When construction GPS is humming along, it feels almost invisible: you load the file, the blade knows where it is, and grade checks become quick confirmations instead of constant rework. But when it’s not working, it can derail a whole shift—operators get frustrated, supervisors lose confidence in the numbers, and the schedule starts bleeding time in tiny, expensive increments.
This checklist is built for the real world: dusty cabs, rushed mornings, changing weather, and equipment that’s doing hard miles. It’s not meant to replace your dealer or manufacturer guidance; it’s meant to help you quickly narrow down what’s actually wrong, what you can fix on-site, and what needs escalation. If you follow it in order, you’ll usually find the issue faster (and avoid “fixing” the wrong thing).
One note before you start: “GPS not working” can mean a lot of different symptoms—no satellites, drifting elevation, no corrections, radio drops, lost design, wrong surface, or the machine “thinks” it’s somewhere else. As you work through the steps, keep your symptom in mind and write down what you see. Those notes will save you time if you end up calling support.
Start by naming the symptom (it’s the fastest shortcut)
Before touching any cables or rebooting anything, take 60 seconds to identify what “not working” looks like today. Is the position jumping around? Is the machine consistently off grade by the same amount? Are you missing corrections? Is the display showing a warning about satellites, base, or RTK? Different symptoms point to different root causes.
Try to capture three details: (1) whether the issue is constant or intermittent, (2) whether it started after a change (new file, moved base, new operator, firmware update, storm), and (3) whether it affects one machine or multiple machines. If multiple machines fail at once, suspect base station, network, or site conditions. If it’s only one machine, suspect hardware, settings, or the specific design file.
Also note whether the problem is horizontal (plan) or vertical (elevation). Vertical issues often come from calibration, antenna height, geoid/model settings, or corrections quality. Horizontal issues can be caused by poor satellite geometry, multipath near structures, wrong coordinate system, or a bad localization.
Quick safety-and-sanity check in the cab
It sounds obvious, but start with the basics that are easiest to miss when the crew is under pressure. Confirm the system is actually in the right mode (GNSS vs total station, 2D vs 3D, guidance vs control). Check that the correct job is loaded and that you’re not looking at yesterday’s surface.
Next, look for any warnings on the main screen: satellite count, PDOP/HDOP, correction age, radio signal, base ID, or “no RTK” messages. These indicators are often the quickest clue. If you can, take a photo of the warning screen—support teams can diagnose faster when they see the exact message.
If the display is laggy, freezing, or randomly rebooting, treat it as a power/connection issue first. A flaky power supply can mimic “GPS drift” because the receiver or display is constantly recovering. If you’ve got a machine that’s been fine for months and suddenly starts acting haunted, power and connectors are prime suspects.
Power, fuses, and the “it’s always the connector” reality
Construction environments are brutal on electrical connections. Vibration, dust, moisture, and repeated cab cleaning can loosen or corrode connectors over time. If your GPS receiver or display isn’t getting stable power, you’ll see random disconnects, slow performance, or complete loss of positioning.
Start with the simplest checks: confirm the machine voltage is normal, inspect the fuse (and the fuse holder), and verify that any inline power adapters are seated properly. If you have a spare power cable, swap it. Cable swaps are one of the fastest ways to isolate whether the issue is the component or the wiring.
Then inspect connectors for bent pins, corrosion, and dirt. Unplug and re-seat them firmly. If you see green corrosion, clean it properly (and don’t just “wiggle it until it works,” because it’ll fail again at the worst time). If the system uses multiple cables—CAN, serial, Ethernet—label them during inspection so you don’t reassemble incorrectly under time pressure.
Reboot sequence: do it in the right order
Rebooting can solve a surprising number of “GPS not working” problems, but only if you reboot the right pieces in a sensible sequence. Randomly cycling one device can leave the system in a mismatched state where the receiver and display disagree about job settings or correction sources.
A practical approach: power down the display, receiver, and any radio/modem components. Wait 30–60 seconds so capacitors discharge and the devices fully reset. Then power up the receiver first (so it can begin tracking satellites), then the correction source (radio/modem), and finally the display/controller. Give it a few minutes to settle—GNSS receivers often need time to reacquire satellites and corrections after a full power cycle.
If you’re on a tight schedule, it’s tempting to reboot and immediately judge the result. Try not to. Watch the satellite count, correction status, and solution type (autonomous/float/fixed). A system that starts in “float” and then becomes “fixed” after a minute may be totally fine—just slow to lock because of sky visibility or correction link quality.
Satellite visibility: the jobsite can block more than you think
GNSS needs a clean view of the sky. On paper, that’s obvious; on a jobsite, it’s easy to forget how quickly conditions change. Tall buildings, tree lines, stockpiles, cranes, and even parked equipment can block signals or create multipath—reflections that confuse the receiver and cause position jitter.
If the GPS is acting up, move the machine to an open area for a test. You’re not trying to finish work there—you’re trying to see if the system behaves normally with a clear sky view. If it suddenly stabilizes, the issue is likely environmental rather than hardware.
Pay attention to the time of day, too. Satellite geometry changes, and a spot that’s “fine in the morning” can become unreliable later. If you’re consistently seeing problems near a structure at certain times, you may need to adjust work sequencing, use a different correction method, or consider alternate positioning (like total station) for tight areas.
Check correction source: RTK radio, network RTK, or base station
Many “GPS not working” complaints are actually “no corrections.” Without RTK or a similar correction source, your accuracy will drop—sometimes enough that grade control becomes unusable. Look for indicators like correction type, correction age, and whether the receiver reports “fixed” or “float.”
If you’re using a local base station, confirm the base is powered, broadcasting, and set up correctly. A base moved even a small distance without updating coordinates can throw off the entire site. Confirm the base antenna is stable, vertical, and not wobbling in the wind. Check the base radio channel and ID settings match the rover (machine) settings.
If you’re using network RTK (cellular), check signal strength and data connection. Construction sites can have dead zones, and modems can fail quietly. If multiple machines lose corrections at the same time, suspect the base/network rather than individual receivers. If only one machine loses corrections, suspect its radio/modem, antenna, SIM/data plan, or configuration.
Localization and site calibration: the silent accuracy killer
When a system is “working” but consistently off in a way that doesn’t make sense, localization is often the culprit. A localization (or site calibration) ties your GNSS coordinates to the project coordinate system. If it’s wrong—or if the wrong localization file is loaded—you can be perfectly “fixed” and still be wrong on the ground.
Common red flags include: the machine matches itself (repeatable) but doesn’t match survey stakes; the offset is consistent across the site; or the error changes direction depending on where you are. Another clue is when one machine is fine and another is off while both claim “fixed RTK.” That can happen if they’re using different localizations or different versions of the job.
To troubleshoot, verify the correct project, coordinate system, and localization are selected. If you have known control points, occupy one with the machine (if your workflow allows) or compare to a rover check. Don’t “tweak” random offsets to make it look right—those quick fixes can create bigger problems later when the design changes or another machine joins the job.
Antenna and mast checks: small movement, big consequences
GNSS antennas are precise instruments living in a world of vibration and impacts. A slightly loose mount, a cracked antenna housing, or a mast that’s been bumped can introduce errors that look like software issues. If your system worked yesterday and now it’s drifting, physically inspect the antenna setup.
Check that the antenna is firmly mounted, level (if your system requires it), and free of damage. Inspect the coax cable for pinches, cuts, or tight bends—especially near hinges and pinch points. A damaged coax can degrade signal quality without fully failing, leading to intermittent issues that are hard to pin down.
If your machine uses dual antennas (for heading), confirm both are connected and recognized. Heading issues can cause the on-screen blade orientation to swing or behave oddly even when position seems okay. In those cases, the operator may describe it as “GPS is going crazy,” but the root cause is often a heading/antenna problem.
Firmware, licenses, and “it worked until we updated” situations
Firmware updates can improve stability and add features, but they can also introduce mismatches—especially if the display is updated but the receiver isn’t, or vice versa. If the problem appeared right after an update, make that your primary clue.
Check firmware versions across components (display, receiver, radio/modem) and verify they’re compatible. Also confirm that required licenses are active. Some systems will operate in a reduced mode if a license expires or fails to validate, and the symptoms can look like a positioning issue.
If you suspect firmware is involved, avoid repeated updates or rollbacks without a plan. Document your current versions first. If you have multiple machines, compare a working unit’s versions/settings to the failing one—differences often jump out when you put them side by side.
Design files and surfaces: when the GPS is fine but the model isn’t
Sometimes the GNSS positioning is accurate, but the design data is wrong, outdated, or loaded incorrectly. Operators experience this as “GPS not working” because the cut/fill numbers don’t match expectations, or the blade keeps chasing an incorrect surface.
Verify you’re using the correct design version and the correct surface or alignment. If multiple surfaces exist (subgrade, finish grade, temporary detour), it’s easy to select the wrong one—especially on busy projects. Also check units (meters vs feet) and vertical datum assumptions if your workflow includes them.
If the surface looks “stepped,” inverted, or wildly offset, the file may be corrupted or exported incorrectly. Re-import the design, or load a known-good version to compare. If your team builds and manages surfaces regularly, it helps to keep a simple naming convention and version history so you can quickly identify what changed.
On projects with complex geometry, high detail, or frequent revisions, the quality of your 3D machine control models can make the difference between smooth production and constant confusion. A clean, well-structured model reduces the chance of operators selecting the wrong surface and makes troubleshooting much easier because you can trust the design layer.
Sensors and calibration: tilt, slope, and blade measurements
Machine control isn’t just GNSS. Many systems combine GNSS with inertial sensors, mast sensors, and implement measurements. If these sensors are out of calibration, you can see errors that look like GPS drift—especially in elevation.
Start by checking whether the system reports sensor faults or calibration warnings. If your machine recently had maintenance (blade work, mast repair, hydraulic changes), calibration should be on your suspect list. Even small changes in mounting points can affect measurements.
Run the manufacturer-recommended calibration routines and verify sensor readings are stable. If you’re seeing inconsistent results, inspect sensor cables and mounts. A loose sensor bracket can produce intermittent errors that come and go with vibration, making it feel like “the satellites are bad” when it’s really a mechanical issue.
Environmental factors: weather, solar activity, and site interference
Not every GNSS problem is on your machine. Heavy rain, wet snow on antennas, and rapid atmospheric changes can degrade signal quality or increase correction noise. You might still get a “fixed” solution, but the vertical performance can get worse than usual.
Electromagnetic interference can also play a role. High-voltage lines, certain site radios, and poorly grounded equipment can introduce noise. If the issue happens in one area of the site and disappears elsewhere, interference or multipath is more likely than a broken receiver.
Solar activity is a less common but real factor. During geomagnetic storms, GNSS accuracy can degrade. If everything checks out—power, connectors, corrections, localization—and multiple machines are acting up, it may be worth checking space weather reports as part of your “big picture” troubleshooting toolkit.
Cross-check with a simple ground truth routine
When you’re under pressure, it’s tempting to trust the screen or distrust it completely. A better approach is to run a quick, repeatable verification routine that gives you confidence either way. Pick a known reference: a benchmark, a control point, or a previously verified spot on the site.
Check the machine’s reported position/elevation at that point, then move away and come back. If the system repeats consistently, it’s likely stable—even if it’s wrong relative to the project (which would point back to localization or design). If it doesn’t repeat, suspect corrections, satellite visibility, hardware, or sensor instability.
If you have access to a survey rover, compare results. This doesn’t need to be a long survey session; even a quick check can tell you whether the problem is site-wide or isolated to one machine. Document the differences and the conditions (open sky vs near structures) so you can make a smart call on next steps.
When only one machine is failing: swap tests that save hours
If you have multiple machines on site, the fastest troubleshooting method is often a controlled swap. Swap one component at a time—antenna, receiver, display, radio/modem—between a working machine and the failing one. This isolates the fault without guessing.
For example, if you swap the antenna and the problem follows the antenna, you’ve found the culprit. If it stays with the machine, move to the next component. Keep swaps organized and take notes so you don’t accidentally introduce a new variable (like a different job file or localization).
Swap tests also help with intermittent issues. If the problem only appears after vibration or after warming up, you may need to run the machine for a bit after the swap. It’s slower than a quick reboot, but it’s much faster than replacing parts blindly.
Settings that commonly get bumped (and how to spot them)
Jobsite workflows involve lots of hands: operators, foremen, survey techs, mechanics. Settings can be changed accidentally—sometimes with the best intentions. A system can be “working” but configured in a way that makes it unusable for the current task.
Common culprits include: wrong correction source (radio vs network), wrong radio channel, wrong antenna height, wrong implement selection, wrong blade width, or an offset applied for a previous attachment. Another frequent issue is selecting the wrong design surface or turning on a display filter/smoothing option that makes the guidance feel delayed.
If your platform supports it, compare settings to a saved template or to a known-good machine. If not, create a simple “golden settings” checklist for your fleet. It’s not glamorous, but it prevents a lot of downtime caused by a single toggled option buried in menus.
Remote support: what to collect before you call
Sometimes you’ll hit a wall and need help. The fastest support calls are the ones where you can share clear symptoms and a few key details. Before calling, gather: screenshots/photos of warnings, receiver status pages, correction type and age, satellite count/PDOP, firmware versions, and what changed recently.
Also note whether the issue is affecting multiple machines, whether it happens everywhere or only in certain site zones, and whether it improves in open sky. If you can export logs from the display/receiver, do it—logs can reveal dropouts and configuration mismatches that aren’t obvious on the main screen.
If you need a hand walking through advanced checks or interpreting what the receiver is telling you, services focused on GPS troubleshooting for construction equipment can often diagnose issues quickly when you provide the right info up front. The key is to treat troubleshooting like a process, not a guessing game.
Preventing repeat problems: small habits that keep GPS reliable
Once you’re back up and running, it’s worth spending a little time making sure the same issue doesn’t return next week. Most recurring GPS problems come from a handful of patterns: loose connectors, inconsistent job files, unclear version control, and skipped calibration after maintenance.
Create a simple daily/weekly routine: quick visual inspection of antenna mounts and cables, confirm corrections are stable, and verify the correct job/localization is loaded. If operators rotate between machines, make sure everyone knows where to check correction status and what the warning icons mean.
On the data side, keep design files organized and communicate revisions clearly. When a new surface is issued, remove or archive old versions from the display to reduce the chance of selecting the wrong one. A little discipline in file management can eliminate a surprising amount of “GPS is wrong” chatter.
A field-friendly checklist you can print or save to your phone
If you want a quick sequence to follow when the machine is down, here’s a practical order that works well on busy sites. Start at the top and stop when you find a clear cause:
1) Identify the symptom: no position, no corrections, drifting, consistent offset, wrong surface, or heading issues. Note whether it’s one machine or multiple.
2) Check screen status: satellite count, PDOP, correction type/age, solution (autonomous/float/fixed), radio/cellular signal, warnings. Take a photo.
3) Verify job/design: correct project, surface, units, and localization selected. Confirm you’re not using an old version.
4) Power and connectors: fuses, power cable, inline adapters, connector pins/corrosion, re-seat all connections.
5) Reboot in sequence: receiver → correction source → display. Wait for lock and fixed solution.
6) Move to open sky: test away from structures/trees/stockpiles to rule out multipath and blockage.
7) Corrections source check: base station powered/steady and set correctly; radio channel/ID; network RTK connectivity and data plan.
8) Physical antenna inspection: tight mount, no cracks, coax undamaged, dual-antenna connections (if applicable).
9) Calibration/sensors: run required calibration routines after maintenance; check for sensor faults.
10) Swap test (if possible): swap antenna/receiver/display/radio with a working machine to isolate the failing component.
If you’re evaluating outside help or better workflows
Some teams only think about workflows when something breaks. But if GPS downtime is becoming a pattern, it might be time to look at the bigger system: how models are built, how files are distributed, how calibrations are tracked, and how the base/network is managed.
Even small process improvements—like a standard naming convention for designs, a shared change log, and a consistent method for verifying localization—can turn troubleshooting from a stressful scramble into a predictable routine. And when you do need expert help, having your house in order makes that support far more effective.
If you’re looking for guidance on setup, modeling, or support options, you can visit website resources that focus on construction positioning and machine control. The best results usually come from combining solid field habits with clean data and a support plan that fits your crew.