RPM Measurement Using Reluctance Sensor

[Airian_007]

Hi,
I need some help regarding signal conditioning of reluctance or inductive sensor which will be used to measure the RPM of a rotating shaft.

I tested the circuit using oscilloscope and a magnet. Whenever I move the magnet towrads or away from sensor I can see a an AC signal of few mV to almost a volt depending upon the distance.
Now I would like to condition this signal into pulses. My idea is to use differential op amp directly to the sensor (I don't know if I need Load resistor or not because I can see signal on scope without load resistor) then amplify the signal and use zero crossing to convert the ac signal into pluses.
Kindly guide me with the good solution or any other ideas.
Thanks.

 

[Visitor]

Do you have a data sheet for the sensor?

 

[Nigel Goodwin]

The classic methods are over-driven amplifiers, or schmitt triggers - there's not usually any need for zero-crossing, just convert the sine wave to a square wave.

 

[Airian_007]

Do you have a data sheet for the sensor?

No I don't have the datasheet but I opened it and its a thin copper wire maybe hundreds of turns

 

[Airian_007]

The classic methods are over-driven amplifiers, or schmitt triggers - there's not usually any need for zero-crossing, just convert the sine wave to a square wave.

I thouth zero crossing is more of a use incade signal is not pure sine wave or not symmetric across zero line. Zero crossing will ensure a pulse whatever the shape of the signal is. What is your thoughtsm

 

[Visitor]

One consideration with these type of sensors – the output voltage is dependent on the rotational speed (and possibly the gear tooth width). Without a load resistor, the output pulses can be pretty high.

 

[Les Jones]

I would suggest a comparator IC such as an LM393 with a small amount of positive feedback to make it behave like a schmitt trigger. This is an example of one used in a tachometer I designed a few years ago. You would not need R15. It might be better to bias the bottom end of your variable reluctance sensor a few hundred mV above ground. R14 sets the trigger voltage. With a variable reluctance sensor you would not need it. Just connect the inverting input to the bottom end of your sensor.

Les.

 

[Nigel Goodwin]

I thouth zero crossing is more of a use incade signal is not pure sine wave or not symmetric across zero line. Zero crossing will ensure a pulse whatever the shape of the signal is. What is your thoughtsm

I would expect the sensor to produce a pretty decent sine wave, it's essentially an alternator. Assuming you have a model number?, google for the datasheet - it may well give examples.

Certainly with such a sensor using zero-crossing detection would never have crossed my mind, and I wouldn't consider it even now you've mentioned it.

 

[Visitor]

Here is a good reference source on variable reluctance gear tooth sensors.

This graph illustrates the point I made earlier. The output voltage of a variable reluctance sensor depends on the gap between sensor and gear tooth, the linear speed of the gear tooth and the shape of the tooth. The graph shows the output for this sensor can reach 10 volts or more. I used these sensors back in the dark ages a few times, and precautions had to be taken to avoid destroying signal conditioning.

This type of sensor has been largely replaced with hall-effect sensors because of the challenges of using them.

 

[Visitor]

Detailed application note: Honeywell Application Note: VRS Industrial Magnetic Speed Sensors.

This application note gives detailed information on calculating output voltage in different situations.

The application note discusses load resistance, which is a factor in limiting output voltage.

Of particular note in this application note: "The output of the 3030 High Output device is 190 Vp-p at 1000 IPS."

 

[alec_t]

Have you considered using an LM2907 or similar IC?

 

[Visitor]

The TI LM1815 Adaptive Variable Reluctance Sensor Amplifier is designed for this application and can handle mV to 120Vp-p.

 

[Airian_007]

Have you considered using an LM2907 or similar IC?

I will go through the datasheet havn't got a chance yet.

 

[Airian_007]

I would expect the sensor to produce a pretty decent sine wave, it's essentially an alternator. Assuming you have a model number?, google for the datasheet - it may well give examples.

Certainly with such a sensor using zero-crossing detection would never have crossed my mind, and I wouldn't consider it even now you've mentioned it.

I will try to analyse the waveform in detail. Yes you are right it should be a sine wave as long as the rotating part is purely balanced or not out of centre. In my application there are chances that the shaft might be out of centre in that case sine wave could be distorted. I will analyse the waveform and will update. Thanks for your suggestions...

 

[Airian_007]

The TI LM1815 Adaptive Variable Reluctance Sensor Amplifier is designed for this application and can handle mV to 120Vp-p.

Its a good option...

 

[picbits]

I've used the LM1815 a fair bit - never had a problem.