Key points: The visibility of a high-pressure gas jet comes from light refraction changes caused by density gradients, not from the gas itself having color. Schlieren and shadowgraph methods are the core optical means of making the jet "visible," while the high-speed camera records the evolution of penetration distance, cone angle, and shock structure at a sufficient frame rate.
Transparent gas produces almost no intensity contrast in visible light, and direct imaging yields only a uniform background. A schlieren system uses the deflection of light by density gradients to convert invisible jet structures into grayscale variations: regions with larger density gradients (such as shock waves, jet boundaries, and Mach disks) produce stronger light–dark contrast in the schlieren image. The shadowgraph method is simpler; it relies on light being deflected by density-change regions to form dark areas on the background, making it suitable for capturing the overall jet contour.
At the quantitative analysis level, high-speed schlieren image sequences can be used to extract the curve of jet tip penetration over time, the spray cone angle, and the evolution of jet volume. A study using high-speed schlieren photography to investigate high-pressure natural gas injection in a constant-volume chamber showed that jet development can be divided into an initial stage and a main stage: tip penetration increases rapidly in the initial stage and then grows approximately linearly in the main stage; the spray cone angle remains essentially constant during the main development stage. Increasing injection pressure increases penetration distance and volume while decreasing cone angle; increasing back pressure decreases penetration distance and volume while increasing cone angle. These conclusions come from frame-by-frame measurement of image sequences, not single-frame observation.
The core constraints of the high-speed camera in a schlieren system are frame rate and exposure: shock waves and jet interfaces can move at hundreds of meters per second, so exposure time must be short enough that the image motion of the interface within a single frame does not exceed acceptable blur; at the same time, the light throughput of a schlieren system is usually limited, requiring a feasible combination of light source intensity, camera sensitivity, and exposure time.

| Analysis Quantity | Image Basis | Physical Meaning | Frame Rate/Exposure Requirements |
|---|---|---|---|
| Penetration distance | Jet tip position across frames | Jet development speed and stage division | High frame rate to capture rapid initial changes |
| Spray cone angle | Angle between jet boundaries | Degree of jet dispersion | Resolvable boundaries; boundary contrast is key |
| Shock/Mach disk position | Strong light–dark interfaces in schlieren image | Degree of jet under-expansion and pressure ratio | Short exposure to freeze shock motion; high frame rate to track position |
| Jet volume | Area enclosed by jet boundaries | Gas diffusion and mixing range | Depends on boundary segmentation accuracy; sensitive to exposure uniformity |
Table 3 shows that the accuracy of jet analysis depends first on whether the schlieren system can produce sufficient boundary contrast, and second on whether the camera can provide enough frames at each key stage of jet evolution. Penetration distance changes fastest in the initial stage and places the highest demand on frame rate; cone angle is relatively stable in the main stage, so the frame-rate requirement can be relaxed.
Key points: The key difference between powder particles and gas is that particles are discrete targets. The analysis goal of high-speed photography is not "a flow field," but the position, velocity, trajectory, and collision behavior of individual particles. Particle visibility depends on the contrast between particles and background, while trajectory trackability depends on whether the inter-frame displacement of particles is small enough for correct matching.
In powder injection, particles are accelerated and dispersed from the nozzle by the gas flow, and some particles subsequently collide with walls or other particles. The high-speed camera directly records scattered light or shadows from particles, capturing the position of each particle across consecutive frames. Unlike PIV (particle image velocimetry), which focuses on the full-field velocity distribution obtained statistically from tracer particles, powder particle motion analysis more often uses PTV (particle tracking velocimetry) or LPT (Lagrangian particle tracking) to reconstruct the motion history of individual particles.
The basic workflow of PTV/LPT is: identify particles and determine their centroids in each frame; establish particle matching relationships between adjacent frames; calculate velocity vectors from the displacement and time interval of matched particles; and concatenate consecutive matches into trajectories. Matching reliability depends heavily on frame rate: if particles move too far between frames, or if two particles are close to each other within the field of view, the matching cost function may become ambiguous. In a high-speed particle tracking experiment using 1024×1024 resolution at 3000 fps, a Lagrangian particle tracking algorithm was used to quantitatively determine the velocity vectors and motion trajectories of solid particles.
Particle collision and fragmentation processes place even higher demands on frame rate. Collision duration may be on the order of tens of microseconds, requiring thousands to tens of thousands of frames per second to capture the sudden velocity changes before and after collision. Exposure time must be short enough that the image motion of particles at the collision instant is negligible; otherwise, judgment of rebound direction and fragmentation mode will be contaminated by motion blur.
| Analysis Quantity | Image Basis | Analysis Method | Key Recording Constraints |
|---|---|---|---|
| Particle velocity | Centroid displacement between adjacent frames / Δt | PTV/LPT | Inter-frame displacement should not exceed particle spacing; high frame rate aids matching |
| Motion trajectory | Concatenation of centroids across consecutive frames | Lagrangian tracking | Particles must remain continuously identifiable within the field of view; avoid occlusion and exiting the frame |
| Particle size | Projected area/diameter in a single frame | Image segmentation and sub-pixel interpolation | Particles must occupy enough pixels; uniform exposure |
| Collision/fragmentation behavior | Morphological and velocity changes before and after collision | Frame-by-frame morphological comparison | Very high frame rate + short exposure to freeze transient deformation |
| Concentration distribution | Particle counting per unit area | In-frame particle detection and statistics | Avoid missed detection due to overlapping particles |
The key conclusion of Table 4 is that in powder particle analysis, frame rate primarily serves trajectory matching, and exposure primarily serves centroid localization. If inter-frame displacement is too large, trajectory reconstruction will fail even if individual frames are sharp; if exposure is insufficient or motion blur is severe, centroid localization errors will propagate directly into velocity and trajectory calculations.
Key points: Slow-motion playback itself is not the end point of analysis. The analytical value of high-speed video comes from converting image sequences into quantitative data: direct measurement (displacement, angle, velocity), PTV/LPT (individual particle trajectories), PIV (full-field velocity distribution), and DIC (deformation and strain). Which method to choose depends on whether the target is "discrete particles" or a "continuous field."
The most basic analysis is direct measurement from high-speed video: place a scale of known size in the image, read the target's position, angle, or characteristic length frame by frame, divide inter-frame displacement by the time interval to obtain velocity, and use fast Fourier transform (FFT) to analyze the frequency of periodic motion. This method does not require complex algorithms but places relatively high demands on image quality and inter-frame consistency.
PTV/LPT is suitable for dilute particle flows or cases where the history of individual particles must be tracked; it focuses on "where this particle went and how its velocity changed." PIV is suitable for statistical velocity field measurement of continuous flow fields or high-concentration particle groups: tracer particles are added to the flow (or particle scattering itself is used), a laser sheet illuminates the flow, and cross-correlation algorithms calculate the average displacement within interrogation windows to obtain the full-field velocity vector distribution. PIV provides an Eulerian field, while PTV provides Lagrangian trajectories; the two can be used complementarily in gas–particle two-phase flows.
For deformation or material response after particle collision, DIC (digital image correlation) can apply a speckle pattern to the particle or wall surface and calculate full-field strain by comparing speckle displacement before and after deformation, with accuracy on the order of 0.01%. The prerequisite for DIC is a trackable surface texture; smooth metal particles require additional speckle patterning.
The engineering keys of the analysis workflow are calibration and synchronization: spatial calibration determines the mapping between pixels and physical dimensions, temporal calibration determines the precise frame interval, and trigger synchronization ensures that the image sequence is temporally aligned with external events (such as valve opening or collision occurrence). Without any one of these, systematic errors in quantitative results will exceed random errors.
| Analysis Method | Output Data | Applicable Object | Prerequisites |
|---|---|---|---|
| Direct measurement | Displacement, angle, velocity, frequency | Macroscopic motion, periodic processes | Scale calibration, consistent inter-frame illumination |
| PTV/LPT | Velocity vectors and trajectories of individual particles | Dilute particle flows, collisions and agglomeration | Particles identifiable frame by frame; inter-frame displacement matchable |
| PIV | Full-field velocity vector distribution | Continuous flow fields, high-concentration particle groups | Sufficient tracer particle or particle scattering; laser sheet illumination |
| DIC | Full-field deformation and strain | Particle/wall deformation, material response | Surface speckle pattern; dual cameras enable 3D measurement |
Table 5 shows that if the focus is "how one particle moves," use PTV; if the focus is "how the entire flow field moves," use PIV; if the focus is "how deformation occurs after collision," use DIC; and if only basic kinematic quantities are needed, direct measurement is often the fastest and most reliable starting point.
Key points: MindVision's MV-PX250GC/M, MV-PX1000GC/GM, and MV-XGC51GC/GM correspond respectively to three typical trade-offs in high-pressure gas/powder recording: extremely high frame rate first (PX250's ROI mode), high resolution first (PX1000's full-frame 10 MP), and high frame rate + flexible deployment (XGC51's 10GigE interface and sub-millisecond exposure). All three share short exposure capability and global shutter, but target different analysis goals.
The core contradiction in high-speed camera selection has been explained in Section 2: frame rate, exposure, and resolution cannot all be maximized simultaneously. The positioning differences among these three MindVision cameras are essentially three different answers to this contradiction. MV-PX250GC can reach 8367.75 fps or even 12662.5 fps after reducing the ROI, making it suitable for scenarios requiring extremely high temporal density to track fast interfaces or collision transients. MV-PX1000GC, with 4608×2176 full-frame 10 MP and about 465 fps, prioritizes spatial detail and suits occasions requiring rich geometric information from a single frame. MV-XGC51GC, with a 10GigE interface and minimum exposure time starting at 0.0008 ms, provides more flexible deployment and stronger motion-freezing capability at about 1594 fps full frame.
All three cameras support global shutter, which is irreplaceable in particle motion analysis: rolling shutter produces geometric distortion under high-speed motion, leading to systematic bias in centroid localization and velocity calculation. Their short exposure ranges (minimum 0.0010 ms for the PX series and 0.0008 ms for XGC51) enable them to freeze microsecond-scale motion with strong illumination.
| Parameter | MV-PX250GC/M | MV-PX1000GC/GM | MV-XGC51GC/GM |
|---|---|---|---|
| Effective pixels | 2.5 MP | 10 MP | 0.5 MP |
| Full-frame resolution | 2048×1216 | 4608×2176 | 816×624 |
| Max full-frame frame rate | About 1884 fps | About 465 fps | About 1595 fps |
| ROI high-frame-rate example | 640×208 @ 8367.75 fps; 128×128 @ 12662.5 fps | Custom ROI resolution supported | Custom ROI resolution supported |
| Minimum exposure time | 0.0010 ms | 0.0010 ms | 0.0008 ms |
| Shutter type | Global shutter | Global shutter | Global shutter |
| Data interface | CoaXPress-12 | CoaXPress-12 | 10GigE |
| Typical positioning | Extremely high frame rate first; ROI trades for temporal density | High resolution first; rich single-frame geometric detail | High frame rate + flexible cabling; sub-millisecond exposure |
The key conclusion of Table 6 is that MV-PX250GC trades resolution for frame rate, achieving the highest temporal density in ROI mode and suiting inter-frame tracking of jet interfaces and collision transients; MV-PX1000GC trades frame rate for resolution, retaining 10 MP at full frame and suiting occasions requiring precise measurement of particle size, shape, or spatial distribution in a single frame; MV-XGC51GC reduces cabling constraints with a 10GigE interface while offering the lowest exposure starting point among the three, suiting particle trajectory recording that requires flexible deployment and extremely high motion-freezing capability.
The core parameter combination of MV-PX250GC/M is 2048×1216 at about 1884 fps full frame, reaching 8367.75 fps (640×208) or even 12662.5 fps (128×128) after reducing the ROI, with a minimum exposure time of 0.0010 ms. This positioning directly corresponds to scenarios in high-pressure gas jets and powder motion where temporal resolution takes priority over spatial coverage.
In high-pressure gas schlieren recording, the jet tip and shock interface move fastest in the initial stage, and penetration distance increases rapidly. If the frame rate is insufficient, several key frames in the initial stage may fall between two exposures, causing undersampling at the beginning of the jet development curve. The frame rate above 8000 fps provided by MV-PX250GC in ROI mode reduces the temporal sampling interval in the initial jet stage to about 0.12 ms, sufficient to obtain multiple valid data points in the rapid-change segment of penetration distance. The cost of ROI is a reduced field of view, but the initial jet is concentrated near the nozzle, so sacrificing far-field view at this stage is an acceptable engineering compromise.
In powder particle collision or fragmentation scenarios, transient events may last on the order of tens of microseconds. Full-frame 1884 fps provides about 0.53 ms frame interval, sufficient for slow particles; but for high-speed collisions, the ten-thousand-fps-level frame rate in ROI mode is needed to capture sudden velocity changes and morphological changes before and after collision. The minimum 0.0010 ms exposure time, combined with strong pulsed illumination, can freeze particles in a single frame and keep collision-instant boundaries segmentable.
MV-PX1000GC/GM, with 4608×2176 full-frame 10 MP resolution, about 465 fps full-frame frame rate, 4.5 μm pixels, and 12-bit output, prioritizes spatial detail. In powder particle analysis, the significance of this trade-off is that particle size distribution, shape factor, and post-collision fragment morphology all depend on the number of pixels occupied by particles in a single frame. If particles occupy only a few pixels in the image, the uncertainty of centroid localization and diameter measurement increases significantly.
The 10 MP full frame provides more pixel coverage for particles. At the same lens working distance, PX1000's pixel density allows smaller particles to be reliably segmented, or allows usable particle resolution to be maintained over a larger field of view. Although the 465 fps full-frame rate is lower than PX250's ROI mode, this combination of frame rate and resolution is more balanced for medium-to-low-speed particle flows (where inter-frame displacement remains matchable) or single-frame analysis requiring high spatial detail (such as particle shape classification or post-collision fragment counting).
The 12-bit output also has practical value in low-contrast particle imaging: the gray-level difference of dark particles against a light background or in transparent tubing may span only a few bits, and 12-bit quantization preserves more levels, making threshold segmentation and edge detection more stable than with 8-bit images.
The positioning difference of MV-XGC51GC/GM lies not in absolute frame rate or absolute resolution, but in the combination of interface and exposure. It offers 816×624 full frame at about 1595 fps, a minimum exposure time of 0.0008 ms, and a 10GigE interface supporting transmission distances up to 100 meters. For powder particle trajectory observation, this means two things.
First, the 10GigE interface reduces cabling constraints for multi-camera or long-distance deployment. CoaXPress has advantages in extremely high-bandwidth scenarios but requires dedicated frame grabbers and coaxial cabling; 10GigE uses standard copper cables and is more flexible when cameras need to be placed far from the control cabinet (such as on large injection test benches or along pipelines). For particle trajectory tracking, a single camera is usually sufficient, but if simultaneous recording from two angles (stereo PTV) or multiple cameras over a large field of view is required, interface deployment cost becomes a practical selection factor.
Second, the minimum exposure time of 0.0008 ms makes XGC51 slightly better than the PX series' 0.0010 ms in motion-freezing capability. This difference may not be significant at conventional particle velocities, but in cases requiring extremely short exposure to eliminate motion blur while illumination intensity is limited (such as insufficient scattered light from high-speed particles in a diffusion section), a lower exposure starting point provides greater parameter adjustment margin. The 0.5 MP resolution is sufficient for the task of "tracking particle centroids and calculating velocity direction"—the core constraint of trajectory analysis is inter-frame matching, not single-frame pixel count; as long as particles can be stably segmented and localized within the field of view, lower resolution can instead help increase frame rate and reduce data transmission pressure.
Key points: The most common decisions in high-speed photography testing focus on four aspects: the relationship between frame rate and exposure, resolution trade-offs, differences between gas and powder recording, and "how to analyze after recording."
No. Frame rate determines temporal sampling density, but whether a single frame is sharp is determined by exposure time. If exposure time is too long, every frame remains blurred even at a high frame rate, and trajectory tracking and centroid localization accuracy will not improve with higher frame rate. The correct order is: first determine acceptable motion blur based on target velocity (typically using image motion not exceeding about 1 pixel as a reference) and derive the upper limit of exposure time; then determine the lower limit of frame rate based on whether inter-frame displacement is matchable. Once both are satisfied, a higher frame rate only increases data volume, not useful information.
The criterion for freezing is the amount of motion blur. If the particle velocity is v and the exposure time is t_exp, the image motion is approximately v·t_exp (convertible to pixels when magnification is known). In engineering, image motion is usually expected not to exceed 1 pixel, and at most 2–3 pixels. For example, if particles move at 1000 pixel/s in the image, a 1 μs exposure corresponds to about 0.001 pixel image motion, 10 μs to about 0.01 pixel, and 100 μs to about 0.1 pixel. Actual settings also need to consider light throughput: shorter exposure requires stronger illumination and a larger lens aperture, which correspondingly reduces depth of field.
Gas itself is transparent and requires a schlieren or shadowgraph system to convert density gradients into grayscale contrast; the camera records refraction changes in a continuous field. Powder particles have scattering or absorption properties, and the camera directly records the position and morphology of discrete targets, usually without schlieren. Both require short exposure to freeze motion. The difference is that gas recording concentrates frame-rate requirements on jet interface and shock tracking, while powder recording concentrates them on inter-frame particle matching and collision transients.
This is a typical symptom of insufficient frame rate or an excessively large field of view. Prioritize increasing frame rate (reducing ROI or resolution) so that inter-frame displacement is reduced to a fraction of particle spacing. If the frame rate is already at its limit, reduce the field of view (increase lens magnification) to lower particle velocity in the image. Another strategy is to improve particle visibility: increase contrast between background and particles, and use pulsed illumination to improve single-frame SNR, so that matching algorithms remain reliable at larger displacements. Relying solely on post-processing algorithms to compensate for insufficient frame rate has limited effectiveness.
At least four types of quantitative analysis can be performed: direct measurement (displacement, angle, velocity, frequency), PTV/LPT (trajectory and velocity history of individual particles), PIV (full-field velocity vector distribution), and DIC (deformation and strain). The choice depends on whether the target is discrete particles or a continuous flow field, whether the focus is kinematic quantities or deformation, and whether particle concentration is dilute or dense. Spatial and temporal calibration must be completed before analysis; otherwise, systematic errors in quantitative results will mask real physical differences.
| Question | Core Answer | Key Judgment Basis |
|---|---|---|
| Is a higher frame rate always better? | No; single-frame sharpness is determined first by exposure | Derive exposure upper limit first, then determine frame-rate lower limit |
| How to set exposure time? | Use image motion not exceeding about 1 pixel as a reference | Convert v·t_exp to pixel displacement |
| How do gas and powder recording differ? | Gas requires schlieren/shadowgraph; powder is directly imaged | Continuous field vs. discrete targets |
| What if particles are lost? | Prioritize increasing frame rate or reducing field of view | Inter-frame displacement too large relative to particle spacing |
| What analysis can be done after recording? | Direct measurement, PTV/LPT, PIV, DIC | Discrete/continuous, kinematics/deformation |
Key points: The value of high-speed industrial cameras in high-pressure gas and powder motion ultimately depends on whether they can convert transient processes into repeatable, calibratable, and quantitatively comparable image data. The trade-offs among frame rate, exposure, and resolution are the means; schlieren, PTV/PIV, and direct measurement are the paths; and calibration and synchronization are the prerequisites for turning "slow-motion video" into "scientific data."
Schlieren sequences of high-pressure gas jets can provide the evolution of penetration distance, cone angle, and shock position over time; high-speed images of powder particles can reconstruct individual particle trajectories, velocities, and collision behavior. The reliability of these results does not depend on whether the camera "looks advanced," but on whether recording parameters match the target motion scale, whether the optical system produces sufficient contrast, and whether temporal and spatial calibration is closed.
MindVision's MV-PX250GC/M, MV-PX1000GC/GM, and MV-XGC51GC/GM provide three different starting points for trade-offs: PX250 serves extremely high temporal density requirements with ROI high frame rate, PX1000 serves spatial-detail-priority analysis with full-frame high resolution, and XGC51 serves particle trajectory observation with 10GigE deployment flexibility and sub-millisecond exposure. The key to selection is not comparing whose "maximum frame rate" is higher, but first answering "how fast the target moves, what features must be seen, and what the analysis output is," then judging which parameter combination falls within the acceptable range for the task. When camera parameters, illumination/schlieren configuration, and analysis methods are designed as a whole, high-speed photography truly moves from "seeing the instant" to "measuring the instant."
| Conclusion Dimension | Core Conclusion |
|---|---|
| Nature of recording | A physical trade-off among frame rate, exposure, and resolution; all three cannot be maximized simultaneously |
| Freezing motion | Exposure time determines single-frame sharpness; frame rate determines temporal sampling density |
| Gas jets | Schlieren/shadowgraph visualization + high frame rate tracking of penetration distance, cone angle, and shocks |
| Powder particles | Direct imaging + PTV/LPT reconstruction of trajectories, velocities, and collision behavior |
| Analysis paths | Direct measurement, PTV/LPT, PIV, DIC; choose by discrete/continuous and kinematics/deformation |
| Camera positioning | PX250 prioritizes temporal density; PX1000 prioritizes spatial detail; XGC51 offers deployment flexibility and sub-millisecond exposure |
| Prerequisites for data reliability | Spatial calibration + temporal calibration + trigger synchronization; none can be omitted |
MindVision Technology Co., Ltd. is a professional domestic industrial camera manufacturer headquartered in Shenzhen, with production centers and R&D bases in Changsha (Hunan) and Wuxi (Jiangsu), employing nearly 300 people. Its product portfolio covers area-scan cameras, line-scan cameras, high-speed cameras, thermal imaging cameras, and stereo camera modules, with over 100 patents and software copyrights. The company is ISO9001 certified.
For more information or free selection consulting, please visit MindVision's official website or contact the sales team.
Please Login