Navigation:  Tutorials > Measurement tutorials > Frequency measurement > DEWESoft 7.0  data acquisition, processing, analyzing and storage software Tacho probe measurement

So, is it easier to measure the encoder or the tacho (one pulse per revolution)? One might state that it is easier to measure tacho signals, but that's not correct. An encoder measurement has a much larger amount of information and therefore the delays in measurements can be significantly smaller than when measuring with one pulse per revolution.

Thus, I changed the setup so that the sensor for all three is set to tacho. If we look at the measurement, we immediately see that the PAD (square green curve) and analog (smooth green curve) measurements have significant delays when compared to the counter measurements.

 

If we look in the Analysis mode, we see that the time delay is approximately half a second and that the signal is also smoothed and doesn't show the real scenario. Therefore we can conclude that for fast responses we should only use counter measurements.

 

But, as usual, there is a trick which enhances the results. For some applications with pickup probes, DAQP-FREQ-A is really useful since it has the ability to cut the DC part of the signal and has variable trigger levels.

A short look at the user's manual of DAQP-FREQ-A reveals that the pin 8 is a conditioned trig out pin. So when the trigger is found, it will produce the transition of voltage on analog. I connected this voltage on another voltage input (so that two analog channels are used for this measurement) and with that we also have the reference signal for the start of a revolution. This is very useful for order tracking or any other dynamic signal analysis.

Additionally, we can make the math function to measure the frequency. The frequency is inverse function of period. Therefore we can enter the following math function:

1/pulsewidth("AnalogCounts">2)

The function says to measure the frequency out of the pulsewidth from the "AnalogCounts" input signal. The pulsewidth function uses real digital values (either 0 or 1), therefore we need to use a logic function to make the signal digital.

 

Thus, when we look at the result, the red curve (which is the analog measured frequency) shows significant measurement delays while the stepped green curve (which is a math signal) has no delay at all. However, it has steps in between two rotations since there is no new information available between the two pulses.

The graph below shows the pulses measured with an analog signal. Those pulses can be used nicely for the advanced analysis of phase angles. The measurement of the pulses is also the most critical item: if the sampling rate is too low, we might miss the pulses and we will never get a good result. The width of the pulses depend on the width of the input pulses if taken from the front connector. We can also make a fixed routing at the back plane, in which case the signal will be enlarged based on the filter settings of the DAQP-FREQ-A. Please refer to the appropriate user's manual for more information on this subject.