Sizing 24VDC and 230VAC Fuses for S7-1500 PLC Systems

Proper overcurrent protection on an S7-1500 station is a three-tier problem: the 230 VAC branch feeding each SITOP, the 24 VDC rail that powers the CPU and integrated I/O backplane, and the 24 VDC field-distribution wiring that lands on 83 sensors and 30 solenoids. Each tier has its own rule. The 230 VAC breaker is sized for the AC wiring, not the load. The 24 VDC supply fuse protects the SITOP and the CPU backplane. The branch fuses inside the marshalling cabinet protect the 24 VDC field wiring between the terminal block and the device. Getting all three tiers coordinated eliminates nuisance trips on solenoid inrush without leaving a short circuit to find a weak link elsewhere in the cabinet.

This article walks through the architecture, the math, and the commissioning checks for a CPU 1512C-1 PN station fed by two Siemens SITOP PSU100M units, with reference values for a 5-station machine that uses ~83 PNP sensors and ~30 24 VDC solenoids/relays.

1. System Architecture and Power Distribution

The reference design has two completely independent 24 VDC power supplies. This is intentional: a field short on the sensor rail must not brown out the CPU, and an output device that latches on a stuck solenoid must not pull the input rail down.

2. Prerequisites, Standards, and Reference Manuals

Before sizing, pull the three documents that govern a Siemens 1500 station:

• S7-1500 Automation System System Manual (entry ID 59191792) - CPU current draw, integrated I/O limits, backplane power budget.
• SITOP PSU100M / PSU8200 manuals - 6EP1332 (2.5 A), 6EP1334 (5 A), 6EP1336 (10 A) input current, inrush, recommended line-side breaker.
• DIN EN 60204-1 (Safety of machinery - Electrical equipment of machines) and NFPA 79 (Electrical Standard for Industrial Machinery) - the two pillars for branch protection on a machine panel.

Tools needed for verification after assembly: a true-RMS clamp meter (Fluke 376/381 or equivalent), a 24 VDC bench supply with adjustable current limit for staging, an insulation tester for the 500 V DC hi-pot between DC- and PE, and TIA Portal with the S7-1500 diagnostic buffer open while you power-cycle.

3. Tier 1: 230 VAC Branch Protection (Upstream of the SITOP)

The 230 VAC breaker does not protect the SITOP - the SITOP has its own input fuse, inrush limiter, and output current limit. The 230 VAC breaker protects the AC wiring between the disconnect and the SITOP terminals. That is the only thing it can do, because the SITOP is the load boundary.

Single-phase apparent power is kVA = V × I / 1000. For the 2.5 A SITOP on a 230 V line:

That is the FLA figure to use in the breaker coordination table. The breaker is then sized to the AC wire, not the load. With 1.5 mm² (16 AWG) AC wiring to the SITOP - the practical minimum on a Siemens panel - the safe continuous current is around 10 A for chassis wiring (10-15 A per DIN EN 60204-1 table 5 with 70°C terminals, derated for bundling), so a 1 A or 2 A breaker is electrically trivial. It exists to give the service electrician a defined disconnect and to clear a dead short if the AC wiring is mechanically damaged.

Rule of thumb that is field-proven across hundreds of panels: choose the breaker for the 230 VAC side by the AC wire gauge, not by the SITOP output. For 1.5 mm² AC, a 6 A breaker is the common maximum. For 2.5 mm² AC, a 10 A or 13 A breaker. The SITOP is downstream.

4. Tier 2: SITOP DC Output Protection

The SITOP 24 VDC output is electronically current-limited and foldback-protected. The 6EP1332-3BA10 will deliver a constant 2.5 A into a normal load, hold 2.5 A on a mild overload for a few seconds, then fold back to a safe value on a hard short. The internal output fuse (typically a 5 × 20 mm glass fuse, F1 on the PCB, 6.3 A slow-blow on the 5 A unit) is a backstop for a failed output rectifier, not a daily-use protection device.

You do not need to add an external fuse on the 24 VDC output of the SITOP itself if the cable run from SITOP to the marshalling terminal block is short and protected (conduit, duct, or cable tray). What you do need is fuses on every 24 VDC distribution conductor leaving the marshalling block.

5. Tier 3: CPU 1512C-1 PN Power, Inrush, and Internal Budget

CPU 1512C-1 PN has the following relevant power figures (from the S7-1500 system manual CPU data sheet):

The 19 A figure that occasionally circulates in student forums is a misreading: it is the per-output pin rating summed across all 32 integrated outputs (32 × 0.5 A = 16 A theoretical, rounded up), not a current the CPU ever draws from the 24 V supply. The CPU does not source 24 V to outputs - the outputs switch the external 24 V supply. The CPU's own 24 V consumption is in the 1-2.5 A band.

Implication for the 24 VDC rail that feeds the CPU:

• If you use a 6EP1332-3BA10 (2.5 A) for the CPU rail, you can power the 1512C-1 PN plus one or two small I/O modules directly. With a stack of 5-6 I/O modules you are at the SITOP's continuous limit and have zero margin for inrush.
• If you have a CPU plus 5+ I/O modules on the same rail, choose a 6EP1334-3BA10 (5 A) or 6EP1336-3BA10 (10 A). Add a 5 A or 8 A slow-blow fuse on the 24 VDC feeder to the CPU backplane to give a defined disconnect point.

6. Tier 4: 24 VDC I/O Group Fusing

This is where the source's question about 1-pole vs 2-pole fuses and 1 A vs 3 A values gets practical. The function of the per-group fuse is to clear a field-device short without dropping the entire 24 VDC rail, and to protect the wire from the terminal block to the sensor. That is the same function the AC breaker performs on a different voltage class.

6.1 Group sizing math for the 83-input station

The 83 PNP sensors are mostly 3-wire inductive proximity switches (e.g. Sick IME12, IFM IFS, Balluff BES) drawing 8-15 mA each at 24 V. Worst-case inrush on a short-circuited sensor lead: the SITOP's 5 A electronic limit, but only momentarily - the group fuse opens first.

Group 1, 2, 3 sensor count: 17 / 16 / 16 / 17 / 17 (five groups, average 16.6)
Continuous current per group: 16 sensors x 15 mA = 0.24 A
Solenoid inrush on same rail: 4 valves x 0.5 A inrush for 50 ms = 2.0 A transient
1 A slow-blow (T1A) per group: passes 0.24 A continuous, opens 2.0 A inrush within 1-2 s, clears 10 A hard short in <50 ms

The choice between 1 A and 2 A fuses depends on how many outputs sit on the same fuse group as the inputs. If outputs are split onto a separate group, 1 A per input group is comfortably correct. If outputs are mixed with inputs on the same fuse, step to 2 A and 3 A to ride through the 500 mA-1 A solenoid inrush.

6.2 Outputs: 8 outputs per fuse

30 outputs at 100 mA continuous = 3.0 A continuous, but with the inrush multiplier for solenoid valves, the realistic peak per output is 500 mA-1 A. Grouping 8 outputs per fuse gives:

8 x 0.5 A inrush = 4.0 A peak for 30-100 ms (decay to holding)
8 x 0.1 A hold = 0.8 A continuous
T3A slow-blow per output group: passes 0.8 A continuous, rides 4 A inrush

For a 4-output group (a typical 1512C-1 PN digital output channel pair), T2A is sufficient.

7. Single-Pole vs Two-Pole DC Fusing

On a 24 VDC control circuit, fuse only the +24 V conductor. The 0 V return is a common bus, usually bonded to PE at one point (single-point ground) and is treated like the grounded conductor of a 120 VAC system - no fuse. This is the same convention as fusing only the hot leg on a single-phase 120 VAC branch.

Two-pole fusing on DC is appropriate in three specific cases:

1. Ungrounded battery systems (e.g. 24 VDC battery-backed control on a railway or marine panel) where either polarity could source current into a fault. Use double-pole fuses or a battery disconnect that opens both legs.
2. PV / solar battery installations where the system is functionally a current source and a ground fault on either polarity is hazardous. Code requires both-poles fused or protected by a double-pole breaker.
3. IT-grounded 24 VDC systems (medical, some process plants) where neither leg is bonded to PE and a single ground fault must not disable protection. NFPA 79 and IEC 60204-1 require ground-fault monitoring on the 24 VDC rail; you still fuse both poles.

For a bachelor-thesis machine on a SITOP-fed, solidly-grounded 24 VDC bus, single-pole fusing on +24 V is correct. Always verify your local electrical code on this - in some EU countries the inspector will insist on a 2-pole MCB on the SITOP AC input, which is a different question (the SITOP itself has only L and N connections, so a 1P+N breaker with N not switched is the common solution).

8. Wire Ampacity - the Real Sizing Constraint

Once you have chosen a wire size for the 24 VDC field wiring, the fuse must protect the wire, not the load. This is the rule in both NFPA 79 article 12 and IEC 60204-1 chapter 12. The breaker/fuse rating cannot exceed the wire ampacity, after derating for ambient temperature, bundling, and conduit fill.

For 16 AWG control wire in a 40°C panel with 4 conductors in a duct, the actual ampacity after derating is closer to 7-8 A. A 6 A fuse is correct. A 10 A fuse violates the wire-protection rule and will not clear a partial short before the wire insulation chars.

9. Field-Device Inrush - Solenoids, Contactors, Sensors

Most nuisance-trip problems in 24 VDC control panels are caused by solenoid inrush, not sensor current. Typical figures:

The 0.5 A inrush then 10 mA hold pattern is real. With 8 solenoids energizing simultaneously in a worst-case scan cycle, a 3 A slow-blow fuse will not nuisance-trip. A 3 A fast-blow fuse will trip every cold start. Slow-blow (T-class, formerly TT) is the right choice for output groups; fast-blow (F-class) is the right choice for input groups where the load is purely resistive.

10. Optional: SITOP Select Diagnostic Module

The 6EP1961-2BA family (SITOP select) is a 4- or 8-channel 24 VDC distribution module with built-in electronic current monitoring per channel. It replaces the fuse strip in larger installations and reports channel status to the PLC or to a SITOP PSU8200 with diagnostics. It costs more than a fuse terminal block but eliminates nuisance trips entirely - a tripped channel can be reset without replacing a fuse.

Catalog numbers (verify with the current Siemens online catalog at the time of purchase):

• 6EP1961-2BA00 - SITOP select, 4 x 10 A channels, 24 VDC, with status output.
• 6EP1961-2BA10 - SITOP select, 8 x 5 A channels.
• 6EP1961-2BA20 - SITOP select, 8 x 10 A channels, with Ethernet diagnostics (SITOP PSU8200 only).

For a thesis machine, the SITOP select is overkill. A conventional fuse terminal block (Phoenix UK 10, Weidmuller WSI, Wago 2002-1981) is the right tool. Reserve SITOP select for installations where you cannot tolerate a downtime event to replace a blown fuse.

11. Verification and Commissioning Checklist

After the panel is wired but before powering the SITOPs, run this checklist:

1. Insulation test. With all SITOPs disconnected, measure 500 V DC between PE and the DC+ rail, then PE and DC-. Reading should be >1 MΩ. This catches a pinched wire or a stray strand at a terminal.
2. Polarity check. With a digital multimeter on diode/continuity mode, verify DC+ lands on the + input of every SITOP, every I/O module, and every sensor terminal. A reversed sensor will not destroy a SITOP, but it will short the SITOP through every sensor's input bridge.
3. Ground bond test. PE to cabinet chassis at <0.1 Ω with a 4-wire milliohmmeter at 10 A. This is a CE-marked machine's required test per IEC 60204-1 chapter 18.
4. No-load SITOP test. Energize each SITOP separately, measure 24 VDC at the output terminal, then at the end of the longest cable run in that distribution. Acceptable droop: <3% (23.3 V at the last sensor).
5. Full-load test with one group at fault. Energize the panel, then deliberately short-circuit one sensor's DC+ to its M12 thread (a common failure mode on a machine). The corresponding group fuse must open within 100 ms and the SITOP's output voltage must remain within 24 V ± 10% on all other groups. If the SITOP itself drops out, the group fuse is too small or the SITOP is undersized for the inrush.
6. CPU inrush test. Power-cycle the CPU 1512C-1 PN 20 times in 30 seconds (cold inrush). The 4 A slow-blow fuse feeding the CPU must not open. TIA Portal should show no diagnostic buffer entries for 24 V undervoltage.
7. PE fault simulation on the DC rail. With the machine off, use a current-limited source to apply a 1 kΩ resistor from DC+ to PE. The 24 VDC rail voltage should remain stable (no insulation monitor should trip) and the SITOP foldback should not engage. This validates that your PE bonding is single-point and that ground-fault current cannot flow into a cabinet ground loop.

Document every test result in the thesis appendix. The S7-1500 has a maintenance log inside the diagnostic buffer (TIA Portal > Online & Diagnostics > Diagnostic buffer) - export it at the end of commissioning and include it.

12. Fuse Schedule Summary Table (For the Thesis Appendix)

13. Common Mistakes on First Panels

• Reading the I/O module's "8 A per group" as 8 A continuous draw. The 8 A figure is the maximum load the module's output drivers can switch, not the current the module draws from the backplane. Always read the "Power dissipation" and "Current consumption from backplane" lines, not the output load rating.
• Using 10 A AC breakers for the SITOP branch. A 10 A breaker on a 1.5 mm² AC wire to a 60 W SITOP is not wrong electrically, but it is wrong functionally - the breaker will not trip on a partial short in the AC wiring until the wire is well into the smoke. A 1 A or 2 A breaker is far more sensitive to insulation degradation.
• Sharing a fuse between inputs and outputs. A shorted sensor will then drop the corresponding output group, and a chattering solenoid will drop the corresponding input group. Functionally separate the two rails at the fuse level even if they share the same SITOP.
• Bonding DC- to PE at two points. This creates a ground loop. If your SITOP sits next to a VFD in the same cabinet, the VFD's high-frequency leakage will induce 50/60 Hz ripple on the 24 VDC rail through the second bond. One bond only, at the SITOP, never at the field marshalling block.
• Skipping the polarity test on M12 sensor leads. A reversed 4-pin M12 PNP sensor passes the bench test but short-circuits when the load draws current. This is the single most common cause of a SITOP foldback on a fresh machine.