Multi-Network Gate SIMs: Why We Stopped Supplying Them

Multi-Network Gate SIMs: Why We Stopped Supplying Them

We used to sell multi-network SIMs for gates and intercoms. We do not any more, and the reason is not commercial. It is that we spent a long time watching how they behave on real installations, and on the two things a gate SIM is actually for — answering a call and opening on demand — a single-network SIM on Vodafone, O2 or EE beats them consistently.

This is an unusual thing for a SIM supplier to publish, because multi-network is the easier product to sell. "Connects to all the UK networks, always picks the strongest signal" is a sentence that closes an order. It is also a sentence that hides how the thing behaves at four o'clock on a wet Tuesday when a delivery driver is standing at your gate.

The two kinds of gate SIM

Almost every SIM you can buy for an intercom falls into one of two categories, and they are architecturally different, not just commercially different.

A single-network SIM is a subscriber of one operator. A Vodafone SIM is a Vodafone customer. Its calls, its texts and its data are handled inside Vodafone's own core network, by the operator whose masts it is using.

A multi-network SIM is, technically, a visitor. It reaches several UK networks through roaming agreements rather than by being a customer of any of them — that is precisely the mechanism that gives it multi-network reach. The consequence is that its traffic is anchored back to its own home core. The mast it is standing under belongs to one operator; the network deciding what to do with the call belongs to somebody else.

That single architectural fact produces both of the problems below.

The delay you can hear

Because a roaming SIM's service is anchored back to its home core, every call and every SMS crosses that boundary. A call to your intercom does not simply arrive at the mast and get handed to the device. It goes to the SIM's home network first, then back out to the visited network, then to the device. Each of those hops is signalling, and signalling takes time before anything rings anywhere.

You do not need a protocol analyser to see this. Fit a multi-network SIM to a dial-to-open gate controller and ring it. You will hear your phone ring two or three times before the module reacts and the gate starts to move.

Now do the same on a Vodafone, O2 or EE SIM. In most cases the gate has already registered the call and started opening before your handset rings at all. The call never gets the chance to ring out, because the network handling it is the network the device is standing in.

Two or three rings sounds trivial written down. Standing at a gate in the rain, holding a phone, wondering whether it has worked, it is not trivial — and it is the single most common "the gate is playing up" call an installer gets for a system that is, electrically, working perfectly.

The same delay applies to SMS-triggered functions, to the intercom calling you when a visitor presses the button, and to any system that uses text messages for status or confirmation. Roaming devices can also be given lower priority than an operator's own subscribers when a cell is congested, which does not help.

The failover that does not fail over

This is the serious one, and it is the reason we stopped rather than simply preferring single-network.

The promise of a multi-network SIM is that if one network has a problem, it moves to another. That is true in the sense that the capability exists. It is misleading about when it happens.

Switching is driven by a heuristic. Depending on the platform, it looks at things like signal quality, location, and the health of the data bearer. It is the data connection that is typically used to infer "this network is working". That inference is the flaw, because a mast can be perfectly healthy for data and impaired for voice and SMS at the same time.

When that happens, nothing triggers. The SIM is registered. Data flows. Every automated check says the connection is fine. And your gate does not answer its phone.

The industry is aware of the mirror-image version of this problem — a SIM that switches profile, reports "Registered", and cannot pass data because the APN no longer matches. Both are the same underlying mistake: registration state is not service state. A device that is attached to a network is not necessarily a device that is doing its job.

There is a second problem stacked on top. Re-evaluation is not continuous. Checks can be hours apart, and on some platforms considerably longer. So even in the case the design does catch, the response is not "instant failover" — it is a switch that happens at the next evaluation, which may be a day or more after the fault started.

What that looked like for us

We are describing this from experience, not theory. We have had multi-network SIMs on live installations where the data side was completely healthy, the SIM was happily registered, and calls and texts were not getting through on that mast. The SIM had no reason to move, so it did not move.

The fix was for us to intervene on the connectivity platform and manually steer that SIM onto a different network. It works, and it is an absurd thing to have to do. It means a customer's gate is dependent on somebody at a supplier noticing, diagnosing, and overriding a system that is reporting itself as fine. No homeowner should be exposed to that, and no installer should have to explain it.

Why a single network is the safer bet

The obvious objection is that a single network is a single point of failure. In theory, yes. In practice, consider what a serious Vodafone, O2 or EE outage actually looks like: it is on the news. It affects millions of people. It is measured in hours, and it is fixed by an operator with every commercial reason to fix it quickly.

That is a very different risk from a quiet, localised, days-long fault that no automated system is even looking for, on a network arrangement where you cannot easily find out who to call.

There is a practical benefit too. When your SIM is a Vodafone SIM, a coverage problem is a Vodafone problem, and it can be checked, reported and reasoned about. When your SIM is a roaming SIM sitting on an unnamed mast, "my gate stopped working" is the beginning of a long conversation.

This is also, not coincidentally, how the major intercom manufacturers think about it. The advice that comes with the hardware is almost always to establish which network has usable signal at the panel and fit a SIM for that network.

So how do you decide which network?

This is the fair challenge to everything above. Multi-network's real appeal was never performance — it was that it removed a decision. You did not have to know which network worked at the gate.

Sometimes you do already know. If everyone at the property has good signal on one network, that is a strong steer. But a gate is usually at the property boundary, often the furthest point from the house, frequently behind a wall, a hedge or a run of trees. Signal at the kitchen window tells you very little about signal at the gate post, and "which network works here" is not the same question as "which network works at my front door".

So we built the tool instead of hiding the problem.

Lynix Signal Intelligence takes the make and model of your equipment and the location it is being installed at, assesses predicted coverage at that location, weighs it against what your specific intercom actually needs to work, and recommends the network most likely to hold a stable connection. It is free, it runs as part of ordering, and the recommendation is yours to take or ignore.

It took six months to build and, as far as we know, there is nothing else like it aimed specifically at gate and intercom installations in the UK. It exists because withdrawing multi-network without answering the question it used to answer would have been no use to anybody.

Where multi-network still has a place

It would be dishonest to say the technology has no use. It has a narrow one.

On older 2G-era equipment in areas where operators are withdrawing legacy coverage, the ability to reach whichever network still carries a usable signal can extend the life of a panel that would otherwise need replacing. The same applies to some specialist deployments where the trade-offs are understood and accepted, and where somebody is monitoring the installation rather than relying on it to look after itself.

That is a considered engineering choice made by a professional for a specific site. It is not the right default for a homeowner's gate, and it is not what we sell to retail customers.

Common questions

Are multi-network SIMs a scam?

No. They do what they say: they can reach more than one network. The problem is the gap between that capability and how it behaves under the specific faults that affect gate intercoms, and how confidently it is sold as a solution to poor signal.

Is a single-network SIM really faster?

For call and SMS set-up, yes, and the difference is large enough to hear. Test it yourself: a dial-to-open on a roaming SIM typically rings two or three times before the gate responds. On a network SIM the gate is usually moving before your phone rings.

What if I pick the wrong network?

Run the Lynix Signal Intelligence check before you order and you are choosing on evidence rather than on a guess. If signal at the panel is genuinely marginal on every network, the answer is an external antenna, which improves matters far more than any SIM can — our support team can advise on this.

Can I change network later?

No. The network is set when the SIM is issued and cannot be changed afterwards, which is exactly why it is worth getting right at the point of order. Plans are 30-day rolling, so you are not tied in.

Which network is best for a gate intercom?

There is no single answer, and anyone who gives you one is guessing. It depends entirely on the masts serving your postcode and where on the property the panel sits. That is the question Signal Intelligence exists to answer.

The short version

Multi-network SIMs sell a promise of resilience that, on gate and intercom hardware, they do not reliably keep. They add a delay to exactly the interaction a gate SIM exists for, and their failover is blind to precisely the fault that matters most: voice and SMS failing while data carries on looking healthy.

We sold them. We stopped. We now supply single-network SIMs on Vodafone, O2 and EE, and we built Lynix Signal Intelligence so that choosing between them is a decision based on evidence rather than a gamble.