Security & Systems · D-05Gauteng & Mpumalanga
We design and install CCTV, access control, electric fencing and ANPR. Then our operators watch the site 24/7 from the Super Ampt Portal, control-room software we wrote ourselves. When a fence zone trips, the matched camera comes up on the operator’s screen and a 15-minute acknowledge clock starts — day or night.
We design and install each system below, and our operators watch the site from the Super Ampt Portal. At handover there is one team behind it and one number to call — not five suppliers to chase.
Coverage planned against a written risk assessment, then kept online under SLA. The Camera Wall reads Hikvision NVRs and Turbo HD DVRs live.
Booms, turnstiles and badge readers, on the same log as the cameras.
Energised, zone-alarmed perimeters on JVA energisers such as the Jumbo 50, six zones and up.
Vehicles captured at the gate and checked against a whitelist. Exceptions raise an alert, and retention is set with POPIA in mind.
Live operators around the clock, working to an escalation plan you sign off before go-live. Every alarm is graded and timed, and its acknowledgement is recorded.
Solar, battery and generator changeover from our Solar & Power division, so the cameras, the fence and the gate stay up through load-shedding.
We wrote the Super Ampt Portal because a control room should not depend on a cloud relay that times out after five minutes, or on an operator switching between a camera app and a fence panel. Camera, fence, access and fatigue events land in one stream, on one screen, graded on arrival.
Every site's plan and cameras, live, in one browser page.
Runs from any PC on the network. No plugin, no client software, no app store.
Click a camera marker on the site's own drawn plan to open that camera.
Talks to Hikvision NVRs and Turbo HD DVRs directly over ISAPI, not through Hik-Connect's cloud relay.
Live status of every fence zone, and the controls to act.
HT voltage, return voltage, armed state and mains/battery for each zone, live.
Alarm, tamper and restore events flow straight into the Portal's alert stream.
Arm, disarm and reset from the Portal, or lock a site read-only. Every command logged with who sent it.
One graded stream of events, each on a visible clock.
Every alert classed P1–P4 on arrival, each on its own acknowledge clock: P1 15 minutes and P2 one hour, day or night; P3 four business hours; P4 the next business day.
A fence alarm cues the matched PTZ camera automatically, and the correlation is stored as an immutable record.
Screens update over a WebSocket the second an event lands. Every acknowledgement is recorded.
One perimeter alarm at 02:14, from trip to closed record. The phone calls follow the escalation plan the client signed before go-live.
Zone 4, north perimeter — alarm
Classed P1 on arrival. The 15:00 acknowledge clock starts on screen.
PTZ 01 cued to Zone 4, automatically
Triggered by the Zone 4 alarm. Matched camera on the operator's screen; correlation stored as an immutable record.
Acknowledged in the control room
Acknowledgement recorded. The operator opens the site's escalation plan.
Escalation contacts called by phone
In the order the client signed off before go-live.
Zone 4 restored and re-armed
Restore event logged; re-arm command audited with who sent it.
Every alert gets a due time the moment it lands, set by its grade. The operator’s screen shows the time left.
The Camera Wall talks to each recorder directly, over the site network or a VPN into it, and its licence never phones home — so a wall in a quarry office keeps running when the uplink drops. The figures below come from our own bench, run against a simulated quarry VPN link.
frames delivered from 302 recorder requests. Viewers share frames, not recorder load.
frame rate the wall loses to one dead camera. 22% before tuning.
plugins, client installs or app-store downloads to open the Camera Wall.
acknowledge clock on every P1 alert, day or night, on the operator's screen.
Frames are shared between viewers: three people watching nine cameras put the load of one person on the recorder.
A built-in diagnostic asks each recorder: can we reach it, which channels does it claim, and which actually return a picture.
The Portal views through a read-only account on each recorder. Recorder passwords never reach the browser.
Security work gets questioned after the fact. The Portal keeps an append-only record of every sensitive action, shows each client only their own sites, and builds its reports from real event and audit data.
Every sensitive action is timestamped and attributed, including every arm, disarm and reset command. Entries are added, never edited.
Operators and managers see and do different things. Permissions are enforced on the server, not hidden in the browser. Each client sees only their own sites.
Event data is kept for 90 days, then deleted on a schedule. The period was set with POPIA in mind.
Five report screens built from real event and audit data, scoped to the viewer's role, read inside the Portal.
An edge service scores haul driver alertness from a camera using a trained model. It only raises a flag for a human. No raw video is stored.
The site's risks are written down first: what needs protecting, where the perimeter is exposed, and what the cameras must see.
Camera coverage, fence zones, gates and backup power are planned from the same drawing, against the risk assessment.
Our own team installs the hardware. Each site's plan and camera markers are set up in the Portal.
Who the control room phones, and in what order. The client signs the escalation plan off before monitoring starts.
Operators watch the site 24/7 from the Portal. Every alert is graded P1–P4 and its acknowledgement recorded from day one.
Next step
Send us the site address, the perimeter and what has gone wrong there before. A written risk assessment comes first, then a priced scope covering installation, backup power and 24/7 monitoring.
Quotes, site bookings and existing-site callouts
Send the scope and the date you need it by
