1
\$\begingroup\$

I am working on a PLC based system that uses High Speed counter cards to pick up a signal from a 24 VDC proximity switch. I am having a problem with short duration 10 to 15 volt spikes being seen on the counter card as a false proximity switch hit. I can see these spikes on an oscilloscope.

I rewired as many of these signals in shielded cable as possible but these errant spikes persist.

The proximity switches I am using are PNP and rated to pass 200 ma. I am considering using a bleed resistor from the proximity switch signal to common to minimize these false spikes.

Any recommendations on how many milliamps/ resistor value I should be using? I have tried a 20K resistor to bleed about 1 ma to ground when the signal is in the low state but I don't know if I should be trying to bleed more current to common?

\$\endgroup\$
4
  • \$\begingroup\$ Why not use a noise discriminator circuit? \$\endgroup\$
    – Ignacio Vazquez-Abrams
    Commented Oct 3, 2013 at 23:28
  • \$\begingroup\$ Can you point me to a link to such a circuit? I am not familiar with this concept. \$\endgroup\$
    – Tinkerer
    Commented Oct 3, 2013 at 23:35
  • \$\begingroup\$ I happened to have the same problem. What is the frequency of the signal you want to count? \$\endgroup\$
    – johnfound
    Commented Oct 4, 2013 at 8:49
  • \$\begingroup\$ The frequency of my signal is only like 10 HZ. What is critical to me is picking up the rising edge quickly and accurately. \$\endgroup\$
    – Tinkerer
    Commented Oct 4, 2013 at 21:13

3 Answers 3

Reset to default
0
\$\begingroup\$

If the transistor is rated at 200mA try a lower value resistor first like 1k ohm. Then try adding a 100nF capacitor across it. Put these components close to the input for best effect. Come back and report what you find.

NB if a 1K resistor is used the power rating will probably have to be 1W. You could try a 2k2 - this will be OK at a 0.6W power rating.

\$\endgroup\$
0
0
\$\begingroup\$

I added a 10K resistor to common on the Counter card. This reduced the occurrence of my spikes but has not eliminated them. The period that these spikes during which the amplitude is above 5v is less than 20 Micro seconds. My system appears to be functioning fine. I am guessing that the counter card has some impedance which may be acting to filter out the spikes. If I run into more issues I will try a lower resistive value and a 100nF capacitor.

\$\endgroup\$
0
\$\begingroup\$

I had a similar problem once - the burst of pulses seemed to correlate with operation of an associated servo drive. In my case, replacing the prox with a different brand fixed it.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.