Convert an AC mechanism to DC

Matching the rated watts is the wrong method, and it is the usual reason this conversion fails twice.

The number on an AC gearmotor is input power at the line. Very little of it reaches the shaft, and none of it describes the four behaviors the mechanism was actually relying on.

Why the watts figure misleads

It is an input figure, from a construction whose efficiency is low by design.

A small AC synchronous or shaded-pole gearmotor is built for cost, robustness and line operation, not for efficiency. A meaningful fraction of the input is spent in the stator and rotor before anything turns, so a nameplate figure of a few watts can correspond to a fraction of a watt at the output shaft. Sizing a DC motor to the nameplate therefore produces a motor several times larger than the mechanism needs, which then draws several times the current the battery budget assumed.

The failure runs the other way too. Sizing to the measured running torque alone produces a motor that cannot break the mechanism away from rest on a cold morning at the low end of the supply, because the AC motor was quietly providing a starting behavior nobody measured.

n_sync = 120 f / p

n_sync
synchronous speed, rpm
f
line frequency, Hz
p
number of poles
The reason the mechanism kept time. Speed was locked to the line, independent of load, until the motor pulled out entirely.

What the AC motor was really giving you

Five behaviors, none of them on the nameplate, all of them now your problem.

Timing
A synchronous motor runs at a speed set by the line frequency and the pole count. Load does not change it. A mechanism that indexed, dosed, timed a cycle or advanced a mechanism per revolution inherited grid-grade timing for free. A DC motor's speed falls with load and rises with voltage, so if timing was a real requirement it now needs closed-loop control, or a mechanical arrangement that does not depend on speed.
Holding
A synchronous motor develops holding torque at zero speed while energized, so a mechanism could stop and stay put with power applied. A DC motor at zero speed produces torque only by drawing stall current, which is a thermal event, not a holding strategy. If the mechanism must hold position, it needs a self-locking or high-ratio train, a mechanical detent, or a brake, and that decision belongs at the start of the conversion.
Direction behavior
A permanent-magnet synchronous motor of this class starts in an unpredictable direction unless a mechanical stop, ratchet or one-way arrangement resolves it. Some mechanisms are designed around that behavior, and others are designed around suppressing it. Either way the DC replacement will start in a defined direction every time, which changes the mechanism's behavior in ways that are easy to miss until the first prototype.
Inherent speed regulation, and an accidental torque limit
There was no controller, no ramp and no overshoot. Torque rose to meet the load until the motor pulled out of synchronism and stalled, which acted as a crude torque limiter that protected the mechanism. A DC motor meeting the same obstruction keeps pulling stall current and heats, so anything the AC motor was protecting now needs a current limit, a slip clutch, or a stop sensor.
Thermal sizing by continuous duty
AC gearmotors of this class are commonly rated and sized for continuous running. If your mechanism actually runs for four seconds a minute, the DC replacement can be far smaller than a like-for-like substitution suggests. This is the one item on the list that usually works in your favor.

The conversion worksheet

Twelve items. The first four are measurements, the middle five are decisions, and the last three are the calculation.

  1. Measure the output torque in the mechanism

    At the output shaft, in the assembly, at the worst point of the cycle. Not from the nameplate and not from the AC motor's rated torque. A torque wrench on the output, or a lever and a spring scale, is enough to get within the accuracy the conversion needs.

  2. Measure the breakaway torque from rest

    Cold, at the worst position, at the worst end of the tolerance stack. This commonly exceeds the running torque several times and it is what the DC design must start against at the low end of the supply.

  3. Measure the output speed you actually need

    Not the synchronous speed the AC motor happened to give. The mechanism has a required speed or a required cycle time, and those are frequently slower than what was installed.

  4. Record the duty cycle in real units

    On time, off time, starts per hour, reversals per hour, and whether it ever drives into a hard stop. This is what makes a small DC motor viable where a continuously rated AC motor was fitted.

  5. Establish the DC supply as it behaves under load

    Nominal voltage, tolerance, whether it is a battery, how far it sags during breakaway, and what current the supply can actually deliver during the start. Design against the worst case, not the label.

  6. Decide what the timing requirement really is

    Within a second, within a percent, or not at all. If speed must hold under a varying load, plan for feedback or a regulated drive now rather than discovering it during validation.

  7. Decide the holding requirement

    Must it hold with power on, with power off, or not at all. Each answer leads somewhere different: a high-ratio or self-locking train, a brake, a detent, or nothing.

  8. Decide what happens at the stops

    If the mechanism drives into a hard stop, choose the protection deliberately: a current limit in the drive, a slip clutch, a position sensor, or a torque-limited train. The AC motor was doing this by accident.

  9. Set the acoustic requirement with a method

    An AC synchronous gearmotor hums at line frequency and its harmonics. A DC gearmotor's noise has a different character even at a lower level, and a measurably quieter unit can still generate complaints. Measure the existing one first and record the spectrum, not just the level.

  10. Fix the envelope and the mounting interface

    AC gearmotor housings are usually larger than an equivalent DC gearmotor, so there is often envelope to spare. That spare length is what buys a lower-loss gear train or a feedback device.

  11. Convert to power, in the right direction

    Take the measured output torque and speed to mechanical power, then divide by the gearhead efficiency and the motor efficiency at that operating point to get electrical input. Working from an input power figure back toward torque is where the original error was made.

  12. Check the thermal case against the real duty

    Root mean square torque over the whole cycle, including the idle segments, against the continuous rating at the enclosure's ambient temperature. Peak against the peak rating separately.

P_out = T × 2πn / 60, P_in = P_out / (η_gear × η_motor)

P_out
mechanical power at the output shaft, W
T
measured output torque, N·m
n
required output speed, rpm
η_gear
gearhead efficiency at that point, as a fraction
η_motor
motor efficiency at that operating point, as a fraction
Both efficiencies divide. Neither is the peak figure from a datasheet; both are the values at the point the mechanism actually runs.

Send the AC part and the measurements

A part number, a photograph of the nameplate, the measured output torque and speed, and the duty. That is a cross-reference we can screen, and the worksheet above is what makes it a real one.

Send it for screening