FDC1004: Greater shield drive than 400pF ?

Hello,

The full scale range of the FDC1004 is 400pF which includes the offset subtraction from the Cap DAC. If you are running into limitations of full scale range, I would recommend one of our resonant sensing-based capacitive sensors like the FDC2214. It similarly has 4 channels, but with better resolution and a larger dynamic range (up to 250nF).  The main disadvantage is that it does not have an integrated shield driver compared to the FDC1004. However, you can use a passive ground layer for shielding the sensor which has the same effect but some sensitivity loss depending on how close the separation is between the sensor and ground. Take a look at this device and let us know if it will work better for your application.

Thanks!

Luke

Luke:

The absence of an integrated shield driver is a showstopper for me since space is a premium. I'm concerned, using the FDC2214, the EMI generated by the external LC tank could be a problem.

Now, does the FDC1004 support floating capacitor sensors, i.e. where the sensor ends are connected to channels A and B or does it only support grounded sensors?

Hi,

Note that if its just the shield driver that will have >400pF then you could potentially connect an op amp capable of driving large capacitive loads to buffer the pin.

For the pin connections the sensor inputs need to be referenced to a fixed voltage potential. This can be ground or an input/shield driver that is out of phase with the sensor connection.  Therefore if you connect CIN1 to a sensor and connect CIN2 to a sensor (where CIN1 and CIN2 are out of phase) then they can be referenced to each other. I would recommend reading some of our application notes that go into much deeper details:

• FDC1004: Basics of Capacitive Sensing and Applications: www.ti.com/.../SNOA927
• Capacitive Sensing: Out-of-Phase Liquid Level Technique: http://www.ti.com/lit/pdf/SNOA925
• Capacitive Sensing: Ins and Outs of Active Shielding (Rev. A): http://www.ti.com/lit/pdf/SNOA926

You can also find other relevant collateral on our application notes page: http://www.ti.com/sensing-products/capacitive-sensing/capacitance-to-digital-converters/technical-documents.html# 

Let me know if this is able to answer your question.

Thanks!

Luke

Thanks for the great references you've provided...

In my particular situaiton, I'm trying to determine liquid height inside a syringe. No sensor(s) is permitted to be inside the syringe as that would contaminate the liquid, so the electrodes would be placed outside the syringe. I was thinking of having my sensor and ground electrodes as parallel plates facing 180 degrees from each other, as they'd be wrapped around the syringe.

Your app notes suggest that an active shield is placed on the opposite sides of both electrodes as a continous plane underneath both electrodes.

Now,:

1. Wouldn't that create a new capacitance between the ground electrode on top and the active shield beneath it? If so,
2. How do I eliminate this extra capacitance? Or
3. Should a ground plane instead be placed under the ground electrode only to eliminate this extra capacitance?

Thanks,
David

Hello,
If you are using an active shield then it should not be a continuous metal covering both CIN1 and CIN2. Instead you should use SHLD1 to cover only the CIN1 portion and SHLD2 to cover only the CIN2 portion and make MEAS1 to be CIN1 - CIN2. This will look like an adaptation of the TIDA-00317 (www.ti.com/.../tida-00317) where the flex would wrap around the syringe. This should give the best solution for both shielding and sensitivity. You could also consider using all 4 sensors in the same phase where a single sensor wraps around the circumference of the syringe and the other channels are spaced at known intervals away from each other. Then a single continuous shield driver on the bottomside of the PCB in the same phase could be used. It won't be generating capacitance between the shield and the sensors if they are in the same phase, which holds them to the same potential. Let me know if either of these solutions would work for you.
Regards,
Luke

Ok, if I understand you correct, you first suggestion would use the Out-of-Phase(OoP) method you talk about to eliminate stray interference, correct? Afterall, the level reading should be the same whether i hold the syringe in my hand or place it on the table somewhere far from me.

So, lets see if the following are correct...

1. CIN1 is connected to Level 1 electrode with SHLD1 plane underneath with no capacitance generated between CIN1 and SHLD1 .

2. CIN2 is connected to the other Level 1 electrode (which is on the opposite side of the syringe, but with same size and dimensions).
SHLD2 plane is underneath with no capacitance generated between CIN2 and SHLD2 .

3. CIN3 and CIN4 are connected to the RL and RE sensors, respectively (doesn't matter which channel has which sensor, i hope).
SHLD1 is underneatth RL and SHLD2 is underneath RE, correct?

4. No electrode or sensor terminal is connected to ground, correct?

In the case outlined above, SHLD1 and SHLD2 drive signals are 180 degrees out of phase with respect to each other as required by the OoP method, correct?

The reason I may seem confused is that the OoP differential method is not mentioned in your datasheet and there's no clear schematic on how I could adapt it for use my situation. All examples have a major ground sensor electrode , which i don't seem to have in my case.

Thanks,
David

Thanks Luke ... I hope all my assumptions still hold if the shield underneath is larger (in width & length) than the sensor on top. The app notes indicate this is the best way to minimize interference from the sides, but that was ground based sensors. Any thoughts?