FDC1004EVM: unexpected capacitance values

user6527039

Part Number: FDC1004EVM

I'm using FDC1004EVM for liquid level sensing,
but the Rigid-Flex PCB used in my application, which also has RE, RL, Level and shield electrodes,
is similar to tidu736a in shape, but only 1/10 in size.

when I connect the PCB to FDC1004EVM and configure it working as differential mode,
the capacitance of each electrode is RE=1.6pF RL=5.3pF Level=6.1pF.
I cannot understand why capacitance of RL is much larger than RE, and close to Level,
even though RL and RE have the same shape and size of electrode, both of whose size is about 1/5 of Level's.
And one more thing confusing me is when I disconnect all the electrodes,
the capacitance values of those channels are different from each other.(I don't remember the exact values.)

Could you explain that for me ?

over 4 years ago

0 John Miller - Sensing over 4 years ago

TI__Mastermind 40295 points

What device settings are you using?
Are you using the TI EVM or your own PCB?
Is the flex circuit attached to a container without air gaps, or is it floating in free space?
Are the trace & layer thicknesses of the flex PCB the same as for the reference design?

0 user6527039 over 4 years ago

Prodigy 70 points

I'm using the TI EVM, but an addition relay PCB is also used to match the pin arrangement of TI EVM and my flex circuit.

The setting of Sensing Solution EVM GUI is as below. (No offset and gain )

CHA CHB

CIN1 CIN4 -->RE

CIN2 CIN4 -->RL

CIN3 CIN4 -->Level

CIN4 disable -->OPEN

The flex circuit is attached to a container closely with double-sided tape.

And the trace is almost the same as tidu736a,

but copper thickness is quite different, which is 30um vs 700um.

0 John Miller - Sensing over 4 years ago in reply to user6527039

TI__Mastermind 40295 points

Are you using CAPDAC settings?
Does the behavior change if you use adhesive instead of double-sided tape?
The general advice is to minimize the gap between your sensors and the container. The double-sided tape may be alerting the capacitance.

0 user6527039 over 4 years ago in reply to John Miller - Sensing

Prodigy 70 points

I'm not using CAPDAC settings.

In fact, the height result got from RL,RL(0),Level and Level(0) is not bad.

I just don't understand the difference of capacitance , especially between RE and RL of the same size.

In my opinion, the gap only biases the result but not alter the difference between each channel, as long as the double-sided tape attach the container and flex circuit firmly.

0 John Miller - Sensing over 4 years ago in reply to user6527039

TI__Mastermind 40295 points

I'm not sure what might be happening here. Let me check wit the team and I will reply with an update tomorrow.

0 John Miller - Sensing over 4 years ago in reply to John Miller - Sensing

TI__Mastermind 40295 points

The different capacitor readings may be caused by ground planes or other objects near the electrodes. Fringing fields as described here and here can also have an effect.
I mentioned adhesive as an alternatives to double-sided tape because tape thickness can vary because of its application and stretching, which could impact the capacitance.

0 user6527039 over 4 years ago in reply to John Miller - Sensing

Prodigy 70 points

Since I have got a reasonable height result calculated from the readings, it would be fine as long as the difference is not caused by inapproporiate usage. Thanks for your reply.

Sensors

Sensors forum

FDC1004EVM: unexpected capacitance values