Encoder/Proximity Switch

Roto pulser--would likely be the case for the prox switch he is asking about.--I believe that a true encoder will know it's physical position at all times were a roto pulser whether optical or inductive can get lost if the power is cycled off. We use a lot of roto pulsers and they do just fine, but there are plenty of conditions that would require a true encoder.

We buy hyd. motors with a built in speed ring. It's essentially a picket fence of high lands and low lands. We've two types of sensors available to us: a single gate that only opens and closes as the lands go by. The other sensor is pair of gates in a similar housing. If installed properly you can detect direction of movement (i.e. if A leads B = CCW) These inputs needs to be fed into a controller for more inteligence. So it is one or two proxis being used as an encoder if so desired.

This is a valid question.

There are three major possibilities here, as follows:

1) Proximity sensor - mounted to detect the passing of one or more projections on the motor shaft. It is usually used to sense the projections on a toothed wheel mounted on the shaft. By counting the number of pulses in a given time (i.e. the frequency), you can read the shaft speed. With one 'tooth', 1 pulse per second (1 Hz) ≡ 60 rpm. With just one sensor, there's no information about rotation direction.

2) Incremental encoder. A disk is mounted on the shaft, with two sensors such that two trains of pulses are produced, out of phase with each other. The pulse rate can be counted (as above) to give the rpm, but by checking the pulse sequence (which one of the two pulse trains is 'leading') you can also determine the direction of rotation. By counting the pulses with the right electronics, you can keep track of the total number of revs (cw or ccw from some starting 'zero' position). If the electronics loses power, you've lost track of the position.

3) Absolute encoder. This is like having a disk actuating a load of switches, so instead of just knowing that the shaft has rotated so many degrees, you know the actual shaft angle. By adding a turns counter, you can determine the number of rotations + the angle through the current rotation from your 'zero' position. This can be maintained during loss of power. Clearly, this can also give you the rpm as above.

There are lots of variations on this theme; you can have a package which uses option (1) with some electronics to produce, say, a 0 - 10V (or 4 - 20mA) signal corresponding to 0 to 1000 rpm (or whatever scaling you need). To hazard a guess, this is probably what your client is looking for.

A tachometer is also a possibility.

Basically a permanent magnet electric motor being driven by the shaft instead of powered. It gives a voltage output proportional to speed.

I think this was hinted at in the pervious post that outlined the other three options.