I was out at a converted bank building on Mostyn Street in Castlemaine a few months back, bluestone walls close to 600mm thick in places, and the gallery owner was showing me a battery report from her existing wireless system. Three sensors had dropped comms in the previous fortnight, all in the same back room. She’d assumed they were faulty units. They weren’t. The wall was the problem.
This comes up more than people expect. Wireless sensors get sold on convenience, no chasing cable through heritage-listed plaster, no drilling into a wall the National Trust has an opinion about, and for a lot of modern buildings that’s a fair trade-off. But stone and old brick behave very differently to a stud wall with plasterboard, and if nobody explains that to the client before the install, you end up with a system that looks fine on the quote and fails quietly six months later.
What thick masonry actually does to a wireless signal #
Radio frequency signals lose strength passing through dense material, that part isn’t controversial. What surprises people is how much worse solid stone and old handmade brick are compared to modern construction. A double-brick cavity wall from the 1960s already knocks a fair chunk off signal strength. A solid sandstone or bluestone wall from the 1880s, often built without a cavity at all, can attenuate a signal by a large enough margin that a sensor sitting perfectly fine at install time starts dropping packets six weeks later when humidity in the stone shifts slightly, or when someone moves a filing cabinet into the signal path.
Iron content in old ironstone and some sandstone varieties makes it worse again. And render over solid brick, common in a lot of colonial-era gallery conversions, adds another layer that most installers don’t account for because from the outside it just looks like a painted wall.
Mounting is a separate problem to signal strength #
So even where the signal holds up, mounting a vibration or shock sensor on solid stone introduces its own issue. These sensors are calibrated to detect a specific range of movement transmitted through the wall or frame, and old masonry transmits vibration completely differently to a timber stud or a modern besser block wall. Stone is dense and rigid, it can either dampen a genuine impact so the sensor under-reads it, or it can carry vibration from an unrelated source, a delivery truck outside, roadworks, even a door slamming three rooms away, and set off false alarms constantly. We’ve written before about how to tell if a vibration sensor is actually calibrated for your frame weight, and the same logic applies to the wall itself, not just the artwork. A sensor tuned for a plasterboard partition is the wrong instrument for a solid stone one.
Battery drain nobody budgets for #
Here’s the bit that catches most galleries out financially. When a wireless sensor is fighting attenuation through a thick wall, it doesn’t just fail outright, it retries. Most wireless protocols will boost transmission power or repeat a signal several times before giving up and flagging a fault. That retry behaviour drains the battery far faster than the manufacturer’s published life expectancy, which is almost always tested in open-air lab conditions, not through 500mm of sandstone. I’ve seen sensors rated for two years of battery life burn through in under eight months in a heritage building, and the client just assumes the product is poor quality when actually it’s working overtime to punch through the wall it’s mounted on.
That matters practically because a monitored alarm system is only as good as its weakest link reporting in. If a sensor’s battery dies quietly and nobody’s doing physical walkthrough checks, you can have a blind spot for weeks without knowing.
Mesh networks help, but they’re not a fix-all #
Mesh topology, where sensors relay signal to each other rather than all reporting directly to a single hub, genuinely helps in older buildings because it shortens the distance any single signal has to travel through masonry. But mesh only works if there are enough nodes in range of each other, and in a gallery with widely spaced rooms separated by solid internal walls, that can mean adding repeater units that weren’t in the original design brief, which pushes the cost up and adds more battery-powered devices that need the same maintenance attention.
I’d argue installers sometimes oversell wireless as the default for heritage retrofits because it’s a faster, cheaper quote to write than a hybrid hardwired system, and then the ongoing reliability cost gets pushed onto the client later in the form of false alarms and dead batteries nobody planned for.
Where hardwired still earns its keep #
For a genuinely solid stone or thick brick building, running a hardwired backbone to key sensor points and using wireless only for supplementary or object-level monitoring tends to be more reliable long term. It costs more upfront and yes, it usually means some cable chasing that a heritage advisor needs to sign off on. But a hardwired sensor doesn’t care about wall density, humidity swings, or battery life, and for critical points like a vault door or a high-value gallery room, that reliability matters more than the neatness of a fully wireless install. AS 2201.1, the Australian Standard covering intruder alarm systems for client’s premises, doesn’t mandate hardwired over wireless, but it does require a documented risk assessment of the site, and wall construction is exactly the kind of thing that assessment should flag before quoting, not after installation when the false alarms start.
Standards Australia publishes the current suite of alarm and detection standards, and it’s worth an installer walking a client through which standard actually applies to their building type rather than assuming a generic residential-grade wireless kit is fit for purpose in a 140-year-old sandstone building. You can check the current standards listing directly through Standards Australia.
What this means for an existing collection #
If you’ve already got wireless sensors in an older building and you’re seeing intermittent faults, dead batteries earlier than expected, or false triggers with no obvious cause, it’s worth having someone map the actual wall construction room by room before assuming the hardware is at fault. A signal strength test at each sensor location, done properly with a spectrum analyser rather than just eyeballing the app’s bar graph, will usually show you exactly where the attenuation is worst.
This ties into the bigger access control and monitoring picture too. We’ve covered elsewhere why most gallery break-ins aren’t what you think, and a lot of it comes back to gaps in coverage that nobody noticed because the system reported as healthy right up until the moment it mattered. A wireless sensor quietly losing comms for three days a month is functionally the same as no sensor at all during those gaps, even though the monitoring dashboard might show a green light most of the time.
For institutions specifically, the National Standards for Australian Museums and Galleries published by Museums & Galleries NSW touch on risk management expectations, and it’s a useful reference point when you’re arguing a budget case to a board for a hybrid system rather than a straight wireless refresh. You can find that guidance through Museums & Galleries NSW. What Australian institutions are actually obligated to do about physical security is a separate question again, and one we’ve addressed in more detail on what Australian museums are actually required to do about security.
A quick word on insurance, and where I’ll stop #
None of this is insurance advice, ArtworkSecurity doesn’t sell or broker insurance and I’m not qualified to give legal guidance on policy wording. But it’s worth knowing that under the Insurance Contracts Act, an insurer generally has to demonstrate that a security failure actually caused the specific loss before a claim can be knocked back on that basis, rather than a missed sensor reading automatically voiding cover. That’s a conversation for a broker, not a sensor installer, and we’ve written separately about why a fine art insurance broker is worth finding early rather than after something’s gone wrong. If your wireless system has been unreliable in an older building, the more useful first step is usually a proper site walk to check where the signal is actually failing, not a call to your insurer.
So if you’re sitting on a heritage building with wireless sensors that keep dropping out, don’t assume it’s a faulty batch of hardware. More often than not it’s the wall doing exactly what 500mm of solid stone was always going to do to a radio signal, and the fix is in the design, not a replacement unit.
– Priya Chandran, Security Systems Editor