Tank Calibration

3 ( 7 5 2 6 3 ( & 7 ,2 1 2/27 Crompton Road Rockingham Western Australia 6169

3 W\/ WG

A CN 063 402 424

Tel: (08) 9528 1888 Fax: (08) 9528 1588 [email protected]

WHAT YOU NEED TO KNOW TO HAVE A TANK CALIBRATED.

1/12


3 W\/ WG

A CN 063 402 424


What you need to know to have a tank calibrated Many people who are in the position of having to have a tank calibrated have little knowledge of what is involved in the calibration. Some may not even know why the tank needs calibrating at all. To help you assess your requirements we have prepared this general outline of tank calibration.

Why do tanks need calibrating at all? If you do not need to know what is in, or what has gone out of, a tank and if spillage or overfilling is of no concern, then you probably do not need to have your tank calibrated. If you require any indication of quantity then you need a calibration of some sort. At its simplest, a calibration of any vessel requires knowledge of the nominal dimensions of the vessel from which the volume may be calculated. For a water tank this may be enough and you probably will not need to call in a specialist. At the other end of the scale, if you are selling a product on the basis of tank levels, a tank can be measured precisely and calculations can be made to take into account factors such as hydrostatic expansion of the tank during filling and thermal expansion of the tank shell in service. For most situations in Australia where you are selling from a tank or where customs duties are levied on tank contents you will need to have a tank calibrated by a suitably certified organisation. Most tank calibrators are certified either by the National Association of Testing Authorities (NATA) or by the local state government trade measurement department. For the purposes of this exercise we shall assume that you have decided that you do require a precise calibration of your tank.

2/12


3 W\/ WG

A CN 063 402 424


What are the options in calibrating a tank? For any tank, whether it is a stationary tank or a road, rail or marine tanker, there are two broad options. It can either be calibrated volumetrically, ie by adding or withdrawing measured volumes of liquid, or by physical measurement of the tank shell and calculation of the volume. In some cases it is clear which method is appropriate and in others either method can be used. On some occasions a combination of both methods may be required. Generally, volumetric methods are suited to smaller tanks or tanks that are irregular in form, and where mathematical models cannot be used. Volumetric methods include the use of calibrated liquid meters and certified volumetric provers. Generally, but by no means always, road tankers are calibrated by volumetric methods, railcars can be done either way, whilst large vertical tanks are calibrated by physical measurement. Food product, brewery and wine vats can be done either way or sometimes by a combination of the methods. We will consider volumetric and physical methods of calibration separately.

Calibration of Tanks for Petroleum and Petroleum Products to ISO, IP and API Standards The International Standards Organisation, the Institute of Petroleum and the American Petroleum Institute have a range of standards to cover the calibration of vertical, horizontal and spherical fixed storage tanks as well as rail tankers and barge tanks. The applicable standards are as follows:

VERTICAL TANKS • • • • • • • •

ISO 7507-1 ISO 7507-2 ISO 7507-3 ISO 7507-4 ISO 7507-5 IP 202 Part II API 2550 API 2555

Strapping method Optical Reference Line Method Optical Triangulation Method Internal Electro-Optical distance ranging method Optical Triangulation Method Section 1 - Vertical tanks Upright Cylindrical Tanks Liquid Calibration of Tanks

HORIZONTAL TANKS • • • •

IP 202 Part II IP 202 Part II API 2551 API 2555

Section 2 - Horizontal and Inclined Tanks Section 3 - Liquid calibration methods Horizontal Tanks Liquid Calibration of Tanks

3/12


3 W\/ WG

A CN 063 402 424


OTHER TANKS & MISC STANDARDS • • • • • • •

API 2552 Calibration of Spheres and Spheroids API 2553 Calibration of Barges API 2554 Calibration of Tank Cars Section 4 Calibration of Spheres and Spheroids Section 5 Calibration of Ship and Barge tanks Section 6 Calibration of Road and rail vehicles ISO 7507-6 Recommendations for monitoring, checking and verification of tank calibration tables (technical report)

Vertical Tanks Although there are differences in the application of some of the above standards there is a lot of duplication. Even where there are differences in the measurement or calibration methods, the difference to the end user is often less than any measurement errors involved, either in the calibration or the measurement of the product. In Australia, being a nominally metric country, it makes sense to use the ISO standard for upright cylindrical tanks. ISO 7507 is essentially a revision of IP 202 Part II, Section 1. The only significant difference is in a calculation that involves temperature correction during tank calibration. The ISO and IP methods have the advantage over the API method chiefly because they are more directly applicable to computerised calculation. There are two methods in common use for the physical (measurement) calibration of the shell of vertical tanks, physical “strapping” and optical strapping. There is a third method available, electro-optical distance ranging, which has not yet seen wide use in this country. The original method is by physical strapping. Strapping is the process of measuring a tank by physically measuring the tank strakes with a calibrated steel tape at several levels on each course of plating. Strapping is a good method for small tanks that are scaffolded or where there is good access to the shell of the tank at all levels. The other method often employed in Australia is the Optical Reference Line Method. This method involves strapping the tank just once at an accessible level, called the reference level. In this method, an “optical plummet”, a device used to establish a true optical perpendicular, is used to measure the radial offset at several levels on each course of plate. This procedure is repeated at approximately four metre intervals around the tank and the circumference at each level is calculated from the sum of the differences between the reference level and the measured level. The optical reference line method has a lot of advantages over strapping. It also has a couple of disadvantages. The advantages of the optical method are: Optical measurement is safer than strapping. There is no need to access the outside of the tank, which for physical strapping is normally carried out from a bosuns chair or with a “cherry picker” type personnel hoist. The optical plummet operator sits at the base of the tank whilst an assistant manoeuvres a magnetic trolley on the tank shell from behind the hand rail on top of the tank. The

4/12


3 W\/ WG

A CN 063 402 424


magnetic trolley is fitted with a scale that can be read through the optical plummet to a resolution of +/- 1mm. Optical calibration is faster than strapping in most cases, with less impact on other operations in the area. A typical 20-metre diameter tank could have the shell measured in only a few hours. There is better traceability of the field data from the calibration. The data from which the tank is calibrated is recorded for later calculation and verification, as are calibration records of the optical plummet. With strapping there is the problem of the tape sagging during measurement and the real difficulty of repeatable measurements whilst the operator is suspended off the side of the tank, trying to maintain 50 newtons of tension on the strapping tape. As a side benefit of the optical method, the data from which the circumferences are calculated will also yield information on the shape of the tank shell. It is a simple matter to process the data to determine the verticality of the tank shell and to determine the “roundness’ of the tank. This is essential information to the engineer who wants to fit floating blankets or roofs to vertical tanks. It is also information that can be useful in determination of any settlement or subsidence of the tank in service. It is simple to optically calibrate a lagged tank. After measuring tank diameters internally, the plummet can be used inside the shell. It is probably possible to calibrate a tank in stronger wind conditions than is safe for strapping although in very strong winds the magnetic crawler used to position the scale on the side of the tank may be blown off the tank. We say only probably here because we do not strap tanks where it is necessary to use a bosun’s chair because of concerns over operator safety.

The optical method has some disadvantages: The equipment used is more expensive than that required for strapping. Optical plummets cost thousands of dollars and are delicate instruments. Even with optical equipment, it is still necessary to use strapping tapes for the reference measurement. It is difficult to use the optical plummet in wet weather, as rain on the lenses makes vision difficult. The existence of scaffolding can prevent the effective use of the magnetic scale used for optical measurements. Strapping is often indicated in this case. Calculation of the tank volume tables is more complex, although with computerised calculation routines, this is no longer a problem.

Other methods of tank shell calibration There are other methods that are being developed for the calibration of tank shells. ISO Standard 7507-3 covers the method of calibrating tank shells by optical triangulation, which requires the use of two theodolites and a laser beam to target the same measurement station. This is a much slower method that the optical reference method, requiring two trained operators

5/12


3 W\/ WG

A CN 063 402 424


and two sets of equipment. It has application in situations where there is no access to tank roofs for hauling a magnetic crawler. ISO Standard 7507-4 & 5 cover electro-optical distance ranging methods, both internal and external. The external method of electro-optical distance ranging requires a minimum of 5 tripods and tribrachs (adjustable mounts) so that the device used can be resited around the tank to access the entire outside of the shell. The method also requires access to the top of the tank shell to fit retro-reflective tape targets. The equipment is fragile, costly and bulky and probably has more application in locations where tank calibration is both local and a daily business. In comparison, a typical optical plummet kit can be fitted into a small suitcase and requires only one tripod. Any of the electro-optical distance ranging methods require that the device be sited on a stable platform. With the current technology, this method is unsuitable for use on tank floors, as currently available equipment will be affected by movement and vibration of the floor with unquantifiable effect upon the data. The use of weights to stabilise the floor, as suggested by the Standard, is both impractical and ineffective. Automated laser systems that are sited on the tank floor to map the inside of a tank cannot easily cope with floating roofs or blankets. We have used air powered magnetic tank crawlers where access has been a problem and we have a 12 volt electric tank crawler under development. There is a safety consideration in the use of these devices to prevent injury to operators from the crawler should it fall from the tank. Suitable protection for the operating technician should be used.

Calibration of tank bottom volumes There are two methods that can be used to calibrate the volume below the dip-plate in a vertical tank: The tank floor profile can be surveyed physically, using one of the following tools: • an engineer’s level or theodolite and staff • a laser plane and survey staff • a water tube or hydrostatic level tool From data obtained the volume can be calculated mathematically. The tank floor can be calibrated volumetrically, using a meter or volumetric prover and water. As before, each method has advantages and disadvantages.

6/12


3 W\/ WG

A CN 063 402 424


The advantages of the physical survey are: Physical survey is faster, cleaner and does not require the availability or require the removal of large volumes of water. A typical physical survey on a 20-metre diameter tank may take only a couple of hours. To fill a 20 metre tank with a bottom volume of 30m3 may take two or three hours to fill and most of a day to empty and clean. A tank of 35 metres diameter and a bottom volume of 100 m3 may take 5 or 6 hours to fill, and a day to empty and clean. The physical survey would still take only a couple of hours. During a physical survey it is simple to measure the precise relative height of all tank internal structures. These heights can be related to the dip plate to give the tank operators accurate figures for when tank structures are covered with product. The data obtained during physical survey can be processed to give the tank owners data on floor profile, tank tilt and settlement. This can be invaluable for comparison with previous or subsequent data to determine tank shell settlement and stresses. Physical survey data from the bottom survey can be related to external measurements during tank hydrostatic testing to determine ground loading response. Physical survey is less intrusive than metered water methods. The tank is immediately available for use after the calibration and does not require any cleaning after removal of the water. (Most water from terminal fire systems will leave a thin layer of mud or sediment on the tank floor).

The advantages of volumetric calibration of tank bottoms are: For extremely irregular tank bottoms, or for tanks with extreme tilts, a volumetric calibration can be more accurate. The accuracy of a physical survey can be improved by use of more measurement points, but under some circumstances, a volumetric calibration may be the best option. If the tank is not safe to enter, it is sometimes possible to calibrate volumetrically if the dip plate is visible from the manhole, or even if it is not, by dipping the tank whilst filling. It is sometimes thought that a volumetric calibration is more accurate than a physical survey because of tank floor sponginess or spring. Our experience and information that we have from overseas is that a water column of 2 metres is required to eliminate 90% of tank floor spring. In reality it not practical to fill any but the smallest of tanks to this level, and floor spring is not a major problem in small tanks. We have performed some practical research on a number of 30 metre diameter tanks in Western Australia, calibrating by both meter and physical survey. We have found the results to be within 1% in all the tanks concerned, which amounted to less than 1000 litres. We have found that it is difficult to determine water levels in a tank to within 3mm during a water bottom calculation, due to meniscus/surface tension effects on the dip plate and to splashing due

7/12

3 ( 7 5 2 6 3 ( & 7 ,2 1

3 W\/ WG

2/27 Crompton Road Rockingham Western Australia 6169

A CN 063 402 424


to the operator being in the tank to observe the filling process. The previously mentioned 1000 litres equates to 1.4mm in a 30 mtr diameter tank. In some circumstances it can be unsafe to work in a tank during water bottom calibration, due to gases from contaminated water. In these circumstances it is not possible to obtain relative height measurements from the water surface to any tank structures such as inlets, outlets etc.

The implications of tank entry for tank bottom calibration In recent years there has been a re-evaluation of the risks associated with confined space entry. Australian standard AS 2865 –1995 applies to any work carried out in a confined space under a breathable atmosphere. This standard applies to the requirements of any tank calibration except for petroleum tanks that have contained a leaded product. Tanks that have contained a leaded product, even if they contained lead free product for several years, are covered under the requirements of the lead manufacturers, (the Octel or Ethyl corporations). These requirements may be reviewed but currently are for air supplied breathing apparatus for both the person entering the tank and the tank watchman. AS 2865 requires that a Confined Space Entry Permit be completed prior to entering the tank. There are also requirements for hazard identification, risk assessment, risk control measures, tank isolation, atmospheric monitoring and training of all persons working on or within the confined space. Confined space entry is one of the biggest industrial killers in Australia and should be taken very seriously. Unfortunately, to calibrate the bottom of any ground based vertical tank (ie, one that is not on legs) with any degree of certainty, it will be necessary to enter the tank at some stage. A water bottom calibration probably requires the shortest tank entry but confined space entry considerations will still apply. With a laser plane or a hydrostatic level device a physical bottom survey will require only one person to enter the tank. Australian standards require that one person be on stand by as watchman while the technician enters the tank.

Calculation of tank capacities for vertical tanks • • • •

Tank capacity tables are calculated from tank circumferences. The following factors are taken into consideration during calculation: Tape rises during strapping (ie. vertical welds or laps, manhole doublers and other obstructions). Shell temperature and tape calibration temperature during strapping. Physical characteristics of the tank shell material (Young' s modulus, thermal expansion factors, etc.)

8/12

3 ( 7 5 2 6 3 ( & 7 ,2 1

3 W\/ WG


• • • • • •

A CN 063 402 424


Difference in temperature between calibration and working conditions. Paint and plate thickness, inside and outside tank. Tank expansion during calibration (if the tank is not empty). Tank expansion from liquid head pressure in service for the density of product stored. Tank tilt Effects of positive and negative deadwood on tank volumes.

What conditions required for a vertical tank calibration? Most fixed roof vertical tanks can be calibrated in one working day, if done by physical measurement. Large floating roof tanks may take up to two days. There are some other considerations: If the tank is new, all the standards require that it should have a full hydrostatic test prior to the calibration. The tank shell coating should be cured, as the strapping tape or the wheels of the magnetic trolley may damage a soft coating. (Some trolleys have narrow alloy wheels that will even damage a cured coating. Our trolleys have soft plastic wheels to minimise this problem). Ideally there should be no other work being performed on the tank, as welding, grinding, hammering and sandblasting all make the shell vibrate, which makes optical measurements difficult. However, we all exist in the real world and if the time window is tight (and it always is) we can usually find some way to work with or around other operators. The tank should be clean and gas free for entry. Most terminals have their own requirements for confined space entry and issue their own work permits. We have our own atmospheric monitoring equipment and we have training in gas free inspections. We will issue our own permits if there is no satisfactory system in operation. If the tank is not clean and gas free for entry, there are alternative options for bottom calibration as discussed elsewhere.

When should a vertical tank be recalibrated? A tank may change its calibration whenever the operating conditions are changed. This could be for a variety of reasons. For example: • •

Change of product density - this will affect the expansion characteristics of the tank, hence the volume. Changes of operating temperature - most tanks are calibrated for a shell temperature of 15 degrees. A large change of operating temperature will alter the tank increment (ie. its ltrs/mm). It is worth mentioning here that we have re-calibrated several large crude tanks in the North West of Australia for a realistic working temperature (in this case 38°C), where the

9/12


3 W\/ WG

A CN 063 402 424


tank had been calibrated for a standard temperature of 15°C. The result was a difference of 22,000 litres in tank volume, which had previously been given away with each tank of crude loaded onto a ship. • •

Any modifications to the tank or the dip plate - i.e., new pipe inverts, fitting a stilling well, floor repairs, floating roof modifications, changes in deadwood, etc. Settlement of the tank with resultant changes in the tank geometry.

It is generally recommended that a tank should be recalibrated at not more than 10 year intervals, especially considering tank settlement. In areas where tank settlement or ground subsidence is known to be a problem, an external settlement survey can be performed without tank entry, to marked positions on the external protrusion of the annular plate. These heights can be referenced to a datum level in the terminal and also to the internal tank calibration data.

The calibration of horizontal tanks Horizontal tanks can be calibrated to either the API or the IP standards. There is little difference in the end result. The calculation routines for both standards have been around for a long time and neither one takes advantage of the ability of computers to perform laborious calculations. The API standard in particular makes extensive use of tables for various corrections, rather than providing formulae. We use a method that is traceable to the API Standard, but exceeds the precision. Through the use of formulae, we can process a greater degrees of tank tilt and greater ranges of head shape with much better accuracy. The standards make several assumptions on the volumes of tank heads, where we are able to calculate volumes precisely. Horizontal tanks can be calibrated either by volumetric methods or by physical measurement. Cryogenic and pressurised tanks can be accommodated.

What is involved in the volumetric calibration of tanks? Most commonly volumetric calibration is performed by adding liquid, usually water, to a vessel in small volumes. The liquid can either be metered in through a calibrated meter or dropped from a calibrated volumetric prover. After the addition of each new volume, the level is measured and recorded. Any corrections required for temperature or meter factor may be applied to the levels and a table or dipstick is manufactured for the tank. No corrections are applied for tank expansion, as this occurs during the filling process. There is a potential problem here for pressurised tanks or tanks holding products having densities different to the measuring liquid. If the tank service conditions for pressure, either hydrostatic or otherwise, are different from the calibrating conditions, there will be a volume error. This can be catered for with a suitable correction table. The problem of similar conditions for calibration and use also applies to tanks that may be installed on different angles from which they were calibrated. If the tank inclination is different from the calibrated condition, there will be errors in the reported volume. Volumetric calibration is ideal for tanks which require dipsticks, as there is no need to apply other corrections, and a dipstick can be made directly from the recorded levels. The process does require a volume of water equal to the tanks capacity to be available during calibration and requires that the vessel be clean enough to allow the water to be disposed of afterward.

10/12

3 ( 7 5 2 6 3 ( & 7 ,2 1

3 W\/ WG


A CN 063 402 424


It can also be quite a time consuming process for a large tank and it may interfere with other work that is being done in the area.

What is involved in the physical calibration of horizontal tanks? Horizontal tanks are generally manufactured in a workshop and transported to site. With physical measurement methods, tanks can be calibrated either in the workshop or after installation, as long as the tank is installed on the designed inclination. Horizontal tanks can be calibrated by either internal or external measurement. External measurement is suited to tanks that are situated above ground, with clear access to the tank shell and heads. Tanks do not need to be empty or clean for external calibration, although it is not generally possible to measure tank deadwood on tanks that are in service. The tank dimensions are determined by a series of circumferences taken along the tank barrel and a by measuring the profile of the heads. We determine the head profile by erecting a reference plane (a string line) across the end of the tank and measuring a series of points across the head. A problem with external measurements on tank in service is that it is difficult to establish the height of the dip plate / striker plate. This is not an insurmountable problem, but it is a source of calibration errors on horizontal tanks. Internal measurement is our preferred method, as it enables more precise measurements to be taken and less assumptions made on the tank construction. A series of internal diameters are measured with an internal tank gauge (a device similar to a large internal micrometer), barrel lengths are measured and the head profile measured with a purpose built head-measuring gauge. Deadwood can be measured precisely, as can the heights of the dip plate etc. Obviously for internal measurement, the tank must be clean and gas free for entry. For either internal or external measurement, the reference height must be measured and if the tank is pressurised, special measurements for tank gauging systems, slip tubes etc., must be taken.

Calculation of tank capacities for horizontal tanks Tank capacity tables are calculated from tank diameters, barrel length and head profile data. The following factors are taken into consideration during calculation: •

Tape rises during strapping (ie. welds or laps and other obstructions).

•

Shell temperature and tape calibration temperature during strapping.

•

Difference in temperature between calibration and working conditions.

•

Paint and plate thickness, inside and out tank.

•

Tank expansion during calibration (if the tank is not empty).

•

Tank expansion from pressure in service.

11/12

3 ( 7 5 2 6 3 ( & 7 ,2 1

3 W\/ WG


•

Tank tilt - this is critical in horizontal tank calibration.

•

Location of dip tube position with regard to barrel.

•

Effects of positive and negative deadwood on tank volumes.

A CN 063 402 424


Notes: In the case of pressurised tanks, tables can be issued for different working pressures or factors can be given for pressures different from the usual working pressure. The same applies for the temperature variations for cryogenic tanks.

What you should look for in a tank calibrator Whilst the mathematics of tank calibration are relatively straightforward, to have a thorough understanding of all the principals and factors involved requires some study. We believe that the person who measures your tank requires a thorough knowledge of the standards, the measurement procedures and the calculations involved ensuring that the correct measurements are taken. NATA is the organisation responsible in Australia for regulating the standards of metrology laboratories. Any tank calibration company accredited to NATA is audited annually to ensure that their quality system and equipment is maintained to the required standards and that the company’s personnel are competent to perform the work. What tends to happen in Australia is that a company may hold NATA registration in one location and use any staff available in another location to perform field measurements. This does not automatically mean that the work performed is shoddy or inferior but it does mean that there is less control over the training of staff, and less traceability over the calibration of equipment and the accuracy of data. The Calibration Company that you choose should also be able to demonstrate knowledge of safe practises in your terminal. In particular, they should be trained in confined space entry procedures and should have the necessary equipment to work safely in confined spaces and at heights. The Calibration Company should be able to offer prompt service. To calculate an average tank calibration table may take eight hours but there is no reason why you should not be able to request a faxed calibration table within 24 hours of the field operator leaving your terminal. You should also be able to expect to receive the finished table within a week. Obviously, we hope that you will come to Petrospection. We believe strongly that tank calibration is an important part of terminal stock control and we treat it accordingly. We have been calibrating tanks for many years throughout Australia, New Zealand and the Pacific region. Petrospection is committed to providing NATA accredited personnel for all aspects of tank calibration. We will not send out inexperienced or unqualified operators to measure tanks. If you choose to use Petrospection for your tank calibrations, the field operative that you see calibrating the tanks in your terminal will be able to discuss all aspects of the calculation of your tables and answer all your questions.

12/12

Tank Calibration

Recommend Documents