The price of an industrial X-ray inspection system bears no relation to the volume of the few parts inside it that actually matter. A single microfocus tube can account for a substantial share of machine cost, and one flat panel detector frequently costs more than the entire mechanical assembly, yet these are precisely the two most fragile items on the machine. The tube is a vacuum-sealed envelope whose ceramic-to-metal joints and sub-millimetre beryllium window carry almost no impact margin. The detector builds its imaging surface on a glass substrate with pixels pitched at 100 to 200 microns, so a single blunt knock can seed an internal crack that stays invisible. An oil-filled high voltage generator that tips over in transit redistributes its insulating oil, and restoring dielectric performance costs far more than repacking the case would have. The design criterion for an industrial X-ray component case is therefore not whether the parts fit, but whether tube vacuum and target position are unchanged, whether the detector imaging face is intact and unscratched, and whether high voltage insulation and lead shielding are still in their as-shipped condition. This article works through each component group and sets out transport attitude, insert materials, the real boundary of radiation compliance, and receiving inspection.
The pattern on site is predictable. The machine is broken into four piles, tube source, detector, cabinet and mechanics, wrapped in stretch film, padded with pearl foam and loaded, often with two layers of foam board laid directly on the case floor. Under a short domestic run in dry weather this usually gets away with it. Introduce sea freight, several transhipments or a long road haul and the failures arrive together: the tube shifts microscopically inside the case, target position drifts, focal spot wanders and resolution falls away; the detector window cover works loose, the imaging face collects dust and packing debris, and fixed dark spots and banding appear once the machine is installed; an oil-filled generator ends up with oil pooled to one side and flashes over along insulating surfaces at first power-up. The harder problem is the evidence itself. Most of this damage is invisible before the machine is energised, so by the time image quality fails, transport damage and installation damage are mixed together and cannot be separated. The sections below follow that chain: why each component is fragile and how fragile, then support and restraint, insert and cleanliness requirements, the practical boundary of radiation compliance, and finally marking, documentation, transport testing and receiving inspection.
Contents
- 1. Why the value of an X-ray machine sits in a few impact-sensitive parts
- 2. Component inventory and protection grading
- 3. X-ray tubes and vacuum assemblies: the physics behind impact prohibition
- 4. High voltage generators and cables: insulation attitude and bend radius
- 5. Detector arrays and collimators: precision faces must not be touched or wiped
- 6. The real boundary of radiation compliance: lead parts, shielding and transport context
- 7. Electrical controls and the image processing unit: moisture, vibration and static
- 8. Mechanical motion parts: transport locks, screw supports and belt slack
- 9. Insert design and sensitive-face isolation
- 10. Moisture control and cleanliness
- 11. Shipping marks, in-case documents and receiving inspection
- 12. Transport testing and packaging validation
- 13. Customization workflow and delivery coordination
- Frequently Asked Questions
- Conclusion and Further Reading
1. Why the value of an X-ray machine sits in a few impact-sensitive parts
Industrial X-ray inspection covers foundry and weld defect detection, solder joint and BGA void inspection in electronics, electrode alignment checks in lithium cells, foreign body detection in food and pharmaceutical output, and delamination checks in composites and tyres. All of these applications are governed by the same physical chain. Focal spot size sets geometric unsharpness, generator stability sets beam hardness and grey level repeatability, the collimator sets the edge of the inspection field, and detector pixel pitch with signal to noise ratio sets spatial resolution and contrast sensitivity. Shift any link and the measurement drifts, usually in the worst possible way: images still appear, but the acceptance threshold is no longer trustworthy.
Three positions on that chain deserve separate treatment. The first is the vacuum envelope of the tube. To hold a high vacuum while accelerating electrons at high voltage, the envelope must be gas-tight, normally ceramic insulating sections joined to metal bodies by brazing or diffusion welding, with a very thin beryllium window at the exit. A gas-tight structure is binary: intact means acceptable, cracked means scrap. A microcrack in a braze or plastic deformation of the window drops the internal vacuum, and the symptoms are flashover during ramping, unstable beam current and random bright pixels in the image, none of which can be repaired on site.
The second is the detector imaging surface. A flat panel detector is an amorphous silicon or CMOS array on a glass substrate, overlaid with a caesium iodide or gadolinium oxysulphide scintillator, with pixels at 100 to 200 microns. Glass has limited bending strength, and a point contact impact seeds radial microcracks that first appear as anomalous response in a few pixels and then extend into dead rows or columns as thermal cycling continues.
The third is high voltage insulation. An oil-filled generator depends on oil distribution and cleanliness for its margin, and it is not a component that can be judged by whether it looks undamaged. Transport protection for radiography equipment is therefore not cosmetic protection at all. It exists to keep three irreversible quantities from changing: vacuum integrity, imaging geometry and insulation margin. Once that is clear, it becomes possible to tell which packaging spend is necessary and which is waste. General selection logic is covered in the instrument case selection guide.
2. Component inventory and protection grading
Breaking an industrial X-ray machine into transportable units normally yields eight to twelve packages. Grading them is not labelling for its own sake; it decides where resources go. Grade A items need purpose-designed case structure and inserts, plus a packing validation test. Grade B items can use a standard case with a custom insert. Grade C items are handled as ordinary electromechanical parts. The table below is the usual teardown list and grading basis, and can be attached directly to a technical agreement.
| Component | Primary failure mechanism | Grade | Key measures |
|---|---|---|---|
| --- | --- | --- | --- |
| Microfocus X-ray tube with beryllium window | Ceramic joint cracking, target displacement, vacuum loss | A | Factory transport bracket or equivalent rigid support, axial location, acceleration recording |
| Oil-filled HV generator | Tip-over redistributes oil, breather flooded | A | Keep as-shipped attitude, upright transport, compare oil sight glass readings |
| SF6 insulated HV generator | Pressure loss in cold conditions reduces dielectric strength | A | Judge against maker temperature-pressure curve, top up before energising |
| Flat panel detector | Glass substrate microcrack, scintillator delamination, ESD damage | A | Original case preferred, desiccant, ESD packaging, no stacking load |
| Line scan detector and collimator | Slit misalignment, window scratching, dust masking | A/B | Dedicated slit cover, clean bag, no wiping |
| HV cable and terminations | Insufficient bend radius initiates partial discharge | B | Dedicated cable drum, coil radius at least six to eight times outer diameter as a working figure |
| Lead shielding and lead glass | Creep deformation, surface marking, glass fracture | B | Upright and independently supported, never a load bearing or packing element |
| Control cabinet and image processing unit | Solder joint fatigue, connector back-out, moisture tracking | B | Transport lock bolts, desiccant, disconnect backup battery |
| Mechanical motion parts, C-arm, column, conveyor | Screw sag, bearing block balls released, belt slack | B/C | Transport locks, intermediate screw supports, carriage retaining fixtures |
| Tube cooling unit and hoses | Oil leakage, hose kinking | C | Cap fittings, coil hoses at generous radius |
After grading, one more step matters: mark the sensitive faces. The beryllium window, collimator slit exit, detector imaging window, high voltage terminations and lead glass clear aperture should each be marked on the packing drawing, with a matching notice fixed inside the case at the corresponding position. The purpose is to carry a no-contact requirement from the design file to the person actually packing the case, because most field protection failures are not design errors but packing operations carried out by someone who was never told which surfaces cannot be touched. How grading translates into insert geometry is described in the custom foam insert design guide.
3. X-ray tubes and vacuum assemblies: the physics behind impact prohibition
X-ray tubes prohibit impact not because they are expensive but because their construction contains two design contradictions that cannot be resolved by adding thickness. The first is insulation against strength. The tube must withstand tens to hundreds of kilovolts internally, so insulating sections are made long and thin to control creepage, and thin-wall ceramics have far lower bending strength than metal. The second is window against gas-tightness. To let low energy X-rays out, the beryllium window is typically only a fraction of a millimetre thick, and that thin metal must hold a vacuum differential while also carrying assembly stress. Neither structure can absorb impact energy, so the energy has to be intercepted by external support structure.
Three practical rules follow. First, the tube should be supported independently of the case, meaning the body does not rest directly on the case floor or wall. Load should pass through the end flanges or the factory transport bracket into structural members, leaving the body located but not compressed. Second, restraint must cover all six degrees of freedom, with contact surfaces chosen on non-machined areas. Axial restraint matters most, because the anode target and bearing position along the tube axis has the most direct effect on imaging quality, and even a fraction of a millimetre of axial slip changes the focal spot to detector distance relationship. Third, acceleration should be recorded rather than assumed. A three-axis acceleration recorder inside the tube case captures actual peaks during transit and gives an objective basis for the arrival decision. As a working figure, makers commonly quote transport shock limits in the region of a ten g half-sine pulse of a few milliseconds duration, but the tube manufacturer's datasheet governs and no general figure should be substituted for it.
Two transport configurations are worth considering. One is to keep the factory packing attitude, reusing the tube's own transport bracket and specified orientation. This is the lowest risk route, particularly for tubes with an oil cooling jacket or an integrated high voltage socket. The other is a purpose-built case for machine relocation that carries the tube with its mounting plate and water or oil connections intact, reducing field reassembly. Both are far more reliable than removing the tube and wrapping it in a single layer of bubble film, which places the most fragile item in the least controlled environment.
One frequently skipped detail is the first power-up after transport. After vibration and thermal cycling a vacuum tube may release absorbed gas, and applying full voltage immediately can cause flashover, which itself leaves marks on electrode surfaces. The accepted approach is staged ramping: run at low kilovoltage and low current for a period, watch beam current and image output for instability, then step up to rated conditions, with the whole sequence recorded. If repeated flashover occurs at the low voltage stage, stop and contact the tube maker rather than attempting to burn the fault out by repeated energising. This step costs half an hour and prevents a tube moving from possibly serviceable to certainly scrap. The structural restraint approach used in tube cases shares its logic with other precision instruments, described in cushion liner design.
4. High voltage generators and cables: insulation attitude and bend radius
The high voltage generator is the component that looks sturdiest and is most easily ruined by attitude alone. An oil-filled unit consists of a steel tank, high voltage transformer, voltage multiplier stack and insulating oil that serves both as dielectric and as coolant. A well designed tank includes expansion volume or a breather to absorb oil volume changes with temperature, and those internal features impose attitude requirements. Prolonged tilting or inversion collects oil to one side, leaves part of the winding exposed above the oil line, and the exposed insulation then meets air and moisture directly while the case vibrates. If the breather port sits against the oil surface in the wrong attitude, the unit may draw oil or spit oil later in service. Transport attitude must therefore follow the maker's instruction, normally upright, with centre of gravity and upright direction marked on the case, and tilt indicator labels fitted inside for long journeys with their readings included in receiving inspection.
For SF6 insulated generators the concern is gas pressure. Sulphur hexafluoride dielectric strength depends directly on gas density, and density varies with temperature, so a gauge reading must be converted using the maker's temperature-pressure curve before it means anything. A low reading after cold transport is normal physics, but energising in that state risks internal flashover from insufficient dielectric strength. The correct order is to let the unit reach ambient service temperature, then check pressure against the specified band, and only then top up and allow settling time before dielectric tests.
High voltage cable is another item frequently sacrificed at the packing stage. Industrial X-ray cable operates at one hundred to more than two hundred kilovolts, and its construction runs from conductor through semi-conductive layer, insulation, semi-conductive screen and metal sheath, with a stress cone or moulded silicone termination spreading the field at the end. The vulnerable point of this structure is bending. Too tight a radius compresses the insulation on the inside of the bend and stretches it on the outside, creating microscopic voids at the semi-conductive interface. Those voids become partial discharge inception points, and partial discharge is a slow failure. Nothing alarms at commissioning; instead, insulation erodes over months to years until breakdown. Coiling must therefore use a dedicated drum or a large diameter former, with a working bend radius of at least six to eight times the cable outer diameter, and no bending within one metre of the termination. Terminations should be fixed separately with protective sleeves and kept clear of the case wall and other hard items.
Before shipment, cable ends and connectors need capping and cleanliness treatment. Electrical grade silicone grease must never be applied near a tube beryllium window, a point developed in section 9. After cleaning, terminations should be fitted with dust caps containing desiccant, and at destination the caps are opened first to check for condensation before any decision to install. After installation, an insulation resistance and partial discharge baseline test compared against the factory record says more about transport damage than any amount of visual inspection.
5. Detector arrays and collimators: precision faces must not be touched or wiped
The detector demands zero contact management more than any other item on the machine. A flat panel detector's active area is a glass substrate carrying a pixel array, protected externally by only a thin layer and a housing window. Its failure mode differs from other precision parts: it is not damaged and then broken, it is knocked and then slowly degrades. A microcrack initially shows as anomalous response in a few dozen or a few hundred pixels, which ordinary black and white test patterns will not reveal. Detecting it requires a flat field image compared against the factory defect map, counting new defective pixels, their locations, and whether they form row or column patterns.
The collimator fails geometrically. Collimators use tungsten or lead slit jaws whose gap width follows the part size and magnification, and the geometric accuracy of that gap sets the field edge and the scatter radiation level. Once a jaw edge is burred or displaced, images show blurred edges or increased penumbra, and this is frequently misdiagnosed as a tube focal spot problem, sending fault finding in the wrong direction. The collimator therefore needs a dedicated cover located on the outer step of the slit rather than on the jaw edge, in a non-hygroscopic engineering plastic or anodised aluminium. Fabric, sponge and tape laid directly over the slit are prohibited, because fibre wound around the jaws is worse than no cover at all.
The shared enemy of detector and collimator is particulate contamination and moisture. Dust on the imaging window produces fixed position image artefacts. In theory it can be cleaned, but the cleaning action carries more risk than the dust itself, because any wiping leaves fine scratches on the protective layer or scintillator, and those never come back. The more robust strategy is to prevent contamination reaching the window at all: fit a rigid cover plate, seal the assembly in a clean bag with desiccant, and carry out packing for clean applications in an ISO 14644 Class 8 environment or better. If dust is found after opening, blow it off with ionised air or dry nitrogen first, and only consider maker-supervised wet cleaning if blowing fails and a precision test confirms a fault.
Line scan and TDI detectors add a directional requirement. Their alignment relationship is set by relative motion between sensor and product, and many models are aligned and locked at the factory. Any transport load that twists the housing can destroy that alignment. Line scan detectors should therefore never be lifted by the housing sides or left unsupported; the bearing surface must be the mounting datum on the base. Related electrostatic and packaging practice is covered in ESD shielding case design and sealed and cleanable case construction.
6. The real boundary of radiation compliance: lead parts, shielding and transport context
Any discussion of transport compliance for industrial X-ray equipment should start by correcting a common misconception. X-rays are produced when electrons strike the target while the tube is energised. A de-energised machine produces no radiation during transport, so a conventional X-ray inspection system is not a radioactive material consignment and does not require radioactive transport documentation. The consequence for packaging design is that the priority is the physical integrity of the equipment, not the continuous effectiveness of shielding. General frameworks for dangerous and special cargo packaging and paperwork are summarised in hazmat transport case compliance, but an X-ray machine should not simply be classified under them.
Only two situations genuinely fall under radioactive material transport rules. The first is equipment containing a radioactive isotope calibration source or isotope radiography source, for example certain thickness gauges, level gauges or gamma radiography sets. Those are handled under GB 11806, the Chinese regulation for the safe transport of radioactive material, and IAEA SSR-6, with packaging type, marking, transport index and documentation determined by a qualified body. The second is shipment of lead waste or lead-bearing items into a destination with specific lead restrictions, where the declaration route must be checked against local law. Both are compliance consultancy matters requiring an accredited opinion, and nothing in this article constitutes a compliance determination.
Lead shielding protection, by contrast, is a purely mechanical problem and needs more care than most teams expect. Shielding on an industrial X-ray machine typically includes lead sheet, lead curtains, lead glass viewing windows and tungsten alloy collimator blocks, with thicknesses from around two millimetres to more than ten millimetres lead equivalent. Three mechanical properties of lead must be remembered. Lead creeps at room temperature, so a loaded lead sheet deforms slowly and must never be used as a load bearing or levelling element. Lead is very soft, so contact with hard items leaves indentations that change local thickness and therefore shielding uniformity. Lead glass has limited strength and fragile edges, and a drop almost always breaks it with no possibility of repair. The corresponding transport measures are to back lead sheet and curtains with timber or plastic frames so the frame carries the load, to stand lead glass upright with soft packing on both faces and restrained movement so it never touches metal, and to pack tungsten alloy collimator blocks separately with protected chamfers.
7. Electrical controls and the image processing unit: moisture, vibration and static
The control cabinet and image processing unit are a modest share of machine weight but a disproportionate share of transport damage, because their failure paths are hidden and concentrated at board level. The first is solder joint fatigue. Heavy components on a control board, such as electrolytic capacitors, relays, transformers and heat sinks, crack at their pin joints under sustained vibration, and at commissioning this shows only as occasional lock-ups or parameter jumps. The second is connector back-out, where board-to-board connectors, ribbon sockets and fibre optic ferrules creep out slightly under vibration and raise contact resistance. The third is moisture. Image processing units and high voltage control boards carry creepage distances, and absorbed moisture lowers insulation resistance until intermittent flashover occurs at power-up. The fourth is static discharge, to which detector interface boards and high speed signal boards are sensitive, with unpacking and installation being the main exposure window.
Four corresponding measures are worth standardising. Disconnect the backup battery before shipment or pack it separately, so it neither keeps supplying the board nor ages faster at elevated temperature. For servo drives, fans and heat dissipation modules with moving parts, fit the transport lock bolts or brackets specified by the maker, because many manuals state explicitly that locking is mandatory before transport and this is the step most often skipped on site. Fix boards and modules by containment rather than clamping, using inserts that surround and limit movement, and avoid lashing boards to metal frames with cable ties, which fret through the solder mask under vibration. For static control, removed detector interface boards and fibre modules go into ESD bags with equipotential bonding, and insert surfaces should sit in the 10^6 to 10^9 ohm range as a working figure, neither fully insulating nor conductive, so that frictional charge does not accumulate.
Moisture control must separate hygroscopic items from ordinary ones. Fibre connectors, high voltage insulators and detector windows belong to the first group and need individually controlled humidity inside sealed packaging, with desiccant and humidity indicator cards; sea freight or humid destinations justify a foil laminate inner package. Indicator cards should be placed at more than one height, because temperature and humidity are not uniform inside a case. Cables and the cabinet belong to the second group, where ordinary desiccant and an outer moisture barrier suffice, but the cabinet interior should be opened and ventilated promptly after arrival rather than left sealed until installation day. Relevant environmental test methods are set out in MIL-STD-810H environmental testing, cited here as a test method reference only and implying no military certification.
8. Mechanical motion parts: transport locks, screw supports and belt slack
Mechanical parts are less delicate than a detector, but they cause damage by dragging other things down with them. A ball screw sagging under its own weight produces positioning accuracy variation after installation, and fault finding usually heads straight for servo parameters, wasting days of commissioning time. The work on the mechanical side therefore comes down to three things: fixation before motion restraint is released, slackening of tension, and preventing bearing blocks from coming off their rails.
Ball screw transport attitude follows clear principles. A horizontal screw longer than about one and a half metres without intermediate support deflects under self-weight, which produces uneven contact between screw and nut and also affects the screw's own straightness. The engineering answer is two or three support points along the span, using soft material that touches only the plain shaft section and never the thread raceway. Vertically mounted screws, as inside a lifting column, do not self-lock; if the column is not mechanically locked for transport, vibration lets the screw rotate slowly and the column creep downwards until it rests on its limit or buffer. Lifting assemblies therefore require a mechanical locking pin or a dedicated support strut that fixes the height.
Linear rail carriages are the other high risk item. Many carriages retain their balls with a plastic cage or stop block from the factory, and if that is removed and the carriage moved without a substitute fixture, balls are released. The number lost is easily overlooked, and the symptom after installation is stick-slip or noise at particular positions. Carriages removed from rails should keep their original retainers or receive a purpose-made sleeve, and should never be allowed to knock against each other in the case. Rotary assemblies with gearboxes, such as a C-arm or turntable, should be fixed at the maker's marked lock position, usually a taper pin hole or a dedicated bolt hole; lashing with wire is not an effective lock.
Belt and mesh conveyors should have their tensioning devices released before transport so the belt or mesh sits slack. Leaving a belt tensioned in storage produces permanent elongation and stress concentration at the splice. Conveyor rollers should each be packed separately, with no stacking load on the roller face. The general protection logic for these items matches that of standard mechanical component cases, described in drive component transport protection.
9. Insert design and sensitive-face isolation
The job of an insert in this application is not to fill space but to do three things: route load into structural members, isolate sensitive faces from everything hard, and separate components with different environmental requirements into compartments. Common materials are polyethylene, cross-linked polyethylene, ethylene vinyl acetate, polyurethane and expanded polypropylene, each with its own working range. The table below gives selection guidance by component type; densities and hardness values are typical bands and the final specification should be checked against actual mass and the acceleration target.
| Insert material | Suitable components | Typical density | Notes |
|---|---|---|---|
| --- | --- | --- | --- |
| Low density PE foam | Cabinets, enclosures, void fill | 25 to 35 kg/m3 | Low cost, poor recovery, not for repeated use |
| XPE or EVA | Medium precision parts, cable drums | 45 to 90 kg/m3 | Good toughness, suits wrap-around inserts |
| Moulded PU foam | Detectors, collimators, irregular shapes | 60 to 120 kg/m3 | Best cushioning, watch outgassing and ageing |
| EPP | Returnable cases, heavy part supports | 30 to 60 kg/m3 | Fatigue resistant, good for multiple trips |
| Timber structure plus soft pads | X-ray tubes, oil-filled generators | not applicable | Timber carries load, soft pads isolate only |
Three design rules are stricter here than in most industries. The first is silicone isolation. Silicone sealants, silicone rubber pads, silicone grease and silicone oils continuously release low molecular weight siloxanes at room temperature, and those molecules deposit as an extremely thin silicone film on beryllium windows, detector windows and optical surfaces. In radiography that film changes absorption and scattering in the low energy band and degrades image quality, and the contamination cannot be removed without damaging the window. Silicone rubber materials are therefore prohibited near tube and detector windows, replaced by silicone-free rubber or foam, and silicone-based lubricants must not be used during assembly.
The second rule concerns bearing surfaces. Insert contact should fall on structural machined faces, mounting bases and foot pads, avoiding seal faces, optical surfaces, high voltage terminations and thin-wall cylindrical bodies. For items such as the tube that arrive with a factory bracket, the insert only needs to wrap and limit movement, not carry primary load, because any sustained static load carried by an elastic material will creep and lose its locating effect.
The third rule is compartmentalisation and environmental separation. The detector and electronics need a low humidity environment, the tube and metal parts need a rust preventive oil film environment, and lead and tungsten items are relatively insensitive but heavy. These three groups should sit in separate compartments with independently controlled humidity, each with its own desiccant and indicator card read separately on opening. Whether those compartments stay isolated over a long haul depends on hinges, latches and seals, whose selection and failure criteria are covered in case hinge, latch and seal selection. A pressure equalisation valve should also be fitted, so that altitude or air freight pressure differentials cannot crush the seal profile and break compartment isolation, as described in pressure equalisation valve configuration.
10. Moisture control and cleanliness
Moisture acts on three different points in an X-ray machine, and each needs its own defence. The first is high voltage insulation. High voltage terminals, insulators and sockets lose surface resistance as they absorb moisture and track across the surface at power-up, so these areas should be capped immediately after cleaning and humidity controlled inside a sealed sub-compartment. The second is the imaging items. Detector windows, scintillators and fibre end faces drift in response when damp, and can develop mould growth, particularly in tropical sea freight where the case cools below dew point overnight and condensation forms preferentially on cold surfaces. The detector compartment therefore needs lower humidity and better temperature uniformity. The third is metal corrosion. Frames, rails, screws and unpainted machined faces need oil film or vapour phase protection, and the effectiveness of that protection can be assessed within the GB/T 10125 salt spray framework, noting that the standard defines the test method only, so acceptance criteria must be agreed separately, normally against voyage duration and destination climate.
A practical configuration follows a three layer, two card, one record structure. The three layers are surface treatment with a displacing rust preventive oil or silicone-free grease on machined faces, vapour phase protection with VCI film or paper for complex shapes that cannot be coated, and an outer barrier of PE or foil laminate for long sea voyages or humid seasons. The two cards are humidity indicators placed at three heights inside the case and shock or tilt labels fixed to the case side near the lift points. The one record is a temperature and humidity logger, one in the detector compartment and one in the high voltage compartment, sampling every fifteen to thirty minutes, so the curve can be exported on arrival to show whether the dew point was crossed and whether peak temperature exceeded component limits. By contrast, a statement that desiccant was placed in the case supports no determination at all.
Cleanliness control is not about bringing the whole case to cleanroom grade. It is about keeping sensitive faces clean and enclosed during transport. Clean the detector window, collimator slit, beryllium window and high voltage terminations in a controlled environment, fit cover plates, then seal the assembly in a clean bag. The main particle sources inside such a bag are the bag material itself and the insert, so inserts should be cleaned and allowed to settle before packing, and shedding materials such as cardboard and low density foam should never touch the clean bag directly. Where the machine itself was assembled in an ISO 14644 Class 8 environment or better, packing should continue at the same grade or at least in an enclosed space with filtered air supply. For sea freight, use treated timber or fumigation-free plywood case construction and verify the wood packaging quarantine rules of the destination country.
11. Shipping marks, in-case documents and receiving inspection
Shipping marks follow GB/T 191, and general packaging requirements for electromechanical products follow GB/T 13384. Radiography equipment should carry at least the keep upright, keep dry, fragile, do not roll, centre of gravity and stacking limit marks. Cases containing an X-ray tube justify additional do-not-invert and no-impact marks, with the component name and model written on the case side for quick verification. Marks also allocate responsibility. If the arrows point the wrong way on arrival or a tilt label has triggered, the damage assessment starts from objective evidence, which matters greatly in a relocation involving several parties.
The in-case document set should be fixed at seven items: packing layout drawing and component number list; the tube maker's statement of transport attitude and permitted acceleration, as a copy; the detector factory defect map and flat field baseline; the oil level or gas pressure record for the high voltage generator with its temperature-pressure curve; the rust prevention and capping record covering oil type, film thickness, list of capped ports and treatment date; desiccant quantity and initial humidity indicator readings; and the serial numbers and positions of shock and tilt labels. These add little cost but they convert the arrival question from a dispute into a checklist.
Receiving inspection runs in two stages. The first happens before the case is opened: check the case exterior, mark orientation, label status and logger curves, and photograph everything before opening. The second happens before energising, using the table below item by item. Any out-of-tolerance finding should be fixed in writing and imagery before any decision to dismantle.
| Inspection item | Method | Suggested criterion |
|---|---|---|
| --- | --- | --- |
| Tube attitude and appearance | Visual plus acceleration recorder data | Attitude matches factory requirement, peak acceleration within maker limit |
| Beryllium window | Visual under oblique light, no touching | No indentation, no oil film, no attached debris |
| Generator oil level or gas pressure | Sight glass or gauge plus temperature conversion | Within the maker specified band |
| Detector imaging face | Flat field image differenced against factory baseline | New defective pixel count and distribution show no abnormal row or column pattern |
| Collimator slit | Cover status and slit visual check | No burrs on jaws, no trapped fibre, cover correctly located |
| HV cable termination | Visual plus insulation resistance | No kinks or dents, insulation resistance matches factory record |
| Control cabinet and boards | Connector seating, desiccant condition | Nothing loose, no condensation, indicator card unchanged |
| Mechanical motion parts | Manual rotation and screw straightness sampling | No stick-slip, no abnormal noise, no permanent screw bow |
Sampling plans and acceptance rules, whether batch sampling or full inspection and what proportion of minor defects is tolerated, are covered in case acceptance and AQL sampling. One point deserves emphasis: inspection must be completed before the machine is energised and image calibration is run. Once calibration overwrites the factory baseline, any later claim of transport damage loses its comparison reference.
12. Transport testing and packaging validation
X-ray equipment is a high value, low volume category, so packaging validation need not be repeated for every case, but each case structure should be validated at least once with a full packing. A sensible test set combines sinusoidal and random vibration to the relevant parts of GB/T 4857 to cover accumulated road and rail damage, horizontal impact to represent shunting and emergency braking, tip-over and limited height drop in defined attitudes to prove lift points, base frame and insert strength, and static stacking load to confirm multiple stacking levels. Distribution cycle simulation can follow the relevant cycle in ASTM D4169, and unitised shipping packaging can follow ISTA 3E. Environmental test methods may be cited from MIL-STD-810H methods 514.8 for vibration, 516.8 for shock and 507.6 for humidity, quoted here as test method references only and implying no military certification.
Setting the criteria is where this industry most often goes wrong. Freedom from visible damage does not prove a packaging design works, so functional criteria are essential. Four are worth mandating: peak acceleration recorded inside the case stays within limits before and after vibration; permanent compression of tube supports and inserts stays below fifteen percent of insert thickness; the increase in detector defective pixels compared with the pre-test flat field baseline stays within allowance; and high voltage circuit insulation resistance is unchanged from the pre-test value. These four convert a claim that the packaging looks fine into evidence that equipment performance is unchanged, which is the actual question validation is meant to answer. Related test procedures appear in GB/T 4857 transport packaging test methods and ISTA transport testing procedure.
Export projects justify one further layer: a full route rehearsal covering packing, lifting, loading, unloading and opening, recording time, personnel, tools and observed displacement at every step. Relocation failures rarely happen on the motorway; they happen during handling and secondary transfer, and those stages can only be exposed by rehearsal. Problems found in rehearsal are the cheapest to fix, and that rehearsal is usually worth more than another round of vibration testing.
13. Customization workflow and delivery coordination
Radiography cases are a textbook non-standard product. The same inspection machine is torn down differently from project to project; some ship as a complete machine with its cabinet, others send the source and detector in separate cases by air. The case concept therefore has to start from the agreed teardown boundary and transport mode before any structure is discussed. A six step workflow is recommended: component survey with a mass and centre of gravity list; marking of sensitive and no-contact faces; confirmation of transport attitude and lifting scheme; insert, support and compartment layout design; sample case validation including empty lifting, loaded vibration and an opening rehearsal; and series production with receiving inspection.
The information a customer needs to provide includes the tube model and its factory transport requirements, the detector model and whether the original case is retained, the generator type as oil-filled or SF6, 2D and 3D drawings of components, individual masses and centres of gravity, whether any isotope source is present, transport mode and destination climate, whether air freight is involved with its large pressure and humidity swings, and destination wood packaging and quarantine rules.
JUNZHIJIA builds radiography and precision electronics cases with compartmentalisation by component grade. It can provide rigid support and axial location for tubes, rigid cover plates and independently humidity-controlled compartments for detectors, attitude restraint and tilt indication for high voltage units, and silicone-free foam systems for inserts, together with matched hinges, latches and seals by model. Under OEM and ODM arrangements, factory marks, component number labels and QR traceability can be applied, and insert material statements, desiccant configuration lists and transport test reports can be supplied with the shipment. This batch is manufactured by JUNZHIJIA. General points on factory evaluation and sampling validation are in how to choose a protective case OEM factory, and case life assessment for repeated trips is in protective case service life.
Frequently Asked Questions
Q: What is the greatest risk to an X-ray tube during transport?
A: Not a drop, but sustained vibration and micro-movement. A drop causes an obvious vacuum leak that is easy to find and act on. Vibration produces a far more insidious failure: the axial position of the anode target and cathode assembly shifts by a fraction of a millimetre, microcracks appear in the ceramic-to-metal joints under alternating stress, and vacuum degrades gradually. After installation this shows up as repeated flashover during voltage ramping before the beam stabilises, drifting beam current readings and random bright pixels, and these symptoms are usually attributed to the high voltage supply or to software parameters, so fault finding starts in the wrong place. Protection should concentrate on rigidly coupling the tube body to the case structure, restraining all six degrees of freedom, keeping bearing surfaces off machined faces, and fitting a three-axis acceleration recorder inside the case so actual transport peaks are recorded as objective evidence for the arrival decision and for any claim. Wrapping the tube in an extra layer of bubble film solves neither the restraint problem nor the evidence problem.
Q: Can an oil-filled high voltage generator be shipped on its side or inverted?
A: In principle no, unless the maker's manual explicitly permits it. The insulating oil serves both as dielectric and as coolant, and the tank normally includes expansion volume or a breather to absorb oil volume change with temperature. Laying the unit down or inverting it collects oil to one side, so part of the winding or insulation sits above the oil line and contacts air and moisture directly while the case vibrates in transit. Once the unit is righted the parts are immersed again, but surface dampness and oxidation that have already occurred do not reverse themselves. The breather is the greater worry: if it sits against the oil surface in the wrong attitude, the unit may draw oil or expel oil in service, contaminating surfaces and lowering the oil level. Ship upright, mark the centre of gravity and upright direction on the case, fit tilt indicators inside, and read the oil level and label status before energising. If container space genuinely requires horizontal shipment, obtain written approval from the maker and follow its drain or bracing instructions.
Q: How can you tell whether a flat panel detector was damaged in transit?
A: Not by eye. A baseline comparison is required. Let the detector settle at a stable temperature, run the maker's offset and gain calibration, then acquire several flat field images across different dose levels and frame rates. Count new defective pixels and record their locations and whether they form row or column patterns, then difference the result against the factory defect map. Three judgement points matter: whether new pixel loss stays within the contractually allowed figure; whether the new pixels form rows, columns or a characteristic regional pattern, which normally points to a glass substrate microcrack or a driver interface fault; and whether noise floor and dark current match factory values. Also collect the indirect physical evidence: whether the window cover plate was still in place, whether condensation appeared inside the clean bag, and whether any label triggered. If something is wrong, secure the flat field image files together with the detector serial number and the environmental records before contacting the maker, and do not keep energising and re-acquiring while the initial state is undocumented, since that only obscures the starting condition.
Q: What special requirements apply to shipping lead shielding and lead glass?
A: Three rules govern it: lead must not carry load, lead glass must not be bent, and neither surface may be marked. Lead creeps at room temperature, so lead sheet and lead curtains must never act as structural members or be used to level other parts. They have to be framed in timber or plastic so the load passes through the frame into the case structure. Lead glass has limited strength and its edges are the most fragile, so it travels upright with soft packing on both faces to limit movement, never touching metal parts and never pressed tight against a case wall where vibration will squeeze it. Surface indentation matters because it changes local lead equivalent thickness, which creates real problems in shielding assessment or dose verification. Finally, if the equipment contains an isotope source, transport falls under GB 11806 and IAEA SSR-6 radioactive material requirements for packaging, marking and documentation. In that case the lead container is part of a regulated package and is handled entirely differently from ordinary shielding, and a qualified body should carry out the work.
Q: Does a de-energised industrial X-ray machine count as dangerous goods?
A: A conventional machine does not. X-rays are produced only when the tube is energised and electrons strike the target, so a de-energised unit emits no radiation and is not a radioactive consignment requiring radioactive transport paperwork. Packaging design therefore targets physical integrity of the equipment rather than continuous shielding. Two exceptions must be separated from this. The first is equipment containing a radioactive isotope source, such as a calibration source on a thickness gauge, a level gauge or a gamma radiography set. Those fall under the safe transport of radioactive material rules, GB 11806 and IAEA SSR-6, as class 7 dangerous goods, with package type, marking, transport index and documentation determined by a qualified organisation. The second is lead-bearing items exported to markets with specific lead restrictions, where the declaration route must be checked against destination law. Both are compliance consultancy matters beyond the scope of packaging design, and the discussion here describes the boundary of the packaging task and is not a compliance conclusion.
Q: Why should insert materials avoid silicone rubber and silicone grease?
A: Because siloxanes migrate and deposit where they do the most harm. Silicone rubber, silicone sealant and silicone grease continuously release low molecular weight cyclic siloxanes at room temperature, and those molecules travel through the enclosed air of a case and form an extremely thin silicone film on every surface they reach. Two locations suffer most on radiography equipment. The first is the tube beryllium window, where the film absorbs low energy X-rays and changes transmission behaviour in the low energy band, and that layer cannot be removed without damaging the window. The second is the detector window and optical components, where the film causes scatter and non-uniform response that appears as a hazy artefact in flat field images. Seals, pads and cushioning materials near these items must therefore use silicone-free rubber or foam systems, and silicone-based lubricants must not be used on threads or sliding surfaces during assembly. The most common way this requirement fails in practice is field repair, where technicians habitually reach for silicone grease on connectors and O-rings, and once it is introduced before packing it stays sealed inside the case acting on those surfaces for the whole voyage.
Q: Should servo drives and fans be removed from the cabinet before transport?
A: Usually not, but the maker's transport locking requirement must be carried out, and that matters more than removal. Servo drives, variable frequency drives, cooling fans and some units with a moving mass are designed with transport lock bolt holes or dedicated bracket locations, and manuals typically state that locking is mandatory before shipment. Without locking, fan impellers and motor rotors hammer their bearings through sustained vibration, and heat sinks and capacitors put alternating stress into the board, with solder fatigue showing up as occasional alarms or parameter jumps at commissioning that take a long time to trace. Removal brings its own risks that are easy to underestimate: modules then need their own packaging, repeated connector mating reduces contact reliability, and reinstallation torque and shield grounding quality are hard to restore. The engineering priority is therefore lock first and consider removal second. Only remove modules when the maker explicitly requires it, or when transport weight must be reduced to suit available lifting capacity, and in that case record every removed position and its refitting requirement in the in-case documentation.
Q: In a machine relocation project, which stage causes the most trouble?
A: Handling and secondary transfer, not the long trunk haul. The trunk leg looks riskier because it lasts longer and accumulates more vibration, but its loading method is normally defined in the packing plan. Handling is where conditions are least controlled: lifting equipment on site does not match the marked lift points, temporary labour has no knowledge of which components are fragile, forklift tines go under the case base or slings are pulled at an angle to save time, and the case is set down on damp ground at an intermediate warehouse. These operations are usually the dominant source of peak acceleration across the whole journey. A relocation should therefore include one complete rehearsal of packing, lifting, loading, unloading and opening, recording time, tools, personnel and observed displacement at each step, with the lifting diagram, the opening sequence with sensitive face warnings, and an exception contact carried in the case and signed off at every handover.
Conclusion and Further Reading
Five things decide whether an X-ray machine case works: a tube restrained in all six degrees of freedom, detector and collimator slit held at zero contact in a dry compartment, high voltage items in factory attitude with cable bend radius above the working minimum, lead parts neither loaded nor bent, and records good enough to support an arrival decision.
Further Reading