Bently Nevada 3500 Module Replacement Checklist: Part Numbers, Compatibility, Configuration and Safe Hot-Swap Practice

A failed monitor in a machinery protection rack is rarely an emergency on its own, but a careless swap can become one. Many spurious trips on large compressors and turbines during maintenance come not from the failed hardware but from the repair itself: a channel left bypassed, a configuration never downloaded, or an I/O module that does not match the field wiring. A disciplined Bently Nevada 3500 module replacement procedure protects both the machine and the production that depends on it.

This guide is for instrument and reliability engineers who maintain 3500 racks and for buyers who must order the right part the first time.

How a 3500 Rack Is Organised

The Bently Nevada 3500 system is a modular, rack-based protection system built to meet the requirements of API 670. A full-size 19-inch rack typically provides fourteen slots for monitors and relay modules, with dedicated positions on the left for the power supplies and the rack interface. Smaller rack variants exist for applications with fewer channels; confirm slot count against the rack datasheet before planning an expansion.

Every functional position in the rack is made of two parts:

• A main module (the monitor board), inserted from the front. It holds the signal processing, alarm logic, LEDs and buffered outputs.
• An I/O module, mounted at the rear directly behind the main module. It carries the field terminations and determines how the transducers are wired: internal or external terminations, and versions intended for use with intrinsic safety barriers or galvanic isolators.

The main module and the I/O module are ordered and identified separately. A replacement main module will only work correctly if the I/O module behind it is the type the monitor expects and the type the field wiring was designed for. This is a frequent source of error during replacement.

Module Map: What Each Part Does

Channel counts and functions below are typical; confirm against the datasheet for the installed revision.

Product pages for the most requested modules: 3500/15 power supply, 3500/22M TDI, 3500/25 Keyphasor, 3500/32M relay module, 3500/33 149992-02, 3500/40M Proximitor monitor, 3500/42M, 3500/45 position monitor, 3500/50M tachometer and 3500/60 temperature monitor.

Reading the Part Number Before You Order

3500 modules are ordered with a base model followed by option fields, for example a format such as 3500/42M-AXX-BXX or 3500/60-01-00. The option digits encode choices such as the I/O module type and the agency approval option. The exact meaning of each digit differs between module types, so the only reliable reference is the ordering information in the module datasheet.

Each physical board also carries its own assembly part number and revision label, such as 135137-01 or 149992-02. That number identifies a specific board, which may be the main module or the I/O module, so record both labels before ordering.

Configuration, Firmware and Software

A new module does not arrive with your machine’s configuration. Channel types, scale factors, full-scale ranges, alert and danger setpoints, delays, trip multiply, filters, relay voting logic and Keyphasor assignments must be restored with the 3500 Rack Configuration Software through the rack interface.

Practical points that save time:

• Keep a current, controlled copy of the rack configuration file for every rack, stored off the engineering laptop as well as on it. Upload and save the existing configuration from the rack before any planned work.
• Check the configuration software version against the firmware of the replacement module. A newer module may need a newer software release to be configured correctly, and an older software version may not recognise newer hardware.
• Where the rack interface has a configuration key switch or equivalent lock, it must be set to allow configuration only for the duration of the download, then returned to its run position.
• After download, upload the configuration again and compare it with the controlled file. Do not rely on “download successful” alone.

If the 3500/22M itself is replaced, treat it as a rack-level change: it carries rack settings and the network parameters used by condition monitoring software, so plan to re-establish and check those links.

Hot-Swap Practice

3500 modules are designed to be removed and inserted with the rack powered. That is a hardware capability, not a permission. Removing a monitor affects the alarm and relay logic that depends on its channels, and the behaviour of the voting logic when a monitor is missing depends on how the rack was configured.

A safe sequence, written to fit within your site’s permit-to-work and bypass management procedures, typically looks like this:

1. Obtain the bypass or inhibit permit for the affected protective functions. Operations must know which machine trips are degraded and for how long.
2. Identify every relay output and every voting group that uses channels on the module to be replaced. Check the relay logic in the configuration, not just the loop drawings.
3. Apply channel or alarm bypasses as required by the procedure, and confirm on the module LEDs and in software that the bypass is active.
4. Remove the main module using proper ESD precautions. Leave the I/O module in place unless it is the failed item.
5. Insert the replacement, wait for it to initialise, and download the configuration.
6. Verify the channels (see the next section) before removing any bypass.
7. Remove bypasses one at a time, confirming that no alarm or danger condition is active on channels that drive trip relays.

Never swap a module with live trip relays and no bypass permit, and never leave a rack in configuration mode overnight.

Post-Replacement Verification

Work through these checks with the machine running if the procedure permits, or at standstill with simulated signals:

• Module OK LED steady and TX/RX LED showing communication activity.
• Every channel reports OK. A not-OK channel usually points to a gap, wiring or transducer fault, or to an incorrect transducer type in the configuration.
• Proximity probe gap voltages are within the expected linear range and close to the values recorded before the swap.
• Direct vibration and position readings agree with the pre-failure trend in the condition monitoring software and with portable readings if available.
• Alert and danger setpoints, time delays and trip multiply settings match the controlled configuration.
• Relay logic is proven by test: simulate an alarm on each channel and confirm the correct relay changes state, with the downstream trip path isolated per procedure.
• Bypass LEDs are off, and the bypass log is closed.

Record old and new serial numbers and firmware revisions in the maintenance system.

Common Mistakes

• Ordering the main module without checking the I/O module type. A barrier-type I/O module cannot be swapped for an internal-termination type without rewiring and hazardous area review.
• Assuming a newer revision is a drop-in replacement. Usually it is, but configuration software support must be confirmed first.
• Forgetting to download the configuration, or downloading an outdated file from an engineering laptop.
• Leaving channels bypassed after the job, which silently removes protection from the machine.
• Mixing TMR and standard modules. TMR racks use dedicated module types and voting arrangements; a standard module is not a substitute.
• Replacing a monitor when the real fault is a transducer. A not-OK channel after replacement points to the probe, extension cable or Proximitor sensor.

Spares Strategy for Oman and GCC Sites

Many protection racks in Oman sit on gas compression, power generation and pumping installations in the interior, hours from the nearest workshop. Lead times for specific modules can be long, and one missing part can leave a machine unprotected or stopped.

A practical minimum spares holding for each site with several racks of the same type is:

• One 3500/15 power supply of each input type in use.
• One 3500/22M TDI, because a failed rack interface takes away configuration access and monitoring data for the whole rack.
• One spare of the most common monitor type, usually the 3500/42M or 3500/40M.
• One relay module matching the type in use.
• One I/O module for each termination style, since I/O module failures are rarer but the wiring dependency makes them hard to source quickly.

Store spares in ESD packaging in a climate-controlled store; warehouses in Sohar, Duqm or Salalah combine humidity cycles with 45 to 50 °C summer ambients that stored electronics tolerate poorly. Label each spare with its full part number, firmware revision and the racks it can serve. Where racks sit in unconditioned shelters, check enclosure cooling, because high internal temperature shortens power supply and monitor life.

Replacement Checklist

1. Confirm the failed item: main module, I/O module, transducer or wiring.
2. Record full model code, option code, assembly part numbers and revisions of both main and I/O boards.
3. Confirm standard or TMR rack and the approval option required.
4. Check configuration software version compatibility with the replacement firmware.
5. Upload and save the current rack configuration before starting.
6. Raise the bypass permit and identify every relay and voting group affected.
7. Apply bypasses and verify them in software and on the LEDs.
8. Swap the module with ESD precautions; leave the I/O module unless faulty.
9. Download the configuration, upload it again and compare.
10. Verify OK status, gap voltages, readings, setpoints and relay logic.
11. Remove bypasses, close the permit and update the maintenance record.
12. Reorder the spare that was consumed.

For background on the protection philosophy behind these racks, including voting, transducer requirements and the role of API 670, see our API 670 machinery protection system guide, and for the transducer side of the chain, our explanation of Bently Nevada 3500 vs 3300 XL proximitor systems.

Sourcing 3500 Modules Through Seven Star LLC

Seven Star LLC supplies Bently Nevada 3500 main modules, I/O modules and associated transducer components for operators across Oman and the GCC. Send us the full part number and option code from the module label, the assembly numbers of both boards and a photo of the rack rear if the I/O type is uncertain, and our team will match the request before quoting. Browse the full vibration monitoring category or contact our team with your list.

Send Seven Star LLC your 3500 module part numbers and rack details, and we will help you source the correct replacement for your machinery protection system.