AD7798 v AD7710 for precise measurements in presence of external vibration

Question The AD7710 works well in the precense of external vibrations, but when I use the AD7798 the performance falls to unacceptable level in the precense of the same vibrations...can you explain ? Answer The AD7710 uses a free-running filter with no chopping so the roll off is quicker than with the AD7798 filter. The AD7798 was designed to use chopping resulting in low offset/low offset drift. However, the roll-off is slower for a given update rate. An external R-C filter is used for anti-aliasing but this filter is required to removed images at multiples of the master clock. To include a filter for vibrations would require a filter with very low bandwidth. The AD7190/2 again uses a free-running filter. So, the filter response will be comparable with the AD7710 response.

AD7710: Serial interface

Question Sometimes the ADC AD7710SQ seems to start in an "unusual mode" after power up. The serial output bits are delayed by one serial clock cycle or the serial configuration input data are read or shifted one bit left (in advance). I use the external clocking mode. As powercycling is not an option how can I reset the chip by software ?? Answer The following procedure is recommended to check if the serial interface is okay. Using the Control register as a shift register: 1 Carry out a read from the Control register. You should read back the default value of 000146hex 2 Write to control register (if SYNC is low then the part won't act on the write operation) 3 Read back value from the control register (MD2~0 may have changed but the other bits should be the same) In addition, to prevent noise etc corrupting the serial interface, we recommend pull ups on the TFS and RFS lines and a pull down on the SCLK line (since a falling edge clocks in data.) In addition, you can use self calibration to reset the interface after power up.

AD7710 input impedance different to what is explained in AN-338?

Q AD7710 input impedance different to what is explained in AN-338? The datasheet for the AD7710 states that the input impedance is ver high. However, in AN338 there is a different statement, the input impedance is gain dependent and can be 288k for Gain =8 and higher, and is never higher than a few MHz. The error we have is consistent with the 288kOhm. Can you confirm that AN388 is correct and give any indication of how much that 288k might vary, for our application 10% variation would not matter (and maybe even if it was more), we just need to know that it will not vary dramatically. A AN-388 is correct, assuming a sampling Capacitance of approximately 11.13pF. However, the AD7710 datasheet is also correct and does state that the gain and frequency influence the input impedance. The AD7710's d.c. input impedance is >1Gohm. The input impedance varies depending on the sampling frequency,fs, and this frequency varies relative to the gain, which is: fs = gain/(fclkin/256) The input impedance, Zin = 1/(Cxfs) and if you refer to Table III, the input impedance is as follows, where fclkin=10MHz and C=11.5pF (i.e. Cint in Figure 7) : Gain=1, fs=39KHz, Zin = 1/(11.5pF*39Khz) = 2.229Mohms Gain=8, fs=312KHz, Zin = 1/11.5pF*312KHz) = 278.7Kohms As per the datasheet, the sampling capacitor's maximum value is 20pF, therefore with Cs=20pF Gain=1 => Zin = 1.6Mohms Gain=8 => Zin = 1.28Mohms

Where can I find the AD7710 evaluation board user guide and schematics, it is no longer on your website.

Q Where can I find the AD7710 evaluation board user guide and schematics, it is no longer on your website. A See the attached AD7710 evaluation board user guide. Attachments: AD7710_EvalBoard_AppNote.pdf