(a discussion document)

Introduction

This discussion of relative merits is offered in an effort to help the potential user of multi level water sampling systems to be aware of the factors important to the performance of a multi level sampling system.Each of the several systems available has advantages that may or may not be important to the application.  Likewise, there are some significant disadvantages, such as cost, for some of the systems compared.  However, it is often easy to show that a multi level sampling system is a good value in the remediation and monitoring application compared to cluster or nested wells.

We build and sell FLUTe systems, we believe that they score very high in this comparison.  However, we are happy to let the facts and the user determine the best design for the application.  No single system is the best for all applications.

Any multi level sampling system on the market claims to be able to obtain samples at discrete levels. Those which do not provide vertical isolation in a hole or well (e.g., the micro purge, diffusion samplers, or bailer methods) are not considered legitimate multi level sampling methods, since they require the absence of any vertical gradient in the hole. Those are better considered minimum purge methods.

A multiple level sampling system must provide isolation of each sampling point.  How well the sampling points are isolated is a strong measure of the system worth.  Other factors are also important, such as:

• 1. Does the method (the procedure and hardware) allow contaminant migration during or before emplacement?
• 2. How easily is the system installed?
• 3. Can it be adjusted in the field to the preferred sampling intervals?
• 4. How easily are samples obtained?Are the samples large enough?
• 5. Are purge volumes large? Should they be?
• 6. How many sampling ports can be installed in one hole?
• 7. Must the hole be cased?
• 8. Does the system seal an uncased hole well? How well?
• 9. Is it removable? Easily? Risk of getting stuck =??
• 10. How much does it cost? Per foot, per port?
• 11. Are head measurements easy?
• 12. Is it reliable? Well built? Chemically inert?
• 13. Can one measure the reliability in place?
• 14. Can it be used at:
  a. shallow depths?
  b. intermediate depths?
  c. great depths?
• 15. Does it function with:
  a. shallow water tables (less than 20ft.)
  b. deep water tables?
• 16. Does it require trained operators
  a. for installation?
  b. for sampling?
• 17. What ancillary equipment is required:
  a. for installation?
  b. for sampling?
• 18. Is it reusable?
• 19. Is it easy to decontaminate?

Since the requirement to case the hole limits sampling to only the screened sections of the hole, the more generally useful system can be used in an uncased hole.  There are many holes that are unstable (e.g., in alluvium, below the water table).  The ability to install a multilevel system in unstable holes is a special advantage. This comparison is limited to the stable uncased hole.  There is also a separate question of the seal quality of a grouted, cased hole in some situations such as a Karst formation.

Since shallow wells are relatively inexpensive, the multi level systems are more useful at depths greater than perhaps 50 ft.  Also, the cost per port is much better than separate wells (e.g., cluster wells) at greater depths.

The first requirement is vertical isolation and therefore hole seal quality.  The next most important feature seems to be the initial cost. A third level of concern is the ease of installation (also a cost factor).  A fourth level of concern is the ease of sampling (again a cost factor).  Since there are so many other variables, the order is not so easy to define. It seems necessary that any system allow head measurements.  In some cases, the head measurements are more desirable than the sampling capability.

The multi level sampling systems considered in this discussion are the Solinst/Waterloo, Solinst/CMT, Westbay, and FLUTe systems.  (FLUTe is a trademark for Flexible Liner Underground Technologies.)  The first three are "pipe" and packer systems, but the first two are sometimes grouted in. The FLUTe system is a continuous packer/liner system.  For more information on these systems, contact the manufacturers.  Nested well systems have been around as long as water wells and are not considered in this discussion.  Beyond 2 sampling intervals, nested wells are very dependent upon the ability to obtain a poured seal with many pipes in the hole.  Furthermore, nested wells are not removable as the list above may be.  One should consider all options including single wells and drive points.

Vertical Seal

The two approaches of the commercially available systems for sealing the hole are straddle packers (the traditional approach) and the continuous liner (the FLUTe approach).  In a pipe, the straddle packers are probably excellent seals.  In a borehole of fractured rock, or alluvium, with vertical flow in the formation, the straddle packer approach is more uncertain, since the flow paths in the medium determine whether the packers are a good seal.  Those flow paths are often unknown.  Also, the straddle packers must be located in a relatively smooth and circular portion of the hole to seal at all. A paper available under Publications addresses the question of packer impedance to flow versus the liner impedance and compares that to other sealing methods.

The FLUTe technique uses a flexible liner (usually a coated fabric) to seal the hole much like a continuous packer throughout the entire hole.  The continuous liner is larger than the hole diameter to assure a seal in the nominal bit-size portions of the hole.  Rough, or washed out, portions of the hole are not well sealed by a continuous liner, but all other portions are relatively well sealed.  Even in the rough walled sections of the hole, the inflated liner offers impedance to vertical flow.  A video is available of the interior of a FLUTe liner sealing a hole in fractured rock (ask for it at ).  In uncased holes, we believe that the continuous seal is better than the discrete packer approach.

The effect of vertical leakage with either type of seal is dependent upon the amount of leakage or the leakage rate.  None of the sealing methods are likely to be perfect or absolute seals.  For head measurements, the vertical leak rate must be much less than the horizontal flow rate at the level of the head measurement.  In other words, the head at the measurement level must be dominated by local conditions. 

For sampling purposes, the vertical seal must be much better in that the introduction of contaminants at the sampling level via vertical flow in the hole is not allowed.  However, it is not necessary for the impedance of any vertical seal of the hole to be greater than that of the geologic material which was removed in the drilling process.  It is certainly not desirable to have the vertical flow in the hole to be such as to allow the vertical spread of contamination preferentially via the hole.

It is clear that the seal required is medium dependent, but it is not good to have a long open hole in an uncased medium even under circumstances of presumably well known medium properties.  Medium properties are usually not well known.  This factor is important in the ranking of multi level system sealing capabilities.  We believe that the FLUTe continuous liner and installation procedures (there are several) are especially well suited to Karst conditions.

Ease of installation

A continuous coiled system is usually easier to install directly into the hole than one made up from a pile of piping and components.  The FLUTe system is shipped on a reel from which it can be directly installed into the hole.  The Solinst/CMT system is coiled in shipment, but must be extended on a clean surface near the well for assembly prior to installation.  The Solinst/Waterloo system and the Westbay systems are assembled from piping and lowered into the well as they are built.  An inherent advantage in a system that is fully assembled in the factory is that the seals of all components can be checked. A system assembled in the field is usually difficult, if not impossible, to leak check in the assembled state in the hole.  Some kinds of leakage may be easy to assess (e.g., the packer integrity), but the sampling system plumbing is much harder to assess in the field.  The possibility of field assembly leaks becomes worse in very cold and unpleasant field circumstances.  The FLUTe system is fully leak checked after assembly in the factory prior to shipment.  Any leakage from the FLUTe liner is easily measured by tagging the water level in the liner after installation.  There are several easy remedies to leakage of a FLUTe system if leakage is observed.

Ease of sampling

There are many methods of sampling used in multi level systems.  The most simple is the bailer approach in which a valved tube is lowered into the tubing to capture a sample.  Next is the "grab sample"  using an evacuated chamber and a needle or valve to allow the water to be drawn into the chamber.  An inertial pump system uses a check valve and motion of the "chamber" (usually a pipe) to catch the water and pump it to the surface.  A peristaltic pump is often used to "suck" the water from the well. (This is not popular with most regulators for volatiles.)  Another method is to force the sample to the surface via a driving gas pressure.  Such pumps are the single and double valve pump systems and the bladder pump system. The bladder pump does not allow the contact of the driving gas with the sample water, and hence, it is less likely to aerate the water sample.  However, the aeration of the sample water can be avoided by the proper procedure and pump design for the single and double valve systems.

In most cases, the system must be purged of the water that is in the sampling system and in the well bore isolated by the system.  The grab sample technique assumes that the water in the well bore is in equilibrium with the formation.  The grab samples are relatively small compared to what exits in  the open well.

The ease of use of the several sampling procedures depends upon several important factors.  One is the rate of recharge in the system from the formation.  Another is the depth of the water table.  Below about 20 ft., the peristaltic pump is not an option.  Below 50 ft., the inertial pumping system is a lot of work.  This is not intended as a discussion of the many pumping methods.  However, it is useful to know that some systems allow many more pump and sample collection procedures than others. 

Since the time spent in the field sampling is a direct labor cost to the system use, the labor-intensive techniques are not attractive.  The same is true for the equipment intensive systems.  If very special training is needed, and/or the samples can be violated by the complexity of the sampling procedure, it is also not attractive.

The relative ranking of the several systems should be done according to the sampling method, since several methods are possible with most of the sampling systems.

The FLUTe system has the relatively unique capability to sample a large volume per stroke from a port located just beneath a deep water table (e.g.,100-1000 ft) in a deep hole.  FLUTe also has a staged pumping design for sampling at several thousand-foot depths with relatively large flow rates.  FLUTe also has the ability to sample all ports simultaneously with a common pumping pressure.  This reduces the sampling labor to little more than that of a single port.

Field adjustment of the sampling interval

Traditionally, the common water well screen can be located wherever the driller and his customer decide to locate it.  That is also possible with some of the multi level systems if the proper parts are in hand for the installation.  The straddle packer systems are built in the field from the parts in hand. 

The FLUTe system continuous liner is not readily built in the field.  More often, the borehole is sealed with a blank liner immediately after the drilling is complete and the proper liner is ordered after the core has been examined and any geophysical logging runs have been made. In fact, many holes can be drilled in this manner before the sampling levels are selected.  The blank liner adds to the cost, but it also minimizes any vertical migration while the sampling levels are being selected. The blank liner is removed and the sampling liner is installed relatively quickly.  This accumulation of holes to be sampled, sealed or otherwise used, allows a great deal of economy when drilling and when bringing the logging trucks to the site for one series of measurements in the several holes. This is possible with the use of the blank FLUTe liner.  The liner is fully reusable in other holes.

Purge volumes

None of the systems considered here have large purge volumes associated with the sampling.  However, only the FLUTe system allows the borehole to be thoroughly purged while the system is being installed. The straddle packer systems have a large volume of water in the borehole between packers (large relative to the sample volume).  This captured volume is subject to alteration by the remaining drilling fluids in and near the hole.  For the FLUTe system, essentially all of the water in the hole is inside the liner and is not in contact with the sample water.  Hence, it does not need to be purged.

Number of sampling ports

The number of sampling ports on the Westbay system is without limit, since a port is simply added to the pipe string with the associated packers for isolation.  The FLUTe system allows as many sampling tubes as will fit in the hole, and so it is hole size dependent.  However, 6 ports were installed in a three-inch hole and 15-20 are possible in a 6-inch hole.  The Solinst/Waterloo system and the Solinst/CMT system are limited to 6 and 7 ports respectively.  The cost increase per port is very dependent upon the system.  An interesting number is the cost per ft per port as a function of depth or number of ports.  These numbers are about $6-7/ft/port for deep FLUTe systems with many ports. The prices are unknown to us for the other systems.  Clearly, these are very competitive numbers for comparison with cluster wells.

Leakage

Leakage is a concern for all of the multilevel systems.  (It is naive to assume that leakage is not a concern with cluster wells.)  Under the best circumstances for each system, it can provide a very good seal of the hole against vertical leakage. Since all of these systems depend upon a pressurized bladder/liner for the seal, the pressure continuity must be a reliable feature of the design.  In other words, the pressure system must not leak (i.e., the joints in pipe sections, coating of liners, the connectors of all tubing.)  We do not know what the fall back position of the several other systems is, but the FLUTe liner is easily filled with a sealing material (e.g., a Bentonite slurry) if a leak should be detected in the basic liner (or the liner is easily removed for repair). Tests have shown that the seal of a small leak can be very effective with a lean Bentonite slurry that does not prevent the easy retrieval of the liner in the usual inverting manner.  Leaks in FLUTe liners are surprisingly rare.

Removable

It is clearly an advantage to be able to remove a multilevel sampling system for a variety of purposes ranging from other use of the hole to repair, or for repositioning of the sampling intervals.  Removal is often associated with the risk of getting stuck. 

By the nature of the everting FLUTe liner installation and inverting liner removal, the entire hole is supported against collapse while the liner is in place.  Even when the liners were incorrectly removed, and the inversion procedure was frustrated, the liners could be collapsed by pumping out all of the water in the liner, and the liner hauled from the hole like a rope.

Straddle packer systems are emplaced and removed by raising or lowering a string of packers in the hole. The chances of a hole slough trapping a packer is much higher than the chance of trapping an inverting liner.

Partially obstructed holes

In some situations, there are partial obstructions in the holes to be sealed.  A surface casing smaller than the drilled hole below the casing is a common situation in domestic wells.  In other situations, the hole may have suffered a partial collapse that leaves the hole at less than its original size.  Only the FLUTe system can be deployed into a hole of half the diameter of the liner, as long as the tubing bundle (controlled by the number of ports) is not too large. 

The FLUTe system also allows the installation of a sampling liner into a hole already sealed by a blank FLUTe liner.  That allows a hole to be sampled anywhere, anytime, without loss of the sealing and supporting function of the original liner.  (called the Duet technique, patented)

These capabilities do not exist for the straddle packer methods.

Reliability

This characteristic has many aspects, but generally it relates to the durability of the system in its intended function.  Is it well designed to stand the mechanical loads expected in emplacement, during its use, and when removed?  Are the components made of durable materials of long life expectancy or of plastics and elastomers that may age too quickly?  Most multilevel systems contain some elastic sealing materials.  Some use o-ring seals, gaskets, and tubing fittings that must endure for the expected lifetime.  Seals assembled in the field and untested are less reliable than those that are assembled indoors under controlled conditions and tested fully assembled in the system.  Underwater, and in boreholes, there is not the assault of UV radiation except at the wellhead.  Plastic components at the surface must be covered.  Is the system likely to suffer decomposition under the attack of some contaminants?  Some contaminants are very aggressive and attack most plastics and elastomers except Teflon, Viton and similar very expensive materials.  In solution at low concentrations, the threat is usually much less.  Is absorption in the system likely to lead to misleading results?  Some designs and some procedures avoid the cost of totally inert systems. Does the sampling procedure violate the sample quality?

FLUTe provides a PVDF (better than Teflon) tubing system. Extensive data is available on the resistance of the FLUTe materials to common contaminants.

Complexity and the requirement that moving seals must always function generally degrade the reliability.

These are all factors that affect the reliability of the system.  Long use does not by itself justify further use.  PVC water wells have been used a long time, but are still not the best solution.  A measurement with a poor hole seal, if not recognized, might be assumed to be a good measurement of wide spread contamination.  A major weakness of all sampling systems is that it is not easy to prove good isolation in a permeable medium. 

Each manufacturer of these multi level sampling systems should be able to provide an answer to the questions of reliability related to hole seal, fabrication procedures, longevity of components, failure rates of pumping systems, and the like.  We have heard of occasional discontent with all of the systems considered here under certain conditions.  The user must assure himself that those problems have been addressed or are not related to the planned use.  It is useful to remember that the traditional water well, with the longest history of use, is also problematic in some conditions such as Karst formations. 

It is noteworthy that cluster wells are entirely dependent upon the seal quality of the grout job of each well, and many grouted casings have been found to leak.

Horizontal holes

Sliding a packer system into a horizontal hole is difficult, if not impossible, unless the hole is smooth, stable and relatively short.  Installation of the everting FLUTe liners into horizontal holes is common for distances to over 400 ft.  The larger the hole diameter, the easier the installation.  Horizontal holes as small as 3" diam. are relatively easy for everting liners.

Relative merits might be scored in a table for each system.  But the ranking is debatable depending upon the explicit situation. Of more value may be the listing of characteristics that should be considered in the selection of the system best suited to one's own situation.  In that table, one can rank each candidate system in all of the above categories without the problem of ill-defined circumstances.

 

Prepared by Carl Keller, of Flexible Liner Underground Technologies, to aid in the comparison of the several methods and to clarify the FLUTe system characteristics.  Comments are welcome.     PH. 888-333-2433.

Top