EVM430-FR6047: VFR + Temperature Calibration

Hello Ashenafi,

Can you attach your generated header files for our experts to review? - Thanks.

Hello Dennis, 

Here it is.

headers after flow and temperature calibration.rar

Thanks Ashenafi,

Let me run this by our expert on calibration issues and see what he suggests.

Okay, Dennis. I hope to hear from you and your team soon.

Hello Dennis,

It is urgent, so please tell the experts to respond to my question raised.

I do have also another related question, The debug plot 1 graph shows 0 temperature even though the option to "Calculate temperature using TOF" is activated.

Hi,

I run your configuration with one meter and debug it in the CCS. It shows the calibration function is working.

I also read the data in the USS GUI, And the calibration function is working. I only changed the USS_GAIN_RANGE and USS_TEMPERATURE_LUT_TRANSDUCER_CONST to meet my meter. 

The debug plot 1 shows the temperature data. 

So, everything is working on my side with your configuration. 

Just for notice, after you download the software with your configuration file. Do NOT use the request update button in the GUI. It will overwrite these configurations. 

Best regards,

Cash Hao

Hello Cash,

Thank you for your response. But, I didn't use the request update button. These are the steps I follow:-

1. upload the code ( with the updated header files) with ccs.

2. connect the EVM to the GUI.

3. I go to the waveform tab, activate the adv calibration table check box, and then start the measurement, the attached error comes here.

Hi,

I did the same process as yours. 

Could you share the screenshot of the Calibration page and the VFR Flow + Temperature page here?

Best regards,

Cash Hao

Here it is.

Do I need to exclude/ Delete the "USS_meterVFRCalibFlow.h" from the project?

Hi, 

These configuration looks good to me. 

It is not neccessary to exclude/ Delete the "USS_meterVFRCalibFlow.h" from the project. Just keep the "USS_meterVFRCalibFlow.h" as the default one. Have you changes the data in "USS_meterVFRCalibFlow.h"?

Best regards,

Cash Hao

Hi,

Great

Ok, So why it isn't working as expected?

First I did only VFR Calibration and test the meter with it and things work as expected, but when the temperature varies there comes a large error. So I start VFR + Temperature Calibration and do the same thing. the only change was the highlighted one in the attached image and the values in calibration header files.

Hi,

I do not have much clue right now, your configuration works on my side. 

Maybe you can try with reinstall the USS GUI again. And even recalibrate the meter again. 

Best regards,

Cash Hao

Hi

I asked after trying what you suggest and other things too. 

Hi,

Recalibrate the meter with less flow points and less temperature points. And add more flow points and temperature one by one. It could tell us when the issue is coming from. 

Best regards,

Cash Hao

Hi cash

I used the default points. ( i.e. 7 flow points and 3 temperature points). 

So when you say with less flow points, can you tell numbers?

Hi，

You can start with 3 flow points with 2 temperature points. 

Best regards,

Cash Hao

Hello Cash,

I hope you're doing well.

I wanted to let you know that the previous issue I mentioned has been resolved.

However, I have a few questions regarding the observations I made during a 17-minute run. Here are the details:

1. I encountered an error message stating "DTOF-Shift value was greater than maxSampleShift." How can I resolve this issue? You can refer to the first picture for more information.

2. I noticed cycle slips in both the time of flight and VFR (Volume Flow Rate). How can I address these cycle slips? Please refer to the second picture for a better understanding.

3. In the debug waveform tab, I also noticed a slip. How can I fix this issue? Additionally, the values displayed in "Debug Plot 2" and "Debug Plot 3" are 14 and 11 respectively. According to the Calibration — Ultrasonic Sensing Design Center User Guide 02.40.00.00 documentation, they should be 1 and 6. What steps should I take to achieve these desired values?

Thank you for your assistance.

Hi,

For the 1st question, you can change the USS_ALG_MAX_SAMPLE_SHIFT into 40/60/80 ... to avoid your issue in the USS_userConfig.h. However, I noticed that you are getting this error randomly, it is not continuously reported. So, change the USS_ALG_MAX_SAMPLE_SHIFT  does not solve this issue.

The USS_ALG_MAX_SAMPLE_SHIFT parameter is used to limit the output of the dTOF result. Assume your ADC sampling rate is 3.6MHz and the transducer frequency is 1MHz. Then with the USS_ALG_MAX_SAMPLE_SHIFT as 4 will allow the dTOF range should be in +-(4/(3.6MHz/1MHz)) cycles, around +-1.11us. If the calculated dTOF result exceed the range, it will report the error 135.

And this error is getting reported when there is an cycle slip happens. So, you have to solve the cycle slip issue first. The cycle slip on the absTOF result usually caused by inappropriate envelope crossing threshold setting. I need the ADC capture data and your advanced parameter configuration page to check on it. 

Best regards,

Cash Hao

Hi, cash

Ok thanks, here attached are the captures you request. 

Also please address question #3.

Hi,

I also need the parameter page. 

For the question #3, I barely uses the Debug waveform page. And when you have solve the cycle issue, the Debug waveform page should also be good. 

Best regards,
Cash Hao

Hi Cash,

I have attached the requested capture page.

Ok, we will check it after fixing the cycle slip issue.

HI,

Emm, the gain setting is 13db which is pretty high. 

Change the envelope crossing threshold from 22 to 30. And I also suggest to change the signal sampling frequency from 4000 to 3600. You can try with this configuration. And if it can not solve the cycle slip issue, we might need to improve the gain setting then. 

Best regards,

Cash Hao

Hi Cash,

Yes, setting the gain to 13.0 dB is necessary to ensure that the ADC capture is in the desired range of +-900 or 1000.

Alright, I will modify the envelope crossing threshold to 30 as you recommended. I initially attempted a sampling frequency of 3600, but it resulted in an unsatisfactory waveform, prompting me to switch back to 4000.

However, I will follow your suggestion and test the configuration you proposed. I will provide you with an update shortly.

Thank you.

Hi,

It is not mandatory to keep the ADC capture in the range of +-900 or 1000. Try to keep the PGA gain lower than 7db is more recommended. 

And the "unsatisfactory waveform" using 3600kHz does not affect the algorithm for calculation. Don't worry about it.

Best regards,

Cash Hao

Hi Cash,

I follow the steps on Ultrasonic Water Meter - Design and Testing, and it says to keep the ADC capture in the range of +-900 or 1000. But I tried with 6.8 db, and the amplitude was around +-450, but the UPS and DNS signals start to slip up and down randomly I think the signal is almost getting lost. 

I also experimented with a sampling frequency of 3600 kHz, but it did not yield any significant changes.

Please answer the following two questions: -

1. After VFR + Temperature Calibration, should I activate/deactivate the Adv. Calibration Table checkbox on the GUI or comment out the colored line from the code provided? When I toggle this option, the VFR values change.

// Initialize the USS algorithms with default parameters
gUssSWConfig.algorithmsConfig->isInitAlgorithms = false;
#if (USS_ALG_VFR_CALIB_MODE!=USS_ALG_VFR_CALIB_OPTION_DISABLED)
// Calibration range feature is defaulted to disabled
// Can be enabled by the GUI using the Advanced Calibration Table check box in Waveforms tab
gUssSWConfig.algorithmsConfig->volumeCalibrationOption = USS_Alg_volume_flow_Calibration_Option_disabled;
#endif

2. After performing VFR and temperature calibration on the meter, I conducted tests and observed a consistent positive deviation of approximately 20 LPH across various flow rate ranges when compared to the reference meter. I'm seeking a solution to address this deviation without the need for another round of calibration. Could you please suggest any possible methods to deduct this deviation effectively?


I appreciate your guidance and assistance.

Dear team,

I am writing to express my concern as it has been five days since I sent my message, and I have yet to receive a response. I kindly request that someone from the team please acknowledge and address my message as soon as possible. I am eagerly awaiting your response. Thank you.

Hi,

Sorry for the late response. I forgot to reply this thread. 

1. activate the Adv. Calibration Table checkbox will enable the calibration function. No need to change the code manually. 

2. Could you share some test data here? I need to find out what causes the 20LPH offset first then come up with a plan to solve it. ADC captures, waveforms, configurations all these will be useful. 

Best regards,

Cash Hao

Hi cash,

It is okay.

1. ok, but my question is, once calibrated. while testing, do I need to activate or deactivate the checkbox? The meter should operate autonomously in any residential area without relying on software. Therefore, there is no requirement for additional adjustments, as it is designed to function remotely and independently from software. 

2. I have previously provided you with the ADC Capture, Configuration parameters, and header files that were generated after calibration. The waveform appears to be satisfactory based on the information provided. However, as per your request, I will send you a sample waveform tomorrow for further analysis and evaluation.

Hi,

1.  Activate or deactivate the checkbox only changes what data is shown in the flow rate chart. In the software, you can find in the USS_Algorithms_Results_fixed_point struct, there are one iq16VolumeFlowRate and one iq16VolumeFlowRateNoCalib. The iq16VolumeFlowRate shows the flow rate data after calibration. And the iq16VolumeFlowRateNoCalib is the data without calibration. If you have already set the calibration parameters correctly, you will see the differences in those two data. 

2. Okay, so the cycle slip issue still exist? I do not recommend you move to the calibration process before the cycle issue solved. 

Ashenafi Abebaw said:

But I tried with 6.8 db, and the amplitude was around +-450, but the UPS and DNS signals start to slip up and down randomly I think the signal is almost getting lost. 

Can you show me this test data?

Best regards,

Cash Hao

Hi Cash,

here I attached the requested captures. 

Hi Cash,

I appreciate it if you respond to my message as soon as possible.

Hi,

Do you want to have a quick meeting to discuss your issue now? Can you give me your email address which I can send you the meeting invitation. 

Best regards,

Cash Hao 

Hi, Cash

Yeah here is my email address ashenafiabebaw02@gmail.com

Sent the meeting link through email. 

Best regards,

Cash Hao

Hi,

Are you able to join the meeting now?

Best regards,

Cash Hao

yeah, i will be there after 5 minutes

Hi Cash,

I hope you are doing well. 

It has been almost two weeks, I kindly request your response to my previous email. I attached all the files you requested there.

Hi,

Could you resend the email again, I couldn't find that email. 

Best regards,

Cash Hao

Hi

I send it

Received.