Electric motors used in model railway locomotives today are typically a sealed, can configuration. By comparison with earlier generations of open-frame motors and wound field motors, they are small, efficient, smooth running and maintenance free.

Figure 1 illustrates such a motor. It includes a high-strength magnet that requires a low armature current to produce a given torque. The armature has a large number of poles giving a smoother torque, particularly at low speed, and self-aligning oil impregnated bearings give low friction at less cost than if ball bearings were fitted.


Figure 1. Typical modern motor with iron armature core.

A development of this is the coreless or, more correctly, the ironless armature motor which is now very commonly available. Figure 2 is a sectional view of a typical coreless motor and Figure 3 compares its construction with that of a conventional motor in which the armature winding is mounted on an iron core. The armature rotates inside the magnets. The coreless armature winding is produced on a former before being encapsulated in epoxy resin. The resin is then cured to solidify it after which the former is removed, leaving the cup shape shown in Figure 3. The stationary magnet is cylindrical and is located inside the armature.


Figure 2. Typical coreless motor.


Figure 3. Comparison of iron cored and coreless armatures.

Coreless motors have the highest efficiency and therefore the smallest size of any of the types used by modellers but they have the disadvantage of being more liable to be damaged by overloading than other types.

When selecting a motor it is useful to have some knowledge of motor theory as statements are often made about amps, volts and magnets which are not necessarily wrong but are often over simplified and can lead to wrong conclusions. These remarks apply equally to cored and coreless motors.

The fundamental relationships which apply to all direct current motors, large and small, are:

Torque is proportional to:

Armature current x (magnetic field strength x number of armature conductors).

An important point to note is that voltage does not feature in this relationship.

No load speed is proportional to:

Voltage ÷ (magnetic field strength x number of armature conductors).

This means that for any given motor, the speed is simply proportional to the applied voltage.

There are constants in these relationships which depend on the type of armature winding but these are not relevant to these notes.

Since the magnetic field strength and the number of armature conductors are fixed by the design of the motor, for any given motor the relationships can be simplified to:

• Torque (and hence tractive effort) is proportional to the current drawn,
• No load speed is proportional to voltage applied.

There are several factors such as internal friction and magnetic losses mean that these statements are not strictly correct, but for the purposes of motor selection they are sufficiently accurate.

The torque produced by a motor can be increased by increasing the current, which in turn means increasing the applied voltage. Doing that increases the resistance loss of the motor. This is sometimes called the I²R loss, where R is the internal electrical resistance of the motor windings, and reflects the fact that the resistance loss increases as the square of the current I. This loss creates heat, and if current flow is excessive the heat is sufficient to burn out the motor.

Modern controllers are commonly fitted with current limiting devices designed to prevent burn out but this should not be assumed. A careful study of the controller and motor specifications is advisable to understand the limiting current of the controller and the maximum current allowable for the motor. Coreless motors are particularly sensitive to overheating and if the manufacturer or supplier recommends a fuse or other means of overload protection, it should be used.

Based on an understanding of the motor characteristics above, an estimate of the speed/torque curve can be obtained by knowing:

• The stalled torque with respect to current
• No-load speed at full voltage.

It is not always easy to find this information. Motor suppliers often do not have it available, but if you know the manufacturer and model number of the motor, that information can often be found on the manufacturer’s web site. Some motors and motor/gearboxes were tested by the Technical Committee and the results are available in the relevant data sheets. The tests were sufficiently accurate and consistent for their purpose, and they were not intended to be regarded as precision laboratory checks on manufacturer’s data and should not be used as such.

Given the stalled torque at zero speed and the no-load speed at zero torque, these two points can be joined by a straight line on a graph to give an adequate representation of the speed-torque characteristic of the motor. Typical results for two motors are shown in Figure 4, and other curves for specific products are given in the data sheets.


Figure 4. Typical motor speed/torque characteristics.

The motors tested had a range of no-load speeds from over 8000 rpm down to 2600 rpm and torques from 35 to 700 gram centimetres, but before this data can be put to practical use it must be converted to track speed and tractive effort. This conversion is the subject of gearing and locomotive performance.

It is customary to refer to motors as ‘12 volt’ or some other value. This is misleading because it means only that the motor will give the stated speed and torque at that voltage and does not necessarily mean that it cannot be used on other voltages.

The applied voltage determines the speed at which a motor will run or, if it is stalled, the current which will flow through it. Thus increasing the voltage will either increase the speed or start it if it is stalled. Hence regulation of the voltage at the motor terminals is the basis of all speed control.

It is permissible to increase the performance of a motor by increasing the voltage provided that both the following conditions are complied with:

1. The current taken is not increased to a value which will cause overheating. This is only likely to occur if the motor is loaded sufficiently that it runs at a very slow speed or stalls. It is stressed that motors can be burnt out at any voltage if they are loaded so much that the designed current is exceeded except for short periods when accelerating or climbing gradients.
2. The no-load speed is not increased to a value which will cause damage.

The voltage of a motor is the value at its terminals. In the case of locomotive motors there is a voltage drop between the wheel and the rail and at other contacts. There may also be a significant voltage drop in the feeder system which could be as high as 3 volts on the long feeders of an outdoor line. Thus the power unit should supply at least 1 and preferably 3 volts above the nominal motor voltage and this value should be maintained up to the full load current of the unit. This is not always the case as the voltage output of some power units falls with load to an extent which significantly affects motor performance.

Due to better magnetic circuit design minimising the variation in torque during each revolution, the ability of small motors to run smoothly at slow speed is now greatly improved compared with earlier types. In older motors the variation was so great that stalling occurred at certain armature positions, a considerable increase in voltage being necessary to re-start rotation. This is termed ‘cogging’ or ‘magnetic locking’.

A flywheel increases the inertia of the drive system and thus minimises the effect of torque variations. It is usually most conveniently mounted on an unused shaft extension of a double ended motor. The wheel should have the largest diameter which can be accommodated in the body and have its weight concentrated in the rim. Accurate balancing is essential to prevent vibration.

With older types of motor the benefit of fitting a flywheel can be significant, but with modern motors used in conjunction with a good supply system the improvement resulting from fitting a practical size of flywheel is likely to be only marginal. The only case for doing so is where the electrical contact between wheel and rail is intermittent due to dirt or some other cause, where a flywheel may keep the train running until the contact is restored.

This article was compiled by the Technical Committee for the Gauge O Guild Manual. It was adapted for the GOGWiki by Nick Baines.