Main dashboard

The dashboard shows live sensor readings grouped by hardware (CPU, GPU, motherboard, storage, network, battery, etc.). SensorPulse reads everything locally via its native SensorPulse Engine; nothing is sent to the cloud.

Header bar

• Polling interval – How often values refresh (e.g. every 2 seconds). Monitoring must be running.
• History – Opens the history window for the selected sensor or the first sensor in the list.
• Tools – Opens the tools hub (thresholds, benchmark, storage, network, etc.).
• Task manager – Detailed process view with CPU/RAM history and actions.
• Start / Stop monitoring – Pauses or resumes polling and live updates. Shortcut: Ctrl+P.

Sensor groups

Hardware groups (CPU, GPU, motherboard …) are shown vertically as expandable sections – individually or all at once. Each group's expanded/collapsed state is saved automatically and restored on the next launch. Group header context menu: Expand all groups / Collapse all groups, pause/resume monitoring for the whole group (permanent license). The Favorites group highlights starred sensors.

Smart tips appear below the header when SensorPulse suggests a useful improvement (e.g. run as administrator for more sensors, enable DB logging for trends, or enable the localhost API for widgets). Dismiss with × or click the suggested action.

• Sensor tile context menu: favorite, pause monitoring, mute alerts, edit threshold, copy sensor ID, open history.

Summary cards

Click a card to filter the sensor list: All, Normal, Warning, Critical. Counts update in real time.

Sensor tiles

• Color indicates severity (green / amber / red).
• Sparkline shows the last minute of values in the current session.
• Min and max since monitoring started are shown under the value.
• Right-click or use per-sensor actions to pause a sensor (permanent license) or open history.

Active warnings & errors

Lists debounced alerts. Click an entry for details. Critical alerts can trigger a tray balloon when the window is hidden (see Settings).

Search

Ctrl+F focuses sensor search. Esc clears the filter or severity filter.

Status bar

Shows monitoring state, elevation (admin), license status, and Settings (gear). Open Settings for database, display, tray, SQL admin, and license options.

Programs & games

Use the Programs tab next to Sensors to browse installed applications from the Windows Start menu in a table.

Table

• Name – Display name of the application.
• Installed – Install date when available.
• Size – Disk space used when available.
• Open – Launches the application via its shortcut.

Games & overlay

• Games (folders named Games, Spiele or Giochi) are detected.
• When you launch a game, the game overlay is enabled automatically if it is currently off.

Search & refresh

• Text search filters programs by name.
• Refresh re-scans the Start menu (after installing new software).

Game overlay (once per session)

The transparent overlay window is created only once per app session. When disabled it is hidden, not destroyed – turning it back on or launching a game shows the same window again.

Settings

All important options in one place – open via the gear button in the status bar. Press F1 while the settings window is open for this help page. Window size and position are saved when you close the window.

Save in DB

• Save sensor data to database – When enabled, readings are written to sensordata.db next to the executable. When disabled, you still see live values and session sparklines, but no new history rows are stored.
• Data retention – Days to keep readings and alerts (1–3650). Saved when leaving the field, closing the settings window, or exiting the app.

Display

• Show inactive sensors – Includes adapters that only report zeros.
• Show active warnings & errors – Toggles the alert sidebar.

Background

• Minimize to notification area – Closing the window keeps the app in the tray. The tray tooltip shows CPU/GPU temperature when available.
• Slower refresh in background – When minimized, sensor polling runs less often. Effect: Lower CPU use on weaker PCs; tray values update more slowly.
• Notify on critical alerts – Tray balloon when a new critical alert appears while the window is not visible.
• Start with Windows – Launches SensorPulse automatically after sign-in.

Game overlay

• Enable game overlay – Transparent always-on-top click-through window with CPU/GPU load, temperatures and FPS (see dedicated chapter).

Hardware access

• Status badges show admin mode and PawnIO installation state.
• Run as administrator – Restarts with elevated rights for more sensors (when not already elevated).
• Install PawnIO – Optional installer for more accurate CPU temperatures (when not yet installed).

Profiles / scenes

• Active profile – switches refresh interval, fan curves, game overlay and engine mode in one click. Profiles: Quiet, Gaming, Benchmark, Night or Custom (your manual settings).

Automation

• Sound on critical alerts – optional alert tone for new critical events.
• Switch to Quiet profile on critical alert – reduces load automatically on critical temperatures.
• Localhost sensor API – JSON endpoints for Rainmeter, browser widgets or scripts (localhost by default). See Localhost API.
• Remote access (Companion app) – optional read-only access from phone/tablet on home network or VPN. See Companion app.
• CSV metric export – periodic export of selected sensors. See CSV metric export.
• Electricity price (cents/kWh) – estimates energy cost from power sensors (current, session, today).

Language

• Language – Six UI languages (DE, EN, FR, ES, IT, PL). Language packs for F1 help appear below the dropdown. The UI switches immediately; after installing a pack, the language appears in the list.

Settings backup

• Export settings – saves thresholds, favourites, group states, profiles and other preferences as a JSON file.
• Import settings – restores a previously exported file (e.g. after reinstall or PC migration).

Management

• SQL admin – Database tools.
• SNMP devices… – Monitor network devices (NAS, router, printer, UPS) via SNMP v2c. See SNMP devices.
• Export / import settings – JSON backup of preferences.
• Factory reset – clears user data (sensor history, settings).

SNMP devices

Optionally monitor any number of devices on your LAN via SNMP v2c – e.g. Synology NAS, router, printer or UPS. Each device has its own template, sensors and extra OIDs. Values appear as separate sensor groups on the dashboard.

Opening the window

Settings (gear) → Management → SNMP devices…. Press F1 in the SNMP window for this help page.

Setup – recommended order

1. Check Enable SNMP monitoring and set the poll interval (applies to all devices).
2. In the device list on the left, select a device or use Add (Duplicate/Remove also available).
3. On the right, per device: choose a template (System, Network, NAS, Synology NAS, Printer, UPS or Custom).
4. Below the template you see a hint and matching fields:
   • Device IP – label depends on template (e.g. DSM IP for Synology, NAS IP, Printer IP, Router IP for Network).
   • SNMP port (default 161) and community (often public – match the device).
   • Display name – optional label in the sensor group.
5. For Network only: pick an interface or use Discover interfaces via SNMP (port index).
6. For Custom: enter OID, name and unit in the Custom section.
7. Enable/disable individual template metrics with the checkboxes.
8. Optionally add any number of extra sensors with custom OIDs per device – use Add sensor; OID/name/unit are editable only when that extra sensor is enabled.

Templates

• System – CPU load, uptime (any SNMP device).
• Network – inbound/outbound traffic on one interface (router/switch).
• NAS – CPU, RAM (generic SNMP NAS).
• Synology NAS – CPU temperature and load (enable SNMP in DSM: Control Panel → Info Center → SNMP).
• Printer – page count, toner level.
• UPS – battery status, runtime, charge.
• Custom – one user-defined OID.

Example configurations (all templates)

The examples below use fictional LAN addresses. Community, port and OIDs must match your device – check the device manual or use Test SNMP.

Test reachability separately

Two independent checks for the configured device:

• Test ping – plain network reachability (ICMP). Result e.g. “Reachable (12 ms)” or “Unreachable”. Does not test SNMP.
• Test SNMP – whether SNMP responds and the community is correct (reads device description).

Effect: Tests do not change settings – they help troubleshooting (ping OK, SNMP fail → community/firewall; both fail → wrong IP or device off).

Ongoing ping sensor

The Availability via ping checkbox is separate from the test buttons: when enabled during SNMP monitoring, an additional Ping sensor appears on the dashboard (round-trip ms or offline).

Dashboard

Active SNMP metrics appear per device in their own group. SNMP polling runs only when monitoring is enabled and at least one device has host/community configured.

Localhost API (Rainmeter & scripts)

Read-only HTTP API returning live sensor data as JSON – for Rainmeter skins, browser widgets on the same PC or your own automation. Default: 127.0.0.1 only (no internet).

Enable

Settings → Automation → Localhost sensor API. Use F1 help: localhost API or Setup guide: localhost API for details.

Endpoints (summary)

• GET /api/health, /api/quicklook, /api/sensors, /api/alerts, /api/history/{id}

Companion app (phone monitoring)

Optional remote access for SensorPulse Companion (Android/iOS, coming soon to app stores): dashboard and alerts from your own phone – read-only, no PC control.

Requirements

• SensorPulse running on Windows PC
• Localhost API and Remote access enabled in Settings
• Phone on same Wi‑Fi or VPN (Tailscale/WireGuard recommended away from home)
• Windows Firewall allows inbound TCP on API port (Private profile)

Pairing

With remote access on you get a QR code (recommended), pairing JSON, LAN IP, firewall status and Add firewall rule in Settings. Scan or paste in the Companion app → test → save.

CSV metric export

Writes selected sensor values to a CSV file at configurable intervals – for Excel or custom analysis.

• Enable CSV metric export, set interval (1–60 min)
• Optional: use same sensors as database logging set
• Target path shown under the option; SensorPulse must be running

Language packs (F1 help)

The app UI supports six languages. German and English F1 help ship with the installer. For French, Spanish, Italian and Polish you can add a language pack later.

In the app

1. Open Settings (gear icon)
2. Select the Language section
3. Below the language dropdown, choose a pack → Download help pack or Install from file… (ZIP saved from sensorpulse.de)

Offline / manual

Download the ZIP from sensorpulse.de/download.html and extract the Help\{language} folder into your SensorPulse install directory (next to SensorPulse.exe). The PDF manual is available separately on the download page (Open PDF link).

History

Opens from the dashboard (History button or Ctrl+H) or from an alert. Shows stored readings from sensordata.db.

Controls

• Sensor – Dropdown of sensors that have data in the selected time range.
• Time range – Presets (e.g. last hour, 24 hours, 7 days) or custom span.
• Search – Filters the alert log list below the chart.

Chart

Spline chart of value over time. Statistics (min / max / average) appear for the visible range. Double-click a point to open the reading detail window with assessment and context hints.

Alert log

Historical alert rows for the same sensor and period. Requires database saving to have been enabled when alerts occurred.

Alert details

Popup when you select an active alert. Shows sensor name, hardware, value, unit, severity, message, first seen time, and occurrence count.

• Open in history – Jumps to the history window for this sensor.
• Close – Closes the popup; the alert may remain active if the condition persists.

Alerts are debounced: a repeat notification for the same condition is suppressed for several minutes unless severity changes.

Open source (MIT)

The community edition is free software under the MIT license. All features – including SQL admin, SNMP, fan control and tools – are available from first launch with no trial, key or payment.

Source code, LICENSE and build instructions are in the public repository. You may use, modify and redistribute SensorPulse – with copyright notice.

SQL admin – query editor

Available in all builds. Works on sensordata.db in the application folder. Protected against manipulation.

Sidebar

Lists tables: readings_cpu, readings_gpu, readings_motherboard, readings_memory, readings_storage, readings_network, readings_controller, readings_battery, readings_psu, readings_power, readings_thermal, readings_other, alerts and settings. Click a table name to insert a sample query.

Editor

• Type SQL and press Run query (or Ctrl+Enter).
• Format – Indents and uppercases keywords for readability.
• Font size – Ctrl + mouse wheel in the editor zooms text (8–32 pt).

Results

Result grid and Messages tab show output or errors. A progress overlay appears while a query runs. Results are limited to 3,000 rows; only the result area scrolls. Allowed: read queries plus INSERT, UPDATE and TRUNCATE. Schema and database commands (e.g. DROP, ALTER, VACUUM) are blocked – protected against manipulation.

Schema (v11)

Readings are split into per-component tables (readings_*) with indexes on SensorIdentifier + Timestamp (text, ISO format). Useful views:

• v_readings_cpu, v_readings_gpu, … – one view per component (no UNION ALL)
• v_alerts_readable – alerts with readable severity name
• v_benchmark_runs_readable – benchmark history
• v_installed_apps – formatted program list
• alerts_fts – FTS5 full-text search for the alert log (kept in sync via triggers)

Example: SELECT * FROM readings_cpu ORDER BY Timestamp DESC LIMIT 100;

SQL admin – automation

Schedule maintenance while SensorPulse is running. Tasks are stored in Settings as JSON (ScheduledTasks.v1).

Task list

• On – Checkbox to enable/disable without deleting.
• Columns show name, action, schedule, last run, status.

Actions

• Delete old data – Purges readings/alerts older than retention (days or hours).
• Backup database – Copies sensordata.db to a folder (default backups).
• Optimize database – Runs VACUUM / optimize.

Schedule

• Every day at a fixed time – Choose hour (0–23) and minute (0–59). Runs once per day at that local time while the app is open.
• Repeat every N hours – Interval from last run.
• On application start – Once when SensorPulse starts.

Buttons

Add task, Edit, Run now, Delete. Save in the editor panel after changing fields.

History – reading details

Opens when you double-click a point on the history chart. Shows sensor, hardware, value, timestamp, severity, and configured thresholds.

Assessment & context

Automatic classification of the reading (normal, warning, critical) with a short explanation. Additional context hints depending on sensor type (e.g. temperature, load).

Tools hub – overview

Opened via Tools on the main dashboard. A progress overlay loads tab data on first open.

F1 jumps to help for the active tab: thresholds, alert history, system, storage, fan control, CPU info, GPU info, benchmark, compression, editor and logs.

Tools – thresholds

Global warning and critical limits per sensor type. Save writes values to the database; Restore defaults resets factory settings.

Use Edit or the dashboard context menu to open the Edit threshold dialog.

Tools – alert history

Chronological list of logged alerts from sensordata.db – not only the live panel on the dashboard, but everything stored in the alerts table.

What can you do?

• Time filter – last 7, 30, or 90 days (display only; data stays in the DB until you delete or retention purges it).
• Clear history – permanently removes all stored alerts after confirmation. The FTS shadow table alerts_fts is cleared correctly as well.
• Export alert history (CSV) – saves filtered alerts to a file.
• Refresh – reloads the list after new alerts or after clearing.
• Double-click an entry to open the alert detail window.

Tools – system

Read-only overview of OS, CPU, RAM, motherboard, and other system facts.

Refresh reloads data after hardware or Windows changes. Effect: Display only; nothing is changed on the system.

Since 1.10: Power plan (active Windows energy scheme) and Disk health (WMI model and status per drive).

Tools – storage

Lists all drives with usage, file system, and type. Below: Show largest files (permanent license) – parallel scan of the top 100 files on the selected drive (NTFS, exFAT, FAT32, ReFS and other Windows file systems) with progress overlay. Double-click opens the location in Explorer. Right-click a row → Compress as ZIP submenu with all compression levels (store, fast, normal, maximum), then choose save location and file name.

Tools – compression

Create or extract ZIP archives with a drive list, folder explorer and selectable compression level (store, fast, normal, maximum). Select files or folders, then Create archive or Extract archive. The default archive name is derived from the first selected item. You can also compress a large file directly from the Storage tab via right-click after a top-100 scan.

Tools – fan control

Adjust fan speeds when your hardware supports software control.

• Set target speed per fan, then Apply on that row.
• Apply all – applies every pending target at once.
• Release (per fan) – returns control for that fan only to BIOS/vendor tool (with confirmation).
• Release all – releases every software-controlled fan at once.
• Auto – temperature-based curves when supported. Save persists settings.

Tools – GPU info

Detailed GPU properties per graphics card (driver, memory, clocks, BIOS version, etc.).

Tools – benchmark

Performance and stress tests with live chart, peak temperature and persisted history.

Layout

Left: test selection, options, start/stop and live values. Right: a wider history table of recent runs with a vertical scrollbar when needed.

Test types

• CPU performance (short) – quick all-core throughput (iter/s).
• CPU sustained load (FFT) – Prime95-like full load using FFT on all cores. Profiles: small FFT (1024) or large FFT (8192, more RAM). Shows peak CPU temperature.
• Memory / disk / GPU compute – throughput or GFLOPS as before.
• Network throughput – local receive/send test on a connected adapter (Mbit/s, live charts).
• GPU 3D stress test – real OpenGL load in the benchmark tab. Optional lite mode. FPS, peaks and SensorPulse points are saved to history.

Results

Choose duration (10–300 s) and press Start test. Duration is remembered next time you open tools. For network throughput, the first connected adapter is pre-selected when no prior choice exists.

For GPU compute and GPU 3D stress: when multiple GPUs are detected (even without opening the GPU info tab), choose GPU for benchmark. Temperature and load metrics use the selected card. The 3D stress test still uses the system default OpenGL adapter.

History lists metric, peak temperature and SensorPulse points. Clear history removes all saved benchmark runs. Double-click a row for run details.

Benchmark – run details

Opened by double-clicking a row in the benchmark history table. Shows time, duration, result, profile/drive, peak temperatures, peak GPU load and SensorPulse points for that run. Close only dismisses the window – the entry stays in history.

Tools – logs

Application diagnostics from logs/application (errors.log, warnings.log, info.log, automation.log). Entries load when you open the tab; level filter and search update the list immediately.

• Refresh – reloads log files from disk.
• Clear logs – empties all four log files after confirmation (cannot be undone).

Task manager

Full process overview with CPU, RAM, thread, and handle data. Header shows totals for CPU, memory, and process count. The integrated task manager the list appears quickly via WMI pre-load; CPU values and detail fields are filled in the background afterward.

Performance details

From the sparkline cards or toolbar you open a separate window (CPU, memory, disk, network, GPU, startup apps) with axis labels, 60-second history and per-core or per-volume mini charts. Startup apps can be enabled/disabled like Windows Task Manager.

Actions

• End task / End process tree
• Suspend / Resume (single process or tree)
• Switch to, Restart
• Copy PID, path, command line; open location; properties

Details & history

Bottom panel: detail fields for the selected process and live CPU/RAM charts. Refresh and Export CSV in the header.

System report

Click System report in the main window header to open this window with a full HTML report of your hardware.

Contents

• CPU, RAM, motherboard – model, cores, clocks and more
• GPUs & drives – all detected graphics cards and storage devices
• Current sensors – live values at the time of creation
• Recent alerts – when stored in the database

Save report exports the HTML file to a location of your choice – useful for support requests or documentation.

Edit threshold

Dialog to adjust warning and critical values for a single sensor type. Lower is worse for sensors where low values are problematic (e.g. voltage).

OK applies the change to the tools list – use Save thresholds there to persist.

Report error

Appears after unexpected errors. Details are saved locally in the error log file.

• Send anonymized crash report – Transmitted to SensorPulse via HTTPS (no mail client).
• Always send automatically / Never send – Persistent preference for future errors.
• Optional comment describes what you were doing.

Game overlay

Transparent, click-through always-on-top window at the top right with CPU/GPU load, temperatures, FPS and optional frametime graph and 1 % low FPS (Glances-inspired). A process rule like Glances AMP enables overlay/profile automatically when the configured process name is running.

The overlay is created once per session; hiding it does not destroy it – turning it back on shows the same window again.