A 64-port pool works on a desk and behaves differently in a rack. The change is not the messaging logic but the physical layer: how the chassis is powered, how the antennas reach signal, and whether the enclosure that protects the equipment is also the enclosure that blocks the network.
This guide covers the engineering that belongs in the deployment plan for a rack-mounted pool at this density: rack space and power budgeting, the antenna routing decision that determines whether the deployment works at all, thermal behaviour under continuous operation, and how SIM handling changes once the equipment is in a data centre rather than an office.
How does a rack pool differ from a desktop pool?
It differs in everything except the messaging logic.
The radio behaviour is the same; the constraints on power, access, thermal envelope and antenna placement are new.
A desktop pool is placed where a person sits. That usually means reasonable signal, accessible ports and ambient cooling. A rack pool is placed where the infrastructure is, which usually means the opposite on all three counts. The consequence is that a design that was validated on a desk can fail in a rack without any component being defective.
The differences that matter operationally are: power, because a rack device draws from a protected supply and the wiring is less forgiving; access, because replacing a SIM in a rack device is an event rather than a task; thermal, because rack ambient temperature is warmer than office ambient; and radio, because a metal enclosure attenuates signal in a way an open desk does not.
Planning each of the four before mounting the device is the difference between a deployment that runs for years and one that produces intermittent faults attributed to the carrier.
How much rack space and power does 64 ports need?
Budget from the loaded figure, and leave room for expansion.
The published unit count is the starting point; the power figure under continuous transmission is the one to design against.
| Input | What to establish | Source |
|---|---|---|
| Rack units | Height of the populated chassis plus any accompanying equipment | Product documentation |
| Steady-state power | Consumption at idle | Product documentation |
| Peak power | Consumption with all modules transmitting | Confirm with the vendor rather than inferring from idle |
| Circuit capacity | Total draw including headroom, on a protected circuit | Your facilities load calculation |
| Thermal load | Heat rejected into the rack, from the peak figure | Converted from peak power |
The distinction between idle and peak is the one that catches deployments out. Modules draw substantially more when transmitting, and a 64-port chassis whose supply or circuit was sized for idle draw will behave unpredictably during a batch. Confirm the peak figure with the vendor rather than deriving it from the steady-state specification, because the ratio between the two varies by module generation.
Leave capacity in the circuit rather than filling it. A pool is frequently expanded, and a circuit at its limit cannot accommodate the next chassis without work in a live environment.
How should antennas be routed?
Bring the signal to the equipment.
Antenna placement decides whether a rack deployment works, and it is the requirement most often treated as an afterthought.
Three options exist, in descending order of simplicity. Internal antennas with an open rack work where the rack is in a room with usable coverage and the chassis is not enclosed in a secondary cabinet. External antennas mounted outside the enclosure re-establish signal where an enclosed rack attenuates it, and require planning for cable routing and for where the antenna physically sits. Remote radio placement moves the radio to a location with good coverage and keeps the SIM estate in the rack, which is the approach used where the room genuinely has no usable signal.
Verify the choice with the actual device at the installed position and with the enclosure closed. A handset that shows adequate signal in the room does not predict what the device will see inside the cabinet, because the enclosure and the chassis both attenuate. Recording the observed registration and signal state per module at installation gives you the baseline against which later degradation can be compared.
Where the room has no usable coverage at all, no antenna arrangement will create it. That is the point at which remote radio placement or a different site becomes the only realistic option, and it is worth establishing before the hardware is purchased rather than after.
How does thermal behaviour affect reliability?
Continuous operation is a different test from a short one.
Equipment that runs cool during a benchmark can behave differently after hours at load in a warm rack.
The mechanism is straightforward: a module that runs warmer than its design envelope may deregister, throttle or behave unpredictably, and the symptom appears as an intermittent network issue rather than as a hardware fault. Because the effect builds over time, it is frequently misattributed to the operator.
Three practices reduce the risk. Leave vertical space between populated units so that air can move rather than being trapped. Confirm the rack intake temperature rather than the room temperature, because the two differ substantially in a populated cabinet. And verify behaviour after a sustained load run at the installed position rather than extrapolating from a bench test. The widely used environmental guidance for data processing environments published by ASHRAE is a reasonable reference when the operating envelope has to be stated in procurement documents, and the environmental test standards published by the International Electrotechnical Commission cover the equipment side.
Where a specification must name a safety or conformity scheme, the certification bodies that operate them, such as UL, publish the scope of each scheme. Confirm which scheme actually applies to the model and market you are buying for, and confirm it with the vendor in writing rather than assuming a family-wide certification.
How should SIM handling change in a data centre?
Treat the SIM estate as controlled inventory.
Access to a data centre is slower than access to an office, which changes the cost of every SIM operation.
Three practices make the difference. Batch operations. Replace or rotate SIMs in planned groups rather than one at a time, because each visit carries the same access overhead regardless of how many cards are changed. Documented mapping. Record which SIM sits in which slot in a system that can be read remotely, so that a fault can be diagnosed without opening the rack. Spares on site. Keep a small stock of known-good SIMs at the facility, because waiting for a card to arrive extends an outage by the delivery time rather than by the repair time.
Where the estate is large enough, separating the SIM layer from the radio removes the access problem entirely. The SIMBANK128 at $1,600.00 provides hot-swapping and dynamic allocation for GoIP gateways, and the SIMPOOL range provides integrated storage from 128 slots at $1,800.00 to 512 slots at $5,400.00, with compatibility stated for the SK gateway range. Where SIM access is the operational pain, moving the estate to a bank is often the change that removes the most work.
How should the pool be segmented on the network?
Separate the management path from the workload.
Three logical paths exist in a pool deployment, and sharing them creates failure modes that are hard to diagnose.
The management path is how administrators reach the host that controls the pool, and it should not be exposed to general traffic. The sending path is the application’s route to the pool, and it should be stable and predictable. The update path is how you apply driver and software changes, and it is the one most often left to chance.
A simple segmentation that works: place the host and the pool on a dedicated segment; allow the application to reach the pool on defined ports only; and reach the management interface through a controlled path rather than an open remote-access service. Current configuration guidance for transport security, such as NIST SP 800-52 Revision 2, is a useful reference when the requirement has to be expressed in a security review.
What should the deployment acceptance check?
Confirm registered channels under load.
A pool verified on a bench and installed in a rack has not yet been verified.
- Registered channel count at the installed position. Compare against the bench figure and investigate any shortfall before go-live.
- Thermal behaviour after a sustained run. Run a defined batch for long enough that the rack reaches steady state, then re-check the registered count.
- Power stability during a batch. Watch the registered count during transmission rather than before it.
- Slot mapping accuracy. Confirm each slot holds the SIM the documentation says it holds.
- Recovery after a power event. Confirm the pool returns to service without manual intervention.
Item two is the one that is usually skipped and most often decisive, because a thermal problem at this density appears only after the equipment has been running long enough to reach temperature. Running it before go-live converts a future incident into a planned adjustment.
Where the deployment also has to satisfy a market or equipment framework, the ETSI standards catalogue covers the network and equipment side of the same specification work.
Conclusion
A 64-port pool in a rack is a facilities project with a messaging component. Power must be sized for simultaneous transmission rather than idle draw, antennas must be planned so that signal reaches the equipment inside its enclosure, thermal behaviour must be verified after a sustained run rather than at the bench, and SIM handling must be treated as controlled inventory with documented mapping and spares on site.
The escalation path is predictable. Where verification at the installed position fails for radio reasons, the choice is external antennas or remote radio placement; where it fails for SIM access reasons, the answer is usually moving the estate to a bank such as the SIMBANK128 at $1,600.00 or a SIMPOOL configuration. The published TYH range runs from the 8-port model at $113.00 to the 64-port model at $579.00, and the TGW-SMS Gateway 64-64 at $1,715.00 covers deployments that need a different chassis class at the same port count.
Verify signal and thermal behaviour at the installed position before go-live. Send your rack environment, coverage conditions and port requirement to service@telarvo.com, or review the published models on the SMS modem solution pages.
FAQ
Will a 64-port pool work inside a closed rack cabinet?
It will work if signal reaches the modules, which depends on the enclosure rather than on the pool. A closed metal cabinet attenuates radio, so verify registration with the device installed and the cabinet closed. Where signal is inadequate, external antennas or remote radio placement are the options; no antenna arrangement creates coverage in a room that has none.
How much power does a 64-port pool need?
Size from the peak figure with every module transmitting rather than from the idle figure, and confirm the peak value with the vendor rather than deriving it. The difference between idle and peak is what causes deployments to fail during a batch, and the symptom usually appears as modules deregistering rather than as a power alarm.
Should the SIM estate be moved out of the data centre?
Access is the reason to consider it. Replacing a SIM in a rack device costs an access visit regardless of how many cards are changed, and that overhead dominates once the estate grows. Separating the SIM layer allows the estate to sit where access is controlled while the radios sit where coverage is adequate, at the cost of one more link in the failure domain.
What is the difference between the TYH and TGW options at 64 ports?
They are different chassis classes at a similar port count, with the TYH 64-port modem pool published at $579.00 and the TGW-SMS Gateway 64-64 at $1,715.00. The right choice depends on how the device is controlled and integrated rather than on the port count alone. Confirm the control interface and management model of each against your sending software before choosing.
{
“@context”: “https://schema.org”,
“@type”: “FAQPage”,
“mainEntity”: [
{
“@type”: “Question”,
“name”: “Will a 64-port pool work inside a closed rack cabinet?”,
“acceptedAnswer”: {
“@type”: “Answer”,
“text”: “It will work if signal reaches the modules, which depends on the enclosure rather than on the pool. A closed metal cabinet attenuates radio, so verify registration with the device installed and the cabinet closed. Where signal is inadequate, external antennas or remote radio placement are the options; no antenna arrangement creates coverage in a room that has none.”
}
},
{
“@type”: “Question”,
“name”: “How much power does a 64-port pool need?”,
“acceptedAnswer”: {
“@type”: “Answer”,
“text”: “Size from the peak figure with every module transmitting rather than from the idle figure, and confirm the peak value with the vendor rather than deriving it. The difference between idle and peak is what causes deployments to fail during a batch, and the symptom usually appears as modules deregistering rather than as a power alarm.”
}
},
{
“@type”: “Question”,
“name”: “Should the SIM estate be moved out of the data centre?”,
“acceptedAnswer”: {
“@type”: “Answer”,
“text”: “Access is the reason to consider it. Replacing a SIM in a rack device costs an access visit regardless of how many cards are changed, and that overhead dominates once the estate grows. Separating the SIM layer allows the estate to sit where access is controlled while the radios sit where coverage is adequate, at the cost of one more link in the failure domain.”
}
},
{
“@type”: “Question”,
“name”: “What is the difference between the TYH and TGW options at 64 ports?”,
“acceptedAnswer”: {
“@type”: “Answer”,
“text”: “They are different chassis classes at a similar port count, with the TYH 64-port modem pool published at $579.00 and the TGW-SMS Gateway 64-64 at $1,715.00. The right choice depends on how the device is controlled and integrated rather than on the port count alone. Confirm the control interface and management model of each against your sending software before choosing.”
}
}
]
}