Inductive Proximity Sensor vs. Hall Effect

Depends on the spec of the device you're using. I looked at two data sheets picked at random, & the respones rates were 5kHz and 350Hz. What does your data sheet say?

I am assuming you are using a monopole to generate the signal, and a frequency reader to read the signal and display the shaft speed. I have observed that in some cases, despite the reader specs state that it can read a certain range of frequencies and input signal tension, both are not true, i.e., higher frequencies require higher signals to allow the reader to recognize it and read. Try to vary the distance between sensor and wheel, and see if you can read speed a little further. If so, this may be the problem. If not, look for those parameters in the equipment documentation and check if everything should work - or not. This is a usually trouble-free device, and work without problems. Should not be really hard to fix, just take a close look... Your target dimensions and frequency sound good to me.

I am using an Altech AIS12F02AN024-2M (12mm unit) rated at 1000Hz. It is connected to a PLC. I input a square wave up to 1500 hz into the circuit and the PLC seems to handle it without problem. Tried adjusting the air gap up to the point where it won't turn on, still seems to run into the same problem.

Any chance you can get an oscilloscope on the signal line, to see how it's failing? (i.e. output always ON, always OFF, or dithering about somewhere in the middle).

You may be able to get a faster response by increasing the electrical load. If I'm looking at the right datasheet, looks to be rated up to 120mA. Chances are your PLC is only drawing about 1/10^th of that (about 24V into 2k = 12mA). What's the PLC input spec?

According to the datasheet, max current = 250 mA. I had a 1.8K pull up resistor in the circuit. Running 13.8V. I will try a 100 ohm pull up to increase the load. Thanks for the suggestion.

I used to use a General Radio photo pickoff for this application. A piece of reflective tape was placed on the shaft. A light shined on the tape to generate a pulse, which was fed to a counter (or to a strobe light for motion studies).

Today you can get a Laser version of that idea. For example:

http://www.usatoolwarehouse.com/usatoolwarehouse/ESI-333.html

I faced similar problem when I used a cheap inductive proximity sensor. In my case the sensor started to count the sensing bolt arbitrarily giving error readings. From internet I came to know that hall effect proximity switches are most suitable for sensing RPM. I am procuring one.

Yes, I gave up on the inductive prox and replaced with a hall effect gear tooth sensor. All is good now.