LVDT excitation and response

Hello all.

I am building a design using an LVDT to measure distance.
I am using the CD-375-250
http://www.macrosensors.com/download...ies-061311.pdf


I am currently exciting the LVDT by taking a crystal and using a clock divider to bring the frequency to roughly 5 kHz square wave.
I then invert this signal using a standard 7404 series inverter.
I will inject the initial signal into the top of the primary of the LVDT, and the inverted initial signal into the bottom of the primary of the LVDT.

The response from both secondaries is a nice sinusoidal. I then take a peak detect of this sinusoid as changes in amplitude.

However, the response from my secondaries is not a linear amplitude change, and the overall change in amplitude is very small. I also notice that the response from one secondary will phase shift left (as viewed on a scope) a little bit before dropping amplitude, and the other secondary will phase shift to the right a bit before dropping in amplitude.
I have tested this LVDTwithout any extra circuitry and it functions as one would expect and LVDT to function.

Does anybody have any ideas as to why my response from my LVDT when connected to the rest of my system is not the same as it is when it is disconnected from the system?

Thanks for your time.

Changing the signal amplitude suggests that you are connecting an inappropriate circuit.

LVDT involves differential evaluation of the output signals. This works much better by forming a differential signal, as suggested in the datasheet and connecting a phase sensitive detector circuit.

Did you consider that a square wave excitation is not a recommended waveform, it creates high null voltages. As in the previous post it is important to understand that a LVDT is a differential output. Here is some info from TI that I have found to be helpful. http://www.ti.com/lit/an/spra946/spra946.pdf Harold Schaevitz Alliance Sensors Group www.alliancesensors.com