FDC2214EVM: FDC2214EVM

Ju-Yong,

Thank you for your post.

It may be the drastic change in capacitance (200nF → 100pF) is causing the device to lose resonance, and the input signal subsequently disappears.

The best place to start troubleshooting is at the device inputs to be sure the waveforms are meeting the device requirements.

For the two capacitance values, please try probing the device input pins with a high impedance oscope probe. Each input pin should have a (10kHz to 10MHz) half-sinusoid with a peak amplitude between 1.2V and 1.8V. Probing the two pins differentially should show a full sinusoid with the same peak amplitudes and frequencies.

One trick that helps minimize distortion of the waveform is to place a 1k leaded resistor between the probe tip and the test point on the PCB.

Once you have snaps of the input waveforms, please post them to this thread, and if everything looks okay, we will go to the next step.

Regards,
John

Hi John,

Thank you for the prompt reply. Here are the photos for the channel input. It seemed Channel 0,1 were similar and Channel 3,4 were similar and the amplitudes are pretty smaller than expected and hard to estimate the frequency. Would these signals are okay for the measurements? 

For photos, I put two signals with (1 V/10 ms) and magnified signals with (200 mV/ 100 ns, 100 mV/200 ns) for Channel 0,1 and Channel 3,4.

Warm regards,

Juyong

Juyong,

If I am interpreting the plots correctly (and I may not be), the amplitudes look a little low compared to the data sheet limits of peak Vmin=1.2V & Vmax=1.8V.
I still think the smaller cap value may be compromising the sensor waveform.
What is the reason for selecting 100pF? Is it a requirement for your system, or would it be okay to use a different cap value?

Board parasitics can make it difficult to compute the best value for an application, so one approach is to step thru cap values until you get the frequency and behavior you are looking for. 

Please give this a try if your application permits it. 

Regards,
John

Hi John,

I also think the amplitude was little lower than I expected (100 - 200 mV), it might be because I connected to computer without power supply. But I wonder why channel 0,1 and 2,3 are showing different signals. 

For capacitance range, 100 pF was used just for checking the board measurement ability for drastic change of capacitance. Our applications will be around 1 nF to 30 nF, so I think it will be fine but still from the analog data, 0.1, 0.2 nF gap should be still well measured. I will try with the thru cap for the measurement!

Thank you so much for your help!

Warm regards,

Juyong 

Hi John,

I also have two quick questions.

(1)Would it be possible to fix the frequency for every measurements? for example, can I fix frequency to 50 kHz and measure different capacitance? it seems the frequency is being changed for every measurement. Should I connect 3.3 V to the voltage pin(J8)? currently I am only connecting to the laptop.

(2) And when I measure 1nF / 1.4nF / 2.1 nF / 3.3 nF / 4.3 nF / 6.6 nF / 10 nF / 22.4 nF / 46.3 nF, before 22.4 nF it showed nice graph for all 4 channels but the capacitance measurement get weird at 22.4 nF and 46.3 nF.

Additionally I tried external capacitance-630 pF connected through breadboard and found the signal stabilization improved a lot. However the value was 11.4 nF or sometimes very small instead of actual value (22.4 nF). Are there any methods I can try for addressing this issues? I need range between 1 nF to 30 nF, but currently I could only measure for 1 nF to 10 nF.

Thank you so much!

Best regards,

Juyong