It is programmed like a PLC. That is where the appeal is and the danger.

An Italian manufacturer sold in Spain through distribution. Two panel families: SmartLiving, the one you see most, and Prime, for larger installations and with more communications. And something worth knowing even if you have come to talk only about intruder detection: the same catalogue has fire detection in it.

That is not a brochure fact: in an industrial unit or a hotel, the fire and the intruder detection share a cabinet, a network and the person who deals with the alert in the small hours. It can be done with two brands too, with more pieces and more things to explain.

It turns up in hotels, large residents' associations, clinics, schools, offices, retail and medium-sized industrial units. It is not the brand for a critical infrastructure specification and it does not present itself as one.

The classic architecture: a panel with its bus, and off the bus the keypads, expanders, key readers, auxiliary power supplies and communicators. The cabled zones are watched with end-of-line, which is what makes cutting or bridging a cable raise an alert instead of leaving a detector dead without anybody finding out.

The bus has the same repeated mistake as always: modules far away, high current draw, voltage drop and intermittent faults that cannot be diagnosed from the keypad. It is sized with numbers. And the wireless side carries a battery with a date, supervision and coverage measured on site.

The keypads matter more than they look: they are the only thing the user will see for ten years. Here there are screens that call the zones by name and voice prompts. A system that says «warehouse door open» instead of «zone 14» is a system people arm, and one that gets armed is one that protects.

You write conditions —if this zone, at this hour, with the system like this, then that— and with them you solve what on other panels is solved with relays: the store arms itself at ten if somebody forgot, the cold room zone does not fire when the compressor starts, the outdoor sounder stays quiet at night but the alert goes out all the same.

The problem arrives two years later. Twenty conditions written by somebody who is no longer around and with no piece of paper saying what each one does are a black box: the next technician studies them with the client waiting, or deletes them and starts again.

So the rule with this brand is that every condition goes into the file in plain language: what it does, why it exists and what happens if it is switched off. It is what keeps the installation in the hands of people who did not build it, including any supplier who comes after us.

What rules is the detector and the place it is put: an infrared looking at a hot air outlet, at a tarpaulin or at a blind that rattles will give you trouble with any panel.

What this house adds is writing the exact condition that zone needs: requiring two detectors to see the same thing, ignoring a zone during the hour the compressor starts, asking for two consecutive triggers. It is finer than raising or lowering a general sensitivity — which is what is not done, because it leaves a deaf alarm.

The last step is looking before you send somebody out. And here we declare what has to be declared: IRIS Neural is the NVMS of Infinity Neural, the other company in this group, which is to say our own product. It adds the image of what set the detector off; it does not replace the panel.

The EN 50131 standards classify alarm systems into security grades, and each grade demands specific things of the equipment, the cabling, the power supply and the tamper detection. The grade is set by the risk of the activity and, very often, by the insurance policy.

There are two things not worth mixing up: equipment being declared for a grade, and an installation meeting that grade. The second includes the whole installation and the file that evidences it. What grade each model from this house supports is read in its documentation, item by item.

The question to ask, of us as well, is what grade is being handed over and that it be handed over in writing. We do not state it here: it depends on the design, and it is asserted by whoever signs the file. And connecting to an alarm receiving centre is another decision, with its own regulations and its own contract: who provides it and on what terms is settled in a quotation, not on a website.

It fits where there are schedules and exceptions: hotels, schools, clinics, residents' associations, units on shifts. There the logic stops being a data sheet and turns into the system being armed every day without anybody having to remember.

It fits especially well where the end user is not technical: a keypad that names the zones and speaks is the difference between a system that gets used and one that gets disarmed the third time nobody understands why it is beeping.

It does not fit if there is a fleet of sites to administer from a single place: this range is not designed to be governed that way. And it does not fit if nobody is going to document the programming — handed over without the list of conditions, it is not an installation, it is a dependency.

Arming and disarming, managing users, isolating a zone because there is building work and launching the scenarios already programmed: the client runs that and should run it.

What they do not touch is the logic. Adding a condition, changing an automatic arming schedule, touching the reporting to the alarm receiving centre or upgrading firmware is done with the integrator present, because a badly written condition does not show: it leaves a zone unwatched in a time band and that appears on no screen. It is discovered on the day they come in at that hour.

And the zone that was isolated «one day» and stayed isolated is the most common way for a good system to stop being any use. Every isolation with an expiry date, the inspection hunting them down, and a file saying what each condition does and who the system calls.