor was installed and operated according to the vendor's instructions. I also certify that the evaluation was performed using methods which meet the requirements of the alternative EPA test procedures as they are applied to interstitial monitors and that the results presented above are those obtained during the evaluation.

H. Kendall Wilcox, President H. Kendall Wilcox, Inc. (Printed Name) (Organization Pertbrming Evaluation)

Independence, MO 64055 (Signature) (City, State, Zip)

September 14. 2015 (816) 195-1991 (Date) (Phone Number)

Interstitial Monitor - Results Form Page 2 of2

4535 Elizabethtown Road 717-664-0600 y Quality and Service Manheim, PA 17545-9410 fax 717-664-0817

Periodic Interstitial Test Procedure

1. Establish a minimum vacuum of 14" of Hg on the tank's interstitial space through the monitor pipe.

2. Hold the vacuum on the interstitial space for the minimum time specified in Table 1. 3. Check the vacuum after the time specified. A decrease in vacuum of less than 5" Hg indicates the

tank is tight. Record the readings on the chart on the reverse side of this page. 4. A vacuum decrease of more than 5" of Hg during test period requires further investigation. 5. If further investigation is required, the tank manufacturer shall be contacted. After investigation is

complete, this test shall be repeated. 6. This test meets the EPA requirements for tightness testing of operating underground storage tanks

containing product.

TABLE 1 Tank Size Vacuum Hold Time

Diameter Volume (Hours) (in) (gal) 48 550 5 48 1000 8 64 1000 7 64 1500 8 64 2000 10 64 3000 14 64 4000 19 72 4000 17 72 5000 20 72 6000 24 84 4000 15 96 4000 14 96 5000 17 96 6000 19 96 8000 25 96 10000 30 96 12000 35 96 15000 43 120 8000 23 120 10000 27 120 12000 32 120 15000 37 120 20000 48 126 10000 25 126 12000 29 126 15000 35 126 20000 45

Stoystown, PA / Monhelm.PA / Waterviiet, NY / Greensboro, NC

Highland Tank Underground Tank Installation Records As required by the NCDENR

Periodic Integrity Testing

Refer to vacuum test procedure and time duration chart on opposite side-Interstitial vacuum test require prior to startup, six to twelve months after startup and every three years thereafter.

Vacuum Established Vacuum test Completed By Conqjany

Vac (in. hg) Time/Date Vac(m.hg) Time/Date

By Conqjany

4535 EBzabethtown Road 717-664-0600 ched Quality and Service Manheim. P A 17545-9410 F A X 717-664-0617

Interstitial Leak Detection Sensor Installation

May 4,2015

Highland Tank's Double walled tanks provide a 3" interstitial monitoring pipe which is mounted vertically through the tank to access the bottom most area of the tank interstice. Tanks arrive from the manufacturer with a presure gauge attached showing the tank's interstice is still holding a vacuum. The interstitial monitoring pipe should also be labeled and have a factory installed vacuum gauge.

1. The Veeder Root Position-Sensitive #794380-323/333 sensor can be installed once the double walled tank has been backfilled. Relieve the vacuum in the tanks interstice by opening the factory installed valve. The reading on the pressure gauge should now read zero.

2. Remove the factory installed pressure gauge and cap.

3. Inspect all connections to the monitor opening to see if they are water and air tight. Use pipe dope for sealing i f needed.

4. Add the underground tank's internal diameter to the length of the riser pipe installed to determine the overall length. Take note of this dimension for later. This length will be used to confirm the sensor is seated at the bottom of the tank interstice.

5. Use Veeder-Root's installation directions to properly prepare the sensor for installation.

6. Use the length detennined in step 4 and measure this distance from the sensor bottom up the sensor cord and mark clearly wifri permanent visible marker. This marker will help show the sensor is sitting at the bottom of the tank interstice.

7. Slowly lower the sensor down the interstitial monitoring pipe till bottom is reached. The sensor has reached the bottom of the tank when slack is noticed in the sensor cord. The permanent visible marker will also be at the top of the monitor pipe riser. This will confirm the sensor is located at the bottom of the tank's interstice.

8. The sensor is now ready to be connected to interface the XLS monitor. Attach the sensor cable to the TLS monitoring system using Veeder Root's installation instructions

9. A warning alaim, "Sensor Out", should appear on the TLS sensor monitor i f the sensor should lose contact with the interstitial tank bottom.

stoystown. PA / Manheim. PA / Waterviiet, NY / Greensboro. NC

Annual Leak Sensor Testing

Veeder Root interstitial leak sensor must be removed and for "sensor out" and "normal" alarm conditions. Sensor must then be re-installed in accordance with the Interstitial Leak Detection Sensor Installation instructions on the opposite side.

SENSOR CHECK DATE BY COMPANY