AD5933EBZ evaluation board issue

I am using AD5933EBZ for my project. I am currently trying to use the calibration resistance to measure for the response of the circuit. The parameters I have currently set are as follows:

Start Freq - End Freq: 30KHz - 50KHz. with 200Hz step.

MCLK : 16MHz

Range 1: 2V p-p 

PGA gain : 1x

Rcal : 1.2K ohm ( according to the application note)

Rfb: 1.5K ohm ( according to the application note)

My gain factor calculation is 3.61E-5 and my impedance values measured are quite perfect for the Rcal. However my real and Imaginary values calculated are incredibly low as compared to what are said to be expected in other questions i read on this forum.  I cannot understand why. Can someone help me understand them? 

This is the sample data set that I have got. Similar values are for the rest of the data set. According to my calculations they are correct.

Frequency Impedance Phase Real Imaginary Magnitude
30000 1200 0 12 14 18.43909
30200 1151.881 1.941486 12 15 19.20937
30400 1151.881 1.941486 12 15 19.20937
30600 1200 0 12 15 19.20937
30800 1200 0 12 15 19.20937
31000 1200 0 12 15 19.20937
31200 1200 0 12 15 19.20937
31400 1200 0 12 15 19.20937
31600 1200 0 12 15 19.20937
31800 1200 0 12 15 19.20937
32000 1200 0 12 15 19.20937
  • Is this the board you are evaluating?

    Re and Im data are indeed way too low - with your settings these should be in thousands. Something is not right: 

    It appears as if your Rcal is not connected to the circuit. Calibration procedure will force "correct value" for Rcal, so it is always better to look at the raw Re and Im data as those are read from the registers and not calculated.

    If you have access to oscilloscope, it would be worth checking the voltage at the output of A2: with your settings, Rcal and Rfb it should be a well-formed sine wave of about 2.5V p-p. Same sine wave (inverted) should be present at the RFB pin of the AD5933.

  • Hi, 

    Yes that is the board that I am using.   When I checked on the oscilloscope, I am getting a Sine wave with DC shift 1.76V and peak to peak of 2V. However, at the Rfb pin I am getting attenuated signal with same DC shift of 1.76V. I now figured out that it is an issue of the AD8606 on the EVM which for some reason isn't giving the inverted output which is expected. So I created an external circuit ( similar to the AD8606) using LM741 and got the expected exact inverted output from it at the Rfb Pin. After this, I gave this LM741 input to the AD5933 and then have got these results which seem correct,

    Frequency Impedance Phase Real Imaginary Magnitude
    30000 1168.357 1.05E-02 637 -1142 1307.644
    30200 1158.932 3.89E-02 650 -1152 1322.726
    30400 1160.514 7.62E-02 654 -1150 1322.957
    30600 1159.351 -1.32E-02 657 -1147 1321.839
    30800 1159.719 4.20E-02 661 -1146 1322.965
    31000 1161.45 8.10E-02 667 -1149 1328.567
    31200 1156.838 9.88E-02 670 -1143 1324.896
    31400 1152.818 7.63E-02 675 -1141 1325.71
    31600 1153.569 9.44E-02 677 -1139 1325.009
    31800 1157.436 -5.85E-02 680 -1137 1324.828
    32000 1164.484 4.10E-02 683 -1131 1321.23

    After this I connected a series RC circuit with 1K Ohm and 4.7nF capacitor to check if it working correctly. However, the result that I am getting is completely unexpected. According to my calculations the Impedance of that circuit should vary between 1677 Ohm and 2128 Ohm. The parameters that I have set are same as the ones initially mentioned. What could be the reason for this?

    Frequency Impedance Phase Real Imaginary Magnitude
    30000 9359.778 -216.18 20 162 163.2299
    30200 9398.231 -216.084 19 162 163.1104
    30400 9419.262 -215.891 18 162 162.9969
    30600 9356.719 -215.423 16 163 163.7834
    30800 9367.668 -215.54 16 163 163.7834
    31000 9426.82 -215.312 15 163 163.6887
    31200 9306.82 -215.505 15 164 164.6845
    31400 9393.778 -215.822 15 162 162.693
    31600 9290.915 -215.164 13 164 164.5144
    31800 9435.122 -215.529 13 162 162.5208
    32000 9352.077 -215.619 13 164 164.5144
  • Something is not quite right still. Before we go further, LM741 is not likely suitable as it requires supply voltage of at least +-10V split supply or 20V unipolar. Unless you power LM741 from some such external power supply.

    If you refer to Fig, 1 from the user guide - it shows the DC operating point for the circuit of Vdd/2, which in your case should be 3.3V/2 = 1.65V - this DC level should be present at the output of A1, A2 and the AD5933 Rfb pin (when active). Yours is 1.76V, which is noticeably higher. Unless your Vdd is 3.52V it might point at both AD8606 having some issues. I would try to make sure that DC operating point is Vdd/2 and that the sine wave is well-formed the output of A1, A2 and the AD5933 Rfb pin.

    As to the series RC: may be C is blocking some of the DC between A1 and  A2 and shifting the CD operating point unfavorably, especially since LM741 may be underpowered, who knows. Besides the point, but the capacitor impedance is at -90 degrees in phase with respect to the resistor, so the expected impedance magnitude equals the square root of the sum of squares, not just the sum of the individual impedances of the resistor and the capacitor. 

  • Hi,

    The output of A1 is at 1.645V (with a sine wave of Vp-p 2V)  and the output of A2 (pin 7) is 3.250V ( Just DC) .

    I also checked the value at pin5 of the AD8606 and that is 1.655V. The value at pin 6 i.e the inverting terminal when I have connected a 1.2KOhm resistor is 1.7V ( with an attenuated sine wave of Vp-p 100mV) which is weird. Also, I am not getting an inverted sine wave at output of A2. I am getting a non-inverted attenuated sine wave at the inverting pin of A2. What can be the reason for this?

    At the output of A2 I am getting 3.3V as the output which is simply the 1.7V from the output of 1.2KOhm and 1.655V of the Vdd/2.Essentially my Vdd/2 Values are okay. But  the AC wave behaviour is something I cannot understand. 

    Also, I understood the error in my calculations of equivalence.

    I was using the LM741 in order to check it is an issue of A2 of AD8606 and I don't intend to use it in my final circuit.

    Awaiting your thoughts for this.



  • Something is wrong with DC operating point of A2. Let's refer to Figure 31. EVAL-AD5933EBZ USB Schematic in the User Guide, there the A2 is designated as U1-B. What is the DC voltage on pin 5 of U1-B? It should be  Vdd/2, but may be R2 is somehow disconnected from the rest of the circuit? When you say "1.2KOhm resistor" are you referring to Rfb? Is Rfb is connected between pins 6 and 7? If yes, the DC on pin 7 should be pretty much equal to the voltage on pin 5. If not - it might be that U1-B is fried or its output somehow pulled to Vdd by some board defect... Hard to tell remotely without seeing the board.
    LM741 is an ancient OPAMP - slow, cannot operate in modern low-voltage circuits, better steer clear of those unless your circuit is powered by a split +/-15V power supply.