EVM430-FR6047: Characterization

Hi Jucetauro,

When you say that you don't have a control for the temperature in the calibration bank, do you mean that you are physically unable to control the environmental temperature during calibration? Or do you mean that you are unable to account for temperature in the device when calibrating? Our devices do have a way to account for temperature. You can see the Flow + Temperature VFR calibration section of the USS Design Center User's Guide to learn more about this.

Generally, I would not expect such abrupt variations with temperature as you've mentioned. Is this temperature the temperature of the overall environment, or the temperature of the water? In general, I would appreciate hearing more details about your experimental set-up.

The transducers themselves are likely more susceptible to changes in temperature, but even then I am surprised that they would react so drastically to slight changes in temperature. Could you relay the transducer model you are using?

As a final question, what error ranges are you looking to achieve in your application? Some error, and some variation is to be expected. I will have to look further into our documentation to see what amount of error is expected with temperature variation.

Thanks for your reply.

We do not have equipment on our calibration bench to vary the water temperature. The data I sent was taken with variations in water temperature between day and night.
The tests were carried out at a constant flow rate of approximately 350 L/h for about 200 seconds. The error graphed corresponds to the error between the accumulated flow rate and the weighing machine. We also record the error of the scale with respect to our electromagnetic flowmeter.
The measurement tube with sensors that we use for the prototype is HS0014-000 from the audiowell company.
The calibration loaded in the firmware is VRF with 5 ranges. The recorded temperature is the average temperature of the test, the same is the temperature sent by the firmware in the USS_Algorithms_Results.temperature structure.
The errors we seek to achieve are less than +-2% in flow rates between 25 to 3100 L/h and less than +-5% between 15 to 25, with pressures between 0 BAR to 10 BAR pressure and 0.1 C up to 50C temperature, according to ISO4064.

Could you take some pictures of your hardware set-up? Showing the MSP, its connection to the transducers, and the pipe? 

Looking at the datasheet for the transducers, it does not seem that the error should vary so much across that temperature range. Of course their error is not zero and neither is the MSP's, but so far I do not believe that these should combine for this error.

Could you also post an image of your VFR configuration? I am wondering what the ranges are for your flow rates.