2.3 Hardware Overview
Four current devices, two output types, one ecosystem. Specifications, capabilities, and honest differences.
Device Specifications
All four current devices use Multi-SLAM (LiDAR + Visual + IMU). The differences are LiDAR configuration, scan speed, output type, and environment scale.
Lixel L3
High-Accuracy Survey, Building to Campus ScaleTwo configurations: 16/120 and 32/120. Both reach 120 m (394 ft). Output formats: point cloud .las, .rcp, .e57, .ply, .ilas; mesh .osgb, .obj; 3DGS .lcc2, .lcc, .ply; plus 3D Tiles (.3tz), .usd, and .jpg imagery. Mounts to a backpack or a standard RTK survey pole, with control points marked directly from the pole; vehicle and drone mounts are in development. The built-in display runs the scan with no phone connected and shows the point cloud, trajectory, coverage, and capture status, with a real-time quality check. AI dynamic object removal and AI noise filtering run on the device during capture.
Lixel L2 Pro
Campus and Multi-Building ScaleThree configurations: 16/120, 32/120, and 32/300. First number = channels, second = max range in meters. See model comparison below. Mounts to a backpack, a vehicle, or a drone (DJI M350 RTK and M300 RTK), and is the only device with drone mounting.
Lixel K2
Lightweight, Indoor and Small-to-Medium OutdoorOutput formats: point cloud .las, .rcp, .e57, .ply, .ilas; mesh .osgb, .obj; 3DGS .lcc2, .lcc, .ply; plus 3D Tiles and .usd. In LCC Studio the K2 runs all three reconstruction modes: Single Scene, Map Fusion (up to 10 sub-scenes per run), and Aerial-Ground Map Fusion. Single-scene length is bounded by the device capture ceiling above and by workstation memory, not by any time limit in the software. Handheld only.
PortalCam
High-Fidelity Visual CaptureCannot produce point clouds and does not output absolute georeferenced coordinates. Processes through LCC Studio only. Control points are used to connect segments during Map Fusion. XGRIDS advises against captures over 30 minutes in any mode, well below the firmware ceiling, because longer captures tend to produce drift and layering artifacts. XGRIDS lists a GPS module for this device but does not document what it does; do not plan georeferenced work around it until that is confirmed.
What the PortalCam accuracy figures mean. The relative and repeat figures above describe internal geometric consistency within a reconstruction. They are not a measurement tolerance, and the PortalCam has no absolute accuracy figure because it produces no georeferenced output. Do not quote them in a proposal or a scope of work. If a client needs a number they can hold you to, the job needs an L3, an L2 Pro, or a K2.
Reading the output rows. The L3, L2 Pro, and K2 all feed LixelStudio and LCC Studio, so all three reach the same export set including mesh. The L3 and K2 rows list formats in more detail because their published specification sheets do; this is a difference between the source documents, not a capability difference between the devices.
Reading the L3 card. The L3's 5 mm (0.2 in) relative accuracy is stated at 1 sigma for a post-processed capture in a 5,382 sq ft (500 m²) indoor test environment. The L2 Pro and K2 figures are RMSE, so treat the two as results of different tests, not a ranking.
No L3 configuration reaches the L2 Pro 32-300's 300 m (984 ft). For work beyond 120 m (394 ft), use the L2 Pro 32-300. The L3 against the L2 Pro in full, including why their accuracy figures come from different tests, is on the L3 buying guide.
Coordinate system constraint. The RTK module on both the L2 Pro and the K2 obtains WGS84, CGCS2000, and ITRF2008 with geodetic height. The narrower constraint is on direct output: the L2 Pro writes absolute coordinates natively only in WGS84 UTM 3-degree zone and CGCS2000 Gauss-Kruger 3-degree zone. If your project requires NAD83, NAD27, or another regional datum, plan for a transformation in LixelStudio on either device rather than expecting it at the point of capture. The L3's real-time coordinate conversion supports WGS84 and CGCS2000 only, so the same applies to it.
L2 Pro Models: 16-120, 32-120, and 32-300
Map Fusion requires matching L2 Pro models. A 16-channel scan cannot be fused with a 32-channel scan. LixelStudio treats them as different device types. Confirm model configuration before combining sessions.
Use Cases by Device
High-Accuracy Survey and Scan-to-BIM
Projects that need survey-grade geometry and a photorealistic deliverable from the same mobile capture. Survey-grade RTK is built in, and one pass produces a true-color point cloud, a mesh, and a 3DGS model. Covers building surveys up to campus and urban block mapping within the 120 m (394 ft) maximum LiDAR range.
- Building, as-built, and renovation and extension surveys
- Scan-to-BIM and as-built records
- Combined indoor-outdoor surveys
- Tunnels and utility corridors, with LixelStudio's stable mapping for low-feature scenes
- Industrial facility documentation and asset inventory
- Digital twin and virtual production work, where the 3DGS model is the deliverable
- Multi-session projects fused by region selection in LixelStudio, without control points
- Sites where RTK drops partway through: Measure Point carries absolute coordinates into the denied area
- Backpack mounting, with vehicle and drone mounts in development
Campus and Large-Scale Projects
Multi-building sites, mixed indoor/outdoor environments, and projects where geometry exists at significant distances from the scanner. Covers hundreds of thousands of square feet per battery.
- Multi-building campus documentation
- Large commercial interiors and warehouses
- Mixed indoor/outdoor on a single scan route
- RTK or PPK georeferencing projects
- Sites where RTK drops partway through: Measure Point carries absolute coordinates into the denied area at 5 cm or better within 50 m (164 ft) of the disconnection point
- Drone-mounted scanning (DJI M350 RTK and M300 RTK), plus backpack and vehicle mounting
- Tunnels and degraded environments (32-300 model)
Lightweight Indoor and Small-to-Medium Outdoor
Single-building and dense interior scanning, surveying and topographic mapping, BIM data capture and verification. The K2's built-in RTK and real-time colorized point cloud reduce on-site verification time and remove the need for an external RTK module. Covers most commercial spaces and small-to-medium outdoor sites within the 100 m (328 ft) maximum LiDAR range.
- Land surveying and topographic mapping
- As-built measurement and quantity surveying
- Interior measurement and floor plan generation
- Stockpile and volumetric measurement
- BIM data capture and verification (under 10 m (33 ft) measurement envelope)
- Single-building documentation with direct geo-referenced output
- Multi-segment 3DGS projects through LCC Studio Map Fusion
Budget for twice the control density of an L2 Pro job on the same site: the K2 maximum control point interval is 50 m (164 ft) against the L2 Pro's 100 m (328 ft).
Visual Documentation
Used when the primary deliverable is a photorealistic 3DGS model, not a measured point cloud. LiDAR with 180° azimuth by 180° elevation coverage, paired with a four-camera array whose combined fisheye field exceeds a full sphere. Weighs 1.9 lbs without tripod. 60-minute battery (vs. 90 minutes for L2 Pro and K2).
- Client walkthroughs and virtual tours
- Visual progress documentation
- Spaces where visual quality matters and measurement is not required
The PortalCam cannot produce point clouds. Do not plan a measured-deliverable project with a PortalCam as the only capture device. Use an L3, L2 Pro, or K2 for point cloud output.
Full Comparison
The L3 against the L2 Pro in more depth, including why their accuracy figures come from different tests, is on the L3 buying guide.
The capture and battery rows are different facts that share their numbers. Maximum single capture is a firmware limit: recording stops there whatever the power situation. Operating time is how long a charge lasts. They coincide at 90 and 60 minutes because the hardware was engineered that way, not because either is a sensible scan length. Neither is a scan target, and on most workstations reconstruction memory binds long before either. Segment planning is on 6.2 Scan Splitting; workstation requirements are on 10.7 Device Specifications.
Accuracy conditions matter. Absolute accuracy of ≤3 cm (1.2 in) RMSE requires RTK or GCP coverage with no continuous unfixed gap exceeding 100 m (328 ft) on the L2 Pro or 50 m (164 ft) on the K2. Relative accuracy of ≤1 cm (0.4 in) applies between points within 100 m (328 ft) for the L2 Pro and within 10 m (33 ft) for the K2. Repeat accuracy assumes two RTK-enabled scans with no disconnection. The L3's 5 mm (0.2 in) is 1 sigma from an indoor test, not RMSE; see Reading the L3 card in Section 1. Its 2 cm (0.8 in) absolute accuracy is RMSE under open sky. Full accuracy guidance is in 10.8 Accuracy Reference.
The L2 Pro, K2, and PortalCam cannot resume a scan from a breakpoint. If a recording stops for any reason, that segment is closed and the next scan is a new one. The recovery path is Map Fusion, which needs overlap you must already have captured. Plan segment boundaries before you start.
RTK
The L3 has a survey-grade GNSS RTK module built in: 0.8 cm + 1 ppm horizontal, 1.5 cm + 1 ppm vertical, with a 4.0 dBi peak-gain antenna. No external attachment is required.
The K2 has the UM980 RTK module built in. No external attachment is required. The K2 outputs absolute coordinates directly with no post-processing.
The L2 Pro supports RTK via an attachable module. RTK is not built into the base scanner. The module must be attached and configured in LixelGO before the session starts, and the RTK type selected in the app must match the module physically fitted.
XGRIDS offers no RTK option for the PortalCam, and the device does not output absolute georeferenced coordinates. A GPS module is listed in the PortalCam specifications, but XGRIDS does not document its function, so do not plan georeferenced work around it. Control points on the PortalCam serve a different purpose: they connect scan segments during Map Fusion. See 7.4 PortalCam Map Fusion.
L2 Pro RTK Module Variants
Two handheld variants are available: Standard and Survey Grade. Both publish 0.8 cm + 1 ppm horizontal and 1.5 cm + 1 ppm vertical accuracy. The ppm term scales with baseline length to the correction source, so a long baseline widens the figure. Both track GPS L1/L2/L5, GLONASS L1/L2, BDS B1/B2/B3, and Galileo E1/E5a/E5b/E6b. The Survey Grade module adds a higher-gain antenna (5.5 dBi against 2.8 dBi) and a tighter antenna axis ratio for better performance under partial canopy, near structures, and in urban environments.
Full setup: 4.2 RTK Positioning. Introduction to when RTK is needed: 4.1 Accuracy Fundamentals. Field thresholds and fault diagnosis: 4.0 Field Guide.
Aerial RTK module. The L2 Pro also supports a drone-specific RTK module for use with DJI M350 RTK and M300 RTK aircraft. It is not used for handheld scanning. A drone mount for the L3 is in development for the DJI Matrice 400. K2 does not support drone mounting.
©2026 Alpine Reality Capture LLC • XGRIDS Pro Guide™ • Site Disclaimer