Aquifer CS-411 is CS-410 sent to utility ground: the same reel, the same hydraulic straightener, the same clamp-and-push stroke drive, the same fibre. What changes is the head it carries, the fluid it pumps and the article it leaves in the ground.
Reel, straightener, stroke drive and entry guide are the same hardware on both programs. CS-411 differs by head kit, fluid package, completion goods and paperwork.
The straightener carries across without modification and the stroke drive is the same patent-pending machine. A long bore asks it for endurance and an accurate stroke count rather than more force.
A rotary cutting head in place of a push-only nose. A fluid package that is the tool rather than a lubricant. Purchased completion goods, and a regime governing what may be pumped into ground somebody drinks from.
Commonality here is mechanical. A part proved on one application is proved on the other, and a spare on the shelf serves both.
Working axial load in the tube stays in the single-digit kilonewton class for most of a long bore, far below what the drive delivers, so the long-bore duty changes what the machine is designed against rather than what it is. The entry guide bridges the same free span, with a surface conductor casing added for returns.
Everything above the ground line is the machine described on the CS-410 page. Everything below it diverges, and that is this page: a long horizontal bore, three services sharing one tube, a fluid package that does the cutting, an article left in the ground, and a water table to be found and respected. The family sits together on the Underground family page, and the buyer's view is who it is for.
Productive aquifers sit in saturated sand and gravel, the hardest ground a push-only head can meet. Face demand there runs an order of magnitude past what a coil tube delivers.
A 100 mm displacement head working saturated sands and gravels at cone-resistance classes of roughly 10 to 30 MPa asks for something like 79 to 236 kN at the face. The tube carries single-digit kilonewtons at the end of a long bore. The shortfall is a factor of 10 to 50, and no pump on the surface closes it, because the limit sits in the tube.
So the head cuts and the fluid clears. Jetting in saturated granular ground is mature practice at low weight on bit, so the mismatch belongs to the push-only head and not to the mission. That one decision is what makes CS-411 a different machine below the ground line while leaving it the same machine above it.
An HDD-class rotary bit of the Tricone or PDC family, turned by fluid power at the head. A coil tube cannot transmit controlled rotation, so nothing at surface turns or holds toolface.
It relieves the face in saturated sand, holds the bore wall open, and carries cuttings out along the annulus. Sizing it sizes the prime mover, the fuel burn and the trailer.
Returns come back to surface settlement, and what comes back is the only continuous read on the bore. A loss of returns while pushing is the precursor everything is watched for.
Fluid, guidance fibre and sensing conductors share one tube bore. On area they fit with room to spare. A completion article does not fit at all.
One tube goes into the hole and carries everything the head needs. On a 38.1 mm outside diameter tube with a 3.2 mm wall the bore is 31.7 mm across, and the signal path takes under five per cent of it.
| Down the tube bore | Area | Share of the bore |
|---|---|---|
| The bore itself, 31.7 mm inside diameter | 789 mm² | — |
| Fibre in a 6 mm protective conduit | 28.3 mm² | 3.6% |
| Two 2 mm sensing conductors | 6.3 mm² | 0.8% |
| Left for fluid | 754.7 mm² | 95.6% |
Class CS-411A, on a 38.1 mm outside diameter by 3.2 mm wall coil tube.
The finding that matters is about the service missing from that table. A water-well screen is a 50 to 100 mm article against a 31.7 mm bore, so nothing that completes a well travels down the tube; the completion is pulled into the bore behind the string instead. Payload through the tube is true for a sensor node and false for every completion article on the mission list.
The fibre gets the same treatment. A line lying loose in a bore carrying water at several metres a second will abrade and fail, so the tube is bought with the fibre and conductor line secured inside it.
Pilot diameter sets the buckling cap, the flow, the pump, the volume that enters the formation and the completion that can be pulled back through.
The obvious lever is the weak one. Clearance between tube and bore wall sets the load at which the tube buckles, and that cap scales as the inverse square root of clearance, so it moves only 43 per cent across a doubling of hole size. The strong lever is flow: a 150 mm hole needs 3.6 times the flow of an 85 mm hole and sixteen times the pump. Only the article that can later be pulled back rewards going bigger.
| Pilot bore | Sinusoidal onset | Helix formation | Flow | Pump shaft | Pull-back article |
|---|---|---|---|---|---|
| Ø75 mm | 7.39 kN | 10.46 kN | 118 L/min | 30 kW | Ø50 mm |
| Ø85 mm, selected | 6.56 kN | 9.27 kN | 163 L/min | 53 kW | Ø57 mm |
| Ø100 mm | 5.71 kN | 8.07 kN | 242 L/min | 116 kW | Ø67 mm |
| Ø125 mm | 4.82 kN | 6.81 kN | 401 L/min | 346 kW | Ø83 mm |
| Ø150 mm | 4.25 kN | 6.00 kN | 595 L/min | 861 kW | Ø100 mm |
The coupled trade at class CS-411A's tube. Pull-back article is the largest completion the pilot supports at the standard 1.5× clearance rule.
An 85 mm pilot is where the five curves cross. A 75 mm hole is better on almost every count and leaves an 18.4 mm radial annulus with no room for a cuttings bed, so it loses on hole cleaning. A 100 mm hole costs 48 per cent more flow and puts 30 per cent more volume into the formation, the wrong direction on the number a regulator asks for. And 85 mm supports a 57 mm completion pulled back on the standard clearance rule, which turns the completion fork into a sequence: a pilot bore, then a reaming pass for the one mission that needs a larger screen.
Push a coil tube along a horizontal hole and it does not stay straight. It goes sinusoidal, then helical, and the helix changes the drag regime rather than adding to it.
Four quantities set the onset: bending stiffness of 1.077 × 10⁴ N·m², 2.75 kg per metre of steel, a buoyed lateral weight of 23.4 N per metre once fluid inside and around the tube is counted, and the radial clearance. What the onset figures do not show is the step that follows. Fully helical in an 85 mm bore, contact force reaches 70 N per metre against 23 N per metre straight, and goes on growing as the square of the load. Published relations for helix formation spread by 29 per cent, so it is carried as a band.
| Axial load | Axial stress | Bending stress in the helix | Combined | State |
|---|---|---|---|---|
| 5 kN | 14.3 MPa | 20.7 MPa | 35.0 MPa | Elastic |
| 15 kN | 42.8 MPa | 62.2 MPa | 105.0 MPa | Elastic |
| 30 kN | 85.5 MPa | 124.4 MPa | 209.9 MPa | Elastic |
| 45 kN | 128.3 MPa | 186.6 MPa | 314.9 MPa | Elastic |
| 60 kN | 171.0 MPa | 248.8 MPa | 419.8 MPa | Elastic |
| 78.6 kN | 224.0 MPa | 326.0 MPa | 550.0 MPa | At yield |
A helically buckled tube in an Ø85 mm bore, against a 550 MPa yield class.
A tube bent into a helix carries bending stress on top of its axial load, and first yield arrives at 78.6 kN. The tube is not empty either: at working internal pressure, hoop stress runs against the compressive state and moves the cap the wrong way, to 72.1 kN at 150 bar. So surface push is held to half the pressurized cap. Helical bending strain there is 0.149 per cent against a 0.28 per cent yield strain, a limit to respect rather than a cliff.
That rule costs less than it looks, because push is not what bounds a long bore. March the axial load from bit to surface through the real profile, with post-buckling wall contact and residual tortuosity in the arithmetic rather than a footnote, and the answer stops responding to force. Doubling the push buys 9 per cent; another 125 kN on top buys 6 per cent more. Past that the march goes singular and no push of any size reaches the bit. Holding to half the yield cap costs about 15 per cent of the lock-up-limited bore.
Two site properties then move the answer by a factor of three between them: the open-hole friction coefficient in saturated sand, and the residual tortuosity of the path actually drilled. Both are read in the ground. The third term is geometric and fixed. A surface-launched lateral has to reach the stratum first, and at a 12° entry with a 145 m build radius to a 30 m target depth that costs about 159 m of hole and 156 m of surface offset before the lateral begins — the same in every class.
The pilot bore names the class, and everything follows it: the tube, the reel, the push the drive is designed against, and the article the finished hole takes.
The plant is a spread rather than a single vehicle: about 8.2 t dry, 11.3 t operating with fluid in the tank, over roughly 22 m² of deck, the reel and its tube taking 57 per cent of the dry mass. That is a tri-axle plus a support truck, or a two-trailer spread, and it is what a mobilization has to move.
Reaction is the check that decides the spread. An 11.3 t machine on ground at a friction coefficient of 0.40 develops 44 kN of sliding resistance, the same order as the push the smallest class is designed for, and a wet civil site does not offer 0.40. Screw anchors or a deadman are load-bearing, not optional.
| CS-411A | CS-411B | CS-411C | |
|---|---|---|---|
| Pilot bore | Ø85 mm | Ø108 mm | Ø149 mm |
| Ream pass, separately permitted | Ø96 mm | Ø149 mm | Ø216 mm |
| Service completion | 57 mm | DN90 | DN140 |
| Design push | 34.3 kN | 110.3 kN | 237.3 kN |
| Reel core radius | 1.00 m | 1.20 m | 1.70 m |
| Reel flange outside diameter | 2.50 m | 3.18 m | 4.73 m |
| Layers on the drum | 5 | 5 | 6 |
| Reel demountable | No | No | Yes |
Design push is the helix-yield cap with a factor of two on it. The largest class travels as its own load, its reel a demountable module.
Yield strain for coiled-tubing steel is about 0.3 per cent. The tube bends onto the drum at 1.9 to 2.6 per cent and unbends off it again, so its life is counted in trips.
Bend strain across the drum is the tube's outside diameter over twice the core radius, stated here as the full range rather than as the amplitude, so the only lever the reel offers is a bigger core, and a bigger core is mass and transport height. Across the three classes the tube grows faster than the drum does, and the fatigue margin narrows as it goes.
| CS-411A | CS-411B | CS-411C | |
|---|---|---|---|
| Inner-fibre bend strain across the drum | 1.91% | 2.51% | 2.61% |
| Trips available | 27.6 | 16.7 | 15.5 |
| Trips required | 15.0 | 15.0 | 15.0 |
| Margin | 1.84× | 1.11× | 1.03× |
A trip is one deployment and recovery of the string.
The largest class clears its requirement by 3 per cent, which is not a margin that survives a lost count. So the fatigue ledger is load-bearing rather than a record-keeping convenience: the count belongs to the tube, follows it between jobs, and is taken from the reel encoder and the injector cycle counter. The straightener adds to the account too, since driving the tube through it forms a travelling plastic hinge at the centre roller.
That accounting also makes a completion left in the ground attractive on the arithmetic and not only the schedule: it spends about one deployment of tube life. Every re-entry to develop a lateral or service a sensor string spends another.
At the reference depth the sand pushes in with about 153 kPa of horizontal effective stress. A static column of fluid offers about 15 kPa against it. The rest comes from circulating.
Formation pore pressure at 30 m is 294.3 kPa and a static column at a specific gravity of 1.05 stands at 309.0 kPa, so the fluid alone is 14.7 kPa to the good. Circulating adds annular friction, and annular friction accumulates from the toe back to the exit, so the support it buys is not evenly spread.
| Support available | Overbalance | Against 152.8 kPa required |
|---|---|---|
| At the toe, circulating | 188.2 kPa | 1.23 |
| At the heel, circulating | 42.5 kPa | 0.28 |
| Anywhere, pump off | 14.7 kPa | 0.10 |
Horizontal effective stress at 30 m in saturated sand, at a lateral earth-pressure coefficient of 0.5.
Overbalance falls by a factor of 4.4 along the hole, so the bore is supported at the toe while circulating and thinly everywhere else, and the heel is the part that has been open longest. Pressure is not the whole story: what holds a horizontal bore open in sand is filter cake, soil arching and the fluid's gel strength working with the overbalance rather than instead of it.
Two operating decisions follow, and they are one decision. Circulation is never stopped with the string in an open lateral. And because a stroke drive that pauses to reset is a drive that stops advancing, long bores make the case for dual-clamp continuous advance, one clamp taking the load while the other resets.
Hole cleaning fixes the flow. Flow, times the length of hole, divided by the rate of penetration, is the volume that leaves the bore and enters the formation.
Class CS-411A runs 163 L/min, set by cuttings-bed avoidance rather than by what the jets would like, which puts 3.45 m/s inside the tube. All three classes solve to the same annular velocity of 0.600 m/s, which is the clearest statement of the bore doctrine here: the pilot is sized by hole cleaning alone and the completion is cut later by a separate pass. Cuttings run at 1.74 per cent of the flow by volume at the feed and nearer 3.5 per cent in the annulus once slip is allowed for, against a 5 per cent guideline. Delivering it takes a 48 to 57 kW shaft across a credible viscosity band, which makes the fluid package the largest power draw on the platform.
Volume lost scales as one over the rate of penetration, because flow is fixed by hole cleaning. Doubling the rate halves the volume that goes into the ground, so penetration rate is written into the operating procedure as an environmental control.
Work in a potable aquifer restricts the fluid to listed products, which in practice means plain bentonite and a short list of listed polymers. The provincial regime decides the rest, before the engineering does.
A weaker fluid suspends cuttings less well, which asks for higher annular velocity, which asks for more flow, which puts more volume into the aquifer. The rheology limit and the environmental limit are one constraint seen from opposite ends.
Nothing in the design transmits and nothing guides on a link. The bore is flown on sensing at the head, carried up the tube on fibre, with a person at the console.
Inclination comes from accelerometers and is referenced to gravity, so it does not drift; its error is a systematic bias rather than something that grows with time in the hole. Azimuth is taken by gyro in every class. A gyro needs no non-magnetic spacing, the constraint that rules a magnetic tool out of a small bottom-hole assembly, and it gives a valid toolface below 5° of inclination, where gravity toolface is undefined. That is the whole of a horizontal lateral.
Its limit is physics. North-seeking works on the horizontal component of the earth's rotation, so azimuth uncertainty scales as one over the cosine of latitude, and a near-horizontal hole is close to the worst attitude for it. The answer is stationary re-references, at an interval short enough that the unsurveyed bow stays small: at 30 m the bow is 0.196 m, 0.65 per cent of the reference cover. Surface locating is not the architecture, and the reason is depth — a walkover sonde reaches roughly 15 to 25 m of cover and the reference aquifer sits at 30 m. Navigation is referenced downhole instead.
Centred in a 4 m water-bearing layer the budget is ±2.0 m. Steering resolution, sensor-to-face extrapolation, station interpolation and the entry tie-in take 0.84 m of it, leaving 1.82 m for the sensor and an inclination accuracy better than 0.12°.
A lateral stays inside a stratum metres thick and hundreds wide, so the budget goes on inclination and tie-in. The exception is a lateral steered to a legal boundary, where azimuth becomes a compliance measurement.
A sand mapped from boreholes tens of metres apart carries its own depth uncertainty, and half a degree of dip walks the target several times the vertical budget. Grade holding is a geological problem before an instrument one.
No radio and no GPS in any Canadian Shield design. The plant stays on the surface and nothing enters the ground but the tube and the head, so nobody goes into the hole. Carriers and conduit tools, never munitions: no warhead, no fuze, no energetic material. Every design in the family is patent pending.
Power at the head splits by function, and that split keeps the fibre architecture intact. Actuation and rotary work run on fluid power off the jetting supply, where the 250 W the head needs is about 1.5 per cent of the 16.3 kW the nozzles already carry. Sensing and telemetry draw 3.5 W and ride the fibre, which deletes a conductor from the tube bore and keeps the signal path all glass. There is no battery downhole, for a reason that belongs to the next section.
Pressure is the easy part of the 55 mm by 300 mm housing: a 5 mm wall clears the mud hydrostatic at the reference depth by a factor of 7.5 and holds 3.04 at a 100 m design case. Jetting erosion sets the wall instead, so the housing is metallic. Heat is easier still: 10.5 W of electronics in 6 °C groundwater moving at 0.6 m/s raises the film by 0.93 K.
The consequence is not a lost asset. It is a permanent steel and electronics installation inside somebody's potable aquifer, so the platform gives up the head and keeps the tube.
Normal retraction runs at about 7 kN. A bore collapsing onto the tube holds it with roughly 5.5 kN per metre of collapsed length at the reference depth in saturated sand, so the pull needed climbs a great deal faster than the length does. What the clamp can grip before it begins to flatten a thin-wall tube is the ceiling on any recovery, and that ceiling is set on the drive rather than in the hole.
Ten to twenty metres of collapsed hole defeats the pull. It is a few seconds of running sand after a pump trip, which is the same reason circulation never stops. So the string carries a weak link ahead of the head: it parts under a pull the tube and the clamp both survive, and it surrenders the head rather than the string, buying back the tube and the fibre. Differential sticking does not govern, at roughly 2.2 kN per metre against the 5.5 above. And the design rule falls out of the arithmetic. If the head can be left in an aquifer, the head is built to be left in one.
A screen pulled back into a mud-filled lateral is a pressure boundary, sized on the case that lasts decades rather than the one that lasts an hour.
Pulled in empty through a mud-filled bore, the screen sees the full mud column of 309 kPa for about an hour. In service, pumped down to near the intake, it sees the full formation head of 294 kPa for decades. The installation load is 5 per cent larger and it happens first, so a fast reading concludes that installation governs. A fifty-year modulus retention belongs to the service case, and once each case carries its own modulus the service case governs.
| The completion, class CS-411A | Figure |
|---|---|
| Screen outside diameter | Ø57 mm |
| Wall | 5.0 mm, SDR 11 |
| Clear bore | 47 mm |
| Installation case, short-term modulus | Safety factor 6.55 |
| Service case, fifty-year modulus | Safety factor 2.41, governing |
Ovality at 0.70 and a slotting knock-down of 0.60 applied to both cases.
Open area is not the constraint: 150 m of 57 mm screen at 5 per cent open area gives 1.34 m² of entrance area, passing 2,403 L/min at the standard 0.03 m/s entrance-velocity limit, against a domestic well's 20 L/min. What the small diameter costs is conveyance inside the pipe. Three hundred metres of the 47 mm bore carries about 89 L/min at 2 m of head and about 150 L/min at 5 m, so the pipe is roughly forty times more limiting than the screen.
Not every bore ends in a well. Where the mission is characterization the record is the product: the head logs conductivity, moisture and strata identification as the bore advances, piezometers and sensor strings are emplaced along the line, and fibre left in a completed lateral is a permanent downhole sensing asset. Sensor nodes of the CS-430 class can later tap a lateral this plant has laid, for power and backhaul instead of battery swaps.
The geometry decides most of it, and the difference is worth stating before a site visit.
Enquiries are screened and nothing here is an offer. The way in is a working session with engineering.
Send the alignment you have in mind, the ground it runs through and the article you need left in it. What follows is a conversation about the pilot bore the ground will take, the class that suits it, the fluid the regime allows and the reaction the site can offer. Export posture is counsel-first, with permits per shipment.
The machine itself is on the CS-410 page, the family on the Underground family page, and fabrication on the manufacturing page.
All Underground family designs are patent pending; international transfer is export-controlled and subject to Canadian government permits taken per shipment, and nothing on this page is an offer.
