Measuring Accurate Relay Pull-in and Drop-out Voltage
Pull-in and drop-out voltage measurements of electromechanical
relays may be taken manually or by automatic test equipment. Manual testing
requires the adjustment of a suitable variable power supply until the
relay contacts are observed to transfer. The rate of voltage application
is not precisely defined or controlled, and may vary from one operator
to the next or even from one test to the next by the same person. However
the voltage is adjusted, though, the operator soon learns where the device
is likely to operate. As that point is approached, the operator will slow
the adjustment to a rate of perhaps a fraction of a volt per second. This
rate of application very nearly approximates the steady state value at
which the relay operates or releases and gives the best measurement result.
Utilizing automatic test equipment to do the same test requires some considerations
if both the speed and accuracy offered by ATE is to be realized. A simple
linear coil voltage ramp profile will introduce significant errors caused
by overshoot due to the finite response time of the relay armature.
Pull-in voltage measured using a fixed ramp rate will always overshoot
while drop-out voltage will always undershoot. The magnitude of error
is expressed simply by:
Verror = RT
where R is ramp speed in volts/second and T is the operate or release
time of the relay at the steady state value. The response time of a relay
is much different at the actual pull-in or drop-out voltage than it is
under normal switched conditions. Refer to the voltage profiles below.
1 shows the resulting error in pull-in voltage measurement caused by the
long operate time of the relay. At a coil voltage of V1, the relay's operate
time is b-a; a being the point at which the armature begins to move and
b the point at which the equipment senses the contact closure. During
this time the coil voltage overshoots by V2-V1.
2 shows the same process for drop-out voltage. At V1 the relay begins
to release, but because of sluggish armature movement it is not detected
until b-a seconds later. The result is in error by V2-V1 volts.
The reason for slow relay response at the steady state operate point is
due to a weak effective spring force on the armature's mass. At voltage
V1 the spring's tension is exactly equal to that imposed by the magnetic
field. Since the armature has mass, its acceleration is slow until the
force of the spring substantially overcomes the magnetic field force.
In the case of drop-out, the effect is dramatic, often resulting in release
times an order of magnitude longer than would be expected from abruptly
removing the coil voltage.
therefore, may be substantial when measuring drop-out voltage or current.
A better way of performing the test is shown in figure 3. In this method
a ramp profile using multiple slopes is used. To shorten the overall test
time, the coil voltage is stepped down to a "must hold" value
and ramped until a change in contact state is detected (not necessarily
the transfer point). At this point the ramp is stopped and the contacts
again checked after a delay. If the relay contacts have not transferred
the ramp is continued at a slower rate until transfer is confirmed
3 shows the measurement error minimized to an acceptable value. With this
ramp profile an excellent trade-off is obtained between test speed and