Here is a short description of the operation of the encoder in the highest resolution setting.

The capacitive sensor consists of

1. Transmitter board with an array of transmitter electrodes in 8 groups

2. Rotor with a modulation pattern of 8 periods (wavelengths)

3. Receiver board

4. Electronics on the transmitter board

The electronics are using a 5 MHz signal that is LF modulated with a 10 kHz signal for sensing the position of the rotor by measuring the phase between the receiver LF modulation and the transmitter LF modulation. This sensing is done 10,000 times per second and gives the absolute position within one period of the rotor modulation (45 deg).  At each of those measurements there is a new determination of the absolute position within that interval.

The absolute position within a wavelength is at each sampling occasion stored in a position register A.

A 2nd order feedback loop updates a second position register B with incremental pulses so it coincides with position register A at the sampling occasions.

The Quadrature output pulses A and B are derived from the same pulses that feed position register B, and there is no way that there would be any pulses missing without being corrected by the feedback between position register B and register A.

If the quadrature signals A and B are decoded in a quadrature decoding circuit such as LS7183 and LS7184 (from LSI Computer Systems, Inc), any interference pulses will be interpreted as a step up followed by a step down and therefore not cause any larger temporary error larger than one increment.
 

The position output at constant speed is without delay, but at acceleration, there will be a lag equal to the time constant multiplied by the acceleration. The following table references the accuracy and time constants for each resolution setting-

There is no reason for our capacitive encoder to lose counts at low speed. The ASIC in the encoder contains a position follower, that is updated for position and speed of movement each T  (T= 100, 50, 25, or 13 µs dependent on resolution setting). The position follower update is done against the capacitive sensor primary position measurement that is absolute within 45 degrees. Therefore, if the quadrature outputs A and B signal are used with a quadrature up-down counters, there is no possibility to loose pulses, as long as the quadrature counter does not lack in response speed.

Warning:  The rumor that capacitive encoders can be missing pulses at low speed does not apply to our encoder.  However, if proper quadrature decoding is not utilized, it is possible that extra increments will be counted.  Here are some examples to watch out for-

·Using only one channel for speed and position measurement. In such case there is no discrimination between up count and down count and any jitter or interference pulses will be added to the count.

·If a microprocessors regular data ports are used for sampling the signals at regular time intervals, it is possible that the microprocessor will miss some pulses or short spikes that occur between such sampling occasions which could cause cause temporary misreading of direction and thereby cause errors in the accumulation of position increments.

Axial

 1. The shaft adaptor can slide in the hub up to + /- 1.5 mm. This slide action presently has a slight friction but it will adjust for this amount of misalignment during assembly.

2. The hub is mechanically limited to move +/- 0.15 mm from center position. If the motor shaft is moving more than that amount the hub will hit its mechanical limitation and further movement will be taken up by the sliding between hub and shaft adaptor.

When mounting the shaft adaptor and sleeve on the shaft according to instruction, there is no axial force on the motor shaft and the shaft adaptor will be placed in the right position based on the tool geometry.

If there are operations on the motor shaft load end after assembly of the encoder that causes the shaft to move axially, this will be taken up by the sliding between shaft adaptor and hub. When the shaft returns to it's nominal position, the hub will be sliding on it's lower limit, which may cause some slight acoustical noise, that disappears after some time. The encoder function will not be influenced by this.

Radial and Eccentricity

The fully assembled (top cover and base) weight for the AMT is as follows:

AMT102= 20.5 g

AMT103= 14.0 g

When the encoder is mounted on a motor with a strong stray magnetic field flux, it is possible for the index pulse to become compromised.  The encoder is based on a capacitive sensor that has 8 cycles per turn.  Each of those cycles gives an index pulse in the interpolating electronics (the fine index pulses).

In order to get just one index pulse, there is a coarse index that is created by a Hall sensor on the electronics board and a small magnet on the rotor.  The Coarse index is gating out just one of the 8 Fine index pulses.

Standard Index Pulse:
 
If there is a strong magnetic stray-field (> 40 Gauss) on the motor that the encoder is mounted on (typically found in stepper motors), it may influence the Hall sensor.  This electromagnetic interference affects the coarse index pulse to the point the index pulse is lost or repeated more than one pulse per revolution.  The outcome is dependent on the polarity of the stray field and its vector alignment with the Hall sensor on the transmitter board.

Magnetic Interference on Index Pulse