FDC1004EVM - Capacitive Liquid Level Sensing - HELP PLEASE?

Hi David,

The capactive level sensing reference design provides a lot of information on level sensing using the FDC10004. Take a look at the design guide - www.ti.com/.../TIDA-00317
Regards,

ChrisO

Chris Oberhauser said:

Hi David,

The capactive level sensing reference design provides a lot of information on level sensing using the FDC10004. Take a look at the design guide - www.ti.com/.../TIDA-00317
Regards,

ChrisO

Hi Chris,

Thank you for your reply, I have read through the design guide. The difference in my case is that I'm trying to measure the volume of liquid flowing through a vessel, whereas the design guide for level sensing refers to fluid being collected in a vessel. Could you provide any assistance as to how I could use the data collected from the FDC1004EVM to achieve this?

Many thanks,

David

Hi David,

To use the capacitive sensing to determine flowrate, you will need to have some section where the height of the water corresponds to the flowrate. You may be able to use an intermediate holding tank in the system to hold the water, or a venturi tube.

If there is no function mapping the flow rate to the liquid height, then you may need to use another technique. We have used the LDC0851 to sense impeller rotation to determine flow rate, or ultrasonic sensing with the TDC1000 is another alternative.

Regards.

ChrisO

i Chris,

Thank you so much for your help. I want to use capacitive sensing as we don't want any parts to come into contact with the fluid that we're measuring. In the system I have a restricted channel that the liquid flows through. It has a cross-sectional area of 360mm (30mm height, 12mm length, 1mm depth) see attached image. How do I calculate flow rate from the liquid height?

Many thanks,

David

Hi David,

So, there are a lot of aspects to this question to which I am not an expert, including the viscosity of the fluid, the dielectric constant of the fluid, the cross-sectional area as a function of the fluid height, and the temperature. You will need to have a function which maps the height to the flow rate as a starting point. These are all very specific to your system design and I don't have a general solution.

Regards,

ChrisO

Hi Chris,

I've managed to find a formula that will give me the flow rate from the cross-sectional area of the channel, which is great. However, I'm having some difficulty with the data from the GUI software that I hope you may be able to help with.

I've been setting the 'New Data Sample Rate' to 1ms (see attached image). 

I timed this test for approx. 10 seconds. The graph and CSV (attached) are only showing 246 measurement samples where i was expecting 10,000. Am I missing a setting somewhere?

Many thanks in advance,

David

time test - 1ms.xlsx

Hi David,

In the graph configuration ( you can see the menu in the image you provided on the top right of the window), select EVM Output rate for the New Data Sample Rate.

That should help, but be aware that there are limits to the maximum sample rate supported by the GUI. The LDC and MCU can process data quicker than the reporting BW to the PC.

Regards,

ChrisO

Hi Chris,

I've just tried this and I'm getting the same result.

data test.xlsx

I've also attached the excel spreadsheet of the data. What the the unit of time between each of the samples? Column "x" has units from 0 to 231, this test was again approx. for 10 seconds...

I really appreciate your help! Thank you!

David

Is anyone else able to help me with this?

Hi David,

With the measurement settings you have (reference count etc.) you the maximum sample rate you cam achieve is about 50 SPS and you are streaming two channels so over 10 seconds you should expect ~250 samples.

Regards,

Varn

Varn Khanna said:

Hi David,

With the measurement settings you have (reference count etc.) you the maximum sample rate you cam achieve is about 50 SPS and you are streaming two channels so over 10 seconds you should expect ~250 samples.

Regards,

Varn

Thank you Varn.

The CSV log files that are created by the software have two columns that I am unsure of (but that I think relate to time in milliseconds), ‘evmSampleDeltaMs’ and ‘logDeltaMs’.

The sum of each of these channels are different, which is confusing. Please can someone explain what these measurements are? Is the device consistently sampling data at regular intervals?

Thanks,

David

I've tried another test for approx. 1 minute. The CSV data log is showing 'evmSampleDeltaMs' and 'logDeltaMs' with the same values which is expected, however the sum of the column is only 29.93 seconds, half of what it should be. Why is this?

Thanks in advance,

David