DC PROX Sens

Status
Not open for further replies.
Location
NE (9.06 miles @5.9 Degrees from Winged Horses)
Occupation
EC - retired
About 10 years ago I installed a DC prox sensor on the bottom of a bucket elevator or Grain Leg. It counts the revolutions of the bottom shaft. Less than 5 counts in 3.57 seconds and the control shuts down feeding augers indicating the leg belt is slipping or the rpm is low for whatever reason. It has worked well.

Problem today is while the indicator light on the sensor itself has a pretty steady pulse the actual output to PICO controller is a repition of pulse..pulse...(skip)...pulse..pulse...(skip) which of course causes the count to be off within a few seconds. Seems like an odd way to fail. Anyone with a thought?
 

jim dungar

Moderator
Staff member
Location
Wisconsin
Occupation
PE (Retired) - Power Systems
About 10 years ago I installed a DC prox sensor on the bottom of a bucket elevator or Grain Leg. It counts the revolutions of the bottom shaft. Less than 5 counts in 3.57 seconds and the control shuts down feeding augers indicating the leg belt is slipping or the rpm is low for whatever reason. It has worked well.

Problem today is while the indicator light on the sensor itself has a pretty steady pulse the actual output to PICO controller is a repition of pulse..pulse...(skip)...pulse..pulse...(skip) which of course causes the count to be off within a few seconds. Seems like an odd way to fail. Anyone with a thought?
What are you actually sensing? Maybe a bolt fell off?
 

Smart $

Esteemed Member
Location
Ohio
... Anyone with a thought?
Yeh... replace it with a new one and observe what happens :cool::grin::roll:

Are you implying that you've rulled out the obvious , and the sensor-to-nut distance has not changed... or someone hasn't replaced the sensor with a different, less sensitive, look-alike model?

I had involvement with a cable cutting saw back in the late 70's, early 80's that used a prox sensor to trigger the saw stroke. Controller board was the usual culprit when things went awry.
 

gar

Senior Member
Location
Ann Arbor, Michigan
Occupation
EE
110214-1758 EST

The sensitivity may have changed over time, likely has. Install a new prox.

Measure the gap from the current unit to the nut.

To test the current prox use a flat steel plate and on perpendicular approach to the plate determine the gap for the change of state. How does this gap compare to the prox to nut gap. My guess is that the setting of the prox to nut should be as small as seems reasonable, and probably less than half of the minimum gap in the test setup. However, the height of the nut, if too short, might require the sensor to be further away.

A Honeywell LCZ260 is an interesting sensor. It is Hall device based and is DC sensitive meaning it can work to zero velocity. It only responses to magnetic material, and thus, not aluminum.

It should work from 4.5 to 26 V DC. The output is an NPN transistor to common. It has an internal 5 K pull-up.

At 5 V excitation I measured the pull-in point at about 0.25" and dropout about0.35". I would run this at a clearance of about 0.063" to 0.10" with a nut height of 0.625" at least.

Operating frequency is 0 to 15 kHz. Good for RPM measurement at 60 teeth on the gear. At 3600 RPM the 60 tooth gear produces 3600 Hz and that is well in the frequency range of the sensor.

The end of the nut or screw head or whatever blob is used to actuate the sensor should not have a hole in it.

.
 

Jraef

Moderator, OTD
Staff member
Location
San Francisco Bay Area, CA, USA
Occupation
Electrical Engineer
If there is any chance moisture is present, most likely it has made its way inside the prox. Even the best prox units that are IP69K will eventually succumb to moisture ingression, it's a fact of life. Almost all plastics that can be used to seal things up are very slightly hygroscopic, meaning they absorb moisture even if only on a microscopic level. But time and temperature changes take their toll, 10 years is pretty good really
 
Status
Not open for further replies.
Top