My aim is to increase the accuracy of phase measurement around 0, -180 and 180 degrees for AD8302. This is because as one can see on page 10 of the datasheet the error is greater than 10 degrees when the phase measurement is around 0 or 180 or -180 degrees.

Normally one would expect with a frequency multiplier type of phase measurement such as the AD8302 the most accurate measurement is around 0 and +/-180 degrees as this is when the absolute value of the dc signal is at its greatest. The 0 degrees should yield a largest possible positive dc signal while +/- 180 should give a largest negative dc signal. This is because with a frequency multiplier cos(wt) x cos(wt+phi) = 0.5 cos(phi) + 0.5 cos (2wt). Then the ac term "0.5 cos (2wt)" is filtered out to give only the DC term that is a cosine function of phase difference (phi) between the 2 incoming signals. In this case when phi is close to +/- 90 degrees the DC term is close to zero (cos(+/-90) ~ 0) hence the error is big.

From numerous previous discussions regarding AD8302 phase measurements such as FAQ: AD8302 Gain and Phase with no Input Signal and AD8302: What is the maximum allowable difference in signal amplitudes in order to obtain a phase measurement? I have seen that there has been a lot of talk about an internal phase shift of 90 degrees in the AD8302. HOWEVER, I also saw in this discussion AD8302 Phase Measurement that Joel mentioned "there is no internal phase shift. There is a 900 mV offset put into the output in the output stage, see Figure 3 on page 16.".

I am confused whether there is an internal phase shift of 90 degrees inside the AD8302 or not. The functional block diagram of page 1 and fig 3 of page 16 seems to indicate that there is no internal phase shift of 90 degrees but merely a 900mV offset that is added to the Vphs (perhaps in an attempt to shift the voltage range from +/- 900mV to 0-1800mV). If this was the case, then the accuracy around 90 degrees should be bad and the accuracy is still best around 0 and 180 degrees. If there is indeed an internal 90 degrees phase shift due to a phase shifter in one of the input channels before the frequency multiplication occurs then that would explain why the accuracy is good at +/-90 degrees and bad around 0 and +/- 180 degrees. The offset of 900mV might also still be there to move the Vphs range from +/- 900mV to 0-1800mV.

In any case my aim is to increase the accuracy of AD8302 at 0, and +/- 180 degrees. I know this can be achieved adding a +90 degrees phase shifter in one of the input of the AD8302. However, it cannot be easily achieved if my frequency varies over a broad range (ie 1 MHz to 300MHz). Moreover I was thinking I could work out the phase difference between 0-360 degrees using 2 AD8302 in parallel and with one of the AD8302 having a +90 degrees phase shifter in front of one of the 2 inputs. Re: AD8302 360 degree phase measurement

My question is, if there is indeed an internal phase shift of 90 degrees (due to a phase shifter?) in the AD8302 before the frequency multiplier, then is it possible to get a version of the AD8302 without the internal phase shift of 90 degrees so that the accuracy is good for 0, and +/-180 degrees? If that is possible it would be very nice to use 2 different versions of the AD8302 (with and without the internal phase shift of 90 degrees) at the same time to work out 0-360 degrees phase measurement with good accuracy.

thanks

Andy

Hi Andy,

Here's the answers to your questions:

1) You cannot fundamentally increase the accuracy at 0, 180 and -180 degrees of phase. The accuracy is what the accuracy is, and is inherent in the XOR style phase detector.

2) As I mentioned in this post: Is AD8302 a lock-in amplifier?, it best not to think of it as a "frequency multiplier/lock-in amp" phase detector.

3) There is no internal phase shift in the AD8302, 90 degrees or otherwise. The limited outputs of the log amps, which drive the phase detector, have the same relative phase as applied at the inputs of the AD8302.

4) The output stage(s), as seen in Figure 3 of page 16, has a 900 mV offset. This is put in there so that, given the default slope of 10mV/degree, the phase detector output will report the full +/- 180 degrees of possible phase information.

5) Using a 90 degree phase shifter is the easiest way of working around the error at 0 and +/-180 degrees. Re: AD8302 360 degree phase measurement is the usually solution we've given out in the past to achieve a 0 to 360 degree phase detection scheme.

Hope this helps,

Joel