What a Control Transformer Is Sized For

Why this matters

A panel runs all day, the control transformer is cool to the touch, and the biggest contactor in the cabinet will not pull in. Somebody measures the secondary, reads it right on the button, and concludes the transformer is fine, so the contactor gets replaced. The new one does the same thing. What the meter measured was the sealed condition, and the transformer was never asked a hard question in that condition. Control transformers are sized against an instant that lasts a fraction of a cycle, and if that instant is not what you sized against, nothing about a day of trouble-free running tells you anything.

Two burdens, not one

An AC magnet coil presents two completely different loads depending on where its armature is.

Inrush burden, armature open. With the air gap open, the magnetic circuit is poor, coil inductance is at its lowest, and current is at its highest. This is the condition at the instant the coil is energized and it persists until the armature closes.

Sealed burden, armature closed. With the gap closed, inductance rises sharply and current falls to a small fraction of the inrush value.

Both directions of that mechanism are worth holding. Open gap gives high current; closed gap gives low current. So a contactor whose armature is jammed part-open by debris, a bent shading ring or a mechanical interference does not draw a little more than sealed, it draws inrush continuously, and it burns its own coil and overloads the transformer while the panel appears to be idling.

What the nameplate is telling you

A control transformer carries two ratings and the second one is the one that matters here.

Continuous VA is what it can supply indefinitely without exceeding its temperature rise. This is the number that keeps it cool, and it is the number a sealed panel is testing.

Inrush VA at a stated secondary voltage retention is what it can supply momentarily while holding its secondary above a given percentage of nominal. The retention percentage is part of the rating and not a footnote: a transformer rated for a given inrush at 90 percent retention is a different device from one rated for the same inrush at 85 percent, and comparing the two on the VA figure alone compares nothing.

For estimating between rated points, treat the secondary drop as roughly proportional to the burden. That approximation holds reasonably over a small control transformer's normal range, where the drop is dominated by winding impedance, and it degrades at large overloads, so it is an estimate to size with and the manufacturer's regulation data is what governs a final selection.

The coil has to pick up, not just see voltage

A magnet coil requires a minimum voltage to pull its armature in, commonly around 85 percent of its rated coil voltage, and the specific figure is a device data-sheet value. NEMA ICS 2 sits behind the common figures and binds through the manufacturer's declaration and your purchase specification rather than on its own authority.

That threshold has a hard edge. Above it the armature snaps closed and the burden collapses to sealed. Below it the armature hangs, the coil stays in its inrush condition, and the transformer stays loaded at the burden that caused the problem. There is no partial success, which is why the symptom is a buzz or a chatter rather than a slow, weak pull.

The primary side is already down before you start

The secondary tracks the primary proportionally. If the line is at the bottom of its normal band, the secondary is at the bottom of its band before a single coil is energized. For a 120 V nominal system, ANSI C84.1 Range A service voltage runs 114 to 126 V, in the edition your utility's service rules reference; that standard binds through the utility's tariff and service rules rather than on its own. Design at the low end. A transformer selection that only works at nominal line voltage is a selection that works in the shop and not in the building.

The gate

A control transformer is adequate for a panel when, with the primary at the low end of its normal band, the secondary voltage retained at the instant of the worst-case pickup is at or above the pickup voltage of the coil that has to pick up. Unit of analysis is the panel at its worst pickup instant, not the panel averaged over a cycle of operation. Both conditions must hold together: the continuous rating must cover the sealed total AND the inrush capability must cover the pickup instant. Passing one and failing the other is the normal case, not an edge case.

The worst-case pickup instant is the largest inrush burden that occurs while everything else that is going to be energized is already sealed. Where two large devices are energized simultaneously by the same contact, both inrush burdens land at once and both go in the total.

Two panels, one transformer, opposite answers

Both panels get the same transformer: 100 VA continuous, rated 300 VA inrush at 90 percent secondary retention. Both have the same coil pickup requirement of 85 percent of rated coil voltage from the device data. Both sit on a line at the bottom of the band, which is 114 V on a 120 V nominal system, or 95 percent of nominal.

Both panels carry the same sealed load: three small relays at 4 VA sealed each, which is 12 VA, plus one contactor at 10 VA sealed, for 22 VA sealed in total. The continuous rating of 100 VA covers that in both cases with room to spare, which is exactly why both transformers run cool.

Panel A. The largest device picking up while the rest are sealed is a contactor with 60 VA inrush. Total at the pickup instant is 22 plus 60, which is 82 VA. That is well under the 300 VA at which the 90 percent retention is quoted, so the actual retention is better than 90 percent. Take 90 percent as a conservative floor anyway: 0.95 times 0.90 gives 85.5 percent of nominal at the coil, against a pickup requirement of 85 percent. It clears, and it clears by more than the floor calculation admits, because the real retention at 82 VA of a 300 VA capability is considerably higher.

Panel B. Same sealed 22 VA, but the largest device picking up is a much larger contactor with 400 VA inrush. Total at the pickup instant is 22 plus 400, which is 422 VA, against a 300 VA inrush rating. Using the proportional-drop estimate above: at 300 VA the drop is 10 percent, so at 422 VA it is roughly 10 times 422 divided by 300, which is about 14.1 percent, leaving about 85.9 percent retention. Multiply by the 95 percent line condition and the coil sees about 81.6 percent of nominal, against a pickup requirement of 85 percent. It fails by about 3.4 points, and it fails on every attempt, at the bottom of the line band, with a transformer that has never been warm.

The trap that kept this hidden. Measure the secondary in Panel B with everything sealed and the burden is 22 VA, so the reading is healthy. Measure it during the pickup and a hand-held meter's averaging will not resolve an event that ends when the armature either closes or gives up. The condition where the transformer is inadequate is the only condition nobody measures, and the reading that "proved the transformer was fine" was taken in the condition where it genuinely is.

Where the answer changes. If the line in Panel B sat at nominal rather than at the bottom of the band, the coil would see about 85.9 percent and the panel would work, intermittently, in a way that tracks the building's load through the day and looks like a haunted panel rather than a sizing error. And if Panel B's complaint were a contactor that drops out during a momentary dip rather than one that will not pick up, the binding constraint is dropout voltage, which sits well below pickup, and a larger transformer fixes nothing because the transformer was never the limiting element in that story.

Fixing the panel that failed

Size the transformer on the pickup instant. This is the direct fix and the only one that needs no change to the control scheme. Select against inrush VA at the retention the coil's pickup voltage demands, at the bottom of the line band, and confirm the continuous rating still covers the sealed total.

Change the coil. Electronically operated or DC-coil contactors present a far smaller and shorter inrush burden than a comparable AC magnet coil. The actual figures come from the device data, and the substitution has to be checked against the control scheme, because a DC coil's dropout behaviour and its response to a control-circuit fault are not the same.

Sequence the pickups. Where two large burdens land at once, an interposing relay can separate them. Treat this as a control design change and not a wiring trick: shifting when a device energizes can change what an interlock is protecting against, and an interlock defeated by a sequencing change is a safety consequence introduced to fix a voltage problem.

Move the primary tap. Where the transformer has multiple primary taps and the line genuinely runs low, the tap table on the transformer is the authority on what is available. This is a de-energized change on the primary side of the transformer, so isolate under 29 CFR 1910.333(b)(2) and prove dead before touching a tap lead.

Why bigger is not free

The obvious response is to fit the largest transformer that will bolt in. That carries a real consequence: a larger transformer has lower impedance, so more current is available on its secondary under a fault. The control components, the terminal blocks and the secondary protection all have short-circuit ratings, and a secondary fault that was previously current-limited by a small transformer's impedance may no longer be. Secondary protection requirements for transformers live in NEC Article 450 and, for motor control circuits specifically, in Article 430 Part VI, both in the edition your authority having jurisdiction has adopted. Size to the pickup instant with margin, not to the largest device on the shelf.

Working on this safely

Reading the secondary during a pickup is energized work, and the fact that the secondary is low voltage does not make the cabinet a low-hazard place: the primary side of that transformer is in the same enclosure, and the arc flash hazard belongs to the enclosure, not to the conductor you happen to be probing. 29 CFR 1910.333(a)(1) permits energized work only where the employer can demonstrate that de-energizing introduces additional or increased hazards or is infeasible due to equipment design or operational limitations, and capturing a pickup transient qualifies where it cannot be captured any other way. Boundaries and PPE come from the risk assessments at NFPA 70E-2021, 130.5 and 130.7, in the edition your employer's electrical safety program has adopted, and the instrument is proved live-dead-live per 120.5. For any work inside the panel with it isolated, follow 29 CFR 1910.333(b)(2); 29 CFR 1910.147 excludes exposure to electrical hazards from work on conductors and equipment in electric utilization installations at (a)(1)(ii)(C), and 29 CFR 1926.417 is the construction counterpart.

One thing not to do while diagnosing a coil that will not pull in: do not push the armature closed by hand. That energizes the load circuit with your fingers in the moving parts of a device designed to close hard, and it bypasses every interlock the control scheme relies on to decide that closing is currently safe.

References

• 29 CFR 1910.333(a)(1) and (b)(2); 29 CFR 1910.147(a)(1)(ii)(C) exclusion; 29 CFR 1926.417 construction counterpart
• NFPA 70E-2021, 130.5, 130.7 and 120.5, in the edition adopted by your employer's electrical safety program
• NEC Articles 430 Part VI and 450, in the edition adopted by your authority having jurisdiction
• ANSI C84.1 voltage ranges, in the edition referenced by your utility's service rules, which bind through the utility's tariff
• NEMA ICS 2 for control device ratings, binding through the manufacturer's declaration and your purchase specification; device data sheets for pickup voltage, inrush VA and sealed VA
• See related: What a Transformer Actually Does; How Control Voltage Differs From Line Voltage; Why Control Power Fails Differently Than Load Power