To evaluate several power supplies starting on one AC branch, assess their combined startup current over time under the actual switching and restart conditions. Normal input current and a single inrush peak cannot establish an allowable unit count. Obtain the exact models’ startup data, include other coincident loads, and compare the aggregate event with the selected protection and switching equipment. Cold start, brief interruption, and warm restart need separate consideration.
Mục lục
Separate AC input inrush from the DC load starting
Input inrush occurs as energy-storage components charge after input voltage is applied. The DC load may start later, depending on the supply and equipment sequence. This manufacturer’s input-inrush explanation describes capacitor charging and several limiting methods. Supplies with the same steady output rating can therefore have different switch-on characteristics.
A 24 V output-current figure describes the DC side. It should not be compared directly with an AC branch breaker’s ampere rating. Likewise, RMS input current derived from efficiency and power factor describes an operating load condition, not the short charging event. Keep both records because the branch must support steady operation as well as startup.
Concept overview
Supply input is energized
Review the charging event
Output load starts or is enabled
Review the receiver demand
Define the switching boundary first. List every supply, drive, transformer, filter, and other device energized by the same upstream action. A cabinet containing four supplies may behave differently when all four receive AC together, when one is added to a running branch, or when an upstream interruption resets the whole group.
Request the waveform and its starting conditions
A useful startup specification identifies the input voltage, temperature, initial energy state, limiting method, and switching conditions. Ask whether the peak is typical or a maximum and whether pulse duration has a defined measurement convention. Width measured at half the peak is not interchangeable with the full charging duration.
| Mục | Record | Why it matters |
|---|---|---|
| Supply population | Full model, revision, quantity and DC load state | Different input stages need separate startup evidence |
| AC source | Voltage range, frequency and representative source/wiring impedance | The actual charging event depends on the source and circuit |
| Starting state | Prior operation, off-time and ambient condition | Capacitor charge and limiter temperature change during an interruption |
| Current event | Peak, waveform, duration definition and repeatability | A peak alone omits the time response relevant to coordination |
| Protection and switching | Exact breaker or fuse, contactor or relay, applicable ratings and manufacturer guidance | Startup withstand, switching duty and fault protection are distinct checks |
| Sequence | Devices per group, timing tolerance and response after a source interruption | The installed sequence may not match the intended drawing |
Include the exact upstream protective function. An overcurrent event and operation of residual-current protection require different investigations. If a circuit already trips, first identify the device, trip indication, and timing; do not presume that inrush is the cause.
Cold start and warm restart can produce different events
In a supply using an NTC thermistor, the limiter’s resistance falls as it warms. The component manufacturer’s NTC application note explains why a warm limiter may provide little current limitation. Check the restart state as well as the cold-start value.
That does not mean the shortest interruption always produces the largest peak. The input capacitors may retain charge while the limiter remains hot; a longer interruption can change both states. Active limiting circuits have their own reset and restart behavior. Ask for the applicable operating sequence and use the off-time range relevant to the machine and permitted by the manufacturer.
For sample qualification, distinguish a start from the defined cold condition, a restart after normal loaded operation, and restoration following the specified brief interruption. Record the prior state with each result. A successful second attempt is not enough to accept the first event, because the starting state may have changed.
Worked example: coincident pulses change the branch event
Hypothetical arithmetic example: four identical supplies each draw an idealized rectangular input pulse of 30 A for 3 ms. Compare perfect coincidence with separated groups whose pulses do not overlap. The calculation excludes normal running current, source impedance, and other loads; the assumed pulses are not WEHO specifications or measurements.
The instantaneous branch current is the sum of the currents present at that instant: ibranch(t) = Σ iunit(t). For the assumed rectangular event, the current-squared time integral is ∫i²dt = I² × t, with time in seconds.
| Assumed sequence | Peak per event | Integral per event | Sum across all events |
|---|---|---|---|
| All four start together | 4 × 30 = 120 A | 120² × 0.003 = 43.2 A²s | 43.2 A²s |
| Two separated groups of two | 2 × 30 = 60 A | 60² × 0.003 = 10.8 A²s | 2 × 10.8 = 21.6 A²s |
| Four separated single starts | 30 A | 30² × 0.003 = 2.7 A²s | 4 × 2.7 = 10.8 A²s |
This illustrates why adding individual pulse-integral values does not represent perfectly coincident pulses. The currents add before the result is squared. It also shows why staggering may reduce the instantaneous demand in this simplified case.
For the real branch, previously started units continue drawing current, pulse shapes vary, and the source influences charging. The summed integrals in the table do not represent a model of breaker heating or recovery between events. Do not use them to claim a permitted supply count or a sequencing delay. Capture or obtain the actual branch event and use the protective-device manufacturer’s method.
Evaluate protection, switching, and sequencing separately
For a fuse, review the selected family’s latest time-current and pulse-withstand information, repetition, temperature, voltage, and breaking-capacity requirements. Obtain the fuse and holder documentation for the actual installation and compare their limits with the expected startup waveform. A pulse integral is one input; it is not a complete fuse selection.
For a breaker, use the actual device’s characteristics and coordination guidance, including its behavior over the event duration. A generic curve letter or rated current does not establish the response to a particular short pulse. Changing the device must preserve the circuit’s required fault protection and conductor coordination.
Check the switch or contactor’s suitability for the capacitive charging duty as well. A current rating for another load type may not describe its making duty with several supplies. For lighting branches, use the LED driver inrush and relay contact-duty review to assess the switching device. This guide addresses the common AC branch and group startup sequence. Any external inrush limiter needs its own continuous-current, thermal, reset, fault, and installation review.
Sequence the AC groups, not just the DC loads
To separate input-charging events, the sequence must act where AC reaches each supply group. An architecture to evaluate is separate group contactors commanded in stages by on-delay timer relays or a PLC. The control provides the sequence; the selected contactors must handle the groups’ actual making duty. Confirm startup, restart and failure behavior for the complete arrangement.
Identify the first group, the later groups, timing tolerance, and the control-power source. The timer or PLC must be available when the sequence needs it. Verify the response to a brief mains interruption, retained control power, and controller reset: a valid cold-start sequence can still restart all groups together under another state. Set the interval from actual startup data and switchgear coordination; the example above provides no universal delay.
Delaying a DC load’s enable signal can reduce initial output demand, but it does not stagger the supplies’ input-capacitor charging when a common AC switch closes. Keep the AC energization sequence and DC load-enable sequence separate on the functional description and acceptance record.
Define an acceptance plan for the assembled branch
Have a qualified test team evaluate representative input and restart conditions using equipment and procedures suited to the mains circuit. Record branch voltage, current, startup order, off-time, protective-device status, and the DC receivers’ startup behavior. Repeat the relevant conditions rather than accepting one favorable switch-on.
Useful acceptance criteria include the intended sequence occurring within its timing tolerance, no unwanted operation of the specified protection, the required DC availability, and no device exceeding its documented switching or operating limits. Keep these AC checks separate from DC branch protection and selectivity.
For a WEHO candidate, request startup information for the full model and revision. A cold-start peak is a useful first input; complete the branch review with pulse duration and relevant restart conditions. The NDR-240 product and specification links earlier in this guide identify one document starting point, not an approved multi-supply switching combination.
Contact WEHO about multi-supply startup. Start with four items: AC voltage range; supply models and quantities; simultaneous or grouped AC starting; and the protective-device model if known. If you have not chosen a supply, include the required output load per unit.
Optional project record: Download the multi-supply startup record (CSV). Use it to record your own inputs and evidence; example entries are illustrative.
