TDC7200: TDC1000, cloclCount

Vyacheslav Serov

Part Number: TDC7200
Other Parts Discussed in Thread: TDC1000, , MSP-ULTRASONIC-DESIGN-CENTER

Hello,

I am using TDC1000 and TDC7200 chip in flow metering. Distance between sensors is about 62mm so I am using standart ToF with blanking. I amusing 10 calibration periods and 4 averaging measure cycles.

Start and Stop signals picture:

My trancdursers are working on frequency of 1Mhz and I am able to see in oscilloscope graph similar to datasheet but seems that calculated time values are not exact. I see problems in zero flow as well (delta ToF is negative all the time). Reading interrupt registers and errors registers not showing any problem (all is good and there is no overflow).

TDC1000 settings:

data 0 42 
data 1 52 
data 2 12 
data 3 4C 
data 4 1F 
data 5 00 
data 6 28 
data 7 00 
data 8 19 
data 9 00

TDC7200 setiings:

data 0 00 
data 1 51 
data 2 00 
data 3 07 
data 4 07 
data 5 08 
data 6 FF 
data 7 FF 
data 8 01

I have marked that read via SPI clockCount periods are not exactly what I see on oscilloscope. For example my clock frequency is 8Mhz and I see that there is about 10-12 periods between clockCount1 and clockCount2, but if I calculate difference received from TDC7200, then it is about 32 periods.

Received by SPI value of clockCount are the next:
clockCount1: 1713
clockCount2: 1746
timeCount1: 2053
timeCount3: 1218

On the picture below there is 2 stop signals on TDC7200 (yellow) and clock (blue).

Also in datasheet is not clear about start of counting clockCount. From the picture seems that every stop is starting clockCount, but in numbers received from TDC7200, seems that it continue to count from the start.

If you need additional information, please, let me know.

Thank you for you help.

BR,
Vyacheslav Serov

over 6 years ago

0 Bharat Aravamudhan over 6 years ago

TI__Genius 11015 points

Vyacheslav,

We have received your question & shall respond to your questions by tomorrow.

0 Bharat Aravamudhan over 6 years ago

TI__Genius 11015 points

Vyacheslav,

The internal ring oscillator starts counting from when it receives the START signal until the first rising edge of the CLOCK and then the CLOCK_Count1 is the total clock cycles counted post that.

Please send me a detailed account of what the exact data you expect to read and what the system actually outputs.

Also please note we suggest an MSP430 based flow meter solution as in the link below for higher accuracy in flow metering.

www.ti.com/.../MSP-ULTRASONIC-DESIGN-CENTER

0 Vyacheslav Serov over 6 years ago in reply to Bharat Aravamudhan

Prodigy 50 points

Bharat,

I am expecting to read some small time oscillation in plus and minus during zero flow (may be picoseconds). But unfortunately I cannot read it. May be I have some wrong setup of my system, but I am not able to diagnose it reading controlling registers in TDC7200 and TDC1000. I am expecting at zero flow TOF between sensors about 48 us, but receiving about 54,226 us.

I perform calibration every measuring with 10 calibration clock periods.

Logic of time calculation is from datasheet:

void calculateTime(float& deltaTime)
    {
        if(s_calCount == 0)
            {
                LOGD("Calibration time is 0. \n");
                return;
            }
        float normLsb = clockPeriod / s_calCount;
        float tof2_s = ((normLsb * (s_timeCount1 - s_timeCount3)) + ((s_clockCount2 >> AVG_CYCLES) * clockPeriod));

        deltaTime = tof2_s;
    }

Also question about clock_count2 to clock_countN. With clock_count1 it is clear, but next STOP signals clock_count are calculated according datasheet picture (between first stop and second stop for clk_count2) or it is counting clk from start signal to N STOP?

May be you could give some advise according to chip setup or in what direction should I look to solve this issue. My testing stand equipped with valve for regulation of water flow and prevent bubbles moving from water tank.

BR,

Vyacheslav

0 Vyacheslav Serov over 6 years ago in reply to Bharat Aravamudhan

Prodigy 50 points

Question about clock_count2 to clock_countN. With clock_count1 it is clear, but next STOP signals clock_count are calculated according datasheet picture (between first stop and second stop for clk_count2) or it is counting clk from start signal to N STOP?

0 Bharat Aravamudhan over 6 years ago in reply to Vyacheslav Serov

TI__Genius 11015 points

Vyacheslav,

This device is not accurate while it measures zero flow, that is the precise reason I recommend you use the MSP430 based device for flow meter.

www.ti.com/.../MSP-ULTRASONIC-DESIGN-CENTER

At zero flow even minute movement in the water pipe provides signal back to the transducer and it interprets it as a valid data.

The MSP430 based device is designed specifically for flow metering.

Sensors

Sensors forum

TDC7200: TDC1000, cloclCount