0
\$\begingroup\$

I have been struggling to solve an issue about signal conditioning an instrument with an analog circuit. First I would like to explain the instrument very briefly:

There is a windmeter which is outputting current pulses proportional to its rotation/windspeed. Here is its datasheet: http://www.synchrodan.hu/letoltes/kriwan/kulsoerzekelok/szel/N292_INT10K_69000248_e.pdf

As you can see in the second page there is a speed-freq. plot. I would like to mention that I only need up to max 1000Hz output(less than 20m/s).

Output of this sensor is current pulse.

Here is the signal output details from the data-sheet:

enter image description here

What it says is, the output frequency of the pulses varies proportional with rotation/wind speed; but the ON time of the pulse remains same around 100us+/-20us. I confirmed it by wiring a 600 Ohm shunt resistor as shown in the wring diagram(I used 24VDC to power) in the datasheet. And observed the output via an oscilloscope.

The output was a pulse train with the pulses jumping from 2.5V to 7.5V. This is because the current pulse is not zero when the pulse is in OFF time. For example at 1000Hz output the pulse train(with a 500 Ohm shunt) will look like:

enter image description here

**So far so good. Up to this was only to introduce the instrument.

**Now the issue was, since the ON time of the pulses were very short and my sampling rate cannot be more than 4000Hz; I was having troubles when sampling these. I don't want to go in details about these. Shortly ON time had to be increased, there is no other option in my case.

Anyway I implemented an analog circuitry using a comparator and a 555 timer.

The circuit increases the ON time of the pulses without changing their frequency.

The simulation works very well and I also tried with a function generator as an input. Here are the results:

Simulation results:

enter image description here

(please left-click to enlarge the picture)

And below I recorded a short video with a 0 to 5V pulse input where I change the duty cycle of the input. As you see, I change the ON time of the input pulse but the output's ON time is longer and its frequency remains the same:

https://www.youtube.com/watch?v=CkZ7nlaaDGU&feature=youtu.be

The problem

This circuit works with simulation(I mimic the current pulse in LTSpice) and the function generator(unfortunately in this case there is no DC offset in the input, I still could not learn how add DC offset if the function generator doesn't have that option).

Long time ago this also worked when the input was the windmeter but I dont remember if it was by chance. But now I'm trying with the windmeter I dont get any signal output just noise. The wind meter output is working I checked several times. Circuit is also working with the function generator. But the circuit doesn't function when I hook up the "current pulse" through a 560 Ohm resistor(so it becomes a voltage pulse input which is from around 2.5 to 7.5V). I use 12V for rails of the opamps and the windmeter uses 24V power suplly.

What could be the problem here? I have spent many hours and no luck..

\$\endgroup\$
  • \$\begingroup\$ Everything's referenced to a common ground, right? \$\endgroup\$ – ThreePhaseEel Jun 14 '16 at 23:23
  • \$\begingroup\$ yes exactly all grounds are connected. \$\endgroup\$ – user16307 Jun 14 '16 at 23:25
  • \$\begingroup\$ Can you give us a scope shot of the function generator output and of the pulses off the 560ohm resistor? \$\endgroup\$ – ThreePhaseEel Jun 14 '16 at 23:26
  • \$\begingroup\$ i added a video when the input is function generator which works fine. the problem is when the input comes from the windmeter's current pulse output where i use a 560 Ohm shunt. right now i dont have the output of the windmeter but i observed it via scope it was a 2.5 to 7.5V pulse train as specified in datasheet. the output of the circuit on the other hand was mess looks like a noise. i dont have photo either with me now. the difference is wind meter is outputting current pulse with offset comparing to function gen. but in LTspice even it is current pulse it works. i will go crazy about this. \$\endgroup\$ – user16307 Jun 14 '16 at 23:31
  • \$\begingroup\$ If you can get the output of the windmeter into a scope shot, that'd be great... \$\endgroup\$ – ThreePhaseEel Jun 15 '16 at 0:03

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.