Deep Hole Drilling for Superconducting MRI Cryostat Cooling Channels: Helium Conservation through Deep-Drilled Magnet Housing Plates

Siemens US8487730 describes deep hole drilling to create integrated cooling channels in MRI cryostat housings, enabling a 50-fold reduction in liquid helium consumption (from ~1 m3 to ~0.02 m3 per scanner). Parallel holes (8-16 mm diameter, 1000-2000 mm length, 2-3D pitch) are deep-drilled in a flat aluminium or stainless steel plate, then the plate is rolled into an annular housing and the longitudinal seam is welded. The housing is helium leak tested at 20 bar (acceptance less than 1 x 10^-9 Pa-m3/s).

Deep-Drilled Cooling Channel Manufacturing Process ​

The manufacturing process for an MRI cryostat with deep-drilled cooling channels involves a sequence of precision operations, each with specific parameters and quality checks. The following table breaks down the complete manufacturing process from raw plate to finished cryostat housing.

The deep hole drilling step is the most critical operation in the process. The plate is clamped to the table of a horizontal gun drilling machine with the plate surface perpendicular to the drill axis. The gun drill enters through one edge of the plate and exits through the opposite edge, drilling a through-hole parallel to the plate surface. The holes are spaced at a pitch of 2-3 times the hole diameter, which provides sufficient material between channels to maintain structural integrity while maximising the number of channels within the available plate width. The straightness requirement of 0.1 mm/m is essential because the holes must not intersect when the plate is rolled into a tube: the curvature of the rolled plate causes the hole axes to follow a curved path (they become parallel to the tube axis but are positioned at different radii from the tube centre). If a hole deviates from straightness by more than 0.1 mm/m, it may break through into the adjacent hole during rolling, creating a coolant leakage path. After drilling and deburring, the plate is rolled using a multi-roll plate bending machine with the rolling axis parallel to the plate length dimension. The rolling operation bends the flat plate into an annular tube, with the hole axes becoming parallel to the tube axis. The number of rolling passes (typically 3-5) is controlled to achieve the specified tube diameter without excessive ovality. The rolled tube is then welded along the longitudinal seam, with the weld designed to achieve full penetration without melting into the adjacent cooling channels. The minimum hole-to-edge distance is 4-5 mm to provide a sufficient heat sink during welding and to prevent the weld pool from penetrating into the channels.

Material Selection and Drilling Parameters ​

The choice of material for the MRI cryostat housing affects the drilling parameters, the cooling performance, and the manufacturing cost. Two materials are used in commercial MRI cryostats: aluminium 6061-T6 and stainless steel 304L. The following table compares the drilling parameters and performance characteristics for these two materials.

Aluminium 6061-T6 is the preferred material for most MRI cryostat housings because of its combination of high thermal conductivity (167 W/mK, approximately 10 times higher than stainless steel), adequate strength (310 MPa tensile), and excellent machinability. The high thermal conductivity of aluminium provides efficient heat transfer from the superconducting coils through the cryostat wall to the helium coolant, reducing the temperature gradient across the housing wall. The drilling of aluminium is straightforward: the high cutting speed (200-300 m/min) allows fast drilling (2-4 minutes per metre of hole), the emulsified oil coolant at 30-50 bar provides adequate lubrication and chip evacuation, and the tool life is excellent (200-400 metres of drilling per gun drill). The chip form in aluminium drilling is well controlled: the chips are short and segmented, which prevents chip packing in the gun drill flute. Stainless steel 304L is used for MRI cryostats that require higher strength (for larger scanner bores) or for applications where the cryostat must operate at higher pressures (for helium circulation at 20 bar). The drilling of stainless steel is more challenging: the cutting speed must be reduced to 60-100 m/min to prevent work hardening, the coolant must be a sulphurised oil at 40-60 bar to provide the extreme-pressure lubrication required for stainless steel, and the tool life is shorter (80-150 metres). The chip form in stainless steel drilling must be carefully controlled: stainless steel produces long, stringy chips that can pack in the flute and cause gun drill breakage. The drilling parameters must be selected to produce segmented chips: chip breakers are often ground into the gun drill tip geometry to promote chip breaking.

Leak Testing and Quality Assurance ​

Leak testing is the most critical quality assurance step for MRI cryostat cooling channels because any helium leakage will compromise the vacuum insulation of the cryostat and increase the helium consumption, negating the benefit of the deep-drilled channel design. The following table compares the leak testing methods used at different stages of manufacturing.

The final acceptance test for MRI cryostat cooling channels is the vacuum helium leak test, which is the most sensitive leak detection method available. The test procedure is: (1) The cooling channels are pressurised with helium gas at 20 bar. (2) The exterior of the housing is evacuated to a high vacuum (typically 1 x 10^-5 mbar). (3) A helium mass spectrometer leak detector is connected to the vacuum chamber. (4) If there is any leakage from the cooling channels, helium will enter the vacuum chamber and be detected by the mass spectrometer. The acceptance criterion is a total helium leak rate of less than 1 x 10^-9 Pa-m3/s, which is equivalent to approximately 1 millilitre of helium per year at standard conditions. Leaks above this threshold will cause gradual loss of helium and degradation of the vacuum insulation over the 10-20 year service life of the MRI scanner. The leak testing is performed after all welding operations are complete (longitudinal seam weld, end manifold welds, and any repair welds). If a leak is detected, the location of the leak is identified using a helium sniffer probe that samples the vacuum chamber gas at multiple positions, and the leak is repaired by localised TIG welding or by applying a vacuum-grade epoxy sealant (for very small leaks). The repaired area is re-tested by the same vacuum helium method. The leak test is specified by the MRI manufacturer and witnessed by the quality assurance department. The test records are included in the cryostat manufacturing documentation and are retained for the life of the MRI scanner.

Frequently Asked Questions ​

How do deep-drilled cooling channels reduce helium consumption in MRI magnets? ​

Conventional superconducting MRI magnets are cooled by immersing the coils in a bath of liquid helium at 4.2 K (-269 C). The helium bath absorbs heat from the magnet and from the cryostat environment (radiative heat input, conductive heat input through supports, and heat generated by current leads). The absorbed heat causes the liquid helium to boil, producing helium gas that must be vented from the cryostat. Depending on the magnet design and cryostat quality, a conventional helium bath MRI system consumes 0.5-1.5 cubic metres of liquid helium per year through boil-off. This represents a significant operating cost ($50,000-$150,000 per year at typical helium prices of $100-$150 per litre) and a logistics challenge (helium gas must be collected, liquefied, and returned to the cryostat periodically). The deep-drilled cooling channel approach, described in Siemens patent US8487730, replaces the open helium bath with a closed-loop cooling system. The superconducting coils are not immersed in liquid helium; instead, they are cooled by conduction through the cryostat housing to the helium flowing through the deep-drilled channels. The channels are connected to a cryocooler (a closed-loop helium refrigerator) that circulates and recondenses the helium, maintaining the magnet temperature at 4.2 K with zero helium boil-off. The only helium consumption is the initial charge (approximately 0.02 cubic metres, or 20 litres, compared to 1 cubic metre for a bath system) and any make-up required during servicing (typically less than 1 litre per year). The 50-fold reduction in helium consumption is achieved because: (1) the closed-loop system does not vent any helium to the atmosphere, (2) the conduction cooling eliminates the heat load from the helium bath structure (the bath vessel walls, the bath support structure, and the bath access ports), and (3) the cryocooler can efficiently recondense the helium at the cooling power available within the MRI system's overall energy budget.

What materials are used for MRI cryostat housings? ​

MRI cryostat housings are manufactured from either aluminium 6061-T6 or stainless steel 304L, each selected for specific design requirements. Aluminium 6061-T6 is the preferred material for most commercial MRI scanners because of its high thermal conductivity (167 W/mK, approximately 10 times higher than stainless steel), which provides efficient heat transfer from the superconducting coils through the cryostat wall to the helium coolant channels. Aluminium also has a lower density (2.7 g/cm3 vs 7.9 g/cm3 for stainless steel), which reduces the weight of the cryostat and the overall scanner weight. The lower elastic modulus of aluminium (69 GPa vs 193 GPa for stainless steel) means that aluminium cryostats require thicker walls to achieve equivalent stiffness, but the weight saving is still significant. Aluminium 6061-T6 has adequate strength (310 MPa tensile, 275 MPa yield) for the 20 bar operating pressure of the cooling channels and for the structural loads from the superconducting magnet. The excellent machinability of aluminium allows fast drilling (200-300 m/min cutting speed) and long tool life (200-400 metres per gun drill). Stainless steel 304L is used for cryostats that require higher strength (for larger scanner bores or higher operating pressures) or for scanners that operate in environments where aluminium is not suitable (for example, scanners located in areas with high chloride concentration in the air, which can cause stress corrosion cracking in aluminium). Stainless steel 304L has higher strength (485 MPa tensile, 170 MPa yield at cryogenic temperature) but much lower thermal conductivity (16 W/mK at room temperature, decreasing to approximately 1 W/mK at 4.2 K). The low thermal conductivity of stainless steel creates a temperature gradient across the cryostat wall that must be managed by careful design of the cooling channel layout and the thermal contact between the channels and the magnet coils.

How is straightness maintained during deep drilling of cooling channels? ​

Straightness of the deep-drilled cooling channels is maintained through a combination of machine rigidity, drill geometry, and process parameters. The horizontal gun drilling machine must have a rigid spindle support system and a guide bushing that supports the gun drill at the entry point. The guide bushing is positioned as close as possible to the plate surface (typically within 5-10 mm) to minimise the unsupported length of the drill at the start of the hole. The guide bushing diameter must match the drill diameter within 0.01-0.02 mm to provide precise guidance without binding. The drill geometry is optimised for straightness: the gun drill has a single cutting edge with a supporting pad (the guide pad) on the opposite side of the drill head. The guide pad rides against the bore wall and provides the lateral support that prevents the drill from deviating. The clearance angle on the cutting edge and the pad geometry are designed to create a balanced cutting force that keeps the drill on a straight path. The drilling parameters are selected to maintain a stable cutting condition: the feed rate must be within the range that produces a consistent chip thickness (typically 0.06-0.12 mm/rev for aluminium), and the cutting speed must be high enough to generate a cutting force that keeps the guide pad in contact with the bore wall but low enough to prevent excessive heat generation and thermal expansion of the drill. The coolant pressure (30-50 bar for aluminium) provides hydraulic support for the drill and helps to maintain the straightness by centring the drill within the bore. The plate must be rigidly clamped to the machine table to prevent vibration or movement during drilling. The clamping must not distort the plate (the clamping force must be distributed evenly across the plate surface). After drilling, the straightness of each hole is verified by inserting a straightness gauge (a precision-ground rod of the same diameter as the hole) and measuring the gap between the rod and the hole wall at multiple positions along the length.

What is the leak testing procedure for MRI cryostat cooling channels? ​

The leak testing procedure for MRI cryostat cooling channels follows a progressive sequence of tests at each stage of manufacturing, culminating in the final vacuum helium leak test. The initial test after drilling is a pressure drop test: each individual hole is plugged at one end and pressurised with air or nitrogen at 10 bar. The pressure is monitored for 10 minutes; a pressure drop of more than 0.1 bar indicates a leak through the hole wall to an adjacent hole or to the exterior. This test detects gross defects from the drilling process (a drill that broke through an adjacent hole, a porosity defect in the plate material, or a crack) but is not sensitive enough to detect the very small leaks that would cause helium loss over the scanner lifetime. After the longitudinal seam welding and end manifold welding, the channel groups are tested by the bubble test (immersion test): the housing is pressurised with air at 10 bar and immersed in a water tank. Any leakage is visible as a stream of bubbles. The bubble test detects leaks larger than approximately 1 x 10^-6 Pa-m3/s. The final and most sensitive test is the vacuum helium leak test, performed after all welding and heat treatment is complete. The cooling channels are connected to a helium supply at 20 bar, and the exterior of the housing is enclosed in a vacuum chamber that is evacuated to high vacuum (1 x 10^-5 mbar). A helium mass spectrometer leak detector is connected to the vacuum chamber. If there is any leakage from the cooling channels, helium enters the vacuum chamber and is detected by the mass spectrometer. The mass spectrometer can detect helium concentrations as low as 1 part per million, corresponding to a leak rate of approximately 1 x 10^-11 Pa-m3/s. The acceptance criterion for the complete housing is a total leak rate of less than 1 x 10^-9 Pa-m3/s, which is equivalent to approximately 1 millilitre of helium per year at standard conditions. The leak test is performed at the maximum operating pressure of the cooling channels (20 bar) and must be maintained for a stabilisation period (typically 30-60 minutes) before the measurement is taken.

How does the plate rolling process affect the drilled channels? ​

The plate rolling process converts a flat plate with straight, parallel deep-drilled holes into an annular tube where the holes become cooling channels parallel to the tube axis. The rolling process has several effects on the drilled channels that must be managed by the manufacturing process. (1) Hole geometry change: when the plate is bent, the holes on the inner radius of the bend (the holes closer to the tube bore) experience circumferential compression, while the holes on the outer radius experience circumferential tension. This can cause ovalisation of the hole cross-section: the holes become slightly elliptical, with the major axis in the radial direction. The ovalisation is typically 0.01-0.05 mm for a 16 mm hole in a 15 mm thick plate rolled to a 500 mm radius, which is within the acceptable tolerance for helium flow. (2) Hole spacing change: the pitch between adjacent holes increases on the outer radius and decreases on the inner radius of the bend. For a 16 mm hole pitch of 40 mm in a 15 mm plate rolled to 500 mm radius, the pitch change is approximately 0.3 mm (outer) and -0.3 mm (inner). This must be accounted for in the hole layout design to ensure that the final hole spacing meets the design specification. (3) Residual stress induced by rolling: the rolling operation introduces a residual stress distribution across the plate thickness that is superimposed on any residual stress from the drilling process. The combined residual stress can cause the tube to spring back after rolling (the tube diameter increases slightly after the rolling load is released) or to distort during subsequent welding. The springback is compensated by over-bending the plate (rolling to a slightly smaller diameter than specified). The distortion during welding is controlled by the welding sequence and heat input. (4) Straightness of the channels after rolling: the channels remain straight relative to the tube axis because the rolling axis is parallel to the channel axes. However, any deviation in the rolling alignment (the rolling axis not perfectly parallel to the channel axes) will cause the channels to follow a helical path in the finished tube. The alignment must be controlled within 0.1 degrees to prevent this. After rolling, the tube is inspected by: measuring the tube diameter and ovality at multiple positions, measuring the wall thickness, and performing a visual inspection of the channel ends to verify that no channel has been crushed or distorted by the rolling process.

The information provided in this article is for general informational purposes only. Data and recommendations are based on published research and industry experience as of 2026.